Method and device for controlling a laser having a threshold current level

FIELD OF THE INVENTION

0001The present invention relates to an image exposure device using a laser; more specifically the present invention is related to controlling the driving current applied to the laser.

BACKGROUND OF THE INVENTION

0002In a conventional image exposure device, as used e.g. in an electrophotographic printer or in an imagesetter, an image is formed on a radiation-sensitive body by exposing the body image-wise by a laser beam from a laser. In a conventional electrophotographic printer, the image on the radiation-sensitive body is a latent image, which is developed into a visible toner image, that is subsequently transferred to an image carrier such as a sheet of paper. In a conventional imagesetter, the radiation-sensitive body usually is the image carrier itself, such as a roll of photographic film; on the film, a latent image is exposed, that is afterwards developed in a processor into a visible image.

0003The laser in the image exposure device may be a semiconductor laser; the laser beam is modulated based on an image signal that corresponds to the image.

0004A semiconductor laser has a threshold current level I_th (expressed e.g. in mA). A laser beam is emitted if the electrical driving current I applied to the laser is larger than the threshold current level I_th. <figref idref="f0001">Fig. 1a</figref> shows two curves, one for a low temperature and one for a high temperature, with corresponding threshold current levels I_thLOW and I_thHIGH; the temperature dependence of I_th is discussed below. Below the threshold current level I_th, the radiation intensity RI (expressed e.g. in mW) of the radiation emitted by the laser is very small, and in fact, the emitted radiation is not coherent, i.e. it is not really laser radiation. Thus, in general only the portions of the curves for I larger than I_th are useful for imaging purposes; therefore the portions of the curves in <figref idref="f0001">Fig. 1a</figref> for I smaller than I_th are drawn as dashed lines. The driving current I that is applied to the laser is modulated based on the image signal, so that the intensity RI of the emitted laser beam is a function of the image signal.

0005The laser threshold current level I_th is not constant, but depends on different factors, such as the ambient temperature, the ageing of the laser, differences between individual lasers. Fluctuations in the threshold current level cause significant variations of the laser beam intensity; as shown in <figref idref="f0001">Fig. 1a</figref>, the threshold current level changes from I_thLOW to I_thHIGH when temperature changes from low to high.

0006In order to obtain good image quality, the intensity RI of the laser beam should depend only on the image signal and should not change with the factors influencing the threshold current level I_th. Therefore, it is customary to control the laser, usually by means of a feedback circuit as known in system control theory. Customarily the required intensity of the laser beam is determined and a signal based on the measured intensity is fed back to the driving circuit of the laser in order to control the laser driving current. In this way, the effect of disturbances, influencing e.g. the laser threshold current level, can be reduced.

0007Two known methods to control the laser are illustrated by means of <figref idref="f0001">Fig. 1a</figref>. Suppose that, at low temperature, an image or a portion of an image must be written at the laser beam intensity corresponding to point B in <figref idref="f0001">Fig. 1a</figref>. In a first method the laser is driven from point O to point B in one step; in a second method the laser is driven from O to B via A in three steps. Both methods are now discussed in detail.

0008In the first method, at the moment the image must be written, the laser starts from zero driving current and zero radiation intensity in point O, and the laser driving current is increased until the driving current corresponding to point B is reached. A disadvantage of this first method is that it takes a long time for the driving current to reach point B, since the operating point of the laser must pass the whole way from O via A to B (see <figref idref="f0001">Fig. 1a</figref>). Because of the long time that is required, the imaging speed is limited.

0009In the second method, the laser is driven from O to B via A in three steps. In the first step, the laser is driven from O to A. Then, in the second step, the laser is kept at point A, until an image, or more precisely, until a non-zero portion of an image has to be written. Finally, in the third step, when a non-zero portion of the image must be written, the laser is driven from A to B. Essential in this method is that, beforehand, the driving current is set to an initial driving current that equals the threshold current level I_thLOW before the image is written. The advantage is that, when an image is to be written, the laser reacts promptly to the change in laser driving current corresponding to a change from point A to point B, thus overcoming the problem of the limited imaging speed of the first method. A disadvantage of this second method is that the laser is kept for some time at operating point A, which means that some radiation is being emitted before the actual image is written. This means that the radiation-sensitive body is exposed by the radiation emitted in A. Therefore, specific countermeasures may have to be taken; in order to avoid that an image is formed by the radiation emitted in A, in an electrophotographic printer for example the electrophotographic parameters such as the cleaning potential may be adjusted (the cleaning potential is discussed e.g. in <patcit id="pcit0001" dnum="EP0788273A"><text>EP-A-0 788 273</text></patcit>). However, these countermeasures have drawbacks (adjusting the cleaning potential, for example, may result in increased carrier loss).

0010Variants of the second method exist: instead of setting the driving current to an initial value that equals the threshold current, the driving current may be set to a value about the threshold current value, or, alternatively, the driving current may be set to a value below the threshold current value.

0011Thus, the problems of the first and the second method discussed above are either the limited imaging speed, or undesired emitted radiation, or both.

0012Patent <patcit id="pcit0002" dnum="US5416504A"><text>US-A-5 416 504</text></patcit> discloses an image exposure device that implements another variant of the second method discussed above (the second method involves driving the laser to B in <figref idref="f0001">Fig. 1a</figref> in three steps: driving the laser from O to A; keeping the laser in A; driving the laser from A to B). In this device, the initial driving current is set about the threshold current level of the semiconductor laser (i.e. the laser is driven from O to about point A in <figref idref="f0001">Fig. 1a</figref>) in two substeps: first the driving current is set to a first value near the threshold current level, using the output of a counter circuit, and then the driving current is adjusted to a second value about the threshold current level by incrementing or decrementing this counter. The counter is incremented or decremented depending on the difference between a reference voltage and a monitor voltage that indicates the actual laser beam intensity and that is obtained as follows: the light from the laser is input to a monitor diode; the monitor diode outputs a current that is converted to the monitor voltage. When the monitor voltage equals the reference voltage, this indicates that the laser beam has reached a predetermined intensity. Finally, in order to write an image, the input image signal is added to the counter output value (the counter output value corresponding to the above second value of the driving current) and the sum is converted into an electrical current for driving the laser. The counter output can only change gradually; the purpose hereof is to stabilise the control of the driving current. The laser beam is monitored only during the time that no image signal is applied to the laser; there is no feedback during exposure of the image. A disadvantage of this device is, as mentioned above, that undesired radiation is emitted which exposes the radiation-sensitive body. Another disadvantage is the high complexity of the circuitry that controls the laser driving current. Yet another disadvantage is that there is no feedback, and hence disturbances - influencing e.g. the laser threshold current level - are not coped with during image exposure.

0013European application <patcit id="pcit0003" dnum="EP0548591A"><text>EP-A-0, 548, 591</text></patcit> discloses an image exposure device that also implements an example of the second method disclosed above.

OBJECTS AND SUMMARY OF THE INVENTION

0014It is an object of the present invention to provide a method for controlling the radiation intensity of a laser beam, that solves the above mentioned problems, by promptly driving the laser to a desired radiation strength and at the same time significantly reducing the emission of undesired radiation.

0015Another object is to provide a method for controlling the radiation intensity of a laser beam, that can be implemented in a simple and cost-effective way.

0016Another object is to provide a method for generating a high quality image.

0017It is another object of the present invention to provide an electrophotographic printer that creates a high quality image on a photoconductive member, such as a drum or a belt, in a fast and cost-effective way.

0018It is still another object of the present invention to provide an imagesetter that creates a high quality image on a radiation-sensitive sheet in a fast and cost-effective way.

0019These objects are realized by the method, the image exposure device, the electrophotographic printer, and the imagesetter having the features recited in the appended independent claims.

0020Specific features for preferred embodiments of the invention are set out in the dependent claims.

0021Some terms used in the claims are defined below.

0022An electrical driving current I "about" the threshold current level I_th fulfils the following relation, wherein RI is the radiation intensity emitted at the driving current I, and RI_max is the maximum value of the radiation intensity RI for the laser that is preferably not exceeded so as not to reduce the laser life: <ul id="ul0001" list-style="none" compact="compact"><li>I > 0.3*I_th and RI < 0.3*RI_max, preferably I > 0.5*I_th and RI < 0.2*RI_max, more preferably I > 0.7*I_th and RI < 0.15*RI_max, most preferably I > 0.8*I_th and RI < 0.1*RI_max.</li></ul>

0023A "substantially zero" electrical driving current is a current for driving a laser that is preferably smaller than 25 % of the threshold current level I_th of the laser, more preferably smaller than 10 % of I_th, most preferably less than 5 % of I_th.

0024A "pixel" is a constituting element of a digital image; a digital image is typically represented by a rectangular matrix of pixels, each having an electronic pixel value. The location of each pixel within the matrix corresponds to a specific location on the image carrier. These locations may be equidistant or not. Each electronic pixel value corresponds to an optical density of the image at the specific location; the electronic pixel values together form the image signal that is used for driving the laser.

0025A "binary pixel" can have two electronic pixel values, customarily represented by 0 and 1; these two values represent a high density and a low density, which may be obtained e.g. by applying toner or no toner, or generating locally dye or no dye, or by keeping or removing silver in a photographic process. In the production of printing plates, 0 may result in an ink repellent zone, whereas 1 may result in an ink accepting zone.

0026A "continuous tone system" is a system wherein multiple density levels may be generated on the image carrier, with no perceptible quantisation to them. In order to achieve such fine quantisation, usually 256 different density levels are required, such that each "continuous tone pixel" may have a value from 0 to 255.

0027A "multilevel system" has a reduced number of density levels. In electrophotography e.g., usually a reduced number of density levels can be generated consistently, e.g. 16 levels; such a system is called a "multilevel system" and the pixels are called "multilevel pixels".

0028A "radiation-sensitive means" comprises a material that is sensitive to electromagnetic radiation. The radiation-sensitive means may be a photoconductive member, such as a photoconductive drum or belt, as used in an electrophotographic apparatus. It may also be a radiation-sensitive sheet, such as a photographic or a thermographic sheet. The electromagnetic radiation may be visible light, but it may also be invisible radiation, such as ultraviolet or infrared radiation.

0029A "radiation-sensitive sheet" may be a cut sheet or a sheet on roll that has a length of e.g. 100 m. The sheet may be made of film, such as poly (ethylene terephtalate) film, it may be made of paper, of (aluminium) plate, or of another radiation-sensitive material as known in the art.

0030In a method in accordance with the present invention, the laser is driven (see <figref idref="f0001">Fig. 1a</figref>) directly from O to B via A in two steps: first the laser is driven from O to A, and then, in a second step immediately after the first step, the laser is driven from A to B. Preferably, the first step is controlled by a first electrical signal and the second step is controlled by a second electrical signal, wherein the second electrical signal is relatively independent from the first electrical signal. The first prior art method discussed above drives the laser from O to B in only one step. An advantage of the two distinct steps in the present invention is that it is possible to take specific measures in the first step so that portion OA of the curve OAB in <figref idref="f0001">Fig. 1a</figref> requires only little time. This provides the advantage that the laser is promptly driven to the desired radiation strength in B. The second prior art method discussed above drives the laser from O to B in three steps while keeping the laser for some time in or near A, until an image has to be written. In the present invention, the driving current is not set to an initial driving current at or near A before the image is written; this significantly reduces the emission of undesired radiation, as discussed below, and is another advantage of the present invention.

0031In the second prior art method, the laser is kept for some time at point A (see <figref idref="f0001">Fig. 1a</figref>), thus emitting undesired radiation; a measure for the time during which undesired radiation is emitted is the time t_keep, defined as follows. Suppose that the image is exposed scan-wise, i.e. as a succession of scan lines (a scan line is shown in <figref idref="f0002">Fig. 2</figref> as the line along arrow B on the surface of drum 15; <figref idref="f0002">Fig. 2</figref> is discussed hereinafter). For each scan line that contains at least two non-zero image portions, the time t_between of the scan line is defined as the maximum time between two successive non-zero image portions in the scan line. The time t_max of an image is defined as the maximum value of the t_between-values of the scan lines of the image (only the scan lines with at least two non-zero image portions). Finally, t_keep is the mean value of the t_max-values of a set of representative images that are to be exposed by the laser. Thus, t_keep as defined above is the maximum time that is typically to be expected between two non-zero image portions; during this time, the laser is kept at point A in the second prior art method and undesired radiation is emitted.

0032Preferably, in an embodiment in accordance with the present invention, the laser driving current is increased to about the threshold current level in a time t_inc that is substantially smaller than t_keep; substantially smaller means that t_inc is smaller than 50 % of t_keep, preferably smaller than 30 % of t_keep, more preferably smaller than 20 % of t_keep, most preferably smaller than 10 % of t_keep.

0033The reduction of emitted undesired radiation is illustrated by <figref idref="f0001">Fig. 1b and Fig. 1c</figref>. In a preferred embodiment of the present invention, discussed hereinafter, the laser driving current I increases in a time t_inc in a substantially linear way from zero to the threshold current level I_th. Suppose that the relation between driving current I and time is linear, and that the relation between radiation intensity RI and driving current I is also linear (i.e. portion OA of the curve in <figref idref="f0001">Fig. 1a</figref> is part of a straight line). Then, the relation between radiation intensity RI and time is also linear, as shown in <figref idref="f0001">Fig. 1b</figref>, and the quantity of emitted undesired radiation, i.e. the (radiant) exposure, is the integral under the curve in <figref idref="f0001">Fig. 1b</figref>, which equals the area 31 of the hatched triangle: <maths id="math0001"><math display="block"><mn mathvariant="normal">0.5</mn><mo mathvariant="normal">*</mo><msub><mi mathvariant="normal">t</mi><mi>inc</mi></msub><mo mathvariant="normal">*</mo><msub><mi>Intensity</mi><mi mathvariant="normal">A</mi></msub></math><img file="EP0989736B1_D0001.tif" /></maths>

0034In the second prior art method, the laser is kept at point A, and t_keep as defined above is a measure for the time during which undesired radiation is emitted. During the time t_keep, a quantity of undesired radiation is emitted at A that equals the area 32 of the hatched rectangle in <figref idref="f0001">Fig. 1c</figref>: <maths id="math0002"><math display="block"><msub><mi mathvariant="normal">t</mi><mi>keep</mi></msub><mo mathvariant="normal">*</mo><msub><mi>Intensity</mi><mi mathvariant="normal">A</mi></msub></math><img file="EP0989736B1_D0002.tif" /></maths>

0035The area 31 is significantly smaller than the area 32, since the area of a triangle is only half of the area of a rectangle that has the same base, and since moreover, as discussed above, t_inc is preferably substantially smaller than t_keep. To get the complete quantity of emitted undesired radiation for the second prior art method, the radiation emitted when passing OA (= the area of a triangle) must still be added to the quantity calculated above (= the area 32 of the rectangle). Also if, in an embodiment in accordance with the present invention, the laser driving current increases in another than a linear way from zero to the threshold current level I_th, e.g. in a parabolic way, the quantity of undesired emitted radiation is significantly smaller than in the second prior art method: the surface 31 of the triangle in <figref idref="f0001">Fig. 1b</figref> then has to be replaced by the surface below e.g. a parabola.

0036The method of the present invention may be implemented in an image exposure device. Preferred embodiments of the method may be implemented in preferred embodiments of the image exposure device, and preferred embodiments of the image exposure device may result in preferred embodiments of the method of the present invention. Some preferred embodiments of method and device are discussed below.

0037In a preferred embodiment in accordance with the present invention, increasing the driving current up to about the threshold current level is achieved by a driving signal "d_fatzero" immediately before the image has to be written. The magnitude and the time duration of "d_fatzero" are determined so that the driving current increases up to about the threshold current level.

0038In another embodiment in accordance with the present invention, the driving current is quickly decreased after an image is written. This may be achieved by applying a signal "d_backbias" as discussed hereinafter.

0039An advantage of a preferred embodiment in accordance with the present invention is that a high quality image can be generated. This may be achieved by keeping the laser driving current well under control, preferably before the image is written (e.g. by means of a signal "d_fatzero", as explained hereinafter), as well as after the image is written (e.g. by means of a signal "d_backbias", as explained hereinafter), and by reducing the emitted undesired radiation.

0040In an embodiment in accordance with the present invention, the exposed image comprises binary pixels. In another embodiment, the image comprises multilevel pixels. In yet another embodiment, the image comprises continuous tone pixels.

0041In a preferred embodiment in accordance with the invention, the feedback circuit is active during exposure. This provides the advantage that the radiation intensity of the laser beam may be adjusted during a scan line (<figref idref="f0002">Fig. 2</figref>, discussed hereinafter, shows a scan line as the line along arrow B on the surface of drum 15).

0042Further advantages and embodiments of the method and of the device in accordance with the present invention will become apparent from the following detailed description and from the drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

0043The invention is described hereinafter by way of example with reference to the accompanying figures, wherein: <ul id="ul0002" list-style="none" compact="compact"><li><figref idref="f0001">Fig. 1a</figref> is a diagram showing the radiation intensity of the laser beam RI as a function of the laser driving current I, and the influence of temperature thereon;</li><li><figref idref="f0001">Fig. 1b</figref> illustrates the exposure of undesired radiation for an embodiment in accordance with the present invention;</li><li><figref idref="f0001">Fig. 1c</figref> illustrates the same for a prior art embodiment;</li><li><figref idref="f0002">Fig. 2</figref> is a schematic illustration of an image exposure device;</li><li><figref idref="f0003">Fig. 3</figref> is a block diagram showing an embodiment of a laser driving circuit in accordance with the present invention;</li><li><figref idref="f0004">Fig. 4a</figref> shows a first image signal;</li><li><figref idref="f0004">Fig. 4b</figref> shows a driving signal and a feedback signal for this first image signal for an embodiment in accordance with the present invention;</li><li><figref idref="f0004">Fig. 4c</figref> shows a driving current for this first image signal for the same embodiment in accordance with the present invention;</li><li><figref idref="f0004">Fig. 5a</figref> shows a second image signal;</li><li><figref idref="f0004">Fig. 5b</figref> shows a driving signal and a feedback signal for this second image signal for the same embodiment in accordance with the present invention;</li><li><figref idref="f0004">Fig. 5c</figref> shows a driving current for this second image signal for the same embodiment in accordance with the present invention;</li><li><figref idref="f0005">Fig. 6b</figref> shows a driving current I;</li><li><figref idref="f0005">Fig. 6a</figref> shows a driving signal and a feedback signal for this driving current I; together <figref idref="f0005">Fig. 6a and Fig. 6b</figref> illustrate how values d_fatzero and d_backbias are determined for a preferred embodiment in accordance with the present invention.</li></ul>

DETAILED DESCRIPTION OF THE INVENTION

0044An embodiment of an image exposure device 50 is shown in <figref idref="f0002">Fig. 2</figref>. A laser beam 11 is emitted by a semiconductor laser 10 and is focused and caused to move by an optical system to scan the surface of a photoconductive drum 15. The optical system, as known in the art, includes a collimator lens, not shown, a polygonal mirror 13 driven by motor 17 and rotating in the direction A and an f-θ lens 14. The laser beam 11 scans the surface of photoconductive drum 15 between the extreme beam positions 11a and 11b and thus exposes a scan line along arrow B on the surface of drum 15. Drum 15 is rotated by motor 18 so that successive scan lines form a latent image 16 on the surface of the drum 15. Monitor diode 12, at the back of semiconductor laser 10, provides a signal f that is proportional to the radiation intensity RI of the radiation emitted by laser 10.

0045An embodiment of a driving circuit for laser 10, in accordance with the present invention, is shown schematically in <figref idref="f0003">Fig. 3</figref>. The embodiment comprises driving means 21, that preferably include feedback means 23 and comparing means 24. The signal f (see <figref idref="f0002">Fig. 2</figref> and <figref idref="f0003">3</figref>) from monitor diode 12 is fed back by feedback means 23 to comparing means 24 where it is compared to a driving signal d. Driving means 21 outputs an electrical driving current I to laser 10 that depends on the difference d - f: preferably, the rate of change of driving current I is proportional to d - f. When d - f < 0, the driving current I will decrease; when d - f > 0, I will increase. Starting from a situation wherein I = 0, which means that the laser does not emit radiation and that thus f = 0, and wherein a driving signal d > 0 is applied, driving current I will increase since d - f > 0. This also causes f to increase, since f is proportional to the radiation intensity of laser beam 11 and since the radiation intensity RI increases with increasing driving current I (see <figref idref="f0001">Fig. 1a</figref>); as f increases, the driving current I increases more slowly, until finally I remains constant when d - f = 0.

0046As shown in <figref idref="f0003">Fig. 3</figref> and discussed below, the driving signal d is based on the image signal 20 and on a plurality of control signals 27. Preferably, the image signal 20 and the control signals 27 are digital signals that are converted to an analog driving signal d via a digital-to-analog converter 25. However, the image signal 20 and/or the control signals 27 may also be analog signals.

0047<figref idref="f0004">Fig. 4a to 4c</figref> illustrate an embodiment in accordance with the present invention wherein the laser driving current I is increased to about the threshold current level I_th by applying a signal "d_fatzero" during a time t_inc that equals the pixel period t_pix, i.e. the time it takes to write a pixel. <figref idref="f0004">Fig. 4b</figref> shows a driving signal d. The driving signal d causes laser 10, as shown in <figref idref="f0003">Fig. 3</figref>, to be supplied with a driving current I, for writing a pixel having an electronic pixel value p. <figref idref="f0004">Fig. 4b and 4c</figref> show the values of d, f and I, that correspond to the specific image signal 20 shown in <figref idref="f0004">Fig. 4a</figref>, as a function of time during a time of five pixel periods t_pix. The image signal 20 is zero during the first two pixel periods, equals p during the third pixel period, and is again zero during the fourth and the fifth pixel periods. The image signal 20 is combined with control signals 27, as shown in <figref idref="f0003">Fig. 3</figref>, to form the driving signal d. <figref idref="f0004">Fig. 4b and 4c</figref> show that d equals d_off during the first pixel period; the driving current I is zero, the laser does not emit radiation and feedback signal f is zero. The value d_off may be zero. How the value d_off is determined, and how the other values d_p, d_fatzero and d_backbias are determined, is discussed below. In the second pixel period, d equals d_fatzero. This causes the driving current I and the feedback signal f to increase since f is smaller than d. The value d_fatzero is determined so that at the end of the pixel period the feedback signal f has hardly increased and the driving current I approximately equals the threshold current level I_th. The third pixel is the actual image pixel. The driving signal d now equals d_p. Since I approximately equals I_th at the start of the image pixel, the driving current I quickly increases to the driving current value that corresponds to the desired laser beam intensity. The laser emits a laser beam and feedback signal f increases until f equals d_p; from that moment on, the driving current I and the laser beam intensity remain constant. In a preferred embodiment, during the fourth pixel period, d equals d_backbias, with approximately d_off - d_backbias = d_fatzero - d_off. In this way, even when f has become nearly zero because the driving current I has become smaller than the threshold current level I_th, the driving current I decreases quickly; the driving current I preferably decreases about as fast as it increased in the second pixel period, i.e. the pixel period of d_fatzero. The idea behind signal d_backbias is the same as the idea behind d_fatzero: passing quickly through the region of undesired emitted radiation, i.e. portion OA of the curve in <figref idref="f0001">Fig. 1a</figref>. If no signal d_backbias would be applied, and if d would equal d_off instead, the driving current would decrease slowly to zero and all the time the laser would continue to emit radiation until f = 0. This would cause the problem of undesired emitted radiation and the additional problem that, because of the slowly decreasing driving current, the driving current for a next image pixel would have to start from some unknown current value between zero and the threshold current level I_th. Both problems are solved by quickly decreasing the driving current during the fourth pixel period, preferably to substantially zero, and preferably in a time not larger than twice the time t_inc it took to increase the driving current up to about the threshold current level I_th, more preferably in a time not larger than t_inc. In the fifth pixel period, finally, d again equals d_off; the image signal 20, the driving current I and the feedback signal f all equal zero.

0048As explained above, <figref idref="f0004">Fig. 4b</figref> shows the driving signal d for writing a single image pixel having an electronic pixel value p. If, instead of writing a single image pixel, several image pixels have to be written one after the other, having electronic pixel values p1, p2, etc. differing from zero, then preferably a signal d_fatzero is applied before the first image pixel and preferably a signal d_backbias is applied after the last image pixel. If two subsequent image pixels, having electronic pixel values differing from zero, are separated from each other by two or more pixels with p = 0, then preferably the procedure of <figref idref="f0004">Fig. 4a to 4c</figref> is simply repeated for the subsequent image pixels. The only remaining case is the one wherein two image pixels having electronic pixel values p1 and p2, differing from zero, are separated by a single pixel with p = 0.

0049<figref idref="f0004">Fig. 5a to 5c</figref> show this remaining case. Between the two image pixels, the driving signal d equals a value d_midzero - determined below - that may differ from d_off, since the driving current I is preferably not decreased to zero but kept at a "standby" value about the threshold current level I_th, so that the driving current I can promptly increase again to write the second image pixel p2. The electronic pixel values p1 and p2, shown in <figref idref="f0004">Fig. 5a</figref>, may differ from each other. <figref idref="f0004">Fig. 5a to 5c</figref> show the image signal 20 and signals d, f and I as in <figref idref="f0004">Fig. 4a to 4c</figref>. During the six shown pixel periods t_pix, the driving signal d subsequently gets the values d_off, d_fatzero, d_p1, d_midzero, d_p2, d_backbias; except for d_midzero, these values are not indicated in <figref idref="f0004">Fig. 5b</figref>.

0050The values of d_off, d_fatzero, d_backbias, d_midzero and d_p may be determined as follows. Preferably, these values are determined before or after an image is exposed, i.e. not during image exposure. Preferably, determination of these values is repeated at regular time intervals. The length of these time intervals may differ for each value, e.g. d_fatzero may be determined more often than d_off. The length of the time intervals depends on the effect that each value has on the laser control system and on the relation of each value with disturbances, such as the ambient temperature and the ageing of the laser; e.g. d_fatzero is directly related to I_th which is very much temperature-dependent; therefore d_fatzero is preferably determined frequently, e.g. every time after a single image is written.

0051To determine d_off, several decreasing values of the (digital) driving signal d may be applied, e.g. d = 10, 9, 8, etc. until the lowest value of d is found that still causes a measurable feedback signal f larger than zero; this lowest value of d is called do. Now d_off = d₀ - off_offset, wherein off_offset > 0 and off_offset < do.

0052Values d_fatzero and d_backbias may be determined as illustrated in <figref idref="f0005">Fig. 6a and 6b. Fig. 6a</figref> shows the driving signal d and the feedback signal f while <figref idref="f0005">Fig. 6b</figref> shows the driving current I as a function of time, represented in the same way as in <figref idref="f0004">Fig. 4a to 4c and Fig. 5a to 5c</figref>. A driving signal d as shown in <figref idref="f0005">Fig. 6a</figref> is applied, wherein d_fatzero successively gets increasing values, e.g. d_off, d_off + 1, d_off + 2, etc. until feedback signal f > f_th, wherein f_th is a very small positive value. The value of f_th may be determined so that the corresponding driving current I equals I_th. It is however not required that I equals I_th; as explained above, it is sufficient that I is about I_th. Since the feedback signal f is very small when I is about I_th, the choice of value f_th is not critical. In <figref idref="f0005">Fig. 6a</figref>, d_backbias may be determined from the equation: <maths id="math0003"><math display="block"><msub><mi mathvariant="normal">d</mi><mi mathvariant="normal">p</mi></msub><mo mathvariant="normal">=</mo><mi mathvariant="normal">p</mi><mo mathvariant="normal">/</mo><msub><mi mathvariant="normal">p</mi><mi>max</mi></msub><mo mathvariant="normal">*</mo><mfenced><msub><mi mathvariant="normal">d</mi><mi>pmax</mi></msub><mo mathvariant="normal">-</mo><msub><mi mathvariant="normal">d</mi><mi>off</mi></msub></mfenced><mo mathvariant="normal">+</mo><msub><mi mathvariant="normal">d</mi><mi>off</mi></msub></math><img file="EP0989736B1_D0003.tif" /></maths> wherein backbias_offset is a value that may be determined so that the driving current I completely decreases to zero from its maximum value about the threshold current level I_th, as shown in <figref idref="f0005">Fig. 6b</figref>.

0053The value of d_midzero may equal d_off. Alternatively, d_midzero may be chosen so that it certainly makes the driving current decrease and that the driving current decreases as slowly as possible; in this way, it is known what is the value of the driving current in the "midzero" pixel. Yet alternatively, d_midzero may be chosen so that feedback signal f has a predetermined value at the end of the "midzero" pixel, i.e. at the end of the fourth pixel period in <figref idref="f0004">Fig. 5a to 5c</figref>.

0054The value of d_p may be determined in two steps: first a value d_pmax is determined, and then d_p is determined by scaling with respect to d_pmax. In the first step, the value d_pmax may correspond to the maximum driving current I_max and the maximum radiation intensity RI_max of the laser; I_max and RI_max are e.g. values that are preferably not exceeded so as not to reduce the laser life. For example, for a TOLD9321 semiconductor laser from Toshiba, I_max = I_th + 15 mA and RI_max = 5 mW; and d_pmax may be determined by successively applying increasing driving signals d, measuring I, and determining the value d_max of d for which I = I_max = I_th + 15 mA. In the second step, d_p is then determined by the formula: <maths id="math0004"><math display="block"><msub><mi mathvariant="normal">d</mi><mi mathvariant="normal">p</mi></msub><mo mathvariant="normal">=</mo><mi mathvariant="normal">p</mi><mo mathvariant="normal">/</mo><msub><mi mathvariant="normal">p</mi><mi>max</mi></msub><mo mathvariant="normal">*</mo><mfenced><msub><mi mathvariant="normal">d</mi><mi>pmax</mi></msub><mo mathvariant="normal">-</mo><msub><mi mathvariant="normal">d</mi><mi>off</mi></msub></mfenced><mo mathvariant="normal">+</mo><msub><mi mathvariant="normal">d</mi><mi>off</mi></msub></math><img file="EP0989736B1_D0004.tif" /></maths> wherein p (with 0 &lt; p &lt; p_max) and p_max are the electronic pixel values to which the driving signals d_p and d_pmax correspond.

0055From the preceding description and drawings, it is clear that a method in accordance with the present invention can be implemented in a simple way that is also cost-effective.

0056Instead of determining the values of d_off, d_fatzero, d_backbias, d_midzero and d_p in the way as explained above, they may also be determined in another way. For example, d_fatzero may be determined by first determining the threshold current level I_th, and by subsequently increasing the driving signal d step by step while measuring I; the value of d where I approximately equals I_th is d_fatzero. First determining I_th may be done by increasing the driving signal d step by step, and measuring the feedback signal f and the driving current I; I_th is the driving current where the relation between I and f changes abruptly (see also <figref idref="f0001">Fig. 1a</figref>).

0057In the embodiment discussed above and illustrated in <figref idref="f0004">Fig. 4a to 4c</figref>, the laser driving current I is increased to about the threshold current level I_th by applying a signal "d_fatzero" during a time t_inc equal to the pixel period t_pix. In another embodiment in accordance with the present invention, t_inc may be smaller than t_pix, or, alternatively, t_inc may be larger than t_pix. As explained hereinbefore, in an embodiment in accordance with the present invention, preferably t_inc is substantially smaller than the maximum time t_keep that is typically to be expected between two non-zero image portions; however, t_inc may be larger than t_pix, equal to t_pix or smaller than t_pix.

0058In a second embodiment in accordance with the present invention, the laser driving current is increased to about the threshold current level by applying a signal "d_fatzero" during a time t_inc longer than t_pix, e.g. equal to 2*t_pix. To determine d_fatzero, the method explained above and illustrated in <figref idref="f0005">Fig. 6a to 6b</figref> may be used, but now signal d_fatzero is applied for a longer time, e.g. for two pixel periods t_pix. The signal d_fatzero need not be constant but may have two or more values, that preferably increase during the successive pixel periods.

0059In another embodiment in accordance with the present invention, no signals d_fatzero, d_backbias, etc. are applied, but the gain of the control loop - as known in the art of system control theory - is adjusted. The laser is driven in <figref idref="f0001">Fig. 1a</figref> from O to B in two steps: in the first step, in portion OA of the curve of <figref idref="f0001">Fig. 1a</figref>, a high gain is used, and in the second step, in portion AB of the curve, a lower gain is used (to discriminate between the two steps, first the threshold current I_th is determined, which may be done by applying successively increasing driving signals as explained above). Because of the high gain in portion OA of the curve, the portion OA is passed through quickly, which, together with the immediate further increase of the driving current in the second step, again provides the advantages of significantly reducing the emission of undesired radiation and of promptly driving the laser to the desired radiation strength. The gain is lower in the second step, because a high gain would make the feedback circuit unstable, since the laser beam intensity strongly increases with increasing driving current above the threshold current level (see also the strongly increasing curve AB in <figref idref="f0001">Fig. 1b</figref>).

0060The present invention is not limited to the embodiments discussed above, but includes other embodiments, such as the ones outlined below.

0061The feedback signal is not necessarily proportional to the radiation intensity of laser beam 11; the relationship may e.g. be parabolic instead of linear. Also, for a radiation intensity of zero, the feedback signal is not necessarily zero but may have an offset value differing from zero. Furthermore, instead of the radiation intensity, another quantity, such as the driving current, may be fed back to the laser control system. Alternatively, more than one quantity may be fed back, such as both the radiation intensity and the driving current. In still another embodiment in accordance with the present invention, the feedback means need not be active during exposure but may only be active when no image is being exposed. In this embodiment, the driving signal d_fatzero may be converted directly, i.e. without using a feedback signal, into an increasing driving current, and driving signal d_p may directly drive the driving current to a desired level; when no image is being exposed, feedback may be used to determine the values of the control signals such as d_fatzero; this determination may be carried out as explained above.

0062As mentioned hereinbefore, the image signal 20 and the control signals 27 are not necessarily digital signals but may be analog signals; thus, the image 16 is not necessarily constituted of (digital) pixels but it may be an analog image as well.

0063In a method in accordance with the present invention, the electrical driving current in a first step increases up to about the threshold current and then in a second step immediately further increases above the threshold current level, without remaining at or about the threshold current level. The increase up to about the threshold current level may be accomplished, as discussed above, by applying a signal d_fatzero, but it is clear to those skilled in the art that numerous other ways of increasing the driving current are possible.

0064A method in accordance with the present invention may be implemented in an image exposure device, as is used e.g. in an electrophotographic printer or in an imagesetter. The electrophotographic printer has a photoconductive member, such as a photoconductive drum or a photoconductive endless belt, that is exposed by the image exposure device. The electrophotographic printer may be a black and white printer, a colour printer, a multi-grey printer. In the imagesetter, a radiation-sensitive sheet may be exposed by visible light, by infrared radiation, by ultraviolet radiation, emitted by the image exposure device.

0065Application of the present invention is not limited to image exposure. The laser beam, controlled by a method in accordance with the present invention, may also be used for other purposes wherein no image is formed. As an example, the emitted laser beam may be reflected, at least partially, by an optical medium and the reflected beam may be analysed.

0066A method in accordance with the present invention may be applied to control a semiconductor laser. The laser may emit visible light, but it may also emit other electromagnetic radiation, such as ultraviolet or infrared radiation. The laser is not necessarily a semiconductor laser; a method of the present invention may be applied for any laser or for any source of radiation that exhibits a threshold behaviour. The threshold behaviour must be such that - see also <figref idref="f0001">Fig. 1a</figref> - the curve showing the radiation intensity as a function of a driving parameter, such as the electrical driving current, exhibits an abrupt change about the threshold value from a low gradient to a high gradient.

0067Having described in detail preferred embodiments of the current invention, it will now be apparent to those skilled in the art that numerous modifications can be made therein without departing from the scope of the invention as defined in the appending claims.

Parts list

0068<dl id="dl0001"><dt>10</dt><dd>laser</dd><dt>11</dt><dd>laser beam</dd><dt>11a,11b</dt><dd>laser beam</dd><dt>12</dt><dd>monitor diode</dd><dt>13</dt><dd>mirror</dd><dt>14</dt><dd>lens</dd><dt>15</dt><dd>radiation-sensitive means</dd><dt>16</dt><dd>image</dd><dt>17</dt><dd>motor</dd><dt>18</dt><dd>motor</dd><dt>20</dt><dd>image signal</dd><dt>21</dt><dd>driving means</dd><dt>23</dt><dd>feedback means</dd><dt>24</dt><dd>comparing means</dd><dt>25</dt><dd>converting means</dd><dt>27</dt><dd>control signals</dd><dt>31</dt><dd>area</dd><dt>32</dt><dd>area</dd><dt>50</dt><dd>device</dd></dl>