Techniques for arrayed printing of a permanent layer with improved speed and accuracy

Array printing technology for permanent layers with increased speed and accuracy

TECHNIQUES FOR ARRAYED PRINTING OF A PERMANENT LAYER WITH IMPROVED SPEED AND ACCURACY

The present invention relates to techniques for array printing of permanent layers with increased speed and accuracy.

Cross-references to related applications

This application claims the US Provisional Application No. 62/059121, "Technology for Array Printing for Permanent Layers with Increased Speed and Accuracy," which was filed on October 2, 2014, on behalf of the first inventor, Michael Baker, and 2014. The priority of each of the US Provisional Application No. 62/021,584, "Technique for Array Printing for Permanent Layers with Increased Speed and Accuracy", which was filed on July 7, the first inventor, Michael Baker. In addition, this application also claims the U.S. Patent Application Serial No. 14/680,960, filed on Apr. 7, 2015, on behalf of the first inventor, Nahid Harjee, for the control of the amount of printing ink for depositing fluids within precise tolerances. "U.S. Invention Application No. 14/340,403, filed on Jan. 3, 2014, to the First Inventor, Nahid Harjee," for the measurement and control of printing ink droplets for depositing fluids within precise tolerances. U.S. Patent Application Serial No. 14/627,186, filed on Feb. Priority and is part of the continuation of these applications.

U.S. Patent Application Serial No. 14/680,960 is the U.S. Patent Application Serial No. A continuation of 14/162525 (now US Patent 9010899, published on April 21, 2015). Thus, U.S. Patent Application Serial No. 14/162,525, the entire disclosure of which is incorporated herein by reference in its entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire all "Technology of Control", and PCT Patent Application No. PCT/US2013/077720, filed on Dec. 24, 2013, on behalf of the First Inventor, Nahid Harjee, for the control of the amount of printing ink for depositing fluids within precise tolerances "Priority." PCT Patent Application No. PCT/US2013/077720 claims priority to each of the following applications: U.S. Provisional Patent Application No. 61/746,545, filed on Dec. 27, 2012, on behalf of the First Inventor, Conor Francis Madigan "Mixed-type mixing"; U.S. Provisional Patent Application No. 61/822,855, filed on May 13, 2013, which is incorporated by reference to the entire disclosure of the entire disclosure of the entire disclosure of the disclosure of the entire disclosure of An inventor Nahid U.S. Provisional Patent Application Serial No. 61/842,351, filed on Jan. 23, 2013, to the U.S. Provisional Patent Application No. 61/ filed on Jan. 23, 2013, on behalf of the first inventor, Nahid Harjee. 857298 "System and method for providing uniform printing of OLED panels"; U.S. Provisional Patent Application No. 61/898769, filed on Nov. 1, 2013, on behalf of the first inventor, Nahid Harjee, provides a system and method for uniform printing of OLED panels U.S. Provisional Patent Application Serial No. 61/920,715, filed on Jan. 24, 2013, which is incorporated herein by reference.

PCT Patent Application No. PCT/US2014/035193, filed on Apr. 23, 2014, to the First Inventor, Nahid Harjee, for the printing of fluids within precise tolerances. "Technology for measuring and controlling ink droplets" and the US invention patent application filed on behalf of the first inventor Nahid Harjee on January 23, 2014 Part number 14/162525 "Technology for the control of the amount of printing ink used to deposit fluids within precise tolerances" is a continuation of the case and also claims to represent the first inventor Alexander Sou- on April 26, 2013. U.S. Provisional Patent Application Serial No. 61/816,696, filed on Jan. 14, 2013, the disclosure of which is incorporated herein by reference. U.S. Provisional Patent Application Serial No. 61/866,031, filed on Jan. 26, 2013, the disclosure of which is incorporated herein to On behalf of the first inventor Nahid The benefit of each of the Taiwan patent application 102,148,330, "Technique for Printing Ink Level Control for Determining Fluids within Accurate Tolerances", is claimed by Harjee.

PCT Patent Application No. PCT/US2014/035193 is a U.S. Provisional Patent Application Serial No. 61/816,696 filed on Apr. 26, 2013, on behalf of the first inventor, Alexander Sou-Kang Ko, which utilizes laser light scattering to measure ink. OLED printing system and method for droplet size, velocity and trajectory", U.S. Provisional Patent Application No. 61/866031, filed on Jan. 14, 2013, by the first inventor, Alexander Sou-Kang, Ko. OLED printing system and method for measuring ink droplet size, velocity and trajectory", and U.S. Patent Application No. 14/162525, filed on Jan. 23, 2014, on behalf of the first inventor, Nahid Harjee, for precise tolerances The priority of the technique of controlling the amount of printing ink in the deposition fluid.

U.S. Patent Application Serial No. 14/627,186 is a continuation of U.S. Patent Application Serial No. 14/485,005, which is hereby incorporated by reference. U.S. Patent Application Serial No. 14/485,005, which is incorporated herein by reference in its entirety to the entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire content On April 10, 2014, the first inventor Eliyahu Vronski was represented. U.S. Provisional Patent Application Serial No. 61/977,939, the entire disclosure of which is incorporated herein by reference in its entirety in its entire entire entire entire entire entire entire entire entire entire entire entire entire content 62/005044 "Layer manufacturing based on the use of halftones to control the thickness of the ink", and U.S. Provisional Patent Application No. 62/019,076, filed on Jun. 30, 2014, the entire disclosure of which is incorporated by reference. "Control of the thickness of the ink-based layer manufacturing" priority.

The priority is claimed for each of the aforementioned applications, and each of the aforementioned patent applications is hereby incorporated by reference.

Some fabrication techniques utilize a printing process to deposit a layer of material on a substrate as part of an assembly process. For example, a plurality of solar panels or organic light emitting diode (OLED) displays can be fabricated together on a common glass or other substrate, wherein the plurality of panels are ultimately cut from the common substrate to produce individual devices. . The printing process deposits a liquid having a solvent suspended or loaded with a material (eg, similar to an "ink"), and another permanent layer is, for example, cured, dried, or otherwise "treated" to become A permanent form that is formed from the material. The liquid system is deposited in a carefully controlled manner for each product such that each deposited layer is closely aligned in position with the underlying layer on the substrate and the desired product location. Such alignment is particularly important in situations where a high degree of manufacturing precision is required, such as in the case where the process is used to fabricate a pattern of dense microelectronic structures or optical structures, each of which is Each layer has a carefully controlled size (including thickness).

Returning to an example of an OLED display for illustrative purposes, on a common substrate Each of the flat devices that are fabricated in parallel is characterized by individual pixel color members that are typically fabricated in a well that maintains the elements that produce light and the fluid of the electrodes. Each deposited layer helps determine the proper operation of individual pixels; the more accurate each deposition process (and the better the alignment and control), the smaller the pixel can be made and at its specific size The more reliable the operation of the finished optical or electrical structure is. For a given panel, pixel-to-pixel variation (including thickness variations) is undesirable because these can create visible defects in the finished product; for example, a typical OLED display (eg, Using as an HDTV screen may involve millions of pixels in a small space, and if the pixels are slightly changed in the liquid deposited via the printing process, this may potentially be seen by the human eye as a brightness Or the difference in color. For such manufacturing applications, precise press control, such as at micron or finer resolution and with a maximum variation of less than 0.5 percent per unit area of fluid deposition, is therefore desirable.

Moreover, in order to manufacture a product at an acceptable consumer price point, maximizing the throughput of manufacturing is desirable. If a given manufacturing device (for example, which contains an industrial printer) spends a significant amount of time on each layer, the conversion becomes a slower manufacturing and thus an increased consumer price point; The price point threatens the feasibility of the process.

According to one aspect of the invention, a method of fabricating a tablet device includes: receiving a first substrate; using a printer to print a liquid onto the first substrate, the liquid system carrying a type that will be formed a material of a permanent layer of the tablet device; and processing the liquid therefrom to form the permanent layer. Wherein (a) the method further comprises detecting a position of a reference on the first substrate and depending on a difference between the position of the reference and an expected characteristic position To detect an error, (b) utilizing the printing machine includes receiving stored data defining where the permanent layer of the tablet device is to be printed, calculating the stored data based on the error to generate Adjusted data, and printing small droplets of the liquid onto the first substrate based on the adjusted data, and (c) calculating the stored data further comprising: specifying a printing of the printing press The individual nozzles of the head deposit individual droplets of the liquid in a particular region of the first substrate in response to the error, in a manner that is a target of the liquid deposited in the particular region The total amount is limited to a predetermined error threshold that falls within a desired amount of the liquid relative to the particular region.

According to one aspect of the invention, an apparatus for manufacturing a tablet device includes: means for receiving a first substrate; means for printing a liquid onto the first substrate, the liquid system carrying a A material that will form a permanent layer of each of the individual products to be fabricated in parallel on the first substrate, the tablet device being one of the individual products, wherein the printing system is implemented as a a portion of the overall printing process of the permanent layer of each of the individual products on the first substrate; for processing the liquid to form each of the individual products after printing of the liquid a device of a permanent layer; and means for separating the tablet device from other individual products on the first substrate after the process. Wherein (a) the apparatus further comprises means for detecting a position of the reference on the first substrate and detecting an error based on a difference between the position of the reference and an expected characteristic position, (b) The means for printing includes means for receiving stored data defining where the permanent layer of the tablet device is to be printed, and for calculating the stored material based on the error, Means for generating adjusted data, wherein printing onto the first substrate is performed in accordance with the adjusted material, and (c) the means for calculating the image further comprises: Means for specifying individual nozzles of a printhead of the printing press to deposit individual droplets of the liquid in a particular region of the first substrate in accordance with the error, in one manner A target total amount of the liquid in the particular region is maintained within a predetermined error threshold relative to one of a desired amount of the liquid for the particular region.

According to an aspect of the present invention, an apparatus for manufacturing a tablet device includes: a robot for receiving a first substrate; and a printer for printing a liquid including a material, the material Forming a permanent layer of individual products on the first substrate, the tablet device being one of the individual products, wherein the printer is for printing the individual products on the first substrate The permanent layer, as part of a printing process as a whole; and a curing mechanism for curing the material to form the permanent layer. Wherein the tablet device is separated from other individual products on the first substrate after printing of the permanent layer, the device further comprising a detecting device for detecting on the first substrate a reference position and detecting an error based on a difference between the position of the reference and an expected feature position; and at least one processor configured to: receive the definition corresponding to the tablet device The data of the place where the permanent layer is to be printed, the data is calculated based on the error to generate adjusted data, so that the printer prints onto the first substrate according to the adjusted data; and wherein the at least A processor system further designating individual nozzles of a printhead of the printer to deposit individual droplets of the liquid in a particular region of the first substrate in accordance with the error, A target total amount of the liquid in the particular region is maintained at a predetermined threshold of error relative to a desired amount of the liquid for the particular region Way, in order to produce the adjusted data.

According to an aspect of the present invention, there is provided a machine comprising stored in a non-transitory state The device of the readable medium on the device, the instructions being executed by at least one processor of a device, the device comprising a robot for receiving a first substrate, and a device for printing a printing machine for liquid onto the first substrate, a curing mechanism for curing the liquid to form a permanent layer of the tablet device, and a position for detecting a reference position on the first substrate Detection device. The instructions, when executed, cause the machine to: detect an error based on a difference between the location of the reference and an expected feature location; receiving the permanent layer corresponding to the tablet device to be printed at the Information on a substrate; the data is calculated based on the error to produce adjusted data; and the printer is printed onto the first substrate based on the adjusted data. Wherein the instructions, when executed, further cause the at least one processor to designate individual nozzles of a printhead of the printer to deposit individual droplets of the liquid on the first substrate in accordance with the error In a particular region, the amount of a target in which the liquid is deposited in the particular region is maintained at a predetermined amount relative to a desired amount of the liquid for the particular region. The manner within the error threshold to facilitate the generation of the adjusted data.

<p>101Substrate</p><p>103Printing area</p><p>105Alignment mark</p><p>107Product area (panel)</p><p>107' first panel</p><p>107" second panel</p><p>107''' Third panel</p><p>107''''Fourth panel</p><p>123xy rectangular coordinate error</p><p>125Printing area</p><p>133 Angle of rotation α</p><p>135Printing area</p><p>145Scale error</p><p>147Printing area</p><p>148 Coverage area</p><p>149Scale error</p><p>153 skew error</p><p>155Printing area</p><p>163Printing area</p><p>201 Flowchart</p><p>203Generate/receive information describing the layers across the substrate</p><p>204Multiple products</p><p>205Location/Dimensions</p><p>206 thickness</p><p>207Prescription Editor</p><p>209dotted line</p><p>211rough (mechanical) alignment</p><p>213Detect actual substrate position</p><p>215Adjustable printing to substrate/panel error (calculation)</p><p>217Uninstall</p><p>218Cure</p><p>219Next substrate</p><p>221 Flowchart</p><p>223New substrate</p><p>225Transfer to position 1</p><p>227Capture benchmark (search)</p><p>229Transfer to location (2...n)</p><p>231Capture baseline (2...n)</p><p>233Judge position (offset, rotation, zoom, skew)</p><p>235sampled version</p><p>236 prescription</p><p>237 bit map</p><p>238Other</p><p>239Linear</p><p>240LSF (curve)</p><p>241Other</p><p>242 indicates the parameters of the conversion</p><p>243 Parallel processing; designated processor/core</p><p>245 Calculation/rasterization</p><p>246Nozzle/waveform data</p><p>247Scan path (raster) adjustment/optimization</p><p>248 consistency check</p><p>249Fence/Boundary Treatment</p><p>251Transfer to the printing press</p><p>253Complete</p><p>261Screen map</p><p>263UI display area</p><p>265Second UI display area</p><p>267 Fourth UI display area</p><p>269 Third UI display area</p><p>301flow chart</p><p>303 Generate converted maps</p><p>305 Overlapping the printed grid on the converted map</p><p>307 pixel processing (for example, P'(x',y')=fn{Pnw(x,y), Pne(x+1,y), Psw(x,y+1),Pse(x+ 1,y+1)})</p><p>308A single pixel mapping</p><p>309weighted mapping</p><p>310Rewind</p><p>311Storage new images in the buffer of the sample</p><p>313 Consistent? /Transfer to the press</p><p>314Anti-aliasing</p><p>315 halftone adjustment</p><p>317Completed</p><p>351flow chart</p><p>353Loading layer position/size</p><p>355 Calculate conversion parameters</p><p>357 produces converted representations</p><p>359According to the conversion parameters, directly calculate the bitmap from the object prescription</p><p>360Affine conversion</p><p>361 Calculate the bitmap from the transformed representation</p><p>363Rasterization/consistency/to the printing press</p><p>365Complete</p><p>401example</p><p>403Original print grid (area)</p><p>404, 405 Printed pixels</p><p>407Print head</p><p>409Nozzles</p><p>410 nozzle</p><p>411example</p><p>412Printing area</p><p>413Offset</p><p>415, 417 examples</p><p>419Printed pixels</p><p>421Layout (example)</p><p>422Print grid</p><p>431Layout (example)</p><p>433Printing area (product geometry)</p><p>435Position error angle α</p><p>441 layout (example)</p><p>443Printed pixels</p><p>445, 447 examples</p><p>451Example</p><p>455Printing area</p><p>457Zoom vector</p><p>465, 467 pixels</p><p>461 layout (example)</p><p>475 skew error</p><p>477 skew adjustment formula</p><p>481 layout</p><p>483Print grid points</p><p>485, 487 print pixels</p><p>501Examples</p><p>503 processor</p><p>505Information</p><p>507, 508, 509 Piezoelectric Transducers</p><p>511, 512, 513 small droplets</p><p>515Small droplet measuring device</p><p>517 memory</p><p>519Sequence (synchronous) signal</p><p>521clock tree</p><p>523, 524, 525 nozzle drive</p><p>527, 528, 529 nozzle</p><p>531, 532, 533 register</p><p>535Sports/Scan</p><p>536Synchronization</p><p>537Rotation of the print head</p><p>541 Circuitry</p><p>543Scratch file</p><p>545First counter</p><p>546second counter</p><p>547 Waveform Generator Circuit</p><p>548Digital to analog converter</p><p>549 counter</p><p>550High voltage amplifier</p><p>551, 553 waveform</p><p>555, 557, 559, 561, 563, 565, 567 signal level</p><p>571Design</p><p>573CPU</p><p>575Ethernet connection</p><p>577Field Programmable Gate Array (FPGA)</p><p>578Soft Processor</p><p>579 Dynamic Random Access Memory (DRAM)</p><p>580FPGA Logic</p><p>581Amplifier</p><p>583Print head</p><p>601 implementation class</p><p>603Soft form</p><p>605 computer image</p><p>607Printer control data (storage media graphics)</p><p>609Manufactured device image</p><p>611Array</p><p>613Portable digital device</p><p>615 TV monitor screen</p><p>617 solar panels</p><p>621Multi-room manufacturing equipment</p><p>623Transmission module</p><p>625Printing module</p><p>627Processing module</p><p>629Input carrier</p><p>631Transmission room</p><p>633 Ambient buffer room</p><p>635 gas enclosure</p><p>636Transmission room</p><p>637Output carrier</p><p>639Nitro Stack Buffer</p><p>641Cure room</p><p>651Printing room</p><p>653Substrate</p><p>655Support table</p><p>657Air support</p><p>659Print head</p><p>661 activity ring</p><p>663x-dimensional movement</p><p>665Nozzle</p><p>667Rotary print head</p><p>669Axis legend</p><p>671 Equipment</p><p>673 processor</p><p>675Print head</p><p>677Substrate</p><p>679Print head movement system</p><p>681Substrate transmission system</p><p>683Image capture device</p><p>685Ink delivery system</p><p>687Print head maintenance system</p><p>689room control subsystem</p><p>690 atmosphere control</p><p>691Small droplet measurement</p><p>692Small droplet volume, velocity, angle</p><p>693 memory subsystem</p><p>694A set of transmit waveforms</p><p>695Substrate change (error) processing</p><p>696Print optimization</p><p>701Flowchart</p><p>703Receive layer data (boundary, thickness)</p><p>705Converted to printed cell data (grayscale data)</p><p>706Development/Grid</p><p>707Adjust grayscale values (boundary processing/error correction)</p><p>709converted to halftone; error diffusion</p><p>710Production of printed images</p><p>713 Error correction</p><p>714 calibration/experience</p><p>715 Waveform adjustment</p><p>716 borrowing of nozzles</p><p>717Scan path adjustment</p><p>718 halftone error</p><p>719Nozzle verification</p><p>720Small droplet measuring device</p><p>721Statistical model</p><p>722Printing/Scanning Project</p><p>725The final printed material</p><p>801 optical system</p><p>803Print head assembly</p><p>805A, 805B print head</p><p>807Nozzles</p><p>808 coordinate system</p><p>809 ink collector</p><p>811Axy motion control</p><p>Light recovery under 811B</p><p>811C supply under the plane</p><p>813 dimension plane</p><p>815Measurement area</p><p>817Light source</p><p>819 Optical equipment for optical transmission</p><p>821Light sensor</p><p>823Light recycling optical equipment</p><p>825 For example, focusing</p><p>827 Non-imaging</p><p>831Program flow</p><p>833Docking print head assembly</p><p>835Investment in small droplets</p><p>837 XYZ activity of one or more optical components around the docked print head</p><p>839n measurement/small droplets, developing statistical reliability for specific quantities</p><p>841Interferometry or shading; near-instantaneous measurement/calculation</p><p>843Dynamic measurement/dynamic update</p><p>845Accuracy<img file="TW201611902A_D0001.tif" wi="30" he="38" img-format="tif" img-content="character" orientation="portrait" inline="no" />10^-2Picoliter</p><p>848Online printing</p><p>849 separate line</p><p>851Method of measuring small droplets</p><p>853 Find the print head reference</p><p>854 Find the corner/corner nozzle</p><p>855Print head configuration information</p><p>856 Snap to the grid</p><p>857Interpolation</p><p>858Loading nozzle</p><p>859Get the grid address of the nozzle</p><p>860Access grid address; search and centering</p><p>862Adjust the relative height of the print head/optics and/or strobe timing</p><p>863Small droplet measurement area</p><p>864 small droplets</p><p>865 Track</p><p>866 small droplets</p><p>867 Track</p><p>868Measurement, speed, trajectory, nozzle xy (nozzle bending)</p><p>869Measurement error processing</p><p>870Storage measurement</p><p>871 Loop</p><p>872Storage statistics</p><p>873 Error handling</p><p>874Update indicator to next nozzle</p><p>875New substrate (interrupted)</p><p>876 Move the small droplet measurement system and perform the next measurement</p><p>881 method</p><p>883 Capture data for each nozzle/nozzle-waveform</p><p>885 memory</p><p>887μ, σ is in range</p><p>889Rejection/abnormal criteria</p><p>891Rejecting small droplets</p><p>892Additional measurements</p><p>893Adjust waveform</p><p>894Adjust timing</p><p>895Re-measurement/storage</p><p>897Next</p><p>901Substrate</p><p>902 Panel products</p><p>903 alignment mark</p><p>905 alignment mark</p><p>906 camera</p><p>907, 908 scan path</p><p>909 Alignment mark</p><p>911 Alignment and detection procedures</p><p>921Examples</p><p>923Substrate/panel prescription information</p><p>925 cached stored samples (for example, object/bit map)</p><p>927Detection geometry (reference)</p><p>928 once</p><p>929 intermittent</p><p>930Repeated</p><p>931Different benchmarks</p><p>932Continuous</p><p>935 Calculation conversion</p><p>937 Find out the information</p><p>938 closest pixel</p><p>939weighted measurement</p><p>940Affine conversion</p><p>941Other</p><p>943 Consistency {quantity/density}</p><p>945Check nozzle data/select waveform, change nozzle</p><p>946Nozzle/nozzle-waveform state (quantity, α, β)</p><p>947Measured at the original location</p><p>948Verification</p><p>951 Well</p><p>953Print grid</p><p>954 nodes (small droplet position)</p><p>955 Error?</p><p>957Error handling</p><p>959Storage; transfer to indicator/print head</p><p>961 cached sample</p><p>1001Flowchart</p><p>1003 Identify well/area; number of original droplets</p><p>1005New number of small droplets (v, α, β, σ_v, σ_α, σ_β) </p><p>1006Different waveforms</p><p>1007 different nozzles</p><p>1008 nozzles available?</p><p>1009Adjustment (depending on necessity)</p><p>1011 Error?</p><p>1012Adjustment steps/recalculation</p><p>1013Adjust the number of scans</p><p>Completed on 1014</p><p>1021Flowchart</p><p>1023For N: Identification Wells</p><p>1025Get nozzle data</p><p>1027simulation</p><p>1029 Error?</p><p>1033Adjustment</p><p>1037Storage (modified bitmap)</p><p>1035Nozzle/waveform data</p><p>1039Change waveform_i</p><p>1041Flowchart</p><p>1045Nozzle/waveform data</p><p>1049Smoothing</p><p>1051 Output; write to memory</p><p>1053Fixed scan offset (grating)</p><p>1055Reevaluation of scan offset (grating)</p><p>1057 Minimize printing time</p><p>1061 Flowchart</p><p>1067Density details (thickness)</p><p>1069 Smoothing variation; adjusting waveforms; adjusting adjacent nodes</p><p>1073 Error?</p><p>1075written into memory; next</p><p>1077Exception</p><p>1081Flowchart</p><p>1083 Expected location</p><p>1085 memory</p><p>1087Detecting related variations (eg, non-substrate variation)</p><p>1088 calibration</p><p>1089Monitoring benchmark</p><p>1090 dead reckoning</p><p>1091 Learned; adapting the algorithm of cache/calculation depending on necessity</p><p>1093Storage relevance (considering the future calculations)</p><p>1095Detecting and correcting errors</p><p>1103Master (supervised processor)</p><p>1105 parallel processor</p><p>1107 core</p><p>1109Inline DRAM</p><p>1111 group (port, array)</p><p>1111A register</p><p>1113 Ultra-wide access path</p><p>1119Sub-panels are assigned to individual cores</p><p>1120Each substrate/panel conversion (sample memory)</p><p>1121 separate conversion operations</p><p>1122Other efficiency</p><p>1123Line type storage (A, B, C, D) / line offset</p><p>1151 Flowchart</p><p>1153Loading the sample (stored by cache)</p><p>1155 Panel/Sub Panel Data</p><p>1157 Find alignment marks</p><p>1159 Panel/Sub Panel Conversion Calculation</p><p>1161Main control configuration range</p><p>1163 Convert each range (each included panel) parameters</p><p>1165 Core/Processor Processing Specified Range</p><p>1167 Print head information</p><p>1169Storage Bitmap/Rasterization</p><p>1171Completion/printing</p><p>Hdirection (z-axis height)</p>

Figure 1A shows the layout of a substrate that will receive a liquid print that will form a permanent layer of individual products at location 107; an integral layer to be used by a printhead nozzle The printed area is represented by dashed squares 103.

Figure 1B shows a printed area 125 similar to the dashed box 103 from Figure 1A, but wherein the substrate is unintentionally offset in position such that, without correction, printing will occur at the wrong location, thus Misregistration can result.

Figure 1C shows a printed area 135 similar to the dashed box 103 from Figure 1A, but with the rotation error of the substrate therein.

Figure 1D shows a printed area 147 similar to the dashed box 103 from Figure 1A, but wherein the substrate misalignment represents a scaling error, such as where the individual arrays of products may have dimensional errors (and locations), respectively. Error).

Figure 1E shows a printed area 155 similar to the dashed box 103 from Figure 1A, but with substrate misalignment (e.g., edge distortion in the alignment process of a leading edge) producing distortion in two separate dimensions .

Figure 1F shows a printed area 163 similar to the dashed box 103 from Figure 1A, but with different types of errors being represented, each error affecting a different portion of the substrate.

2A provides a flow chart 201 relating to alignment of one or more printed deposition regions with respect to substrate locations.

2B provides another flow diagram 221 relating to alignment of one or more printed deposition regions with respect to substrate position.

2C is a screen diagram from a recipe editor and panel definition software that is used to define the layer geometry for each product design in an array (ie, the array will utilize one A common substrate is fabricated and layer deposition will be repeated for a series of substrates).

Figure 3A provides a flow chart 301 relating to a particular method of adjusting a "pattern" to fine tune printing based on detected errors.

Figure 3B provides another flow chart 351 relating to a particular method of adjusting a "pattern" to fine tune printing based on detected errors.

Figures 4A-4I are used to discuss the handling of alignment errors and, more specifically, to be explained Used to specify or reassign printhead nozzle firing decisions to fine tune the process of printing.

4A shows a layout of a print head 407 relative to a region 403, wherein the presence or absence of a position in a "print grid" indicates that a small droplet will be used by the print head, respectively. It is emitted in a particular area (e.g., from nozzle 410 to position 404) or blocked in a particular area (e.g., when traversing position 405 from nozzle 410). For example, FIG. 4A can be a section of a "bitmap" (ie, it represents a desired portion of a substrate when the printhead and substrate are moved relative to each other during a scanning motion). Nozzle launch decision). It is noted that, as used herein, the term "bit map" refers to the data for each nozzle, regardless of whether the data for each nozzle consists of a single bit or a plurality of bits; For example, in one embodiment, 4 bits of data may be utilized (i.e., it represents sixteen possible values, one representing a decision not to transmit the nozzle, and the other representing 'can be transmitted' 'The different pre-programmed nozzles drive the waveform to the nozzle. Obviously, many alternatives are also possible.

4B shows a layout of the original print grid 403 from FIG. 4A, but wherein the substrate is misaligned relative to the desired print area, and wherein different nozzles can be utilized to compensate for this without having to adjust the scan path Alignment error; it is noted that in the depicted example, the misalignment may not be "tidy" aligned with the nozzle spacing or "pitch", for example, when the print head and substrate are moved relative to each other, A larger or smaller number of nozzles of the printhead and a different set of nozzles may now overlap the desired print area 412.

Figure 4C shows a layout 421 in which a printed grid can be seen overlying a desired print area 412 from Figure 4B.

4D shows a layout 431 associated with one of the regions 403 of FIG. 4A, but where it can be seen that the desired printed area or product geometry 433 is relative to substrate error or other misalignment. This area is rotated.

4E shows a layout 441 in which a printed grid can be seen overlying a desired print area 433.

Figure 4F shows a layout of the original printed grid 403, but with a detected scaling error associated with the particular substrate, i.e., as reflected by the desired printed area 455.

Figure 4G shows a layout 461 in which a printed grid is seen to be overlaid onto the desired print area 455 from Figure 4F.

Figure 4H shows a layout of the original printed grid from Figure 4A, but with a detected skew error 475, for example due to some type of distortion or linear advancement error affecting the substrate. What is actually required is to print along the contour of a parallelogram (represented by shape 475).

Figure 4I shows a layout 481 in which it can be seen that the printed grid is overlaid onto a desired print area 475; wherein the printing system utilizes the print grid of Figure 4I and is calibrated using the techniques presented herein, the print can be Aligned accurately (i.e., represented by the shape 475).

Figure 5A shows a circuit that is used to specify or adjust nozzle drive waveforms (i.e., electronic drive signals) to/for different nozzles of a printhead.

Figure 5B shows a drive circuit that can be utilized to store each nozzle of data for a corresponding printhead nozzle, wherein the stored data enables a programmable waveform definition (and thus enables a variable amount) , droplet trajectory, drop location, velocity, or definition of other droplet parameters for each nozzle).

Figure 5C shows an example waveform and is used to help illustrate how the waveform can be benefited. Use to change the nozzle droplet parameters by programmable adjustments.

Figure 5D shows another example of a drive circuit for a nozzle of a printhead.

Figure 6A is a diagram showing a series of illustrative examples of a hierarchy, product or service that may be independently embodied in the techniques described herein; for example, the techniques presented herein may be implemented in the form of a software (according to component symbol 603), or as press control data (printed on a substrate to control a printer according to component symbol 607), or as a product made according to these techniques (eg , as exemplified by the symbol 613, 615 or 617).

Figure 6B is a schematic view of a manufacturing apparatus including a printing press.

Figure 6C is a plan view showing a printed area within a manufacturing apparatus; the printed area may optionally be contained within a gas enclosure, i.e., such that printing occurs in a controlled atmosphere. .

Figure 6D is a functional block diagram of a manufacturing apparatus including a printing press.

FIG. 7 is a flow chart 701 providing a measurement used to describe the effects of individual nozzle changes and/or nozzle-drive waveform changes, and associated compensation techniques. In one embodiment, printing is not only adjusted for positional errors, but variations from nozzle to nozzle are also considered in this adjustment to allow for printing to be produced within carefully defined tolerances.

Figure 8A is a diagram showing an illustration of a small droplet measurement system capable of measuring the amount of small droplets for each nozzle of a large printhead assembly. For example, such a system can be utilized to measure the change from nozzle to nozzle just cited.

Figure 8B is a process diagram showing various processes and options associated with the measurement of droplet details for each nozzle to achieve a highly reliable understanding of the desired droplet characteristics.

Figure 8C is a flow chart showing an embodiment associated with small droplet measurement.

Figure 8D shows a flow diagram associated with nozzle verification; in other words, in one embodiment, (in addition to other purposes described herein) measurements can be utilized to verify that the nozzle is acceptable or unacceptable.

Figure 9A provides a block diagram relating to a raster process; more specifically, Figure 9A is useful to illustrate how the results of small droplet measurements can be correlated with detected substrate errors (e.g., position, Integration of rotation, scaling or skewing errors to provide accurate printing. A shaded area (907) represents a single scan path and a blank area (908) represents another scan path.

Figure 9B shows a flow diagram associated with printing on a substrate as part of a process.

Figure 9C shows another flow diagram associated with printing on a substrate as part of a process.

Figure 10A shows a flow diagram associated with printing on a substrate as part of a process.

Figure 10B provides another flow diagram associated with printing on a substrate as part of a process.

Figure 10C provides a flow chart 1041 relating to antialiasing for geographies to be accepted, for example to test whether "adjusted" press control data still conforms to one or more desired criteria ( For example, th₁<img file="TW201611902A_D0002.tif" wi="30" he="39" img-format="tif" img-content="character" orientation="portrait" inline="no" />the amount<img file="TW201611902A_D0003.tif" wi="27" he="41" img-format="tif" img-content="character" orientation="portrait" inline="no" />Th₂And further adjust the press control data if it does not meet the one or more desired criteria.

Figure 10D provides a flow chart 1061 for incorporating the updated nozzle-small droplet data into the printer control data to mitigate substrate-printer variations.

FIG. 10E provides yet another flow chart 1081 for incorporating the updated nozzle-small droplet data into the printer control data to mitigate substrate-printer variations.

Figure 11A is a block diagram showing an environment for parallel processing.

Figure 11B is a diagram showing how the calculation and/or rasterization of a printed image is in a flat state. A block diagram executed in a row-processed environment.

The subject matter defined by the scope of the appended claims may be better understood by referring to the following detailed description, which should be read in conjunction with the accompanying drawings. The description of the one or more specific embodiments is set forth below in order to enable the invention to be able to Instead, exemplify its application. Without limiting the foregoing, this disclosure provides several different examples of techniques for fabricating a thin film for each of a plurality of products (or other array of products) for a substrate, as A part of an overall repeatable printing process. The various techniques can be embodied as software for performing the techniques, in the form of a computer, printer or other device for performing such software, having control data for forming such a layer (eg, a printed image) The form, either as a deposition mechanism, or in the form of an electronic or other device made by these techniques (e.g., having one or more layers produced in accordance with the techniques). Although specific examples are presented, the principles described herein can be applied to other methods, apparatus, and systems.

This disclosure provides techniques, processes, equipment, devices and systems that can be utilized to make products faster and more reliably via a printing process, as well as products made according to such processes.

An assembly line process uses a printer to print (i.e., deposit) small droplets of liquid onto each of a series of substrates. The liquid system is then cured, dried, or otherwise treated to form a permanent layer of material (i.e., a permanent film). Each substrate is used to make one or more products, and the layer will typically be formed for carrying by the substrate Every product. Once the printing is completed, the substrate is advanced and a new substrate is loaded, so the process is repeated. For example, the process of curing the liquid can occur in some embodiments in the original position in the assembly line, or in another location, and typically the printing and processing is performed within a controlled atmosphere. To minimize contamination by particles, oxygen or moisture. In a particular application, the printer is used to deposit an organic light emitting diode ("OLED") display panel or a single layer of solar panels. For example, in the fabrication of high-resolution OLED television screens, such processes can be utilized to deposit one or more light-generating layers for each pixel of a display (ie, each discrete light-generating layer) element). Although the described techniques can be applied to many types of materials as well as many types of products other than flat panel displays, the use of a printer and associated processing has been found to be particularly useful for deposition without the ease of utilizing other processes. The deposited organic material layer, and many of the examples presented herein, will therefore focus on these materials. As with other types of products and processes, alignment in printing and layer thickness must be achieved with a high degree of accuracy to enable small, reliable electronic components with low variations and high manufacturing throughput (eg, micron scale) ) becomes easy.

Due to various process difficulties associated with different substrates, assembly line equipment, manual processing, and numerous other factors, each substrate may be in position, rotation, scale as the substrate is moved into and through the printer. A slight change in skew, or other dimensions. Some forms of error may be specific to the substrate (eg, warpage of a substrate or edge nonlinearity), while other forms of error may represent repeated systematic errors (eg, repeatable due to the system itself) Under the motion error, a substrate is erroneously edge-guided through the printer. Regardless of the source, since the printing process is intended to print the same product on each of the substrates in the series, it is generally desirable to detect and mitigate positional errors.

Thus, in one embodiment, a detection mechanism is used to detect fiducial on each substrate as the substrate approaches the printer. These references are used to identify product locations, although there are variations between different substrates or other deviations from the expected product position, orientation, and/or size. The detected product location is compared to the expected location and is used to identify the error to facilitate (software) conversion or adjustment of the press control data (ie, to modify the data or how it is calculated) Figure to perform printing) to accurately align the layer to be printed with any product size below. As used herein, "position" or "product position" should be interpreted broadly to include any skew, scale, orientation, etc., even if these items are not listed separately when referring to "location". .

It is noted that in a conventional printing process, an assembly comprising one or more print heads each having a nozzle is being scanned relative to one or more of the sweeps in one or more of the sweeps. The printed substrate is transferred for transmission. Each sweep span defines an "in-scan" dimension of the substrate, and after each sweep, the printhead and/or substrate is attached to a "cross-scan" of the substrate. "The size is reset to prepare for a subsequent sweep. Each print head can have hundreds to thousands of nozzles, each nozzle for ejecting a small droplet of liquid, wherein the liquid system is similar to an ink and contains a treatment that will be cured or otherwise To form the material of the desired permanent layer. For example, under a known technique, the liquid is a monomer or polymer, and an ultraviolet curing process is used after printing to treat the deposited liquid to form a permanent layer.

In a more detailed variation of this first embodiment, some mechanisms can be utilized to reconcile rapid printing, accurate printing, and precise layer thickness and alignment goals.

In a first embodiment, the error is identified (eg, one in place) Set, rotate, skew, or linear or non-linear offset on the scale), the new nozzle firing decision can be based on a deviation from the detected deviation from the ideal position, and with an offset Or by changing the manner of pre-planned raster sweep between the printhead assembly and the substrate. In other words, since the printing time (and thus the manufacturing throughput) is directly related to the number of rasters required to cover all areas of the substrate that will receive the liquid, the number of raster sweeps and their individual positions are Not changed in an embodiment, but potentially new nozzles and/or drive parameters are individually specified in a manner that allows printing to occur to deposit the desired droplet density and pattern from the printhead assembly. , but with a nozzle designation and/or drive parameters (eg, drive waveforms) that are converted or calculated relative to the original press control data to facilitate alignment with any of the underlying product layers, or It is specified exactly in the way that the desired product is geometrically aligned. It is noted that the manner of such processing is optional, that is, in other contemplated embodiments, the raster sweep is re-evaluated to possibly utilize different offsets of the printhead assembly relative to the substrate (eg, , different advance modes in the direction of the scan and/or the number of different scans or sweeps).

Depending on the environment, the decision of a simple "changing" nozzle launch may not always produce the desired deposition parameters. For example, in a situation where the printing system is used to deposit light-generating elements in the "well" of the fluid, and the well system contains the deposited liquid until the liquid is cured, the error may not be Neatly aligned with the nozzle spacing or nozzle firing timing such that a linear displacement at the nozzle firing specification may place too much or too little deposited droplets in a pixel well. To address this, in an embodiment, an "anti-aliasing" process is used to test the expected position of each pixel well in accordance with one or more processors acting under the control of the indication logic (giving the detected Error) relative to the expected amount of liquid, and in case the expected total amount deviates An ideal value or range of acceptability exceeds a threshold amount, then selectively re-viewing the "offset" nozzle designation and adjusting one or more of the number of nozzles that will emit small droplets into the pixel well. This then helps to ensure that the exact amount of liquid is deposited in the desired location.

In another embodiment, the small droplets produced by each nozzle are empirically measured at the original location and the measured changes between different nozzles, or based on The droplet characteristics are used to store the droplets from a nozzle, and the droplet characteristics are stored and taken into account in the nozzle designation. For example, in a first variation, individual nozzles may be generated based on whether empirically measured droplet parameters for the nozzle (and for the applied drive parameters) are consistent with acceptable droplets. Standard, and verified as pass/fail. Thus, if the positional error and associated alignment means that an unverified nozzle should not be used for deposition, the indication logic can use the verification data to specify the nozzle so as to alternatively use other nozzles or different small Droplet ejection parameters (eg, a different drive waveform to control a selected nozzle). In a second variation, each given nozzle (and each of the plurality of alternate drive waveforms that can be utilized to drive the given nozzle) is measured for an expected amount of droplets, trajectory And/or the droplet drop position, and this data is then taken into account in the error handling or initial nozzle designation. A simple example is given. If two adjacent nozzles are dependent on a total of 20.00 pL of liquid, they will be dependent to deposit 9.97 and 10.03 picoliters (pL) of droplets into a fluid well, respectively, and the substrate. The error means that two other nozzles will be used to perform the deposition, then (a) a non-adjacent nozzle that is expected to produce 9.90 and 10.10 pL of small droplets can be designated to perform the deposition (keeping 20.00 pL) The desired amount of total), and/or (b) nozzle drive details (eg, drive waveforms) for one or more nozzles may be adjusted to produce a small fluid having an adjusted amount or trajectory characteristic from a preselected nozzle. Drop to maintain the desired total amount in the well of the fluid The tube has detected errors. These types of processing are also optional and not required for all embodiments.

It is noted that the printer control data can take many forms depending on the embodiment. In one embodiment, the "prescription" information describing the desired size (eg, including thickness) for the desired layer of each product can be stored as a cached sample (cached). Template), and then calculate the graph in an adjustable way that takes into account the error of the execution phase. In a second embodiment, the information herein may be partially pre-processed and stored in an object (vector) or other representation to serve as a template for processing cache storage for each substrate. When a position or alignment error is detected for each new substrate, the pattern is captured in the final printer control data of the calculation and modified as appropriate (in other words, the nozzle as appropriate) Launch decisions, raster sweeps, and associated timing). Another example is given below. In another embodiment, the prescription data is received for a product (and/or array) and is calculated to be a bitmap representing the printing decision; the bitmap is substantially Is an array of nozzle emission decisions or equivalent data at each node of a printed grid that is representative of the nozzle on the substrate during a scanning motion of the print head relative to the substrate Triggering at a discrete location or not emitting a small droplet. This information can also include variable drive waveform definitions or selections for each nozzle. The bit map is cached as the same version and is fetched during the execution phase, modified via direct processing on the bit map to cause the print to morph to match the error, and then used to control printing. Of course, other examples exist, that is, under each change, the detected error is considered in some way into the calculation process, in order to be customized in consideration of the detected error. Print to achieve alignment of the layers.

Under various embodiments, in order to make the height accuracy accurate without increasing the printing time The printing becomes easy, and the misalignment detected by the substrate (or any individual product represented thereon) in any given manufacturing iteration can be corrected, at least in part, by software. This correction then reduces the need for time consuming, very high precision mechanical alignment or repositioning, and the need for a very high precision alignment mechanism. This reduces the cost and increases the throughput of manufacturing without sacrificing accuracy and reliability. In a well-thought-out application related to the printing of arrays of multiple large OLED TV screens, a single large substrate may contain 6-8 HDTV screens (eg panels or sub-panels, which will be interchangeably Use); For a successful process, what is currently thought is that a layer to be printed on the substrate should be no more than 90 seconds per substrate. In some embodiments, this maximum printing time is expected to be 45 seconds or less. Since screens or panels involve millions of pixels, alignment should be accurate for the manufacture of high-quality displays; therefore, it takes less than a few seconds (for example, 2 seconds) to detect And a process for correcting misalignment or other errors in the substrate and/or product to print the liquid onto the substrate. The disclosed embodiments make this speed easy to manufacture (and throughput of manufacturing) and assist in producing products that are smaller and more precisely aligned.

Many manufacturing applications are quite robust to the rough mechanical alignment of the products being assembled. For the disclosed embodiment, a processor (ie, a machine) that functions under the control of one or more software (instructions or indication logic stored on non-transitory machine readable media) The fine-precision alignment performed is usually to adjust the millimeter to sub-millimeter error in position offset, rotation offset, skew, scaling error, or other distortion (eg, from nanoscale to number) One hundred microns, or smaller). It is noted that for certain features in applications where very dense precision structures are used (eg, HDTV applications have millions of pixels) and/or where thousands of print head nozzles are involved, error handling may A lot of computing resources are needed. Such as As mentioned above, in one embodiment, it is desirable to utilize hardware logic and/or software logic to perform error compensation within a few seconds of, for example, 2 seconds or less. In order to make this goal easier, in some embodiments below, hardware design and program for relying on parallel processing (eg, multiple processors or a multi-core processing) that allows this calculation to be generated quickly are proposed. The thread specifies the technology. For example, to predict an embodiment that will be discussed further below, a supervised processor or group of processors can detect errors and determine a recipe for converting or converting the original prescription information (or other cache storage). Formulation of the sample material). The formulation over the substrate can be linear or can vary regionally (e.g., non-linear or intermittent). The supervised processor then assigns a section of the layout and an associated affine to each core of a multi-core processor for processing. In one design, each core has its own dedicated memory for data processing and manipulation; the supervised processor identifies each core and associated modification algorithms (eg, affine transformation) The modified layout section and providing this information to the individual cores, and the plurality of cores then respectively perform their configured work to contribute to completing the converted output, which overall represents the adjusted printing for the substrate Machine control data; this converted output is suitable for immediate printing. The memory for processing can also be designed to provide direct memory access (DMA), if desired, to speed up the processing of the cached stored pattern material and its adaptation to printing and use in printing. For embodiments that use tens to hundreds or more cores, this essentially reduces the processing time required for reprinting the desired printing job. Other designs are also possible; for example, instead of specifying a different layout for each core, different mathematical operations (or different sequential programs) can be assigned to each core. As this example should clarify, almost any function or division of processing can be provided to maximize efficiency or reduce processing time. Relative to other techniques described herein, one For example, the parallel processing environment just discussed should also be considered as an option.

The present disclosure provides hardware (ie, circuitry or circuitry) implementations that facilitate highly accurate printing in an assembly line type process, and may be substituted for such hardware or additionally The software technology that is being used. In general, the features discussed below can be mixed and matched (or otherwise optional) and can vary depending on the implementation. For example, an embodiment discussed below provides a hardware that is pre-stored for each of a number of nozzles (eg, hundreds to tens of thousands of nozzles) that can be utilized for a printer (eg, 16 Customizable nozzle drive waveforms. Each waveform is previously selected to produce a slightly different expected droplet parameter (eg, amount, trajectory, etc.) that provides a range of possible small droplets that can be generated from the nozzle; the system These waveforms are selected in advance to provide a selection range, and the various waveform selections are programmed in advance to drive the circuitry for each nozzle. Thus, during the execution phase, the system only needs to select one of the waveforms. In one embodiment, there are sixteen such selections (e.g., one representing a "zero" waveform with no decision to transmit, and fifteen variable drive waveforms). In another embodiment, a predetermined waveform can be programmed asynchronously (ie, in advance), and a binary "trigger" is then applied to "start" regardless of which one was recently programmed into a The default waveform. As such, these various features are optional and not required by all embodiments, and the various disclosed features may be utilized in any desired combination or permutation as appropriate to the embodiments. All such combinations and permutations and any such combinations and permutations are contemplated by the teachings of the disclosure.

It is expressly contemplated that an embodiment may include an apparatus that includes instructions stored on a non-transitory machine readable medium. Such indication logic can be written or designed in a manner that has some structure (structural characteristics) such that when the instructions are finally executed, it They cause one or more general purpose machines (eg, a processor, computer, or other machine) to function as a special purpose machine having the inevitable execution of the instructions on the input operands in accordance with the instructions. Work to take a specific action or to produce a specific output structure. As used herein, "non-transitory machine readable medium" means any physical (ie, physical) storage medium, regardless of how the material is stored on the medium, which is unrestricted. The following includes random access memory, hard disk memory, optical memory, a floppy disk or CD, server storage, volatile memory, and other entities in which instructions can be continuously retrieved by a machine. mechanism. The machine readable medium can be in a standalone form (eg, a program disc or solid state device) or embodied as a laptop, portable device, server, network, printing A portion of a larger mechanism of one or more devices of the machine, or other group. According to an embodiment, the instructions may be implemented in different formats, for example, to become metadata, to be a Java code, or a scripting when it is called to be effective to cause a certain effect. a code written in a particular stylized language (for example, as a C++ code), as a set of instructions for a particular processor, or in some other form; the instructions may also be by the same processor or different processors, Or the processor core to implement. Throughout this disclosure, various procedures will be described, any of which can be implemented substantially as instructions stored on non-transitory machine readable media, and any of which can be Used to make products, for example, using microelectronics, micro-optics, "3D printing" or other printing processes. Depending on the product design, such a product can be manufactured in a form suitable for sale or as a preparatory step for other printing, curing, manufacturing or other processing steps that will ultimately result in a finished product. For sale, distribution, export or import. Also according to an embodiment, the instructions may be executed by a single computer, and in other cases, may be distributed on a decentralized basis. The storage and/or execution, for example, utilizes one or more servers, network clients, or special purpose devices. Each of the functions mentioned herein with reference to the various figures can be implemented as part of a combined program or as a separate module that is stored together in a single media representation (eg, a single floppy disk) ), or on multiple individual storage devices. The above content is also true for the printed image or the printing machine control data, that is, the prescription information generated according to the method described herein, and the same version or the result of processing the prescription information or pattern can be stored in the non-temporary. State machine readable media for temporary or permanent use on the same machine, or for use on one or more other machines; for example, press control data may be utilized by a first machine Generated, and then stored for transmission to a printer or manufacturing device, such as via an internet (or other network) for downloading, or manual transmission (eg, via a transmission medium such as a DVD) ) for use on another machine.

Furthermore, reference has been made above to a detection mechanism and a reference that is recognized on each substrate. In many embodiments, the detection mechanism is an optical detection mechanism that uses a sensor array (eg, a camera) to detect an identifiable shape or pattern on a substrate. Other embodiments are not based on an array, such as a line sensor that can be utilized to sense a fiducial when a substrate is loaded into or within the printer. It is noted that certain embodiments rely on dedicated patterns (eg, special alignment marks), while other embodiments rely on identifiable substrate characteristics (which include the geometry of any previously deposited layers). Each one is a "benchmark." In addition to utilizing visible light, other embodiments may rely on ultraviolet or other invisible, magnetic, radio frequency or other forms of detection of substrate details associated with the intended printing location.

Now that some basic embodiments have been introduced, this disclosure will now continue. Continue to explore more detailed implementations. Figures 1A-1F are used to depict various principles associated with press control data adjustment in a repeatable process.

FIG. 1A depicts the layout of a hypothetical substrate 101. More specifically, the substrate comprises a plurality of alignment marks 105, a printed area 103 defined relative to the alignment marks, and an area 107 corresponding to some of the products that are to be formed, or have been There is a substrate layer that is deposited as part of the process of product formation. For purposes of illustration, it can be assumed that the substrate 101 will be used to form a plurality of products 107, and that each "panel" will be cut from the substrate to form a separate solar panel or display device, but the present invention is not This is limited. In one embodiment, each panel 107 will be a separate OLED panel and the substrate 101 is a large piece of glass.

In spite of the presence of multiple products, it is desirable to utilize a printer in a unitary printing process to eject liquid onto the array or substrate 101; the liquid system carries a material such that the deposition of the liquid and its After curing or other processing, the material and/or liquid will become a permanent part of each of the resulting products, and the layer will have a well-defined thickness. In an alternate embodiment, the layer thickness is imparted by controlled droplet deposition, wherein the amount or density of liquid deposited per unit area is used to construct the layer thickness. In other words, the liquid system has limited diffusion, which results in blanket-covered liquid coverage without undesired holes or gaps, or the liquid system is otherwise deposited or hardened in a manner that would be geometrically limited, all Both are in a way that would otherwise produce the thickness of the explicit plan; this process is generally referred to herein as "halftone", even though the deposited fluid is usually colorless and is not used to produce any type. Hue (ie, "half", "mixed", or other). It is noted that in a typical embodiment, the printer prints the entire substrate (ie, in the array) Each product of a layer in the column), and the substrate is then transferred from the printer to an individual curing chamber in which the liquid will be cured or hardened, all in a controlled atmosphere present Medium (for example, a nitrogen or other non-ambient air atmosphere that prevents the liquid from being exposed to moisture, oxygen, or other forms of undesirable particles). It will be assumed for this example that the prescription material is generated in advance, which causes the printer to deposit the desired material at discrete locations within the boundaries of the overall printed area 103. Briefly, the prescription data for each product describes the layer thickness and dimensions for the product, as well as any desired details such as corner arcing or edge buildup profiles, and for The prescription data for the array describes where the layer of each product (i.e., the product prescription material) will be placed relative to the substrate. When the prescription material is processed or graphed, it will instruct an inkjet printer to deposit liquid onto each individual product area 107 on a reproducible basis, which will be reused for many subsequent similar Substrate. The prescription material (or any processed version thereof) is stored or "cached" in memory for use as the same version, which will be used to make the printer on the substrate The liquid is deposited on each of the substrates in the series. Figure 1A shows a single substrate of these substrates (with a number of products or "panels" thereon) that represent an idealized deposition relative to the substrate.

However, as noted earlier, virtually every array or substrate may be in its structure or its position, rotation, scale, skew, or other layer that affects the deposition of one or more of the products. There is non-uniformity in the alignment of the alignment; for many processes, such distortion may be tolerable, but for applications such as microelectronics or where very precise alignment is necessary Such alignment problems may be limited to feature size, product defects, or increased manufacturing time and/or cost. What is needed is to avoid and/or compensate for these problems, and to provide printing that is automatic and as fast as possible while maintaining accuracy.

To achieve this effect, in one embodiment, the hardware logic and/or the indication logic is in a manner that does not ideally require adjustment of the programmed scan path of the printer (eg, only the nozzle data is adjusted). The detected errors are taken into account to adjust the nozzle emission data; however, this is not necessary for all embodiments. The references (eg, alignment marks or optical identification of the substrate) are optically detected as each new substrate is loaded, advanced, or placed in the printer. The associated imaged data is then used by a processor to compare the actual position of the substrate (and its panel) to the expected position; based on the deviation, the printed image is adjusted to allow printing and panel layout information (and any previous The deposited layer) is aligned. The alignment corrections performed using these techniques are typically on the micrometer scale or finer (e.g., they correct for errors in sub-millimeter positions, and in some embodiments are corrected for 100 microns or even substantially smaller locations). Error).

Figure 1B shows a situation in which a substrate is indeed out of alignment with respect to an ink jet printer; in this case, whether or not due to edge error, mechanical positioning error, difficulty in substrate processing, or other problems, the misalignment system The possibility of an alignment error between the layers of each manufactured product is produced. It is noted that the errors depicted are exaggerated on a scale for ease of illustration. More specifically, an xy rectangular coordinate error (represented by component symbol 123), if left uncorrected, would result in an individual product layer (107) being offset from the desired printed area 125 on the substrate. Shift; this offset may cause problems with any of the product layers below (or subsequently deposited or fabricated). The techniques disclosed herein are used to adjust printer control data such that the ink deposition is shifted in a manner consistent with the alignment error, such as the printing process being matched in a substrate and/or panel position. The way the error is offset (or is a transition). This detected error is generated or adjusted by a method that matches such errors. Control data and be corrected. As noted earlier, if the prescription material is pre-processed, the optional technique can be utilized to re-specify any previous nozzle firing pattern in a manner that is dependent on the error. It is noted that in an embodiment, this re-specification typically involves not only simply shifting all nozzle firing decisions by a vector amount, i.e., retaining accurate liquid fill for the desired area, nozzle firing Decisions can also be re-evaluated taking into account variations between different nozzles and mismatches with the location of the deposited area of interest (eg, a pixel well) and the printed grid, and are directed to other factors (eg, as in The anti-aliasing is discussed further below. In one embodiment, this reassignment is performed by a process that operates at the level of the bit map or vector representation, which is effectively based on a printed network associated with any previously processed or calculated data. The individual nozzle firing decisions are converted by grid points (eg, according to the same version of the bitmap) and/or nozzle drive parameters. Other adjustment techniques such as scan path variations (e.g., adjusting the offset of the print head and changes in rasterization) are also possible.

Fig. 1C represents a case in which one substrate has another form of error represented by the angle of rotation α (element symbol 133). If left uncorrected, this error will cause the individual product layers (107) to be out of alignment with other product layers (e.g., with respect to the desired print area 135 of one of the substrates). Using the techniques discussed herein, this error can be corrected by detecting the misalignment and, for example, using hardware, software, or a combination thereof to account for the adjustment of the rotation in the map of the printer data. Similarly, if a pattern print image such as a bitmap has been generated, the correction can be optionally applied directly to a backup of the image by calculating the new directly from the preprocessed bitmap. The nozzles emit designation and/or drive parameters (eg, without having to regenerate a printed image from the prescription data) to perform. As in the previous example, the adjustment system produces a printing process for each substrate, wherein despite the error, the deposited layer is exactly as desired. The product is geometrically aligned; the printing can be performed almost immediately.

It is noted that whether the error is taken into account with the preprocessed data (eg, the same version or a bitmap) or the press data that was applied directly to the original calculation from the prescription information is an implementation decision. For an OLED device with millions of pixels (and millions of possible small droplet deposition points per substrate), the adaptation of each substrate of a previously calculated bit map is supported. Hardware functionality may not be computationally feasible; in some embodiments, directly calculating the printer data from the prescription data in a manner that is detected by a consideration may be faster, but In other embodiments, the use of a one-bit map (or other pre-processing) may be more efficient. For embodiments that further use the parallel processing features presented below, a larger number of processing options are available.

Figure 1D represents yet another type of error that is depicted as being linear across the entire substrate or product array, in this case, it represents a scaling problem. For the purposes of this example, it should be assumed that a substrate system is slightly larger or smaller than desired; for example, it may be due to temperature changes that the substrate has shrunk or expanded slightly in a manner that causes the error to be at least partially uniform or linear. Therefore, a simplified example is provided below, for which it is assumed that all dimensions and positions of the sample image should be adjusted k₁X,k₂Y, where k₁And k₂It is a scalar quantity, and where X and Y are the right angle coordinates on the substrate. It is noted that a situation in which the scaling error is only in one dimension (eg, the X dimension) presents a simplified situation, where k₂=1. If left uncorrected, the scaling error will result in an offset error, and the individual product layers (107) are slightly smaller (149) for the desired footprint (148) for each panel; similarly, the overall printed area It can be seen that it is slightly smaller (145) than the desired printed area (147) for the substrate. In order to correct this error, the scaling is based on k₁And k₂It is corrected. Similarly, in an implementation In this case, this can be done by taking the variable k₁And k₂Considerations are made to produce converted press control data (eg, with potentially different nozzle launch decisions and mappings directly from prescription data or from pre-processed pattern data). Anti-aliasing (as will be discussed below) and/or nozzle droplet detail may also be optionally applied/considered as part of this conversion to ensure that the deposited liquid is in a given location or unit area. The density of the regions (i.e., they will determine the thickness) corresponds to the desired layer size. As in the previous example, the adjustment system produces a substrate-based printing process in which the deposited layer is exactly aligned to the desired product geometry despite the substrate position error, and wherein the desired deposition fill, density, and amount are Accurate; it is noted that the particular adjustment associated with FIG. 1D may result in a substantial change in the designation of the printhead-scan nozzle, and depending on the embodiment, may potentially affect the number of scans (which are implemented according to And the nature of the error). In other words, in one embodiment, as mentioned earlier, the nozzle firing decision is redesignated according to the techniques disclosed herein (eg, without changing position or number of scans); in another embodiment Given the data and errors of each nozzle-small droplet, the scan path can be re-optimized so that different scan paths and a larger or smaller number of scans can be utilized. Regardless of the embodiment, printing can be performed almost immediately as planned, where the substrate error is independently considered by reference to each substrate (or a portion of a given substrate on a partial basis) The dynamic adjustment of the position error is corrected in software, hardware or both.

It is again noted that all of the errors depicted in these figures are visually exaggerated for illustrative purposes, that is, if an array of regions (eg, a glass substrate) is assumed to have a width of a few meters On a few meters long scale, the coarse mechanical alignment is usually accurate to the millimeter scale or better. Thus, in many embodiments, the disclosed techniques are used Correcting relatively small misalignments (eg, tens to hundreds of microns or even smaller).

Figure 1E represents yet another type of error, for example, where individual product areas or integral substrates are subject to skew. Here, the error can be seen as an error in the Y position (the direction related to the horizontal level of the drawing page), which is changed according to the X position on the substrate (related to the vertical direction of the drawing page); For example, this type of error may be caused by a substrate edge error or by a transmission error that uniquely affects the particular printer or manufacturing device. In the figure, all dimensions and positions are visible as skew X' = X, Y' = fn {X}, where X' and Y' are adjusted coordinates, as represented by graph 153. Similarly, it is noted that in addition to the panel skew error, a skew condition will actually result in an offset error in the position of each panel 107 (this is seen in the figure as a panel in a column relative to the lower The horizontal offset of a column). If uncorrected, the skew error will cause the individual product layers (107) to be deformed relative to the desired footprint. In order to correct this error, the printed image is again adjusted/transformed with software to advance the advance-skew panel printing in order to match the errors below. As with the previous examples, the conversion performed may include more than just a simple shift nozzle firing decision, and may be characterized, for example, by an additional process to ensure that the proper amount of ink is deposited in the appropriate target area; as previously discussed, This can optionally result in a substantial change in the designation of each printhead-scan nozzle and potentially affect the number of scans. Similarly, the printing is performed almost immediately, wherein the substrate error is corrected by software for immediate adjustment of the pattern image or the calculation of the prescription information for the particular substrate. of.

Figure 1F represents a hypothetical situation in which the errors mentioned may affect the individual panels independently of each other, and wherein such errors may be additional or alternative to errors in substrate position. For example, Figure 1F depicts a situation in which the desired print area is relative to Field 163, a first panel 107' is visible as skewed, and the second, third and fourth panels (107", 107"' and 107"'') are visible respectively relative to the other The rotation, offset, and scaling errors of panel 107; it is noted that because the scaling error of panel 107"" only affects this panel, the offset of the location is depicted as being small to none. As taught herein, the error of each panel is individually corrected by adjusting the graph that will be used to print the given array or substrate. In this regard, for any given panel or It is a partial error that may be complex, where the correction is one or more of the conversion techniques that are modeled as one or more of the offset, rotation, skew, and scaling errors of the position; this may be for one The panel of the group, a single panel, a portion spanning multiple panels, or a portion of any given panel, is done on an independent basis. For cached storage of a previously calculated version of the pattern image For the example, for each A new substrate (eg, every 45 seconds), a backup of the cached image can be processed to reduce alignment, orientation, or other errors, as necessary to re-specify nozzle firing decisions without The particular substrate mechanically changes position; for each subsequent substrate (eg, every 45 seconds), an instance of the cached stored plate printer control data can be loaded again and used to correct The error in positioning or alignment that is unique to the new substrate produces data that can be immediately applied to the calculated graph of the print.

There are techniques for fine alignment of the press control data to match the deviation of the substrate or panel from the intended location. First, a contemplated embodiment overlays a printed grid onto the actual (i.e., detected) substrate and/or panel position; as previously indicated, the printed grid can be Characterized by a "horizontally separated" node representing the nozzle spacing on the printhead, and a "vertically separated" node representing the digital timing at which the nozzle can be emitted, such as when the printhead and substrate are "scanned" relative to each other "The distance of each micron". For being here Each node in the covered space, the system (i.e., the indication logic executed on one or more processors) determines whether another nozzle should be emitted at that point to deposit a small droplet of liquid. For the covered space, the emission decision at each node is the original "prescription" material for an ideal substrate, and the position at the node deviates from one of the perfectly aligned substrates. One of the deviations of the point function. In one embodiment, wherein the printer control data is "pre-calculated" to form a one-dimensional map (i.e., wherein the nozzle firing decision has been specified in a hypothetical uniform alignment), Each emission decision of the offset printed substrate or panel position of the printed grid can be mapped to an equivalent point in the original bitmap by a transition. An example is provided in which if a given substrate is "rightward" offset by a position of 31.0 microns, then the decision for each node in the printed grid can be determined by first identifying the original The same node position in the bit map is then calculated by identifying a point at the "leftward" 31.0 micron at that location, and finally transmitting the details associated with the node closest to the shifted point. As will be discussed further below, in such an embodiment, particularly where liquid will be printed in the pixel well, such a process may result in too much or too little droplets of the liquid being deposited at the well boundary. Thus, an adjustment procedure such as anti-aliasing (discussed below) can be utilized to mitigate this possibility. In a second embodiment, each of the transmission decisions in the covered grid may be a weighted function from one of the pre-calculated patterns of the transmission decisions (eg, for the original bit) The figure "to the left" is offset by a weighted average of the emission decisions of the four closest printing nodes of a point of 31.0 microns. Of course, those of ordinary skill in the art will appreciate many variations and possible adjustment mechanisms. Other embodiments may utilize the input alignment data or the measurement of the nozzle parameters at the original location, which is then taken into account, taking into account the calculations of the printer control data for the given substrate, Anti-aliasing, small droplet density (halftone Patterning and other processes; each of these helps to create a more reliable deposition by using one or more additional procedures to ensure that the appropriate amount and/or density of liquid is printed on In the discrete regions of the substrate. In some embodiments, the number of scan paths or the offset of the print head across the scan can be re-examined to optimize printing time. Various combinations of the disclosed techniques and their associated benefits will be apparent from the following discussion.

Figure 2A is used to describe the processing of some examples. More specifically, Figure 2A shows a flow chart 201 in which certain functions are depicted as being executed as part of an offline process (i.e., above dashed line 209) or as being depicted during the execution phase. The period is executed (ie, the line is part of the process, under the dotted line 209). According to the component symbol 203, a computer (e.g., a stand-alone or part of a manufacturing apparatus such as an industrial or manufacturing printer) first receives information describing the desired deposition across a layer of the substrate. As indicated on the right side of Figure 2A, some variations are considered. For example, an embodiment (204) contemplates the fabrication of many products as part of an array of deposition. Referring briefly to FIG. 1A, for a particular process, the layers of each of the illustrated 15 panels may be identical in width, length, and thickness; for example, a single prescription (representing these panels) One of them can be electronically loaded and arranged by hardware logic or indicator logic, or both, to produce a total of printed data having a total print position for the intended print position of 15 individual products. Printed images. Not all embodiments need this. For example, a TV screen having a first size can be presented by a panel of a first horizontal column, and a TV screen having a second size can be presented by a panel of a second column, both of which are On the same substrate; each panel in this example will have different manufacturing details. Regardless of whether multiple products are represented by each substrate, The method can include receiving information such as the desired location of each layer (e.g., based on the location of the "upper left" corner of component symbol 205 and the length and width dimensions, and the desired layer thickness based on component symbol 206). In one example, a "prescription editor" software can be utilized to define layer parameters for a substrate carrying one or more products (see, for example, the discussion of Figure 2C below). The computer then stores information representative of the desired layer for use as the same version in each substrate in which a plurality of substrates are processed, i.e., in a continuous operation as part of an assembly line process. The stored information is pre-mapped as a bitmap or other form to represent an "idealized" substrate (ie, a perfectly placed substrate on which all products are Printed perfectly, or in some way related to a predetermined substrate/product position.

During the execution phase, when each new substrate is loaded into the printer, or is received, the substrate is roughly set by mechanical means; for example, according to component symbol 211, this step can be via " Roughly "providing the substrate to a handpiece of a desired printing position or performing it via an edge guide. According to component symbol 213, the printer then utilizes a sophisticated detection system to detect the actual substrate position; in one embodiment, this is performed using a high precision camera that is imaging a reference a region (in the substrate or a known pattern on the substrate), wherein the camera is accurately positioned relative to the printing system, and wherein the position detection soft system is used to perform image processing for identification The exact location of the reference is, for example, to the nearest micron. Advantageously, the fiducial can be a two-dimensional pattern, or a set of patterns, which enables the determination of the rotational orientation and/or scale and/or skew of the substrate; the more complex the reference (or identifiable independence) The greater the number of features, the greater the accuracy and/or quantity and/or complexity of alignment or positional problems that can be corrected. An example of this is provided below, an embodiment (see Figure 9A, which is also discussed below) Each panel on a substrate will have at least two dedicated alignment marks that enable error detection or error reduction for each panel of a single panel; however, this is not true for all embodiments. necessary. Regardless of the detection mechanism, the computer (and/or one of its processors) is then derived between (1) the detected panel position, size and orientation and (2) the expected panel position, size and orientation. The deviation, and it is necessary to adapt the printing job to align printing with any underlying layers and/or product geometry (215). According to the symbol 217, the panel is then unloaded from the printer and selectively cured according to the component symbol 218. As indicated by symbol 219, a new substrate can then be loaded for a new print job.

As an example of a printing process that represents an assumption for OLED or solar panel fabrication, a particular layer that will be "printed" (and then cured or otherwise processed) is typically used for one of the panels. The layer thickness may be on the order of submicron to hundreds of microns; information defined by the array may be received and used to determine the print position for each panel based on a prescription for a particular panel that is attached to the location . For example, in a configuration with 15 panels per substrate, array-defined information can be received to indicate that a first panel "prescription" is replicated in each column of the front two columns of the substrate. Times (ie, at precise locations), and a second unique panel prescription is replicated five times in a third column of panels (ie, at a precise location, assuming a second prescription is different In the first prescription). The array definition in this example will indicate each panel location and the specific recipe used for that location. It is noted that an example of a uniform continuous layer for any given product is merely illustrative; in some embodiments, the layers within each product can be patterned (this is relative to one A uniform coating, such as a film encapsulation for a particular product or panel, or thickness can be varied within a layer defined by the particular panel formulation. Continue with fifteen panels, three columns For example, the product of the third column may potentially be characterized by a different layer thickness, wherein a denser halftone system is used to "construct" a thickness that is greater than the thickness produced by the panel for the first two columns. .

2B is a flow chart 221 of an exemplary printing process for each of a series of substrates. More specifically, when each new substrate in the series is received (223), a robotic hand transfers the substrate to a first position (225); this is one of the expected reference positions. The substrate is placed in a manner that is within a field of view of the camera. In an embodiment, one or more cameras may be utilized to capture the reference position based on the component symbol 227. For example, a process discussed below in connection with FIG. 9B uses a standard camera to first identify the coarse location of the reference, and the substrate and reference system are then re-set for necessity to utilize a second high resolution. The image capture of the camera, that is, the micrometer-scale position used to identify the reference. Figure 2B depicts a number of optional programs 229 and 231 in dashed lines, i.e., one of the substrates is reconfigured to allow for the capture of multiple references; as indicated, the more reference and alignment features used, the more The greater the ability to measure the nonlinearity of the substrate or individual panels. In this embodiment, the substrate is controlled by a precision robot to efficiently transfer the reference for each panel to the camera, but in an alternative embodiment, the camera can be mounted to a motion system And it can be transmitted relative to a stationary substrate to find a reference, or as in the case of a separate motion axis, both the substrate and the camera can be moved. As indicated by the text on the left side of the figure, another embodiment uses a row of scanners to image the substrate and/or its fiducial during loading or other transport of the substrate; still other embodiments are characterized by A camera is mounted to the printhead (i.e., for very frequent "continuous" error monitoring and correction), and yet another embodiment uses a non-visible detection mechanism. Regardless of the number of benchmarks and / Or, if the location is identified and any errors are identified (233), the stored pattern is retrieved (235) and graphed. 2B depicts that the stored template can be in the form of a prescription (236), a bitmap (237), or have some other form (238) (eg, an object, vector, or other format). Offset, misorientation, and/or other errors may be linearly (239), using a least squares fit (LSF) method (240), or using some other type of error estimating mechanism (241). determine. For example, an embodiment uses two references per panel; the fiducials can have a complex shape (eg, like "cross" as depicted in Figure 9A) and are used to determine across a given Panel panel rotation, corner offset, and X and Y scaling errors. The press control data can then be adjusted or calculated from the sample using linear or affine transformation. For an LSF method, nonlinear errors within a panel or across a substrate can be detected and modeled by a polynomial in one or more dimensions, for example, in software. The supervised processor operating under control controls a curve to the detected error (eg, it utilizes an error measure associated with multiple references). Adjustments are then performed according to the polynomial equation or an affine transformation to specify printer grid emission decisions and associated nozzle parameters. Equipped with a mechanism for mapping (242) the pattern to the actual substrate position, a supervised processor then optionally assigns processing work to an additional processor (eg, a core in a multi-core processor) . In one embodiment, a multi-core processor with one or more processing cores is used to segment the computational (or translation) work to speed up processing; in another embodiment, hundreds. Core systems are used, which further speeds up the process. This parallel processing is optional, as is meant by the use of the function block 243 of the dashed line. Regardless of whether parallel processing is used, the system continues to calculate the printer control data in a manner that depends on the position, rotation, skew, and scale of the detected substrate and/or panel. And raster scan (245) is specified as appropriate. In one embodiment, as indicated by a first optional step (see the right side of Figure 2B), the latest expected droplet detail for a given nozzle waveform is The supply (eg, the mean and standard deviation of the small droplet volume, and the two-dimensional droplet trajectory for each unique waveform of each unique nozzle). This information is used in the calculation process in a manner that is calculated to produce a precise amount or to ensure a uniform distribution of ink; as will be discussed below, in one embodiment, this information The system is continuously re-measured (eg, between substrate printing) in order to develop a robust measurement population and to account for changes in the properties of the deposited liquid (eg, it is viscosity, ambient temperature, and others) a function of the factor). In an embodiment, as indicated earlier, the scan path is not adjusted in response to poor alignment, but only the nozzle/small droplet emission parameters are changed. In another embodiment, the rasterization is re-evaluated and re-optimized according to component symbol 247; for example, when the nozzle emission data changes, performing fewer scans may be feasible (or conversely, increasing Scanning for optimal control over the amount and location of deposited droplets may be desirable). Component symbol 247 is indicative of an embodiment in which printhead offsets from scan to scan can be re-evaluated as necessary. Depending on the component symbol 248, the consistency check can also be performed on the printer control data that has been partially adjusted from a spatial perspective to resolve the error; for example, in one embodiment, the printer control data is first based on the Detector The measured substrate-to-printer changes are graphed, and then a software program is called to identify the well of the fluid (or other discrete areas of the substrate) and "check" the ink density and fill in those areas. - If the conversion will produce too much or too little deposited ink, the software program smoothes or otherwise adjusts the printer control data to eliminate the error, thereby maintaining a total ink fill and ink density Within the required norms. In yet another embodiment, a continuous layer (eg, an encapsulation layer) The "edge" is evaluated after error adjustment to maintain a desired edge profile (249); in other words, due to the diffusing nature of the deposited fluid, the ink density is increased/decreased at the edge of the layer up to the desired layer boundary. The primary uniform layer thickness may be desirable, and the edge features may be evaluated or adjusted after spatially converting the printer control data to resolve the error error processing in such an embodiment. The edge treatment as referred to represents another type of consistency check and is widely discussed in the aforementioned U.S. Provisional Application Serial No. 62/019,076, which is generally incorporated herein by reference. It is noted that although this provisional application discusses the use of the techniques mentioned for the encapsulation layer, these teachings can be applied to reinforce in any deposited layer, patterned or otherwise layered accuracy. Many variations of these principles will be considered by those skilled in the art. Finally, according to the component symbols 251 and 253, when the processing is completed, the printer control data containing the calculated graphics of any raster scan information is transmitted to the printer, and the method ends. As noted earlier, in one embodiment, for each substrate or panel processed by the system, it is generally desirable that each substrate perform this process in about two seconds or less.

As noted earlier, layer information can be received, stored, and graphed in a variety of forms. Figure 2C provides an example of a "prescription editor" that can be utilized to generate and edit a description of a layer in order to define "ideal" printing. Such material can be stored (with or without some form of pre-processing) so that the resulting pattern data will be adjusted or transformed for each new substrate.

More specifically, Figure 2C shows a screen 261 of a software user interface ("UI") that provides panel prescription definition or editing. Such a UI can be utilized at an earlier design time by integrating with a printer or manufacturing system (e.g., as part of a computer aided design system) or in some other way. As if by the screen As represented by 261, a software application advantageously simulates the layout of a substrate or printed area and provides a mechanism for editing the layout. The UI display area 263, seen on the left side of the screen map, contains a variety of different panel sizes that are embodied in the six depicted panels. In this example, each of the three different panel sizes will have an individual prescription. The second UI display area 265, which is found at the bottom right of the screen map, provides a window for array definition and allows confirmation for inclusion of "Sub-panel1", "Sub-panel2", "Sub-panel3", The right-angled coordinate position of each of the six panels depicted by "Sub-panel3H1V0", "Sub-panel3H0V1", and "Sub-panel3H1V1" and one for a corresponding prescription (for each panel), In this case, it is achieved by the names such as "Sub-panel3"; the latter three panels (Sub-panel3H1VO", "Sub-panel3HOV1", and "Sub-panel3H1V1" all use the same prescription. ("Subpanel3"), but characterized by individual horizontal and vertical offsets from the substrate. The top one of these definitions is visible as being expanded (ie, because it is "selected"). As such, the second UI display area 265 provides graphical selection and removal of any given panel (right side), and deployment of data for any desired panel (eg, as depicted for panel1). And processing; for example, Figure 2C shows the top panel "Sub-panel1" Defined to have a size of 825 by 500 mm such that its upper left corner is placed 50 mm from the left and top edges of the printable substrate area, respectively, and has selective corner arcing (ie, via a Selected "corner radius" value). The expected location of one or more references may be indicated by a third UI display area 269; as indicated earlier, the printing system is effectively defined in relation to such a reference, such that printing (and for any given The printed area of the panel can be based on the expected reference position (related to the printer or print transmission path) and any actual (ie, Detective) for that particular panel The measured deviation between the reference positions is calculated. As indicated by the screen map 261, different reference or reference features can be specified, such as to detect scaling errors or skew. If desired, in other embodiments, the third UI display area 269 can be configured to accept a reference shape or definition of substrate characteristics, the software then attempting to match to be identified on each substrate during the execution phase. A detected (imaging) shape. Finally, the UI can be seen to provide a fourth UI display area 267 that allows for the definition of substrate size and thickness. Although in one embodiment, the printed layer thickness is common across all panels (eg, for each layer of an individual panel), not all embodiments must be so; In contemplated embodiments, the thickness can vary depending on the panel and can be affected by different local halftone densities. For each layer, printing is performed, for example, by multiple scans as a printing job across the entire substrate, and the substrate and its wet ink are subsequently transported to a curing chamber to create a permanent layer. This post-printing curing process is also typically performed in a controlled atmosphere to prevent contamination by particulates, oxygen or moisture. Other layers may be added after the deposited liquid is cured or hardened.

As indicated earlier, layer data, for example, defined by a prescription editor, can be stored in a number of forms suitable for the embodiment, for example, becoming a meta-graph, becoming the original prescription material, utilizing a vector representation. Or in some other way. In one embodiment, the prescription data is used to generate a grayscale image (eg, an array of 8 bit values for each location on the substrate), wherein each of the images The grayscale value represents a density of ink to be applied in the corresponding substrate position, and wherein each grayscale value is used to generate a local halftone (i.e., according to the printed grid) to construct the thickness of the layer. . An example of this would be helpful; in one embodiment, the grayscale value of eight bits of a matrix is utilized with a value between "0" and The columns of "eight" values between "255" are defined by columns and rows. Each value (eg, "235") corresponds to a unit area of the substrate and is mapped to a particular layer thickness. For example, the value "235" may (depending on the process and material) be mapped to a thickness of 7 microns in one of the completed layers. If all of the grayscale values in the matrix share the same value, then this may be mapped to a fairly uniform layer, but the values can be changed to vary the thickness across a panel, filling in the geometry below. (for example, in the pores of the structure below the layer), modify the edge accumulation, or for any desired effect. The designation of the grayscale value can be "dead-reckoned" to the desired thickness (which utilizes a variable selected in a manner that adapts the grayscale value to the desired thickness based on ink and process parameters) Partial halftones, or can be calibrated to the desired thickness in advance (which utilizes a halftone with a fixed relationship to the grayscale value). In one embodiment, the correlation between the grayscale value and the thickness is measured after the formation of the test layer (eg, after curing or drying), and thus the deposited ink density is closely related to the final Layer thickness. Based on the measured data (or other feedback), the software can then map the droplet density to a desired grid spacing, for example using the following formula: desired thickness = h x {quantity_Droplet/(spacing_{Within the scan}× spacing_{Cross scan}In this formula, the spacing within the scan represents the spacing between droplet opportunities in a first direction of relative motion between the printhead and the substrate, the spacing of the cross-scans being representative of The interval between droplet opportunities in a direction perpendicular to this first direction (or independent of this first direction), and the parameter h (multiplied by 100) is a grayscale value in percentage. In an embodiment, this relationship may vary over time, and thus ink details may be continuously re-measured for dynamic factors such as process or temperature, for specific machine or ink details, Nozzle aging, or other factors to develop robust empirical data, where weighting The department emphasizes recent measurements. It will be apparent that many examples are possible, and as indicated by this example, many different types of processing can be performed to generate a "pattern" from the prescription material, or based on the detected error during the execution phase. Calculate the printing work from the same version.

Figures 3A-B and Figures 4A-G are used to discuss various charting procedures. As indicated in the above various embodiments, in one embodiment, the error of linear distortion representing the entire substrate or panel can be detected and corrected, while in another embodiment, the entire substrate or any panel is represented. The error of the nonlinear distortion can be detected and corrected.

3A presents a flow chart 301 in which an alignment between the detected panel layout and the printed grid is first defined (303) in order to adjust for detected errors. This mapping is generated to adapt an instance of a stored "pattern" (or original layer data) to the detected distortion. For example, referring briefly back to FIG. 1B, in the event of an offset of the substrate position, such a mesh will be used to define pixels associated with the actual panel position in a manner corresponding to the offset vector 123. The mode of transmission (ie, adaptation to error). In the case of the example of FIG. 1C, the mapping will identify or indicate a portion of the printed mesh having sufficient dimensions to represent the printed image rotated by angle a; in the case of the example of FIG. 1D, the mapping Is of sufficient size to receive by (k₁)X and (k)₂) Y-scaled printing. Roughly, the conversion is defined to encompass a desired printable area of a particular operation of the array or substrate of the product; for example, if the substrate distortion increases or decreases the substrate size relative to the stencil image, The printed grid may then be characterized by more or less nozzle locations representing areas that are enlarged, reduced, or displaced. The conversion is then used to calculate the printed material in a manner that is misaligned with any substrate (e.g., panel). For example, if a pre-calculated template is stored for repeated print runs, the template can be adapted to any detected one using one of the printed images. The manner in which the positional or alignment error is run, and is converted in a manner corresponding to the alignment (and therefore corresponding to the error) (ie, rotated, scaled, offset, skewed, etc.) ). The conversion is performed by identifying a portion of the printed grid corresponding to the detected panel position (305), and then utilizing the identified grid points, but for the detected errors The method is adjusted to modify the cached stored sample data. For example, in one embodiment, as referenced by element symbol 307, if the pre-processing produces a P(x,y) emission decision for each point on the printed grid, then for the printed grid The modified printer control data that matches the detected product location node can be determined based on the error vector of a location and the analysis of one or more of the pre-processed transmission decisions. Such dependencies may be formed (308) with respect to a closest node of the pattern data, either because of an error vector, or because the misalignment or position error may be a multiple of a non-integer of the grid spacing, It is therefore offset (309) for a function weighted according to one of the 2, 3, 4 or a different number of transmission decisions from the template. In an embodiment, the emission decision is based only on a single node of the printed grid represented by the original template (ie, offset according to the error vector), and in a second embodiment, It is based on the weighted average of one of the four closest nodes, and is based on P'(x',y')=fn{P_Nw(x,y),P_Ne(x+1,y),P_Sw(x,y+1),P_Se(x+1,y+1)}, where P_Nw(x, y), P_Ne(x+1, y), P_Sw(x, y+1) and P_Se(x+1, y+1) represents grid point decisions for the northwest, northeast, southwest, and southeast, respectively, and wherein x and x' and y and y' are related based on the detected error. Clearly, this is just an error compensation algorithm, and many alternative algorithms will be easily appreciated by those skilled in the art.

For example, under a given error, if a hypothetical grid point {P'(x',y')} (which represents a particular grid point for one of the converted panels) corresponds to a point that is closest The cache storage Each of the specific NE and SE grid points (printed pixels) of the pattern is less close to a specific corresponding NW and SW grid point (printing pixel), and if these NE, SE, NW, and SW points are respectively Corresponding to the emission decisions of (1, 1, 1 and 0), the printed grid point {P'(x', y')} may be based on a transmission decision weighted according to the error distance from the four nodes (1) a function of 1, 1, 1 and 0), which is assigned a binary transmission according to {0.40*(1)+0.40*(0)+0.10*(1)+0.10*(1)=0.60&gt;0.50} Decision (1 or 0) {for example, specify a "1" transmission decision to P'(x', y')}. As such, this mathematical relationship is merely an example, and many algorithms can be utilized to specify nozzle firing decisions; as indicated, in a well-thought-out embodiment using a cached stored pattern, The emission decision is based only on the closest single printed pixel from one of the converted images, with a software anti-aliasing process being used to process the nozzle firing decision to ensure a consistent number of small liquids in the converted image. drop. In an embodiment, as mentioned, an affine conversion system is used. Regardless of the calibration method, when the error-adjusted press control data is generated, it is stored in the memory as a new or custom printed image (311), where the completed graph is The printer control data is transmitted to the printer (313) for execution. After completion of the layer (317), the system is then ready for a subsequent layer (e.g., using the same conversion), or a new product run or substrate.

It is noted that the component symbol 310 refers to a dashed (optional) wrap process that is used to ensure consistency of deposition. For example, if a particular point P'(x', y') falls within a fluid well that is used to hold a liquid that uses a layer that produces light as a display, and a weighted error function needs to be dependent on The patterning pixel may be shifted (i.e., rewinded) in a pattern printed pixel other than the well image associated with the template image, in order to alternatively be associated with the unshifted printing well Other printed pixels. This rewinding procedure can also be optionally arranged in a parallel process. Each core or processor in the environment performs it, for example, by borrowing printed pixels from other processors in order to maintain local consistency. Other examples are also possible. For example, the component symbol 313 indicates that in some embodiments, the template image is shifted or distorted according to an error, and a software resident daemon is then called to process the predetermined area of the substrate ( For example, fluid wells), and the test is within a critical value. One such procedure (i.e., anti-aliasing, 314) involves simply testing the number of droplets (or the expected amount from the sum of the stored data representing the average amount of individual droplets), To ensure that the total liquid after the error shift still meets the required standards. In another variation, nozzle data such as verification data, expected droplet volume or orbit, or other data may be tested by such resident programs to ensure consistency of the desired total values, 315 (For example, to make the uniformity of the layers easy), where necessary to perform adjustments to the press control data to provide consistency of deposition. As noted, many different processes can be utilized.

FIG. 3B depicts another flow diagram 351 that is used to describe the processing of a stored template. More specifically, the information representing the desired layer is first loaded from the template (353). In one embodiment, the information includes original prescription data. In a second embodiment, the information includes a bitmap representing nozzle firing commands, and in a third embodiment, the information is Some other format is used to represent it, for example, in the middle of the two forms. The system then calculates the conversion parameters (355) required to map the print represented by the template to the actual (detected) panel or substrate position; these parameters are then applied to produce a converted representation. (357). As indicated by the optional process block 359, in one embodiment, the map of the printer control data can be converted directly on the prescription data (ie, stored as the template) during the execution phase. Parameters to be executed. As indicated, in one embodiment, one or more imitations Shot conversion can be utilized (360). The converted representation is then output as a calculated bitmap (361) and used to generate raster data (363). The converted representation can be checked for consistency of ink density per unit area or total amount, or for other quality control reasons (363). After any correction, this data is then output to the printer (363) and the method ends (365).

4A provides an example 401 of a conversion of a printed grid that is used to illustrate an example and a bitmap that is applied to a cache. A printed grid for an ideal (ie, error-free) "bitmap" printed image is referenced by element symbol 403; each "square" in the printed grid (eg, according to Component symbol 404) represents a printed "pixel" defined by a grid point, wherein an "X" in the square indicates that a printing nozzle will be moved relative to a substrate below when a print head is moved. , a small droplet is emitted at the corresponding grid point. Similarly, it can be seen that a "square" or pixel 405 that is empty (i.e., without "X") indicates that a printing nozzle will not be emitted at that location. The top end of Figure 4A shows a representation of a printhead 407 in which individual triangles (e.g., identified by symbol 409) represent each nozzle. As the print head moves relative to the substrate, each given nozzle will pass through a series of printed pixels (discrete dropletlet emission points), each of which is at a different point in time. For example, when the nozzle 410 moves downward with respect to the pattern page, it will pass through an area corresponding to the printed pixel 404 (where it is emitted), and then an area corresponding to the printed pixel 405 (where It has not been launched). The bitmap 403 is predefined and stored in memory as the same version for a repeatable process.

Figure 4B presents an example 411 in which an example of a product is considered to be at a position offset from the pattern, e.g., due to an error in position in a substrate. Because of this offset (413), the area of the print that should receive the exact alignment is represented by the symbol 412. If the nozzle firing command provided by the desired printed image 403 remains unadjusted, then printing in a manner that is geometrically misaligned with the underlying product will occur. Therefore, what is required is to adjust the pattern so that the printing takes place in the correct position. In this regard, the pattern from Figure 4A can be distorted during the execution phase (e.g., by displacement in position) to dynamically align with another substrate, panel, or other product area so that To reduce the error.

Figure 4C provides an example 421 of a portion of the printed grid 422 that covers the detected product location. It is noted that in this embodiment, this is the same printed grid as represented by Figure 4A, but with nodes corresponding to the location of the detected product. Likewise, each depicted node defines a printed pixel that the printhead 407 will pass in its movement relative to the substrate, and each "square" (eg, printed pixel 419) corresponds to a printhead. Nozzle and a nozzle launch time. In one embodiment, the particular nozzle is preset (e.g., the scan associated with the template will also be used for that particular product instance), but this is not required for all embodiments. However, it is noted that the offset (413) of this position may not be exactly aligned with the inter-scan spacing of the printhead nozzles or the spacing across the scans. Thus, the soft system effectively covers the detected product location with the printed grid 422, and the launch decision for this overlay is based on a shifted version of the original template image based on the offset 413. Depending on the offset 413, each of the printed pixels in this example may correspond to (cover) up to four printed pixels of the original stenciled image.

For example, Figures 4B and 4C show two examples in which a printed pixel 419 for the detected product location corresponds to (at a given error 413) in four/two stencils. A point in the middle of the brush pixel, as represented by element symbols 415 and 417. It is noted that for various embodiments, any number of nearby pixels can be applied in the weighting function; in one embodiment, only a single "closest" pixel is used, in a second In an embodiment, "two" adjacent pixels may be utilized (eg, according to element symbol 417), while in a third embodiment, a different number of nearby pixels (eg, "9") may be Utilizing, that is, the disclosed techniques should not be considered to be limited by a weighted average of the four pattern printing pixels. The two printed pixel instances 417 function to target any number of pixels, whether or not they are exceeded, less than or equal to two, and whether or not a proxy is based on a weighted average of a plurality of pixels. A decision for the printed pixel 419 is made based on the referenced pixel group, and once each grid point visible in Figure 4C is performed, this effectively validates the material from Figure 4A. The conversion "becomes the actual product location. Once the process is complete, rasterization (e.g., scanning plan) is then performed as appropriate, and the printing system is then executed based on the detected product location. The above procedure accurately maps any ink density represented by the original stencil image to the shifted product geometry as detected during a particular print run.

Figures 4D and 4E provide examples 431 and 441 of bitmap processing in the event of rotation of an unplanned product (array, product, substrate or panel), respectively. As can be seen in Figure 4D, according to component symbol 435, it is assumed that a substrate or a portion thereof is positioned at an error angle a. For proper printing, the print nozzle firing pattern should therefore also be distorted according to this same value, as represented by contour 433. Figure 4E highlights a printed grid area corresponding to the detected product/panel position. It is noted that although the desired print area is rotated relative to the substrate, the print head 407 and the nozzle (e.g., nozzle 409) will define small droplets, respectively. A regular printed grid where the emission can be made. Similarly, for each grid point (or printed pixel) of the printed grid, the soft system shifts and overwrites the original pattern image and then weights any covered print pixels of the original pattern. In order to obtain a weighted pixel emission decision for the grid points. This operation is for example referenced to the printed pixel 443 in Figure 4E and is referred to with reference to four template grid points 445 or related (exemplified) processing of two pattern grid points 447. Similarly, the two grid point examples 447 are cited to simply indicate that alternative embodiments may use any number of references to the cached stored image.

Figures 4F and 4G provide examples 451 and 461 of processing respectively depicting events in scaling errors. Similarly, the component symbol 403 is a template for a particular layout, and the component symbol 455 represents a footprint that should receive the print to properly overlay a new layer with the desired product based on a scaling vector 457. Area alignment. In such an example, a new printed grid is effectively defined by an arithmetic multiplication process based on the scaling vector. It is noted that in addition to being effectively scaled in size, the print grid may also use variable offsets depending on the substrate position. Assuming that there is an increase in both dimensions of the footprint for a product or panel to be printed, the effect is as if the pixel system of the covered template image is increased in size (eg, comparison) The dimensions depicted by the pattern image pixels 465 and/or 467 as shown in Figure 4G are the same as the printed image pixels 445 and 447 depicted in Figure 4E.

Figures 4H and 4I show graphical representations of skew compensation. Component symbol 475 represents a footprint that should receive the print to properly align a new layer with the desired product footprint in accordance with a skew adjustment formula 477. It is noted that the amount of skew varies depending on the height of the target being twisted. If the compensated error corresponds to a uniform skew across the substrate, the amount of distortion in the "Y" dimension (crossed from left to right) is based on the amount obtained from the edge of the substrate. The value of the "X" dimension changes. In this example, the conversion of the X coordinate is 1 (i.e., unchanged), but it is noted that, in general, this result will vary depending on the orientation of the skew. The template is scaled according to a scaling vector, by an arithmetic multiplication process according to equation 477, and is effectively overlaid against a printed grid corresponding to the detected product scale; The emission value of each printed grid point of the covered grid is calculated according to Equation 477 (ie, the reciprocal of the scaling), for example, according to a weighting criterion based on 1, 2, 4 or other quantities Overlapping or nearby printed pixels; this is represented in Figure 4I by the matching depicted by printed grid dots 483 against printed pixels 485 or 487 overlying the weighted positions of the pattern; likewise, It is noted that each grid point represents a possible one of the nozzles of the print head 407 (e.g., referenced by the symbol 409) when a print head 407 is moved vertically relative to the pattern page. Launch location.

It is noted that in the depicted example, printed pixels having edges adjacent to the footprint of the printed grid are not necessarily by a consistent number of pattern printed pixels associated with other portions of the printed grid ( For example, 4) are weighted. In order to avoid the underrepresentation of small droplets in a boundary region of adjacent new coverage areas, an edge grid point may advantageously use an algorithm (eg, rewinding from pixels on the opposite side of the coverage area) To ensure proper droplet density. In the case where the footprint (e.g., 475) depicted therein is representative of the boundary of the desired layer, an edge process (e.g., Fencing) as previously referred to may be applied if desired. To ensure a well defined edge. Other technologies and changes are also possible.

It is mentioned earlier that in one embodiment, a bitmap can be composed of grayscale values (for example, 8 bits or other multi-bit values), where each value represents an ink amount. Or the desired thickness. The treatment just described can also be measured by simply measuring the distance and gray. The order size is used to weight each grayscale value of the template and is applied to such grayscale values. For example, again using the example of the grid point {P'(x', y')} assumed by one of the detected product geometries, according to the standardized overlapping area, it is mapped to the same version of the NE and SE networks. 40% of each of the grid points (printed pixels), and 10% of the NW and SW grid points of the pattern, and if the NE, SE, NW, and SW points have (235, 235, 150, and 0, respectively) For the gray-scale decision, the assumed grid point {P'(x', y')} can be based on {0.40*(235)+0.40*(235)+0.10*(150)+0.10*(0)= 203} is assigned a grayscale value of 203. If desired, this measurement can also be annotated for individual print grids, directly converted by software via comparison to any desired threshold (eg, if 203&gt;Th, then "1" = emission) Become a binary launch decision. It is noted that these relationships are merely examples, and that many algorithms can be utilized to specify nozzle firing decisions in a custom printed image; as noted earlier, in at least one contemplated system The details of the droplets measured by each nozzle and each drive waveform can be considered in this program (for example, the actual situation is a pair of nozzles applied to P'(x', y'). Small droplets (eg, amount = 12.4 pL), assuming a given print head scan offset can be dependent on specifying a different nozzle to instead emit small droplets for P'(x', y') Depending on whether or vice versa, the scan can be reworked to facilitate the use of a different print head offset, and thus a different nozzle is used for P'(x', y')). Other changes are also possible.

In the context of a method for converting a printed image, this disclosure will now discuss some of the circuits associated with printing and printing heads used in an industrial printer to make a product or Multi-layered.

In the figures discussed above, a simplified print head (e.g., as referenced by reference numeral 407 in Figures 4A-I) is depicted as having a relatively small number of nozzles, such as 20 Around. But in fact, for a typical manufacturing operation, especially for large-scale products, a print head may have a very large number of nozzles (for example, thousands) arranged in multiple columns; A plurality of such print heads are mounted to a common assembly, for example such that all of the thousands to tens of thousands of print nozzles are used to spray material across a relatively wide width of the substrate. This structure provides high quality ink droplet delivery at a highly accurate position resolution. Each print head is expected to deposit a nominal amount of ink per droplet (eg, ten picoliters, or 10.00 pL), but in practice, this number may vary from nozzle to nozzle because of the nozzle position. The droplet velocity, as well as the droplet ejection trajectory, may vary. These changes, if left unchecked, can potentially cause defects in a deposited layer that may be brought to the quality of a finished product.

In an embodiment, in order to more accurately fabricate the desired layer, the droplet details for each nozzle are measured and used to develop a specific for each of these parameters for each nozzle's performance. Statistical models; repeated measurements help to effectively average the measurement errors and develop a fairly accurate understanding of the mean and standard deviation of each parameter. Various techniques are used to address the variations mentioned above; in general, these techniques make full use of these variations based on the average of each nozzle or each droplet measured and The associated standard deviation provides accurate droplet deposition for the project, with the goal of achieving specific ink fill, ink fill density, and small droplet distribution. Halftone (i.e., ink density) and/or pattern adjustment may incorporate such variations (or be corrected for such variations) such that the failed nozzles and/or the droplet location and/or amount The changes can be corrected. It is noted that "halftone" as used herein refers to varying droplet density (eg, the number of droplets of printed grid nodes per group) to affect layer thickness, ie, even " The hue itself is not utilized (ie, the "ink" Or the deposited liquid is usually colorless). In the technique described below, a plurality of (alternative) electrical drive waveforms are made available for each nozzle to provide a function of varying/transmitting a small droplet amount and position selectivity of a target. (It contains at least one option that approximates an ideal target droplet size and position). These statistical measurements can be performed for each such waveform and for each nozzle to provide high accuracy in the plan; these measurements can be updated or re-executed over time In order to consider changes in ink properties (eg, viscosity), consideration of temperature changes, nozzle blockage or aging, and other factors. In the following paragraphs, the measurement functions used to provide this understanding are introduced. In the earlier discussion of the use of a printed grid, it is simply reflected that each grid point will be associated with a nozzle of a print head; the measurement function just indicated can be applied to this process. This allows, for example, a faulty nozzle to be assigned a launch decision in one printhead pass (and/or to cause any desired droplets to be redispersed to other nozzles). In addition, the difference in the "X" axis position of a small droplet can be corrected by the use of the indicated alternative nozzle drive waveform or by tuning or amplifying a preselected waveform to facilitate changing the droplet The timing is to better position the droplets or change their amount. Additional details of these practical aspects are provided in the patent application filed earlier, which is hereby incorporated by reference.

Figures 5A-5D are used to introduce control of nozzle firing and drive waveform selection.

Usually, the effects of different drive waveforms and the amount of small droplets produced are measured in advance. In one embodiment, up to sixteen different drive waveforms are then stored in a 1k static random access memory (SRAM) per nozzle for each nozzle for later effective use. Provides a discrete amount of variation as selected by the software. In these different drivers With the waveform available, each nozzle then stylizes the data for a particular drive waveform, and each droplet is indicated with respect to which waveform to apply. The information set by this configuration is individually stored by a printer (or control processor) from the launch decision that will be applied to a substrate.

FIG. 5A depicts one such embodiment, which is generally labeled component symbol 501. In particular, a processor 503 is used to receive information defining the desired fill level for each target area for a particular layer of material to be printed. As represented by component symbol 505, this material can be a layout file or a bitmap file that defines the amount of droplets for each grid point or location address. A series of piezoelectric transducers 507, 508, and 509 are small droplet sizes 511, 512, and 513 that produce associated ejections, which are each determined according to a number of factors, including nozzle drive waveforms and different Manufacturing variations between printheads. During a calibration operation, given the specific ink that will be used, each of a set of variables is tested for its effect on the amount of small droplets, including variations between different nozzles and different drive waveforms. Use; if desired, the calibration operation can be done dynamically, for example, in response to changes in temperature, nozzle clogging, nozzle aging, or other parameters. This calibration is represented by a small droplet measurement device 515 that provides measured data to the processor 503 for managing the printing schedule and subsequent printing. In one embodiment, the measurement data is calculated during an operation that is actually spent in minutes, for example, for thousands of nozzles no more than thirty minutes and preferably shorter (eg, for number Thousands of print head nozzles and potentially dozens of possible nozzle emission waveforms are used as an off-line process. In another embodiment, such measurements may be updated with different subsets at different points in time (eg, between successive substrates when the substrate is being loaded and unloaded in an assembly line process). The way of the nozzle is repeatedly executed, changing In other words, it is executed for each nozzle. A non-imaging (e.g., interferometric) technique can be used selectively, which potentially produces dozens of small droplet measurements per nozzle, covering tens to hundreds of nozzles per second. This material, along with any associated statistical models (and means), can be stored in memory 517 for processing layout data or a printed image (e.g., bitmap data) 505 as it is received. In one embodiment, the processor 503 is part of a computer at the far end of the actual printing press, and in a second embodiment, the processor 503 is associated with a manufacturing mechanism for the product (eg, one) For systems that make displays, or integrated with a press.

In order to perform the transmission of the droplets for the depicted embodiment, a set of one or more timing or synchronization signals 519 are received for use as a reference, and these timing or synchronization signals are passed through a clock tree 521 for use. The drive waveforms for the particular nozzles (527, 528, and 529, respectively) are distributed to each of the nozzle drivers 523, 524, and 525. Each of the nozzle drivers has one or more registers 531, 532, and 533, and the registers receive multi-bit stylized data and timing information from the processor 503. Each nozzle driver and its associated register receive one or more dedicated write enable signals for the purpose of programming the registers 531, 532, and 533, respectively (we_n). In one embodiment, each of the registers includes a corresponding amount of memory, including a 1k static RAM (SRAM) to store a plurality of preset waveforms, and a programmable register. Choose between those waveforms and other ways of controlling waveform generation. Data from the processor and timing information are depicted as multi-bit information, and although this information can be provided to each nozzle via a series or side-by-side bit line (as seen in Figure 5A) The parallel signal representation, as will be seen in Figure 5B, discussed below, in one embodiment, the connection is in series).

For a given deposition, print head or ink, a processor selects a set of sixteen drive waveforms for each nozzle, which may alternatively be applied to produce a small droplet; It is noted that this number is arbitrary, for example, four waveforms may be used in one design, and four thousand may be used in another design. These waveforms are advantageously selected for each nozzle to provide the desired change in the amount and/or position of the output droplets, for example such that each nozzle has at least one that produces a near ideal droplet amount (eg, The waveform of an average small droplet amount of 10.00 pL is selected, and a range of carefully considered amount changes is provided for each nozzle. In various embodiments, the same set of sixteen drive waveforms are used for all of the nozzles, although in the depicted embodiment, sixteen possible waveforms are individually prior to each nozzle. Defined by ground, each waveform is characterized by an individual small droplet amount.

During printing, in order to control the deposition of each droplet, the data of one of the predefined waveforms is selected to be programmed into each of the individual registers 531, 532 or 533 for each nozzle. in. For example, given a target droplet amount of 10.00 pL, the nozzle driver 523 can be configured to be configured by writing data into the register 531 to set one of the sixteen waveforms corresponding to sixteen different One of the small droplets. The amount produced by each nozzle will have been measured by the small droplet measuring device 515, wherein the amount of droplets and associated distribution of each nozzle (and each waveform) is borrowed. It is temporarily stored by the processor 503 and stored in the memory to help generate the desired target. This same process can be performed for small droplet positions or trajectories. The processor can program the scratchpad 531 to define whether it wants the particular nozzle driver 523 to output one of the sixteen waveforms selected by the processor. The processor can also program the register for a given scan line to The delay or offset of each nozzle utilizes the emission of the nozzle (eg, to align each nozzle with a grid traversed by the print head to correct for errors including velocity or trajectory errors, and for Other purposes); this offset is achieved by a counter that adjusts the use of that particular nozzle (or transmit waveform) by the number of programmable timing pulses for each scan. . To provide an example, if the result of the small droplet measurement indicates that a particular small droplet system tends to have a lower than expected velocity, the corresponding nozzle waveform can be triggered earlier (eg, by reducing for use in pressure) a time-out before the effective actuation of the signal level, and which is advanced in time; conversely, if the result of the small droplet measurement indicates that the particular droplet has a relatively high velocity, Then the waveform can be triggered later, and so on. Other examples are also clearly feasible - for example, a slow droplet velocity may be increased in some embodiments by increasing the drive strength (i.e., the signal of the piezoelectric actuator used to drive a given nozzle). The level and associated voltage are offset. In one embodiment, a synchronization signal that is spread across all of the nozzles is generated at a defined time interval (e.g., 1 microsecond) for synchronization purposes, and in another embodiment, the synchronization signal is It is adjusted relative to printer motion and substrate layout, such as incremental movement of each micron between the printhead and the substrate. High speed clock (φ_Hs) running thousands of times faster than the sync signal, for example at 100 megahertz, 33 megahertz, etc.; in one embodiment, multiple different clocks or other timing signals ( For example, the strobe signal can also be utilized in combination. The processor also programmatically defines a grid spacing value; in one embodiment, the grid spacing is common to the entire group of available nozzles, although this is not required for each embodiment. . For example, in some cases, a regular printed grid can be defined where each nozzle is "per 5 microns" launched. This printed grid can be unique to the printing system, substrate, or both. Thus, in an alternative embodiment, a printing grid can be used It is defined for a particular printer in which the sync frequency or nozzle emission pattern is used to effectively convert the printed grid to match a previously unknown substrate layout. In another contemplated embodiment, a memory system is shared between all of the nozzles, which allows the processor to pre-store some of the different grid spacings (eg, 16) that are shared between all of the nozzles. So that the processor can (on demand) select a new grid spacing, which is then read out to all nozzles (eg, to define an irregular grid). For example, in an embodiment in which the nozzle is used to emit for each color member well of an OLED (eg, to deposit a layer of a particular color), three or more different grids The interval can be continuously applied in a cyclic manner by the processor. It is obvious that many design alternatives are feasible. It is noted that the processor 503 can also dynamically reprogram the scratchpad of each nozzle during operation, i.e., the sync pulse is applied as a trigger to initiate any setting in its register. The programmed waveform pulse, and if the new data is received asynchronously before the next sync pulse, the new data will be applied at the next sync pulse. In addition to setting parameters for the generation (536) of the sync pulse, the processor 503 also controls the start and speed of the scan (535). In addition, the processor controls the optional rotation of the printhead (537). In this manner, each nozzle can simultaneously (or simultaneously) transmit at any time (i.e., at any "next" sync pulse) using any of the sixteen different waveforms for each nozzle. And the selected transmit waveform can be dynamically swapped between any of the sixteen different waveforms during the single scan.

Figure 5B shows additional detail of a circuit (541) used in such an embodiment to generate an output nozzle drive waveform for each nozzle; the output waveform is represented in Figure 5B as "nzzl-drv.wvfm" ". More specifically, the circuit 541 receives the synchronization signal, and carries a serial A single bit line of data ("data"), a dedicated write enable signal (we), and the high speed clock (φ)_Hs)input of. A register file 543 provides data for at least three registers that carry an initial offset, a grid definition value, and a drive waveform ID, respectively. The initial offset is a programmable value that adjusts each nozzle as mentioned to align with the beginning of a printed grid. For example, given an implementation variable of multiple print heads, multiple columns of nozzles, different print head rotations, nozzle firing speeds and modes, and other factors, the initial offset can be utilized to align each nozzle The droplet pattern is with the beginning of the print grid to account for delays and other factors. The offset can be applied differently across multiple nozzles, such as to rotate a printed grid or halftone pattern relative to the substrate layout, or to correct substrate misalignment. Similarly, as mentioned, the offset can also be utilized to correct for aberrant speed or other effects. The grid definition value is a number representing the number of "counted" sync pulses before the programmed waveform is triggered; in the case of an implementation of a printed flat panel display (eg, an OLED panel), it will be printed The target area therein is assumed to have one or more regular intervals associated with different print head nozzles, which correspond to a regular (fixed spacing) or an irregular (multiple spacing) grid. As mentioned previously, in one embodiment, the processor maintains its own sixteen items of SRAM to define up to sixteen different grid intervals, which can be read out for use as needed All nozzles of the scratchpad circuit. Thus, if the print grid spacing value is set to 2 (e.g., every two microns), then each nozzle can theoretically be fired at this interval. The drive waveform ID is representative of a selection value used to select one of the pre-stored drive waveforms for each nozzle. In one embodiment, the drive waveform ID is a four-bit selection value, and each nozzle has its own dedicated 1k byte SRAM to store up to sixteen preset nozzle drive waveforms. It is stored as 16×16×4B items. In short, each waveform can be Consisting of sixteen discrete signal levels for sixteen items of each waveform, including four bytes representing a programmable signal level, where the four bytes are Represents a two-tuple resolution voltage level and a two-tuple programmable duration that is used to count the number of pulses at that high speed clock. Each programmable waveform can therefore consist of (0 to 1) discrete pulses, up to sixteen with programmable voltages and durations respectively (eg, having a clock equal to 33 megahertz) Discrete pulses of duration of 1-255 pulses).

Component symbols 545, 546, and 547 are an embodiment of a circuit that demonstrates how a given waveform can be generated for a given nozzle. A first counter 545 receives the sync pulse to begin a countdown of the initial offset, which is triggered by the beginning of a new line scan; the first counter 545 counts down in micrometer increments And when it reaches zero, a trigger signal is output from the first counter 545 to a second counter 546; this trigger signal essentially begins the emission process for each nozzle of each scan line. The second counter 546 then implements a programmable print grid spacing in increments of microns. The first counter 545 is reset in connection with a new scan line, and the second counter 546 is reset using the next edge of the high speed clock after its output trigger. When the second counter 546 is triggered, it activates a waveform generator circuit 547 that produces the shape of the selected drive waveform for a particular nozzle. As represented by dashed squares 548-550 found under the generator circuit, the generator circuit is based on a high speed digital to analog converter 548, a counter 549, and a high voltage amplifier 550. According to the high speed clock (φ_Hs) is being timed. When the trigger from the second counter 546 is received, the waveform generator circuit retrieves the pair of signals represented by the drive waveform ID value (signal level and hold) And a predetermined analog output voltage is generated based on the signal level value, wherein the counter 549 is effective to maintain the DAC output for a duration according to the counter. The appropriate output voltage level is then applied to the high voltage amplifier 550 and output as the drive waveform of the nozzle. The next number pair is then latched from the scratchpad 543 to define the next signal level value/duration, and so on.

The circuit depicted provides an effective means of defining any desired waveform based on the information provided by the processor 503. If it is necessary to match the geometry of the printed grid or to mitigate a nozzle with an abnormal speed or angle of flight, then any particular signal level (eg, defining a first "zero" signal relative to the offset of the synchronization The associated duration and/or voltage level can be adjusted. As noted, in one embodiment, the processor determines a set of waveforms (eg, 16 possible waveforms per nozzle) in advance, and which will then be used for each of these selected waveforms. Writing into the SRAM of the driver circuit for each nozzle, wherein a predetermined selection of the programmable waveform that is applied in response to a firing decision is then written by writing a four-bit drive waveform ID It is achieved by going to each nozzle register.

The use of multiple signal levels to form a pulse train is further discussed in relation to Figure 5C.

In other words, in one embodiment, the waveform may be predefined as a sequence of discrete signal levels defined by digital data, wherein a drive waveform is generated by a digital to analog converter (DAC) of. The component symbol 551 in FIG. 5C refers to a waveform 553 having discrete signal levels 555, 557, 559, 561, 563, 565, and 567. As noted for this embodiment, each nozzle driver includes circuitry that receives and stores up to sixteen different signal waveforms, each of which is defined as a series of up to sixteen signals. Level, each signal level is expressed as a multi-bit voltage and for a duration. In other words, the pulse width can be effectively varied by defining different durations for one or more signal levels, and the drive voltage can be waveformd in a manner selected to provide a fine droplet size change. Forming, for example, wherein the amount of droplets is metered to provide an increment of a particular amount, for example, in units of 0.10 pL. Thus, under such an embodiment, the waveform shaping system provides the function of modifying the amount of small droplets to approximate a target droplet size; for example, using the techniques exemplified above and other specific droplet sizes and When the positions are combined, these techniques make it easy to accurately fill each target area. In another embodiment, a preset waveform may be applied, wherein the optional further waveform shaping or timing is applied as appropriate to adjust the droplet volume, velocity and/or trajectory. In yet another example, the use of an alternative to the nozzle drive waveform provides a mechanism to plan the amount such that further waveform shaping is not necessary.

FIG. 5D shows yet another design 571. (for example, a master computer) A CPU 573 transmits (adjusted for error) printed material to a printing module having a plurality of print heads (eg, one of six different print heads, each The print head has hundreds of printing nozzles). At the printing module, an Ethernet connection 575 receives the data from the CPU and provides the data to a field programmable gate array ("FPGA") 577. A custom "soft processor" (578) within the FPGA processes the data as appropriate and writes the data to memory (ie, dynamic random access memory or "DRAM") ) 579. Unlike the embodiments described above, in the depicted embodiment, a waveform is written to the memory for use by one or more nozzles; the particular format in which the material is written or stored may be an implementation. decision making. For example, in one embodiment, the soft processor 578 writes a particular waveform for each nozzle as a series of drive levels (eg, as discussed above in relation to Figure 5C); other FPGA logic 580 is then appropriate time This information is read and provided to amplifier 581 and ultimately to the associated print head 583 and associated nozzle driver. In one embodiment, the depicted FPGA 577 and DRAM 579 is the component of each printhead (e.g., the material is individually sent to each printhead), but this need not be the case for all embodiments. It is noted that a function of logic 580 and amplifier 581 are to place the appropriate waveforms in a parallel form, that is, they can be read together in parallel for each nozzle at the appropriate time (as necessary for design) Or as appropriate.) In one embodiment, the DRAM 579 can be optionally configured to have individual memory for each nozzle. It is noted that in this embodiment, instead of using a trigger, each nozzle receives one of sixteen waveforms (ie, one of the waveforms is a flat waveform or a zero drive signal, It states that a particular nozzle will not be fired). The particular drive waveform can be dynamically supplied by the CPU 573 or written in advance (i.e., dynamically changeable), wherein a 4-bit value of an execution phase is used to provide nozzle emission. The decision and associated waveform selection (and triggering the output of the waveform from the DRAM 579 or logic 580). Other alternatives are also possible.

Further details regarding one possible embodiment of such a device will now be presented under the nozzle control circuit as described so far as can be used in an exemplary manufacturing apparatus. As mentioned earlier indirectly, one embodiment of the technology described herein is to fabricate a flat panel device in an array wherein the devices are then cut from a common substrate. In the following discussion, a system for performing the example of such printing will be described, and more specifically, it can be applied to electronic devices (eg, smart phones, smart watches, tablets). The manufacture of solar panels and/or display devices in computers, computers, televisions, monitors, or other forms of displays. The manufacturing technology provided by the disclosure is not limited to this special The application is applicable and can be applied, for example, to any 3D printing application as well as a wide range of other forms of products.

Figure 6A represents a number of different implementation levels, the entire system of which is indicated by element symbol 601; each of these levels represents one possible discrete implementation of the techniques described herein. First, the techniques as described herein may be in the form of instructions stored on non-transitory machine readable media, as represented by graphics 603 (eg, for controlling a computer or Is an executable instruction or software of a printing press). Second, according to computer image 605, these techniques can also be optionally implemented as part of a computer or network, for example, in a company that designs or manufactures components for sale or use in other products. . Third, as exemplified by a stored media graphic 607, the techniques described earlier may have the form of a stored printer control command that, for example, becomes a material, and acts on the material in accordance with the above discussion. In this case, it will cause a printer to make one or more layers of a component depending on the amount of ink or position used to mitigate alignment errors. It is noted that the printer instructions can be sent directly to a printer, for example, on a LAN; in this context, the storage medium graphic can represent (without limitation) in a computer or printer Or its accessible RAM, or a portable medium such as a flash drive. Fourth, as represented by a manufacturing device image 609, the techniques described above can be implemented as part of a manufacturing apparatus or machine, or in the form of a printing press within such apparatus or machine. It should be noted that the particular depiction of the fabrication device 609 represents an exemplary printer device that will be discussed below in relation to Figure 6B. The techniques described above may also be embodied as a component of a fabricated component; in Figure 6A, for example, a plurality of such components are depicted in the form of a flat device having a semi-finished array of arrays 611 that will Separated and sold for inclusion in the final elimination In the fee product. The device depicted may, for example, have one or more layers or encapsulating layers or other layers that produce light in accordance with the methods described above. The techniques described above may also be embodied in the form of an end consumer product, for example, as referred to with a display for a portable digital device 613 (eg, such as an electronic tablet or smart phone). A screen, television display screen 615 (eg, OLED TV), solar panel 617, or other type of device.

Figure 6B shows a fabrication apparatus 621 that is contemplated to be utilized to apply the multi-chamber of the techniques disclosed herein. In general, the depicted device 621 includes a plurality of general modules or subsystems including a transmission module 623, a printing module 625, and a processing module 627. Each module maintains a controlled environment such that printing can be performed, for example, by the printing module 625 in a first controlled atmosphere, and is, for example, another deposition process such as inorganic encapsulation deposition. Alternatively, other processing of a curing process (e.g., material for printing) can be performed in a second controlled atmosphere. The device 621 uses one or more robots to move a substrate between the modules without exposing the substrate to an uncontrolled atmosphere. It is possible to use other substrate transport systems and/or specific devices and control systems adapted to the processing to be performed for the modules within any given module.

Various embodiments of the transmission module 623 can include an input carrier 629 (i.e., a chamber that provides buffering between different environments while maintaining a controlled atmosphere), a transmission chamber 631 (also useful) A robot for transporting a substrate, and an ambience buffer chamber 633. Within the printing module 625, it is possible to use other substrate transport mechanisms, such as air-floating stations, for supporting a substrate that are stable during a printing process. In addition, an xyz motion system such as a split shaft or a gantry motion system can be utilized with respect to at least one print head relative to The substrate is precisely positioned and provides a y-axis transport system for transport of the substrate through the printing module 625. It is also possible to use a plurality of inks for printing in the printing chamber, for example by using individual print head assemblies such that, for example, two different types of deposition processes can be placed in the control module in a controlled atmosphere. carried out. The printing module 625 can include a gas enclosure 635 that houses an inkjet printing system having an inert atmosphere (eg, nitrogen) and other means of environmental conditioning (eg, temperature and pressure). , the total amount of gas (constituency) and the presence of particles to control the atmosphere.

Various embodiments of a processing module 627 can include, for example, a transfer chamber 636; the transfer chamber also has a robot for transporting a substrate. In addition, the processing module can also include an output carrier 637, a nitrogen stack buffer 639, and a curing chamber 641. In some applications, the curing chamber can be utilized to cure, bake, or dry a monomer film to form a uniform polymer film; for example, two well-thought processes include a heating process and a UV radiation curing process.

In one application, the device 621 is adapted for mass production of liquid crystal display screens or OLED display screens, such as the fabrication of one array of, for example, eight screens on a single large substrate. These screens can be utilized for television as well as display screens for other forms of electronic devices. In a second application, the device can be utilized in mass production of solar panels in much the same manner.

The printing module 625 can advantageously be used in such applications to deposit an organic light-generating layer or to assist in protecting the encapsulating layer of sensitive components of the OLED display device. For example, the depicted device 621 can be loaded into a substrate and can be controlled during the encapsulation process by an interruption that is not exposed to an uncontrolled atmosphere to back and forth the substrate. Move between various rooms. The substrate can be loaded via the input carrier 629. A robot disposed in the transmission module 623 can move the substrate from the input carrier 629 to the printing module 625, and after the printing process is completed, the substrate can be moved to the processing module. 627 for curing. By repeatedly depositing subsequent layers, each of the controlled thickness, total encapsulation can be established to suit any desired application. It is again noted that the above techniques are not limited to the encapsulation process, but it is also noted that many different types of tools can be utilized. For example, the configuration of the device 621 can be changed to set the various modules 623, 625, and 627 in different side-by-side configurations; further, additional, fewer, or different modules can be utilized.

Although Figure 6B provides an example of a set of joined chambers or fabricated components, it will be apparent that many other possibilities exist. The techniques described above can be utilized with the apparatus depicted in Figure 6B, or indeed to control a process performed by any other type of deposition apparatus.

Figure 6C provides a plan view of the substrate and the printer as it may appear during the deposition process. The printing chamber is generally indicated by the symbol 651, the substrate on which the substrate to be printed is substantially indicated by the symbol 653, and a support for transporting the substrate is substantially by the symbol 655. To indicate. Roughly speaking, any xy coordinate of the substrate is reached by a combination of movements including x-dimensional and y-dimensional movements of the substrate by the support (eg, utilizing The air bearing support, as represented by element symbol 657, and the x-dimensional movement of one or more print heads 659 along the "slow axis" of a traveler 661, as Represented by arrow 663. As mentioned, the air float table and the substrate transport infrastructure are used to move The substrate, and optionally, provides control of deskew removal along one or more "fast axes." The print head is visible with a plurality of nozzles 665, each of which is individually controlled by a mode of emission derived from the same plate (eg, to follow the "slow axis" at the print head. When moving from left to right and from right to left, printing of the line corresponding to the grid point of the printer is achieved). Where the relative motion between the one or more print heads and the substrate is provided in the direction of the fast axis (ie, the y-axis), the printing system depicts a printer grid that typically follows individual columns The width of the point. The print head can also be advantageously adjusted to vary the effective nozzle spacing (e.g., according to component symbol 667 by rotating the one or more printheads). It is noted that a plurality of such print heads can be utilized together, which are oriented as needed with an x-dimensional, y-dimensional and/or z-dimensional offset relative to each other (see Figure 6C). The axis legend (669) in the middle. This printing operation continues until the entire target area (and any border areas) has been printed with ink as needed. After depositing the necessary amount of ink, the substrate is processed by evaporating the solvent to dry the ink (eg, using a thermal process), or by using a curing process (eg, a UV curing process).

Figure 6D provides a block diagram showing various subsystems of a device (671) that can be utilized to fabricate a device having one or more layers as indicated herein. Coordination over the various subsystems is provided by a set of processors 673 that operate under the instructions provided by the software (not shown in Figure 6D). As indicated earlier, in order to dynamically perform the conversion to correct substrate or panel errors, in one embodiment, these processors include a supervised processor or a general purpose CPU, and a set of parallel processing. Additional processor. In a well-conceived embodiment, these additional processors have a multi-core processor or a graphics processing unit (GPU, for example, having hundreds or more cores). Each core system is specified A portion of the printed image is converted to distort the printed image to match substrate error; if the error is to be independently corrected in a panel or other product-by-product manner, each core may advantageously be assigned a single product or A segment or portion of the panel (ie, such that the particular core performs only a single conversion operation on all of the specified image data). For all embodiments, parallel processing and/or the form of this appointment is not required. During a process, the processors feed data to a printhead 675 such that the printhead ejects various amounts of ink based on the firing commands provided by the halftone printed image. The printhead 675 typically has a plurality of inkjet nozzles arranged in a column or array, and associated reservoirs that are capable of injecting ink in response to actuation of piezoelectric or other transducers; such transducers are A further nozzle is caused to eject a controlled amount of ink by an amount determined by an electron emission waveform signal applied to the corresponding piezoelectric transducer. Other launching mechanisms can also be utilized. The print head applies the ink to a substrate 677 corresponding to various xy positions of the grid coordinates as represented by the halftone printed image. The change in position is achieved by both a printhead motion system 679 and a substrate transport system 681 (e.g., such that the print depicts one or more widths across the substrate). In one embodiment, the printhead motion system 679 moves the printhead back and forth along a movable loop that provides stable substrate support and "x" and "y" dimensional transmission of the substrate. (and rotating), for example for alignment and removal of skew; during printing, the substrate transport system provides relatively fast transmission in one dimension (eg, in relation to the "y" dimension of Figure 6C), while the printing The head motion system 679 provides relatively slow transmissions in another dimension (e.g., in relation to the "x" dimension of Figure 6C), such as for print head offset. In another embodiment, a plurality of print heads can be utilized, with the primary transmission being transmitted by the substrate transport system 681. An image capture device 683 can be utilized to find any benchmarks and assist earlier The described alignment and/or error detection functions.

The apparatus also includes an ink delivery system 685 and a printhead maintenance system 687 to assist in the printing operation. The print head can be periodically calibrated or subjected to a maintenance process; for this purpose, the print head maintenance system 687 is used to perform proper priming, ink or gas removal during a maintenance sequence. , testing and calibration, and other operations, depending on the specific process conditions. Such a process may also include, for example, individual measurements of the parameters of the small droplet size, velocity, and trajectory, as in the PCT patent application (PCT/US14/35193), which is incorporated herein by reference. The arguer, and as referred to by component symbols 691 and 692.

As previously described, the printing process can be performed in a controlled environment, in other words in a manner that exhibits a reduced risk of contaminants that may degrade the effectiveness of a deposited layer. For this effect, the apparatus includes a chamber control subsystem 689 that controls the atmosphere within the room, as represented by functional block 690. The process variations as mentioned may be performed in the presence of an ambient nitrogen atmosphere (or another environment having a particular selected gas and/or controlled to exclude inertness of the undesirable particles). Spraying of deposited material. Finally, as represented by the symbol 693, the device also includes a memory subsystem that can be utilized as needed to store halftone pattern information or halftone pattern generation software, pattern print image data, And other information. For example, the memory subsystem can be utilized as an operational memory for conversion of a previously generated printed image in accordance with the techniques described above to internally produce a print that determines the emission (and timing) of each small droplet. Machine control instructions. If part or all of such an algorithm is performed elsewhere, and the operation of the device is to manufacture a device layer based on the received printer command, the received command may be stored in the The memory subsystem 693 is used during the printing process and/or processing as appropriate. As indicated by reference numeral 694, in an alternate embodiment, individual droplet details can be altered by changes in the emission waveform for any given nozzle (eg, to correct nozzle anomalies). . In one embodiment, as described earlier, a set of alternate transmit waveforms may be selected in advance and made available for use by each nozzle in a shared or dedicated manner, optionally in combination. Substrate change (error) processing 695 is utilized. As noted, although some embodiments use a preset scan path (i.e., despite errors), where the compensation for errors is exploited with different nozzles and/or drive waveforms for certain nozzles. Achieved, but in another embodiment, print optimization (696) is performed to re-evaluate scan path details and potentially improve deposition time.

FIG. 7 is a flow chart 701 that provides another process related to certain of the processes discussed. As in the earlier example, according to the component symbol 703, the data representing the layout of a desired layer is first received. This data indicates the boundaries of the layers to be deposited and provides sufficient information to define the thickness of the entire layer of interest (eg, for a given panel). This material may be generated on the same machine or device on which the process 701 is executed, or it may be generated by a different machine. In one embodiment, the received data is defined in accordance with an xy coordinate system and the information provided is sufficient to calculate the desired layer thickness at any of the represented xy coordinate points, for example, The land designation indicates a single height or thickness that will be applied to the entire layer, which is consistent with the earlier examples of x-micron y-micron by z-micron. According to component symbol 705, this material can be converted to a grayscale value for each print cell (or each printed pixel) in a deposition area that will receive the layer. If the print cell region does not inherently correspond to the xy coordinate system that matches the layout material, then the layout data is converted (eg, by targeting multiple seats) Punctuation to average thickness data and/or by interpolation to obtain a grayscale value for each printed pixel. This conversion may be based on preset mapping information generated, for example, using a relation or equation. According to component symbol 707, the grayscale values can be optionally adjusted to produce a uniform layer (or for other desired effects). An example is provided below. If compensation is required for the height of the microstructure that will be located below the desired layer, then an optional technique is to add an offset to select the grayscale value at a particular location. "lifting" the layer of interest to effectively planarize a top surface of the deposited layer. In an embodiment, gray scale value processing may also be utilized to correct for abnormalities emitted by the nozzle (eg, in the direction within the scan) to deposit more ink (eg, if a particular nozzle or nozzle group is generated) Insufficient ink volume) or less ink (for example, this particular nozzle or nozzle group produces too much ink). According to function block 714, such an optional process can be based on a calibration process and/or empirically determined data. According to component symbol 709, the grayscale values are then converted to a small drop density pattern (e.g., a particular halftone pattern), and a bitmap is then generated based on component symbol 710. As indicated by the figure, error diffusion can be relied upon to assist in improving the uniformity of the layers.

Figure 7 illustrates the use of a collection of nozzle/waveform error correction processes 713 that are optionally applied to assist in ensuring uniformity and accuracy in the deposited layers and to adjust the detected panel or substrate position. Error. Such uniformity may be important to the quality of the device, whether to ensure accurate alignment with any of the underlying product layers, to develop sufficient encapsulation to create a barrier to water/oxygen, or It is a high quality light-emitting or light-guiding element that provides a display panel, or for other purposes or effects. As mentioned above, according to the component symbol 714, a calibration process or empirically determined (by-point estimation) data can be utilized to correct for errors in grayscale values, or for nozzle or nozzle waveform details. Or for substrates or The change in panel position. Alternatively, as represented by component symbol 715, individual nozzle drive waveforms can be programmed or adjusted to correct for errors. In another embodiment, as indicated earlier, the nozzles can be verified or judged to be acceptable (719), wherein each nozzle is judged to meet a minimum droplet generation threshold, or is unacceptable. Not used. According to the symbol 716, if a particular nozzle is already unacceptable, but is temporarily selected for use, a different nozzle (or repeated passage of an acceptable nozzle) may be provided to provide proper droplet deposition. It is utilized to deposit small droplets that would otherwise be printed by the failed nozzle. For example, in one embodiment, a print head has nozzles arranged in columns and rows such that if a nozzle is abnormal, a different redundant nozzle can be utilized to deposit a particular grid point. Small droplets. Also optionally, such a problem can be considered and used to adjust a scan path, for example, by offsetting the print head in such a way that the desired droplets can be deposited using a different nozzle ( Where the print head is adjusted in position to allow this, or to increase or decrease the number of scans. In Fig. 7, this is indicated by the symbol 717. Many such alternatives are feasible. As represented by reference numerals 720 and 721, in one embodiment, each nozzle is previously calibrated using a small droplet measuring device (720) that repeatedly Measure the droplet parameters (to develop the distribution of each nozzle or each drive waveform), where the error in the position of the nozzle and/or the average of the amount, velocity and trajectory of the nozzle droplets Next, and with an understanding of the variation of each of these parameters for each nozzle expected, the software then establishes a statistical model (721) for each nozzle. As mentioned, this information can be utilized to determine whether a particular nozzle (and/or small droplet) is qualified/verified, or to select a nozzle that will be used to generate each individual droplet, and Or adjust or customize the same version of the printed image for each new substrate or other product array. Each such measurement/error correction procedure can Considered in the printing program (722), which includes adjustment/customization of the printed image, and is considered to be any scan path plan and/or print grid calculation, ie, the press control data is generated and / or update to optimize the printing process while ensuring the desired layer properties. Finally, according to component symbol 725, the final printed material is then generated for transmission to the printer during the execution phase.

Figures 8A-8D are techniques used to generally describe the measurement and verification of small droplets for each nozzle.

More specifically, FIG. 8A provides a view depicting one of an optical system 801 and a relatively large printhead assembly 803 (ie, it is represented by a nozzle plate of an individual printhead 805A/805B). Each print head has a large number of individual nozzles (e.g., 807); in a typical embodiment, there are hundreds to thousands of nozzles. An ink supply source (not shown) is in fluid connection with each nozzle (e.g., nozzle 807), and a piezoelectric transducer (also not shown) is used in an electrical control signal based on the nozzle. Control the ink that ejects small droplets. The nozzle design maintains a slightly negative pressure of ink at each nozzle (e.g., nozzle 807) to avoid overflow of the nozzle plate, wherein an electrical signal for a given nozzle is used to activate the corresponding piezoelectric A transducer pressurizes ink for the given nozzle and thereby discharges one or more droplets from the given nozzle. In one embodiment, the control signal for each nozzle is typically at zero volts, wherein a positive pulse or signal level at a given voltage is applied to a particular nozzle to eject a small liquid of the nozzle Drops (one droplet per pulse); in another embodiment, different modified pulses (or other more complex waveforms) can be utilized between different nozzles. However, with respect to the example provided by Figure 8A, it should be assumed that measuring the amount of small droplets produced by a particular nozzle (e.g., nozzle 807) is desirable, with one small droplet being The print head is sprayed downward (i.e., in the direction "h", which represents the z-axis height associated with a three-dimensional coordinate system 808) for collection by a spittoon 809. It is noted that in a typical application, the size of "h" is typically on the order of one millimeter or less, and that there are thousands of nozzles (eg, 10,000 nozzles) in an operating press. The individual droplets are individually measured in this way. Therefore, in order to accurately optically measure each small droplet (i.e., a small droplet derived from a particular one of the thousands of nozzles in the environment of a large printhead assembly, as described earlier) Within the measurement window of approximately millimeters, certain techniques are used in the disclosed embodiments to accurately position the components of the optical component 801, the printhead assembly 803, or both relative to each other, For optical measurement.

In one embodiment, these techniques utilize a combination of: (a) at least a portion of the xy motion control (811A) of the optical system (e.g., within plane 813 of the dimension) to accurately measure a measurement region 815 is positioned in close proximity to any nozzle that will produce a small droplet for optical calibration/measurement; and (b) light recovery under the plane (811B) (eg, thereby allowing the measurement region to be easily set to Any nozzle adjacent, despite having a large printhead surface area). Thus, in an environment having an example of a printing nozzle of about 10,000 or more, the motion system is capable of positioning at least a portion of the optical system at, for example, about 10,000 each adjacent to the printhead assembly. The discrete paths of the discharge paths of the nozzles; in one embodiment, a continuous motion system or a system with more sophisticated positioning functions can be utilized. As will be discussed below, the two considered optical metrology techniques include shadowgraphy and interferometry. In each technique, the optical device is typically positionally adjusted such that precise focus is maintained over the measurement area to facilitate capturing small droplets in flight (eg, in the case of shadowing to facilitate effectiveness) Ground the shadow of the small droplet). Noticed that one Typical droplets may be on the order of microns in diameter, so the optical placement is typically quite accurate and presents challenges in terms of the relative positioning of the printhead assembly to the metrology/measurement area. In some embodiments, to assist in this positioning, an optical device (mirror, cymbal, etc.) is used to orient a plane 813 from the measurement region 815 for sensing the dimension. The lower light capture path allows the measured optical device to be placed close to the measurement area without interfering with the relative positioning of the optical system and the print head. This allows for an effective position in a manner that is not limited to the deposition height h of the millimeter level in which a small droplet is imaged, or the large scale x and y width occupied by a carefully inspected print head. control. Under interferometric-based small droplet measurement techniques, individual beam systems incident on a small droplet from different angles produce interference that can be detected from a point of view that is substantially orthogonal to the optical path. a pattern; therefore, the optical device in such a system is at an angle of about ninety degrees from a path that is offset from the source beam, but also in a manner that utilizes light recovery under the plane to facilitate measurement of small droplet parameters. Capture light. Other optical metrology techniques can also be utilized. In yet another variation of these systems, the motion system 811A is optional and advantageously formed as an xyz-motion system that allows for selective selection without moving the printhead assembly during small droplet measurement The input and disengagement of the small droplet measurement system. Briefly, in an industrial manufacturing facility having one or more large printhead assemblies, it is contemplated that in order to maximize the normal working time of manufacture, each printhead assembly will be "stopped" from time to time. One or more maintenance functions are performed in the service station; given the large size of the print head and the number of nozzles, it may be desirable to perform multiple maintenance functions on different portions of the print head at a time. To this effect, in such an embodiment, it may be advantageous to move the metrology/calibration device around the printhead rather than moving the printhead around the metrology/calibration device. [This also allows for the input of other non-optical maintenance procedures, for example, if desired In other words, it is related to another nozzle. In order to make these actions easy, the printhead assembly can be optionally "parked", wherein the system identifies a particular nozzle or range of nozzles that will be the subject of optical calibration. Once the printhead assembly or a given printhead is stationary, the motion system 811A is acted upon to move at least a portion of the optical system relative to the "stopped" printhead assembly to accurately The measurement region 815 is positioned at a position suitable for detecting a small droplet ejected from a particular nozzle; the use of the movement of the z-axis allows the optical device to recover light from the plane of the printhead Very selective input below makes it easy to replace other optical calibrations or additional maintenance operations beyond optical calibration. It may be further noted that the use of an xyz-motion system allows for a selective investment of a small droplet measurement system independent of the test or test equipment used in a service station environment. It is noted that this configuration is not required for all embodiments; other alternatives are also possible, such as where only the printhead assembly is moved and the measurement assembly is stationary, or where the printhead is The docking of components is not necessary.

In general, an optical device for small droplet measurement will include a light source 817, a set of optional optically transmitted optical devices 819 (which are necessary to direct light from the light source 817 to the measurement). a region 815), one or more photo sensors 821, and a set of recovered optical devices 823 that direct light from the measurement region 815 for measuring the droplets to the one or more Light sensor 821. The motion system 811A is optionally provided with a direction in which the light that is allowed to be measured by the small droplets is from a measurement area 815 around the ink collector 809 to a position below a plane, while also providing a container (eg, a set) Ink 809) moves together any one or more of these components and ink collector 809 in a manner that collects the ejected ink. In one embodiment, the optically-transferred optical device 819 and/or the optically-recovered optical device 823 are oriented using a vertical dimension that travels parallel to the droplets. To direct the mirror of the light to and from the measurement region 815, wherein the motion system moves each of the components 817, 819, 821, 823 and the ink collector 809 in a unitary unit during the droplet measurement This setting provides an advantage that there is no need to recalibrate the focus relative to the measurement area 815. As indicated by reference numeral 811C, the optical device for optical transmission is also used to selectively supply source light from a location below the plane 813 of the dimension of the measurement region, for example, where light source 817 and light Both sensors 821 direct light on both sides of the ink collector 809 for purposes of measurement, i.e., as generally depicted. As indicated by reference numerals 825 and 827, the optical system can optionally include a lens for focusing purposes and a photodetector (eg, for non-imaging techniques that do not rely on the processing of a "picture" of a multi-pixel) ). It is again noted that the optional use of the optical assembly and the z-motion control of the ink collector allows for optional insertion and disengagement of the optical system at any point in time when the print head assembly is "parked". And the measurement region 815 is positioned close to any nozzle. Such docking of the printhead assembly 803 and xyz-motion of the optical system 801 are not required for all embodiments. For example, in one embodiment, laser interferometry is used to measure droplet characteristics, wherein the printhead assembly (and/or the optical system) is within or parallel to the deposition plane The plane is moved (e.g., within plane 813 or parallel to plane 813) to image small droplets from various nozzles; other combinations and arrangements are also possible.

Figure 8B is a flow chart showing a procedure associated with small droplet measurement for certain embodiments. This program flow is generally indicated by the symbol 831 in FIG. 8B. More specifically, as indicated by reference numeral 833, in this particular process, the printhead assembly is first docked in a service station (not shown) such as a printer or deposition apparatus. A droplet measuring device is then placed (835) into the printhead assembly, for example by an optical Part or all of the system moves through from below a deposition plane to a selective input in which the optical system is capable of measuring the position of individual droplets. Depending on the component symbol 837, this movement of the components of one or more optical systems relative to a docked printhead can optionally be performed in the x, y and z dimensions.

As mentioned previously, even a single nozzle and associated nozzles emit a drive waveform (ie, a pulse or signal level that is used to eject a small droplet) may produce a slight change between droplets. Small droplet volume, trajectory, and velocity. In accordance with the teachings herein, in one embodiment, a small droplet measurement system as indicated by element symbol 839 obtains n measurements of a desired parameter for each small droplet to derive relevant parameters. The statistical confidence of the expected nature. In one embodiment, the measured parameter may be an amount, and for other embodiments, the measured parameter may be flight speed, flight trajectory, nozzle position error (eg, nozzle bow), or other A parameter, or a combination of multiple such parameters. In an embodiment, "n" may vary with each nozzle, while in another embodiment, "n" may be a fixed number of measurements that will be performed for each nozzle (eg, "24" In another embodiment, "n" refers to a minimum number of measurements such that additional measurements can be performed to dynamically adjust the statistical properties of the measurement of the parameter or to improve reliability. Obviously, many changes are feasible. For the example provided by Figure 8B, it should be assumed that the small droplet volume is being measured in order to obtain an accurate average representing the amount of small droplets expected from a given nozzle, and a rigorous The confidence interval. This average value can be specified by a processor taking into account appropriate measurement data, weighting, or other means. This system enables an optional plan for the combination of small droplets (which utilizes multiple nozzles and/or drive waveforms) while reliably maintaining an expected target in a target region (ie, relative to a composite The distribution of the composite ink fills the average of the small droplets. As with matching As indicated by blocks 841 and 843, interferometry or shading is the optical measurement procedure that is thought of as ideally measuring or calculating the instantaneous or near-instantaneous energy (or other desired parameter). With such rapid measurements, it is possible to constantly and dynamically update the amount of measurement, for example to consider the properties of the ink (including viscosity and composition), temperature, power supply variations, and other factors. The change over time. Constructed at this point, shading is typically characterized by, for example, using a high resolution CMOS or CCD camera as a light sensor mechanism to capture an image of a small droplet; although the droplet can be in multiple locations Accurately imaged in a single image capture frame (eg, using a strobed light source), but image acquisition typically involves a limited amount of time from a large printhead assembly (eg, with thousands of nozzles) Imaging of a sufficient population of small droplets can take hours. Interferometry based on multiple binary photodetectors and detection of interference pattern spacing based on the output of such detectors is a non-imaging technique (ie, it does not require image analysis) and thus is more than shadow Or a technique that is many times faster (eg, 50 times) to produce a measure of the amount of droplets; for example, under a printhead assembly of 10,000 nozzles, it is expected for the thousands A large measurement population of each of the nozzles can be obtained in a matter of minutes, making it possible to perform small droplet measurements frequently and dynamically. As indicated earlier, in an alternate embodiment, small droplet measurements (or measurements of other parameters such as trajectories and/or velocities) may be performed as a periodic intermittent process, Where the droplet measurement system is based on a plan or between substrates (eg, when the substrate is being loaded or unloaded), or stacked to other components and/or other printheads for maintenance The program was put in. It is noted that for embodiments in which alternative nozzle drive waveforms can be used in a specific manner per nozzle, a fast measurement system (eg, an interferometric system) is readily tolerated for each Nozzles and for each of the alternate drive waveforms for the nozzle The development of statistical populations, thereby making it easier to combine small droplets of small droplets produced by various nozzle-waveform pairs, as mentioned earlier. According to the symbol 845 and 847, the target deposition area is determined according to the target deposition area for each nozzle (and/or by pairing for each nozzle-waveform) to an accuracy of better than 0.01 pL. It is possible to plan a very precise combination of small droplets, where the composite fill can also be factored to a resolution of 0.01 pL, and wherein the target amount can be maintained at a specified error of 0.5% of the target amount. (eg, tolerances) or better; as indicated by reference numeral 847, the measurement population for each nozzle or each nozzle-waveform pairing is planned in one embodiment for each Such nozzles or nozzle-waveform pairings produce a reliability distribution model, that is, a 3σ confidence (or other statistical measure, such as 4σ, 5σ, 6σ, etc.) relative to the allowable small droplet tolerance. Wait). Once sufficient measurements have been taken for the various droplets, the filling of combinations involving those droplets can be evaluated in the most efficient manner possible and used to plan printing (848). As indicated by divider 849, droplet measurement can be performed intermittently between the effective printing process and the measurement and calibration procedure; note that in order to minimize manufacturing system downtime Time, such measurement is typically performed while the printer is being assigned to other processes, such as during substrate loading and unloading.

Figure 8C depicts another embodiment of a method of small droplet measurement, which is generally represented by the symbol 851. When a print head is installed, or the small droplet measurement system is calibrated to correct the positional offset, a calibration routine can be performed to accurately match the small droplet measurement to a given print. Head nozzle. In a typical embodiment, the alignment process is performed by means of an "upward facing camera" or other imaging device that takes an image of a printhead from below, in other words, from a substrate View upwards The nozzle plate identifies one or more print head references or alignment marks (853). In an embodiment, the camera (imaging system) may be the same device as for small droplet measurement, but it may also be a separate imaging device. For a print head having an example of 1024 nozzles, for example, 256 nozzles arranged in four columns, the reference on the nozzle plate (not confused with the substrate reference) is used to determine the nozzle plate The offset from the grid system corresponding to the imaging device and the skew of the rotation. It is noted that, in an embodiment, the reference may be a specific nozzle (854), such as a nozzle closest to the corner of the print head (eg, nozzles 1 of the first and fourth columns) And 256); other mechanisms can also be utilized. In a typical embodiment, the printhead configuration data (855) is loaded into the system by software and used to identify the nozzles at the corners and is used to map the addresses of all nozzles to the A grid 856 of the imaging system in which interpolation is relied upon to initially estimate the position of each nozzle (857). An example is provided below. In one embodiment, the system soft system is designed to accommodate different print heads having different nozzle configurations, and for this effect, the system soft system loads the print head configuration data to identify the number of columns. The presence of a reference, if any, the number of nozzles in each column, the average vertical and horizontal offset between the columns and rows of nozzles, and the like. As mentioned, this information enables the system software to estimate the position of each nozzle on the printhead. In a contemplated system, the calibration procedure is performed once a printhead is changed, but is not executed between print processes; in a different embodiment, the calibration procedure is performed each time The small droplet measurement system is executed when initialized, for example, where each new measurement is performed between two printing operations.

During manufacture, nozzle (and nozzle-waveform) measurements can be performed in turn, which moves forward a range of nozzles, with each interruption being between substrate printing operations. Whether to invest in re-measuring all nozzles, or measuring in such a way, the same basic of Figure 8C The program can be utilized for measurement. For this effect, according to the component symbols 858 and 859, when the small droplet measuring device is put into a new measurement (which is followed by a previous measurement or a substrate printing operation), the system software can The next nozzle to be measured is identified (eg, by loading an index of "nozzles 2, 312" that point to the "312th" nozzle for a second printhead). In the case of the initial measurement (for example, in response to a new print head installation, or a recent boot, or a periodic program such as a daily measurement procedure), the indicator will point to A first nozzle of a print head, for example, "nozzle 2, 001". The nozzle is associated with a particular imaging grid access or a nozzle that is queried from the memory. The system uses the provided address to advance the small droplet measurement system (e.g., the ink collector and measurement area referenced earlier) to a position corresponding to the desired nozzle position. It is noted that in a typical system, the mechanical throw associated with this movement is quite accurate, i.e., to a resolution of about micrometers. The system at this point selectively searches for nozzle positions for the desired micro-resolution position and finds the nozzle based on image analysis of the print head within a small micron distance of the estimated grid position and Centered on its location (860). For example, a zigzag, spiral or other search mode can be utilized to search for a nozzle at the desired location. [It is noted that in an embodiment, the procedure can also be performed manually by an operator by adjustment of the printer. A typical spacing distance between nozzles may be on the order of 250 microns, while the nozzle diameter may be on the order of 10-20 microns. Once the nozzle of interest is identified, the soft system emits a small droplet from the nozzle in question and confirms that the nozzle in question is actually emitted according to the small droplet measurement system (which then confirms the nozzle) identity). Figure 8C shows that this procedure is performed each time a new nozzle is identified for measurement (e.g., each time the droplet measurement system is moved), but in some embodiments, in one embodiment It is also possible to perform this measurement once during offline configuration. (eg, in the case where the droplet measurement system grid is very dense) to store the grid position for each nozzle, and then only when the print head is replaced or in response to error processing This location is updated. In systems where the mechanical portion of the droplet measurement system and/or the position of the print head are not very accurate, the function of estimating and searching for each nozzle may be when the nozzles that are carefully inspected are changed at any time. advantageous. It is noted that, as indicated by the symbol 861, in one embodiment, the function of estimating and searching is in each of three dimensions (xyz), the small droplet measuring device (and its associated optical equipment) are aligned with the carefully inspected printhead nozzles.

The precise z position of each nozzle (distance relative to the small droplet measurement area) is then adjusted (862) to ensure consistent droplet measurement and/or image capture. For example, it has been previously mentioned that a small droplet measurement system typically calculates each droplet by multiple measurements and calculates it by distance (eg, relative to a centroid of each droplet image). These parameters are used to determine the velocity of the droplets and the flight path. Various parameters may affect the correct droplet measurement, which is included in the gating timing (eg, for a shadow-based small droplet measurement system) in which the droplet imaging system Uncorrected alignment errors with the nozzle plate, nozzle process corners, and other factors. In one embodiment, various statistical procedures are used to compensate for such errors, for example, in a manner that normalizes gated emissions associated with small droplet measurement locations across all small droplets; for example, if A hypothetical printhead has 1,000 nozzles, and the system can normalize the z-axis offset relative to the printhead by picking an average offset, which is the average droplet position. The average offset produces a minimum positional error (average across the 1,000 nozzles or subsets thereof) when centering on a desired number of droplets in the measurement region. Similar techniques can be applied to an interferometric based system or other small Drop measurement system.

Figure 8C shows a hypothetical channel 863 and a hypothetical channel through which the two small droplets 864 and 866 pass, along the individual hypothetical tracks 865 and 867. Several things should be noted about the examples provided in this figure. First, the measurement of velocity and trajectory can be seen by measuring the same small droplets multiple times (three times in the case of small droplets 864 and 866, respectively). Such a requirement can be utilized to properly position the measurement region relative to the strobe (or imaging source) transmission by changing the timing of the strobe (or source), the change being used to transmit the correlation The driving waveform of the droplets, changing the z-axis position of the droplet measuring system, and/or changing one or more of the z-axis positions of the printhead. For example, when the gating is repeatedly fired (in the case of a shadow-based system during a single exposure), if there are three small droplet images expected for a single small droplet system, but only two It is observed that the measurement area is misaligned in height and is adjusted in relation to the measurement area to effectively redefine where the droplet position is captured until three exposures are obtained. Of course, this assumption is only an example, so other implementations may measure more than 3 gated droplet exposures, or less than 3 exposures. It is also noted that the deviation of the trajectory 865 relative to the trajectory 867 may be due to statistical variations in the manner in which the small droplets are generated, and thus may alternatively be used to create a small droplet trajectory representative of the average ( The statistical model of the standard deviation in each of these dimensions, in terms of α and β angles. As should be appreciated, although the small droplet measurement area 863 shows a two-dimensional droplet depiction (eg, according to a yz plane of the pattern page), the trajectory angle associated with the x-axis can be The change in the size of the droplets in the appearance of a given image frame in a plurality of gated exposures is indicated, indicating that the droplets are approaching or moving away from the pattern paper represented by Figure 8C. Plane; similar interference mode variations apply to interferometry In the case of law-based technology.

The design represented within measurement 863 can also be utilized to measure the curvature of the nozzle rows. In other words, for example, if the small droplets 864 and 866 are assumed to originate from a common exact nozzle position, the reverse trajectory does not correspond to the small droplet measurement area (ie, related to the pattern page from the left To the right of the expected y-axis center alignment, the nozzle in question may be offset at its y-axis position relative to other nozzles in the same column or row. As implied by earlier discussions, such anomalies may result in deviations from idealized droplet launches, which can be considered in a well-planned combination of small droplets, ie, as early as It is preferred that any such column "bending" or individual nozzle offsets are stored and used as part of a printed scanning program, wherein the printing system utilizes each individual in a plan The difference in nozzles is not the average of those differences. In an optional variation, the same technique can be utilized to determine irregular nozzle spacing along the x-axis, although for the depicted embodiment any such error is attributed to the small fluid In the correction of the drop velocity deviation (for example, any such interval error can be corrected by the adjustment of the nozzle speed). To determine the y-axis curvature of a nozzle that produces small droplets 864 and 866, the individual tracks 865 and 867 and other metrology tracks of the same nozzle are effectively inversely drawn (or mathematically applied) and Used to identify the average y-axis position of one of the specific nozzles that were carefully examined. This position may be a deviation from one of the expected positions of such a nozzle, which may be evidence of the curvature of the nozzle row.

As previously described and as contemplated by this discussion, an embodiment establishes a statistical distribution for each nozzle for each measured parameter, such as for volume, velocity, trajectory, nozzle Bending, and potentially other parameters (868). As part of these statistical processes, individual measurements may not be considered or utilized to identify errors. The following In some instances, if a small drop measurement is obtained with a value that has been removed from other measurements of the same nozzle so far, the measurement may represent a launch error; in one embodiment, if an abnormality is reached A point that exceeds a statistical error parameter, the system discards this measurement. If no small droplets are seen at all, this may be that the small droplet measurement system is at the wrong nozzle (wrong position), or has an emission waveform error, or a carefully inspected nozzle is Invalid evidence. The measurement error handler 869 is employed to make appropriate adjustments, including any new or additional measurements, as necessary. According to component symbol 870, each measurement is advantageously stored and utilized to establish an appropriate statistical distribution, wherein the system then performs a loop to perform measurements for additional small droplets from the same nozzle until obtained The full robustness of the measurement error is up to now. This loop (871) is seen in Figure 8C to indicate execution until n small droplets are obtained for each nozzle or for each nozzle-waveform pairing. When a sufficiently robust distribution has been obtained, the system calculates (stores) and specifies the desired statistical parameters (eg, the mean, standard distribution for each measurement parameter) to a given nozzle (872) and executes Any appropriate error handler 873 (eg, verifying the nozzle just measured, or treating the nozzle or an associated waveform as invalid), and then moving to the next transmit waveform or next as appropriate Nozzle (874). In other words, the system software can identify the next nozzle to be measured (step 874) after a measurement distribution for a given nozzle or nozzle-waveform pairing has been completed. , for example by updating the address index) after And then according to the symbol 876, as the case may be, to move the droplet measurement system and perform the next measurement. Alternatively, depending on the symbol 875, if the time has elapsed and the system is being called to print another substrate as part of a manufacturing line, the system is updated based on the newly acquired data (if any). Any scanning operation stores the address of the "next" nozzle and returns to substrate printing (875). In an embodiment, the printing operation is completed Thereafter, during an expected interruption (or maintenance downtime), the system retrieves the stored nozzle address and details of the droplet measurement and continues from where it left off.

It is noted that although not individually emphasized by Figure 8C, the depicted measurement procedure will typically be performed for waveforms that can be utilized in each of the nozzles. For example, if each nozzle has four different piezoelectric drive waveforms that can be selected, the internal program loop 871 of Figure 8C will be substantially repeated 4*n times; if a particular implementation requires A statistical distribution of waveforms is based on 24 small droplets, and there may be 96 such measurements for a nozzle (24 for each of the four waveforms), each of which is It is used to develop statistical averages and spread measurements for each of the droplet velocity, trajectory, and amount, as well as for the estimated nozzle position (eg, for the purpose of evaluating nozzle bending).

Figure 8D provides a flow chart for nozzle verification. In one embodiment, a small droplet measurement system is performed for each nozzle and for each waveform applied to any given nozzle, for any of the small droplet volume, velocity, and trajectory and/or Or each to generate a statistical model (for example, distribution and average). Thus, for example, if there are two waveform choices for each of the 12 nozzles, there will be up to 24 waveform-nozzle combinations or pairings; in one embodiment, for each parameter (eg, amount) measurement This is done for each nozzle or waveform-nozzle pair that is sufficient to develop a robust statistical model. It is noted that, despite such a plan, it is conceptually possible that a given nozzle or nozzle-waveform pairing may result in an exceptionally wide distribution that should be specially treated, or an average of sufficient anomalies. This particular processing applied in an embodiment is conceptually represented by Figure 8D.

More specifically, a general method is indicated by the symbol 881. The data generated by the small droplet measuring device is stored in the memory 885 for After use. During the application of method 881, this data is recalled from memory and data for each nozzle or nozzle-waveform pairing is extracted and processed individually (883). In an embodiment, a normal random distribution is established for each variable to be verified to be described by an average value, a standard deviation, and a number of small droplets (n) to be measured, Or use the same measurement. It is again noted that other distribution formats (eg, Student's T, Poisson, etc.) may also be utilized. The measured parameters are compared to one or more ranges (887) to determine if the associated droplets are actually usable. In one embodiment, at least one range is applied to cause the droplet to fail for use (eg, if the droplet has a quantity that is sufficiently large or small relative to the desired target, the nozzle or nozzle - Waveform pairing can be excluded from short-term use). An example is provided below, if a small droplet of 10.00 pL is desired, then a nozzle or nozzle-waveform associated with a small droplet is indicative of a deviation from this target by, for example, 1.5% (eg, &lt;9.85 pL or &gt;). 10.15pL) can be excluded from use. Range, standard deviation, variation, or another distribution measurement can also be used, or alternatively used. For example, if a small droplet statistical model with a narrow distribution (eg, 3σ &lt; 1.005% of the average) is desired, small droplets with measurements that do not meet this criteria can be excluded. It is also possible to use a set of sophisticated/complex criteria that take into account multiple factors. For example, an average of anomalies in combination with a very narrow distribution may be feasible, for example, if the distribution of the measured average (eg, anomalous) μ (eg, 3σ) is within 1.005%, Then a related small droplet can be utilized. For example, if it is desirable to use a small droplet with a 3σ amount within 10.00 pL ± 0.1 pL, then a nozzle-waveform pairing with a mean value of 9.96 pL with a 3σ value of ±0.8 pL may be excluded. However, one nozzle-waveform pairing that produces a 9.93 pL average with a 3σ value of ±0.3 pL may be acceptable. It is obvious that many possibilities are feasible according to any desired rejection/abnormal criteria (889). Note that This same type of processing can be applied to the flight angle and speed of each droplet, that is, it is expected that the flight angle and velocity of each nozzle-waveform pair will exhibit a statistical distribution, and based on measurements and The statistical model derived from the small droplet measuring device, some small droplets may be excluded. For example, a small droplet with an average velocity or flight trajectory outside of 5% of the normal value, or a variation in velocity that is outside a specific target may be hypothetically excluded. use. Different ranges and/or evaluated criteria can be applied to each of the small droplet parameters measured and provided by the reservoir 885.

It is noted that small droplets (and nozzle-waveform combinations) can be processed and/or disposed of in different ways according to the criteria 889 for rejection/abnormality. For example, as mentioned, a particular droplet that does not meet a desired norm may be rejected (891). Alternatively, it may be feasible to selectively perform additional measurements for repeated operations of the next measurement of a particular nozzle-waveform pair; for example, if a statistical distribution is too wide, then for that particular nozzle - Waveform pairing is particularly feasible to perform additional measurements to improve the tightness of a statistical distribution through additional measurements (eg, variation and standard deviation are based on the number of measured data points). It is also possible to adjust a nozzle drive waveform according to component symbol 893, for example, using a higher or lower voltage level (eg, to provide a larger or smaller speed or a more consistent flight angle), or A waveform is reshaped to facilitate the generation of an adjusted nozzle-waveform pairing that meets the specified norm. Depending on the component symbol 894, the timing of the waveform can also be adjusted (e.g., to compensate for the average velocity of the anomaly associated with a particular nozzle-waveform pairing). As an example (mentioned earlier), a slow droplet can be emitted at an earlier time relative to other nozzles, and a fast droplet can be emitted later in time to compensate for Fast flight time. Many of these alternatives are feasible. Finally, according to component symbol 895, any adjusted parameters (eg, emission time, waveform voltage level or shape) may be stored, and optionally, if desired, the adjusted parameters may be applied to re-measure one or more associated small droplets . After each nozzle-waveform pairing (modified or otherwise) is verified (passed or rejected), the method proceeds to the next nozzle-waveform pairing according to symbol 897. As such, the particular droplet detail is advantageously taken into account in the emission instructions for exporting the printed grid to ensure uniformity (at least locally) over any of the converted deposition parameters. In an embodiment, the nozzle details are weighted into a conversion calculation for each grid point. In another embodiment, the emission decision of the printed grid points is made in a weighted manner based on the merged pattern image overlays (as discussed above), wherein a second program is used for picking, redistributing Or to adjust the emission decision made for the grid points corresponding to an abnormal droplet or nozzle; in other words, the nozzle emission decision can be made in a first conversion procedure, and then a second error correction procedure can be Applied to take into account the details of the nozzle or droplet. Other alternatives are also possible.

Through the use of precision mechanical systems and alignment techniques for small droplet measurement systems, the disclosed method allows for very high accuracy measurements of individual nozzle features, including for the mentioned parameters (eg, The average droplet size metric for each of the amount, velocity, trajectory, nozzle position, droplet drop position, nozzle bend, and other parameters. As should be appreciated, the techniques mentioned make it easy to achieve a high degree of uniformity in the process of the OLED device process in particular, and thus enhance the reliability in the finished product. By providing control efficiency, particularly with regard to the speed of small droplet measurements, the stacking of such measurements relative to other system processes, and the inclusion of alignment error correction procedures, the teachings presented above assist in providing a faster and more An inexpensive process that is designed to provide flexibility and precision in the process.

Figures 9A-9C are used to depict the execution of an example for a product array The printing process of the stage is similarly exemplified by one or more flat plates on a substrate.

Figure 9A depicts a substrate 901 in which some of the dashed squares represent individual panel products. A such product as seen at the bottom left of the figure is indicated by the symbol 902. Each of the substrates (in a series of substrates) in an embodiment has some alignment marks, such as represented by element symbol 903. In one embodiment, two such indicia 903 are used for the entirety of the substrate, which enables adjustment of substrate position offset, rotation error, and scale error, while in another embodiment, three or More such markers 903 are used to make adjustments to skew errors easier. In yet another embodiment, each panel (eg, any of the four depicted panels) is provided with an alignment mark for each panel, such as indicia 905. These alignment marks can be utilized to perform independent error processing for each panel or each product; likewise, a sufficient number or density of such indicia (eg, three or more per panel or other product) The system is capable of compensating for nonlinear errors. These indicia may be additional to, or in lieu of, substrate reference 903 relative to substrate reference 903. In yet another embodiment, the alignment marks are replicated at regular intervals (e.g., as represented by ellipse 909) regardless of panel position. Regardless of which design is used, one or more cameras 906 are used to image the alignment marks to facilitate detection of the errors just cited. In a contemplated embodiment, a single immovable camera is used and the transport mechanism of the printer (eg, a robotic and/or airborne airfloating mechanism) moves the substrate to be in the single camera. Each alignment mark is sequentially positioned in the field of view; in a different embodiment, the camera is mounted on a motion system for transmission relative to the substrate. In still another embodiment, as discussed below, low magnification and high magnification images are taken out, the low magnification image is used to roughly locate a reference for high resolution magnification, and the high magnification image It is used to identify an accurate reference position based on a press coordinate system. think Earlier discussion, in one embodiment, the printing mechanism of the printing press controls movement to within about one micron of the desired position; the system is thus capable of accurately tracking the position with software and can calculate the correlation Errors in press-based or substrate-based coordinate systems.

In a typical embodiment, printing will be performed to deposit a given layer of material on the entire substrate at a time (i.e., where a single printing process provides a layer for multiple products). To illustrate this, Figure 9A shows two exemplary scans 907 and 908 of a printhead along the long axis of the substrate; in a split-axis printer, the substrate is typically moved back and forth (e.g., at In the direction of the depicted arrow, wherein the printer advances the printhead in position between scans (i.e., in the vertical direction associated with the page). It is noted that although the scan path is depicted as being linear, this is not required in any embodiment. Moreover, although the scan paths (eg, 907 and 908) are depicted as being adjacent and mutually exclusive in terms of the area covered, this is not necessary in any embodiment (eg, the print head is deemed necessary) Instead, it can be applied in sections with respect to a printing width. Finally, it is also noted that any given scan path is typically in a single pass by printing a layer for multiple products over the entire printable length of the substrate. Each pass is a nozzle firing decision based on the printed image (distorted or corrected), and each firing decision applies a selected stylized waveform to produce a desired amount of droplets, trajectories, and speed. Advantageously, the processor on the plate of the printer (as previously described) performs measurement and verification of both nozzles/small droplets and updates of parameters, details of each substrate or each panel The detection, the same version of the correction, and the stylization of the nozzle emission data (see, for example, the "data" from Figure 5B and the "drive waveform ID"), which will be related to the binary nozzle emission decision (see for example The "trigger (emission)" signal from Figure 5B is used. Once the printing is completed, the substrate and the wet ink (ie, the deposited liquid) The liquid that can then be transported for the deposition is cured or processed into a permanent layer. For example, briefly returning to the discussion of FIG. 6B, a substrate can be "ink" applied in a printing module 625 and then transported to a curing chamber 641, all without interrupting the controlled atmosphere ( That is, the controlled atmosphere is advantageously used to prevent contamination by moisture, oxygen or particulates.

Figure 9B depicts an example of an alignment and detection procedure 911 for a manufacturing operation that is again fabricated using a tablet device. It is noted that many alternative programs are possible, and Figure 9B is merely an example. The process uses a three-part method of operation in which the new substrate is first mechanically substantially aligned, wherein the alignment marks are then optically measured, and finally the software program corrects subtle (virtual) pairs. quasi. When a new substrate is loaded, a robotic arm first places the substrate over the top lift pins of the printer, which is used to advance the substrate to a vacuum holder. The holder moves the substrate while the substrate is supported by an air floating platform, and the substrate is advanced to a position in which it is substantially aligned and is in an appropriate position for optical measurement (also That is, to find the benchmark). It is noted that the system is programmed for coarse substrate detail such that a camera control system (or other imaging device) is activated and/or activated as necessary to image an expected baseline therein. Area). First, a low magnification image is captured and the substrate is reset, after which a higher magnification image is used to accurately detect the reference position. During any stage of optical metrology, a search procedure (eg, spiral) is used to find each benchmark as appropriate. The system then proceeds to measure a second reference (and/or a higher order reference as the case may be), such as to assist in identifying the precise position, orientation, skew and/or scale of the substrate, as appropriate; As pointed out earlier, this procedure can also be applied for each panel, or indeed for each of all benchmarks, in order to obtain a position/distortion that is appropriate for the resolution of the associated error correction procedure. News. The detected reference parameters (eg, position, size, shape, rotation, skew, or associated reference point) can be compared to the expected parameters of such a reference to detect errors. As noted, any detected error may involve one or more of translation, rotation, scaling, or skewing errors, or an overlap of multiple such errors. Based on the detected errors, the press control data is created, which corrects for any errors so that the print will be in the "right place." As indicated on the right side of the figure, the system soft system corrects the error by computing on the template based on the determination of the alignment mark. As indicated earlier, the template is captured, and the example of this template is then manipulated (calculated) via the aforementioned error correction procedure (or another equivalent procedure) as appropriate. The situation produces a nozzle firing decision. The printing department is then executed.

Figure 9C provides a diagram that is used to illustrate one of the substrate error corrections with software. In an embodiment generally indicated by the symbol 921, a data representative of a layer to be printed, for example according to a substrate or panel prescription or both, is first retrieved (923). As indicated by component symbol 925, this information is typically maintained in the same version, where a backup or "instance" of the material is captured, adapted for error, transmitted to the printer, and then Discard (ie, avoid contamination of data from earlier substrate processing during the next substrate process). These processes are shown in various forms in Fig. 9C.

In the case where a backup of the template is available, the system detects the geometry of the substrate according to component symbol 927; as indicated by component symbols 928-932, the detection procedure can be performed once (eg, at Executing intermittently before printing, or when a new substrate is loaded (eg, the printing process can be interrupted, or the reference capture can be for each of the plurality of subdivided portions of the substrate, for example It is generated for each panel, it can be repeated, or it can be executed continuously. As explicitly indicated by the symbol 931, in an embodiment A number of different reference systems are used to allow detection and correction of different types of linear and non-linear errors. According to component symbol 935, once the error has been detected, the system then calculates a transition. For example, in one embodiment, the error is linear across the entire substrate, such that a simple linear equation is derived and used to calculate the cached stored pattern to facilitate the production of printer control data. In other embodiments, the error may be modeled by a quadratic or other polynomial in a discontinuous (e.g., region dependent) manner, or in some other manner. In a parallel processing embodiment, a master or supervised processor makes this decision and then assigns processing to the discrete core or processor. Based on component symbol 937, the system then proceeds to find the appropriate data from the captured instance of the template to modify, and/or use to specify nozzle firing decisions to mitigate errors. According to the symbol 938-941, and as previously discussed, in an embodiment in which the pattern is in the form of a bitmap having nozzle emission data, the system can adapt the error based on the pattern for one A "closest" pixel of the error vector position, a weighted measurement of one of the plurality of pixels from the template, or in some other way. According to the symbol 940, an affine transformation can be applied, wherein a program performs substantially matrix operations on a "tile" of printed grid points in order to obtain new emission decisions; for example, such conversions The pattern data can be weighted by offset, rotation, and other factors to obtain data for the "converted space" corresponding to the actual substrate position (and orientation, skew, etc.). Other programs can also be executed according to the component symbol 941. Under the nozzle designation, the system can then optionally invoke the post-positioning compensation process to correct for errors in the fill area or ink density for discrete areas of the substrate. The options and types of processing are represented in Figure 9C (these will also be discussed generally below in relation to Figures 10A-E). For example, component symbol 951 depicts a fluid well that will hold one of the light-generating components of an OLED display panel, which is printed grid 953 Covered, and one of the specific nodes or droplet positions is indicated by the symbol 954. If the positional error causes too much or too little droplets within the boundaries of the well 951, an anti-aliasing procedure can be applied to check the number of small droplets that will fall within the well; for example, given With the small droplet density represented by the template, it is possible that the rotation of the printed grid changes the number of droplets that will fall within the well - the anti-aliasing program detects this problem And the nozzle specification is modified (945) from the position compensation program so that the correct number and/or amount of small droplets will be deposited within the well. As indicated on the right side of the figure, via the symbol 946-948, droplet detail and/or nozzle verification can also be considered in this procedure. For example, if a small droplet for node 954 has stored a small droplet parameter that indicates an error in position (eg, the expected emission trajectory has an error that would cause the small droplet to fall outside of the depicted fluid well) , β), then the problem can also be corrected by software (945), which checks the details of other "nodes" that fall within the well to determine which ones should cause the launch decision to be changed to facilitate compliance. The required fill in parameters. Similarly, if a verification procedure is used and a launch decision is assigned to an invalid (or unqualified) nozzle by the position compensation procedure, the problem can be detected by the software and Correction (945). As noted earlier, in yet another embodiment, the droplet parameters can be measured at the original location and periodically updated to provide a robust current data set that takes into account the changing conditions. In general, step 945 corrects the fill and sawtooth errors to ensure that the liquid transfer will conform to the details required for deposition. If an error is detected or cannot be resolved according to the program 955, an exception handling routine (957) is called, which is usually to adjust or recalculate the scan path in order to resolve the problematic error, or Recalculate the data. If the error has then been completely mitigated, then according to component symbol 959, the output data is then stored and/or transmitted to the printer and printhead as appropriate. It is noted that, in an embodiment, The system soft system further detects the associated repeated errors between the substrates, and then can optionally update the cached stored pattern (ie, according to 961) to "know" the repeatable error And the template is adapted to the error. For example, it is possible that the error in the position in a given system is caused by an unintentional protrusion in a system edge guide for the substrate, which causes a substrate to be in motion. A particular position is either slightly "tilted" relative to a particular position; these and other repeatable errors are known to the system and are used to reduce the error handling for each job. This technique will be further discussed below in relation to Figure 10E.

10A-10E are used to illustrate different types of error compensation techniques that can be applied, that is, can be applied in a variety of nozzle emission decisions, at a particular node for the print grid. The adjustment of the droplet detail is and/or as a compensation procedure after positioning.

More specifically, FIG. 10A shows a flow chart 1001 relating to droplet designation and/or droplet adjustment in a region; for example, such techniques can be applied locally to halftones (ie, To deposit a fluid that will spread out to provide a blanket overlay at a calibrated thickness, or to apply a total fill within a fluid well (eg, as discussed above in relation to Figure 9C) ). According to component symbol 1003, the method first identifies the area or well being considered, and the desired number of droplets or the desired amount of fill. Based on the component symbol 1005, the system then retrieves statistical averages and variations of the droplet parameters for each nozzle and/or nozzle drive waveform; this data provides the expected droplet parameters for a desired degree of confidence. Understanding. It is noted that, as represented by the parameters v, a and β, in one embodiment, the system draws from the memory a representative amount of small droplets and an expected two-dimensional droplet drop position. Reference Number, and variations of these parameters. The system software then proceeds to simulate the fill to determine whether the expected ink density (or the aggregated amount) meets a predetermined threshold; in one embodiment, the system simply determines that the predetermined threshold is met. a combination of qualified droplets, and which in turn selects one of these qualifying combinations (i.e., in an optimization while accomplishing this for multiple regions over a scan width traversed by the printhead) The way of ability). In another embodiment, the system simply selects some small droplets, and if the combination would be expected to produce a result outside of the predetermined threshold, then the combination of droplets is adjusted. In accordance with the component symbols 1006, 1007, and 1008, in one embodiment, the system can select a different drive waveform for a given nozzle (eg, one of the sixteen pre-programmed waveforms described earlier). ), or it can choose a different nozzle, or check to see if a particular nozzle is available. According to the component symbol 1009, the system adjusts the nozzle designation (which includes selection information of the drive waveform) as necessary, and the printing process is then carefully checked again for errors (1011, such as preset scan paths and scans). The appropriateness of the number). If there is an error, the system can adjust the offset of the print head for a given scan path (1012) and/or the number of adjustment scans (1013). If there is no error, the method ends (1014) where the specified data is output to the printer as previously discussed.

Figure 10B provides another flow chart 1021 for pattern adjustment, this time as an error corrected step. Based on the component symbol 1023, for each of the N regions or fluid wells, the system soft system performs, for example, identifying the well location based on the detected error data. At step 1025, the soft system examines the emission decisions for the printed mesh nodes that fall within the region of interest, and takes an average (μ) and standard deviation or other distribution for each such parameter. Measure (for example, σ) (1026). Simulation of the total amount or quantity distribution used in the area (1027) Thereafter, based on component symbols 1029 and 1031, the system then determines whether the amount or distribution meets a predetermined threshold, such as a minimum threshold (Th1) and a maximum threshold (Th2). If the amount or distribution does not meet the predetermined thresholds, as indicated by the component symbols 1033, 1035, and 1036, the system soft system adjusts the droplets and/or waveform designation until the amount or The distribution does match the preset thresholds. In an embodiment, since a plurality of waveforms are available for each nozzle and have been previously selected to produce a carefully considered change in the amount of droplets, the error can be simply by specifying a new emission waveform to the Specific print grid nodes are compensated without changing the scan path; if the waveform is different from another shot decision for the same print head nozzle, then the system decides to schedule between nozzle shots to select one Different "preset" waveforms are used for this particular nozzle (i.e., see the "trigger (emission)" signal depicted in Figure 5B). Based on component symbol 1037, the system then stores the associated nozzle data as appropriate. As indicated by the optional program block 1039, if no appropriate waveform has been programmed for a particular nozzle, in one embodiment, the system can add a waveform, or Stylize a selected waveform to try to improve the details of the emission that can be obtained from a given nozzle.

Figure 10C shows a flow chart 1041 in which a nozzle/waveform data is taken into account in an example of the pattern to account for substrate or panel errors. More specifically, as indicated by the component symbols 1043-1047, the system can retrieve nozzle drive details corresponding to a one-dimensional map (e.g., a pattern having a bitmap pattern). For example, again utilizing the emission designation of one of the new printed mesh nodes is based on an example of a weighted average of four printed pixels corresponding to the same version, if the four designated nozzles are expected to produce 10.00, respectively. 10.50, 10.00 and 10.20pL droplets, this information can be considered and used to select a nozzle Or a nozzle waveform to deposit small droplets onto a misaligned substrate. Returning to the example of fluid well 951 from Figure 9C, any printed mesh node (e.g., 954) that falls within the depicted well and is designated for launch can potentially be used to obtain a fluid well for use. The amount is aggregated, and the system can effectively weight the four printed pixels in this hypothesis and select a new grid point (eg, nozzle or launch time) and associated drive waveform (if multiple waveforms are available) Utilizing) to generate a small droplet having an expected amount parameter corresponding to a weighted combination of the four printed pixels. According to the component symbol 1049, the software can apply any smoothing as needed (for example, adjustment of adjacent printed pixels as necessary, or subsequent droplet selection), and also output data (including one) The selected drive waveform is written to the memory; this output data will then eventually be transferred to the press to control printing. According to the component symbol 1053, the system can assume a fixed scan mode (i.e., and specify the nozzle and drive waveform in a manner that merely corrects the error), or in another embodiment, it can be re-examined and re-most Optimized scanning (rasterization), for example by changing the scan path, print head offset, number of scans, or other details, all with a view to minimizing printing time; these optional features are by component symbols 1055 and 1057 are indicated.

Figure 10D shows yet another flow chart 1061 relating to halftone (i.e., drop density) adjustment. With respect to this embodiment, it should be assumed that it is desirable to control the ink droplet density, for example, using deposited ink to provide blanket coverage over an area of the substrate, but such that the ink droplets are administered one One of the desired layer thicknesses is deposited (i.e., given the limited diffusion characteristics of each droplet). Such a procedure is especially useful for creating other layers of the barrier layer, encapsulation layer, or layer thickness over a substantial area. According to the component symbol 1063, it is assumed that some parts of the material from the template are to be calculated. The figure compensates for errors in the substrate and/or panel position. According to the component symbols 1067, 1069 and 1071, the soft system in this embodiment draws the expected droplet size, position and variation as well as the desired layer parameters (e.g., thickness and halftone density required to produce the desired thickness). Or ink density). The system then proceeds from an available population (eg, a nozzle that will pass over the particular area under a particular scan path) to be evaluated to facilitate representation by the original pattern data. The same ink density is used to select the nozzle and drive the waveform. For example, if it is desired to create a five micron thick encapsulation layer, then the system (a) identifies the desired ink density that will pass through the nozzles above each region, producing the desired thickness, each Nozzle and droplet detail for each waveform, (b) applying a mathematical function to select the nozzle and drive waveform that will produce the desired density (i.e., it will then also alternate to nozzle selection for adjacent regions) And it then outputs the press control data. It is noted that in planning halftone or density patterns, the system software typically processes the process deposits to deposit a uniform fluid density, for example, if a given nozzle detail is not consistent. (eg, 10.00 pL), in the case of small droplet distribution, "light" droplets (eg, 9.00 pL) are utilized to balance "heavy" droplets (eg, 11.00 pL). Similarly, according to component symbol 1073, the output details are again carefully checked for errors (eg, a scan path does not produce the desired result), and if no errors are found, the data is output to the printer (1075). Or, conversely, if an error is found, an exception procedure is called (1077, for example to re-evaluate rasterization). Other alternatives will be considered by those skilled in the art.

Finally, FIG. 10E provides a flowchart 1081 for a template that is stored for updates such as repeatable errors as described above in relation to FIG. 9C. For each new substrate, the method depicted compares the deviation from the expected location (1083) and has been stored The past error in the memory according to the component symbol 1085. Based on component symbol 1087, the system software attempts to detect correlations in the error (eg, relative to unique substrate errors). Depending on the component symbols 1088-1090, the correlation sought may be based on some calibration (ie, to set an expected substrate norm), continuous monitoring of the reference for successive substrates, or based on navigation Bit speculation. According to the component symbols 1091 and 1093, the system effectively "learns" the pattern in the error, for example, according to a regression software, a neural network, or other adaptive program, and then modifies it as appropriate. This version and update the history of the deviation data. Finally, according to component symbol 1095, the soft system corrects any uncorrected errors (e.g., unique errors) of the current substrate and outputs the data to the printer as previously described.

As indicated earlier, it is generally desirable to perform printing quickly in order to minimize processing time and increase throughput; in one embodiment, a substrate having an order of two meters wide and long can have one at each The layer is deposited uniformly over its surface in less than 90 seconds; in another embodiment, this time is 45 seconds or less. Therefore, it is advantageous in such an application that the calculation and error reduction of each substrate (or each product) is performed as quickly as possible.

Figures 11A-11B are used to introduce a parallel processing architecture to make this processing speed easy. As should be apparent, under precise manufacturing (for example, multiple TV screens printed one layer at a time, each TV screen has millions of pixels), a detailed print grid software adjustment to produce press control data may It will take a few seconds. The architecture and procedures presented in Figures 11A-11B are used to reduce this time, ideally to two seconds or less.

More specifically, as shown in FIG. 11A, a cached stored print image, stored prescription material, or other suitable source material representing an ideal print is used. A backup system is first captured. A master or supervised processor 1103 will have been made aware of the configuration of any of the substrates, and thus will have an understanding of the number of products (panels) and their individual expected locations. The supervised processor 1103 receives the alignment data and calculates an error (e.g., independently for each panel or product). Panel processing or error compensation at a sub-panel level is optional, i.e., the disclosed techniques can even be applied to situations where an instance of the printed image is uniformly linear throughout a particular substrate. The system also includes a set of parallel processors 1105, in one embodiment a multi-core processor or a graphics processing unit (GPU). A GPU typically contains a large number of cores 1107 (e.g., hundreds to thousands), each core (or multiple cores therein) being utilized in this embodiment to perform parallel processing. The supervised processor 1103 is responsive to the definition of the panel, the desired error correction, and other factors to make a decision as to how many parallel threads or programs should be used to perform the conversion of the instance of the pattern print image. With respect to this procedure, the supervised processor subdivides the overall printed area for the substrate as appropriate, assigns any appropriate conversion parameters to the appropriate core or processor, and stores (caches) the pattern print An example of an image in an embedded DRAM 1109. It is noted that the embedded DRAM advantageously comprises a plurality of banks, banks or arrays 1111 each having a register 1111A for one or more of the cores 1107 for error correction. And the calculation of the control data of the printing press (which includes the conversion and coverage of any source data). The embedded memory system is advantageously constructed for parallel access such that access by one core or processor does not limit the bandwidth of another core or processor. For this effect, the architecture may be characterized by an ultra-wide access path 1113 (eg, characterized by hundreds of data access lines, as the case may be) such that each core can access memory in parallel Body 1109. It is noted that, in an embodiment, the manner in which the source material is stored is determined by the processor or core designation such that each core or The processor has access to the data it needs for conversion and to a write area for the converted press control data. In a contemplated embodiment, the supervised processor 1103 does not necessarily use each available core or processor, but rather specifies a core or processor to process discrete products (eg, panels) or A panel or a mutually exclusive part of a product. For example, if a hypothetical panel has 15 panels and the system contains 31 parallel processing units, the supervised processor can specify 2 parallel processing units, each unit handling half of a panel. The supervised processor will, depending on the situation, direct the portion of the pattern to the embedded DRAM for storage, and will write any conversion parameters (eg, skew algorithm) to each memory bank or The array is in the appropriate register 1111A (i.e., for each core or parallel processor) and will then wait for a signal from each core or processor whose processing has completed. The output is then a set of converted substrate-based printer control data that is warped to adapt the print to any substrate-based or product-based change to accurately set the layer of material in question to be deposited. . It is noted that, as represented by the component symbols 1115 and 1117, the supervised processor and each of the individual cores or processors are each stored by instructions stored on non-transitory machine readable media. Managed, it controls the individual supervised processors, cores or processors to perform specific processing functions as already described earlier. Furthermore, it is noted that the embedded DRAM can be designed to provide a form of direct memory access (DMA) such that the calculated data can be accessed by a processor core (when available) It is written to the memory and is unloaded by a separate device (i.e., when the processor core performs an error to adjust other data to correct the position).

Figure 11A shows some additional implementation options on the right side of the figure. First, as noted earlier, each panel can be assigned to one or more individual cores or processors. That is, as represented by the component symbol 1119. As in a typical process, the sequence of substrates will use the same prescription or pattern, and the configuration represented by option 1119 is typically a fixed cost (eg, the supervised processor 1103 typically does not need to be targeted The individual substrates recalculate the designation or change the data storage parameters. Second, according to the component symbol 1120, given the fixed designation, the supervised processor can optionally specify individual conversions to each core; The examples generally discuss offset, rotation, scale, and skew correction separately, but in a typical implementation, the transitions can be complex, depending on the situation, involving any combination of these corrections. The device calculates the appropriate conversion and then specifies the calculated conversion to the appropriate processor or core, which then performs its conversion on all of the affected template data that has been designated. As represented by component symbol 1120, In an embodiment, each conversion is provided to a different processor or core. According to component symbol 1123, in an embodiment, The processed sample material or the stored material may be striped. In other words, it is pointed out earlier that a conversion may involve a (converted) printed image or a representative panel printing. The weighting of the emission decisions of the four overlapping pixels on the image data. According to the options represented by the component symbol 1123, the printed image pixel data can be stored in a manner that facilitates this processing, for example by storing the northwest, Pixel data in the northeast, southwest, and southeast are stored in the same memory array, in individual memory arrays (but at the same address), or in other ways. There are many possible memory storage techniques. It can be utilized to speed up the retrieval and processing of data (eg, as is conventionally done for graphics and game programs to speed up the calculation of images). The same type of memory program can be utilized in this embodiment. As also indicated by element symbol 1123, in one embodiment, a variable (per memory array) line offset can be utilized, for example, if printed Pixel data is stored in a given column of, the The system automatically and quickly extracts the data needed for the next set of processing. For example, the cached stored image data can be stored in the DRAM (1111) in a line manner (eg, in a column of each of the two memory arrays), wherein the line offset system Incrementally used to capture relevant printed pixel information for two memory arrays. Many alternatives and variations of this process are also possible.

It is noted that the configuration just described is not the only feasible. For example, instead of specifying the layout of the substrate to each core, the supervised processor or other master 1103 can separate the conversion operations, specifying different programs to the individual cores (1121). In an embodiment, a core may be designated to perform work related to an affine transformation, and a different core may be designated to perform another work. As indicated by the component symbol 1122, almost any task assignment can be achieved between the multiple cores, whether parallel or sequential; due to the detection of errors in each substrate or panel and It may be desirable to print quickly so as to maximize the throughput of manufacturing, so any efficiency in acceleration can potentially be applied to multiple available processors or cores as long as it achieves this goal.

Figure 11B is a flow chart 1151 showing a process associated with an execution phase. In this example, it will be assumed that once the error has been measured, the plurality of cores are each assigned to apply an individual layout of one or more affine transformations to the overall printed area (i.e., the substrate). As indicated by component symbol 1153, the pattern data is first loaded or cached for storage. The supervised processor utilizes such source material (1155) as appropriate to identify the associated alignment mark or reference (1157), and it then directs the printer's camera system to the appropriate coarse coordinates. The printer's camera system transmits accurate measurement information, and the supervised processor measures the source information 1155 obtained. The supervised processor then considers the detected cloth The conversion calculation (1159) of the panel and/or substrate is performed as the case may be, and then the processing range and conversion parameters are assigned to each core, processor or parallel processing thread (1161) as the case may be. And 1163). For a repeating substrate prescription, the supervised processor can automatically analyze the subset of the received template images to a memory dedicated to a particular core, processor or thread. Each core, processor or thread then begins to convert (1165) its specified material to convert or distort the pattern in such a way that the printed details conform to the detected substrate or panel layout. As part of this process, given the details of any print head, droplets or nozzles, reasonably newer print nozzles and/or droplet data are available (1167) to ensure completion. The converted press control data will provide accurate deposition. It is noted that in other embodiments, this adjustment (ie, depending on the details of the print nozzle) may be by a dedicated core, processor or thread (ie, with the execution of the print grid operation) The core, processor or thread is separate) or can be executed by the supervised processor. Once any conversion consistent with the particular characteristics of the nozzle/small droplet has been performed, the converted printer control data can be unloaded from the memory, rasterized (1169) and appropriate instructions for printing Together they are transferred to the printer (1171).

Considering the various techniques and considerations described above, a process can be performed to mass produce a product quickly and at a low cost per unit. In the manufacture of display devices such as flat panel displays, these techniques enable rapid printing of each panel, with multiple panels being selectively fabricated from a common substrate. By providing fast, repeatable printing techniques (for example, using common inks and printheads between different panels), the letter printing can be substantially improved, for example, by reducing the printing time of each layer to none. A small part of the time that will be required under the above technology, and at the same time all ensure that the filling amount of each target area is Within the specifications. Going back to the example of a large HD TV display, each color component layer used for large substrates (for example, 8.5 generation substrates, which is about 220 cm × 250 cm) can be accurately and reliably in one hundred and eighty seconds or more. Printing is short, or even 90 seconds or less, which represents a substantial process improvement. Improving the efficiency and quality of printing is well prepared in order to significantly reduce the cost of manufacturing large HD television displays, and thus reduce the cost to end consumers. As noted earlier, although display manufacturing (and especially OLED manufacturing) is an application of the techniques described herein, these techniques can be applied to a wide variety of processes, computers, printers, software, manufacturing equipment. And the terminal device, and thus is not limited to the display panel. In particular, it is contemplated that the disclosed technology can be applied to any process in which a printer is used to deposit a plurality of layers of a layer as part of a common printing operation, including but not limited to any microelectronics, micro Optical or "3D printing" applications.

It is noted that the described technology provides a number of options. In one embodiment, the misalignment or distortion of the panels (or each product) can be adjusted in a single array or on a single substrate, one by one. A printer scan path can be programmed with subsequent adjustments/adaptations based on one or more alignment errors such that a scan path across the two panels has different firing commands for each substrate, albeit in common Printed material (for example, the rotation or adjustment of the material used for a panel may vary from print to print). Optionally, this information can be instantly adjusted from a source template (eg, a bitmap representing a binary transmission decision). In other embodiments, the print area and/or scan path may be increased or completely re-planned as the substrate is different, despite having common printer source material. The described techniques can be utilized to fabricate OLED panels, such as 2, 4, 6, or a different number of panels as part of a single printing job. After manufacture, these panels can be separated and Applied to individual products, such as the manufacture of individual HDTV displays or other types of devices. By performing fine alignment (eg, sub-millimeter alignment) with software, the disclosed technique is based on less emphasis on precise mechanical positioning, with each product or each of the following in place. The manner in which the substrate is misaligned or deformed provides more accurate product fabrication and accurate placement and alignment of deposition errors.

The specific terms and symbols are set forth to provide a thorough understanding of the disclosed embodiments. In some instances, these terms and symbols may be used to refer to specific details that are not necessary in the embodiments. The terms "example" and "embodiment" are used to describe an example, rather than a preference or a requirement.

Various modifications and changes may be made to the embodiments described herein without departing from the broader spirit and scope of the disclosure. For example, features or characteristics of any of the embodiments may be applied at least in combination with any other embodiment or in place of the corresponding features or characteristics. Therefore, for example, not all features are shown in the drawings and the drawings, and, The technology should be assumed to be an element that can be utilized as a feature of any other schema or embodiment, or a combination thereof. Accordingly, the specification and drawings are to be regarded as