Differential dual floating gate circuit and method for programming

Description of equivalent WO 2004064115 A2

DIFFERENTIAL DUAL FLOATING GATE CIRCUIT AND METHOD FOR PROGRAMMING

0002FIELD OF THE INVENTION This invention relates to a method and circuit for an accurate voltage reference, and more specifically to a differential circuit that uses a feedback loop and dual conduction of tunnel devices to accurately program a desired charge level on a floating gate.

0003BACKGROUND OF THE INVENTION Programmable analog floating gate circuits have been used since the early 1980's in applications that only require moderate absolute voltage accuracy over time, e.g., an absolute voltage accuracy of 100-200mN over time. Such devices are conventionally used to provide long-term non- volatile storage of charge on a floating gate. A floating gate is an island of conductive material that is electrically isolated from a substrate but capacitively coupled to the substrate or to other conductive layers. Typically, a floating gate forms the gate of an MOS transistor that is used to read the level of charge on the floating gate without causing any leakage of charge therefrom.

0004Various means are known in the art for introducing charge onto a floating gate and for removing the charge from the floating gate. Once the floating gate has been programmed at a particular charge level, it remains at that level essentially permanently, because the floating gate is surrounded by an insulating material which acts as a barrier to discharging of the floating gate. Charge is typically coupled to the floating gate using hot electron injection or electron tunneling. Charge is typically removed from the floating gate by exposure to radiation (UV light, x-rays), avalanched injection, or Fowler-Νordheim electron tunneling. The use of electrons emitted from a cold conductor was first described in an article entitled Electron

0005Emission in Intense Electric Fields by R.H. Fowler and Dr. L. Νordheim, Royal Soc. Proc, A, Vol. 119 (1928). Use of this phenomenon in electron tunneling through an oxide layer is described in an article entitled Fowler-Νordheim Tunneling into Thermally Grown Si0₂ by M. Lanzlinger and E. H. Snow, Journal of Applied Physics, Vol. 40, No. 1 (January, 1969), both of which are incorporated herein by reference. Such analog floating gate circuits have been used, for instance, in digital nonvolatile memory devices and in analog nonvolatile circuits including voltage reference, Vcc sense, and power-on reset circuits.

0006Fig. 1A is a schematic diagram that illustrates one embodiment of an analog nonvolatile floating gate circuit implemented using two polysilicon layers formed on a substrate and two electron tunneling regions. Fig. 1 A illustrates a cross- sectional view of an exemplary prior art programmable voltage reference circuit 70 formed on a substrate 71. Reference circuit 70 comprises a Program electrode formed from a first polysilicon layer (polyl), an Erase electrode formed from a second polysilicon layer (poly2), and an electrically isolated floating gate comprised of a polyl layer and a poly2 layer connected together at a corner contact 76. Typically, polysilicon layers 1 and 2 are separated from each other by a thick oxide dielectric, with the floating gate fg being completely surrounded by dielectric. The floating gate fg is also the gate of an NMOS transistor T0 shown at 73, with a drain D and a source S that are heavily doped n+ regions in substrate 70, which is P type. The portion of dielectric between the polyl Program electrode and the floating gate fg, as shown at 74, is a program tunnel region (or "tunnel device") TP, and the portion of dielectric between the polyl floating gate fg and the poly2 erase electrode, shown at 75, is an erase tunnel region TE. Both tunnel regions have a given capacitance. Since these tunnel regions 74,75 are typically formed in thick oxide dielectric, they are generally referred to as "thick oxide tunneling devices" or "enhanced emission tunneling devices." Such thick oxide tunneling devices enable the floating gate to retain accurate analog voltages in the +/-4 volt range for many years. This relatively high analog voltage retention is made possible by the fact that the electric field in most of the thick dielectric in tunnel regions 74,75 remains very low, even when several volts are applied across the tunnel device. This low field and thick oxide provides a high barrier to charge loss until the field is high enough to cause Fowler-Nordheim tunneling to occur. Finally, reference circuit 70 includes a steering capacitor CC that is the capacitance between floating gate fg and a different n+ region formed in the substrate that is connected to a Cap electrode. Fig IB is a schematic diagram that illustrates a second embodiment of a floating gate circuit 70 that is implemented using three polysilicon layers. The three polysilicon floating gate circuit 70' is similar to the two polysilicon embodiment except that, for example Erase electrode is formed from a third polysilicon layer (poly 3). In addition, the floating gate fg is formed entirely from a poly2 layer. Thus, in this embodiment there is no need for a corner contact to be formed between the polyl layer portion and the poly2 layer portion of floating gate fg, which is required for the two polysilicon layer cell shown in Fig. 1 A.

0007Referring to Fig. 2, shown at 20 is an equivalent circuit diagram for the voltage reference circuit 70 of Fig. 1A and 70' of Fig. IB. For simplicity, each circuit element of Fig. 2 is identically labeled with its corresponding element in Figs. 1A and IB.

0008Setting reference circuit 70 to a specific voltage level is accomplished using two separate operations. Referring again to Fig. 1 A, the floating gate fg is first programmed or "reset" to an off condition. The floating gate fg is then erased or "set" to a specific voltage level. Floating gate fg is reset by programming it to a net negative voltage, which turns off transistor T0. This programming is done by holding the Program electrode low and ramping the n+ bottom plate of the relatively large steering capacitor CC to 15 to 20V via the Cap electrode. Steering capacitor CC couples the floating gate fg high, which causes electrons to tunnel through the thick oxide at 74 from the polyl Program electrode to the floating gate fg. This results in a net negative charge on floating gate fg. When the bottom plate of steering capacitor CC is returned to ground, this couples floating gate fg negative, i.e., below ground, which turns off the NMOS transistor T0.

0009To set reference circuit 70 to a specific voltage level, the n+ bottom plate of steering capacitor CC, the Cap electrode, is held at ground while the Erase electrode is ramped to a high voltage, i.e., 12 to 20V. Tunneling of electrons from floating gate fg to the poly2 Erase electrode through the thick oxide at 75 begins when the voltage across tunnel device TE reaches a certain voltage, which is typically approximately 1 IV. This tunneling of electrons from the fg through tunnel device TE increases the voltage of floating gate fg. The voltage on floating gate fg then "follows" the voltage ramp coupled to the poly2 Erase electrode, but at a voltage level offset by about 1 IV below the voltage on the Erase electrode. When the voltage on floating gate fg reaches the desired set level, the voltage ramp on poly2 Erase electrode is stopped and then pulled back down to ground. This leaves the voltage on floating gate fg set at approximately the desired voltage level.

0010As indicated above, reference circuit 70 meets the requirements for voltage reference applications where approximately 200mV accuracy is sufficient. The accuracy of circuit 70 is limited for two reasons. First, the potential on floating gate fg shifts down about lOOmV to 200mV after it is set due to the capacitance of erase tunnel device TE which couples floating gate fg down when the poly2 Erase electrode is pulled down from a high voltage to 0N. The amount of this change depends on the ratio of the capacitance of erase tunnel device TE to the rest of the capacitance of floating gate fg (mostly due to steering capacitor CC), as well as the magnitude of the change in voltage on the poly2 Erase electrode. This voltage "offset" is well defined and predictable, but always occurs in such prior art voltage reference circuits because the capacitance of erase tunnel device TE cannot be zero. Second, the accuracy of circuit 70 is also limited because the potential of floating gate fg changes another lOOmN to 200mN over time after it is set due to various factors, including detrapping of the tunnel devices and dielectric relaxation of all the floating gate fg capacitors.

0011An analog voltage reference storage device that uses a floating gate is described in U.S. Patent No. 5,166,562 and teaches the uses of hot electron injection for injecting electrons onto the floating gate and electron tunneling for removing electrons from the floating gate. The floating gate is programmed by controlling the current of the hot electron injected electrons after an erase step has set the floating gate to an initial voltage. See also U.S. Patent No. 4,953,928. Although this method of programming the charge on a floating gate is more accurate than earlier analog voltage reference circuits including a floating gate, the level of accuracy is still on the order of 50mV to 200mN.

0012Prior art floating gate storage devices have sometimes used dual conduction of Fowler-Νordheim tunnel devices, i.e., wherein both the program and erase tunnel elements in a floating gate device are caused to conduct simultaneously in order to provide the coupling of charge onto the floating gate. However, this method has only been used in digital circuits to program the floating gate to either a "1" condition or a "0" condition to provide memory storage. The precise charge on the floating gate in such applications is not of concern and so is not precisely controlled in such circuits. According to the prior art, such dual conduction digital programming of a floating gate is considered to be a less efficient and desirable way than generating electron conduction through a single tunnel element to control the level of charge on a floating gate. Known disadvantages of dual conduction digital programming of a floating gate include the fact that a larger total voltage is required to provide dual conduction and tunnel oxide trap-up is faster because more tunnel current is required. An example of a prior art analog nonvolatile floating gate circuit that uses dual conduction of electrons for adding and removing electrons from a floating gate is disclosed in U.S. Patent No. 5,059,920, wherein the floating gate provides an adaptable offset voltage input for a CMOS amplifier. In this device, however, only one Fowler-Nordheim tunnel device is used. The electrons are injected onto the floating gate using hot electron injection, while Fowler-Nordheim electron tunneling is used to remove electrons from the floating gate, so as to accurately control the charge on the floating gate. This means of injecting electrons onto the floating gate is used because the charge transfer is a controlled function of the voltage on the floating gate. Another example of a prior art dual conduction floating gate circuit is disclosed in U.S. Patent No. 5,986,927. A key problem with such prior art devices is that they do not compensate for common-mode voltage and current offsets, common-mode temperature effects, and mechanical and thermal stress effects in the integrated circuit. Applications that require increased absolute voltage accuracy generally use a bandgap voltage reference. A bandgap voltage reference typically provides approximately 25mN absolute accuracy over time and temperature, but can be configured to provide increased accuracy by laser trimming or E² digital trimming at test. While a bandgap voltage reference provides greater accuracy and increased stability over the prior art voltage reference circuits discussed above, a bandgap voltage reference only provides a fixed voltage of about 1.2N. Therefore, additional circuitry, such as an amplifier with fixed gain, is needed to provide other reference voltage levels. Moreover, prior art bandgap voltage references typically draw a relatively significant current, i.e., greater than lOμA.

0013What is needed is an analog programmable voltage reference circuit that can be quickly and accurately set to any analog voltage without the need for additional amplification and that provides improved stability and accuracy over time and temperature as compared to prior art voltage references. It is also desirable that the improved stability and accuracy be obtained in a voltage reference circuit that draws significantly less current than prior art voltage references.

0014SUMMARY OF THE INVENTION The present invention is directed at addressing the above-mentioned shortcomings, disadvantages, and problems of the prior art. The present invention comprises a floating gate circuit, comprising: a) a first floating gate for storing charge thereon; b) a first steering capacitor coupled to said first floating gate for controlling the charge level on said first floating gate as a function of an input set voltage that is coupled through said first steering capacitor to said first floating gate during a set mode; c) a second floating gate for storing charge thereon; d) a first circuit coupled to said second floating gate for controlling the charge level on said second floating gate during said set mode; and e) a feedback circuit coupled between said second floating gate and said first floating gate for causing the charge level on said first floating gate to be modified during said set mode until the voltage on said first floating gate is a predetermined function of the voltage on said second floating gate, and such that at the conclusion of said set mode the difference in charge level between said first and second floating gates is a predetermined function of said input set voltage.

0015The floating gate circuit generates a condition at the end of the set mode wherein the voltage on the first floating gate is approximately equal to the voltage on the second floating gate. In addition, program and erase tunnel devices for the first floating gate operate in dual conduction during the set mode for modifying the charge level on the first floating gate, and program and erase tunnel devices for the second floating gate also operate in dual conduction during the set mode for controlling the voltage on the second floating gate. The floating gate circuit can then be configured in a read mode to operate as a voltage reference circuit having a negative feedback loop for causing the floating gate circuit to reach a condition such that a reference voltage is generated that is a predetermined function of the input set voltage, and is preferably approximately equal to the input set voltage.

0016The present invention is also a method for setting a reference voltage in a i < dual floating gate circuit, said method comprising the steps of: a) causing an input s -,e-.t- ' voltage to be capacitively coupled to a first floating gate during a set mode for controlling the charge thereon; b) causing a predetermined voltage to be coupled to a second floating gate during said set mode for controlling the charge level thereon; c) causing the charge level on said first floating gate to be modified until said dual floating gate circuit reaches a condition during said set mode such that the voltage on said floating gate is a predetermined function of the voltage on said second floating gate, and such that at the conclusion of said set mode the difference on charge level between said first and second floating gates is a predetermined function of said input set voltage. An object of the present invention is to provide a method and circuit for generating a voltage reference that has an improved accuracy and stability over the prior art voltage references.

0017A key advantage of the present invention is the improved initial setting accuracy over prior art floating gate voltage references by more than a factor of 100. Another key advantage of the present invention is that, without the need for using laser trimming or E² digital trimming, the present invention has an improved accuracy over bandgap voltage references of a factor of 10 to 50 while drawing less power by a factor of more than 10. Moreover, a voltage reference of greater than or less than 1.2 volts can be set using the present invention without the need for additional amplifiers.

0018Another advantage of the present invention is that, after a high voltage set mode, the invention allows for a controlled ramp down sequence to ramp down the voltages at each floating gate erase and program electrode such that, when voltage and current sources are completely shut down in the circuit, a more accurate voltage is set on the floating gates. Another advantage of the present invention is that the voltage at the erase electrode that is coupled to the reference floating gate is controlled during the ramp down sequence by shutting off the negative voltage charge pump while allowing the feedback circuit to remain active. Another advantage of the present invention is that tunnel current is used to self-discharge the voltage at each program electrode.

0019Another advantage of the present invention is that any voltage and current offsets coupled through the tunnel devices to the corresponding floating gates and any offsets in the transistors used in the floating gate circuit are common mode because of the use of dual floating gates, and do not change the voltage differential produced by the floating gates, and thus do not change the programmed voltage reference at the output during a read mode.

0020Yet another advantage of the present invention is that changes in the threshold of transistors in the floating gate circuit due to changes in temperature are also common-mode as the result of the use of two floating gates and therefore have greatly reduced effect on circuit operation.

0021Another advantage of the present invention is that changes in source or drain voltages due to bias current, temperature, or power supply voltage variations, which are capacitively coupled to the floating gates, are common-mode and therefore have greatly reduced effect.

0022Another advantage of the present invention is that mechanical and thermal stress effects are also common-mode due to the use of two floating gates and so their effects are also greatly reduced.

0023BRIEF DESCRIPTION OF THE DRAWINGS The forgoing aspects and attendant advantages of the present invention will become more readily appreciated by reference to the following detailed description, when taken in conjunction with the accompanying drawings, wherein:

0024Fig. 1 A is a schematic diagram that illustrates a cross-sectional view of a prior art programmable floating gate circuit formed from two polysilicon layers; Fig. IB is a similar prior art floating gate circuit formed from three polysilicon layers; Fig. 2 is an equivalent circuit diagram for the reference circuit illustrated in Figs. 1 A and IB;

0025Fig. 3 is a circuit diagram of a differential single floating gate circuit according to the present invention for high precision programming of a floating gate; Fig. 4A is a circuit diagram of a differential dual floating gate circuit according to another embodiment of the present invention;

0026Fig. 4B is a combined schematic and block diagram illustrating a single floating gate circuit coupled to the dual floating gate circuit of the present invention, during a set mode; Fig. 5 is a flow diagram illustrating a method for setting a floating gate using the single floating gate circuit;

0027Fig. 6 illustrates various voltage waveforms vs. time for a specific implementation of the method of Fig. 5;

0028Fig. 7 illustrates various voltage waveforms vs. time for a specific implementation of the method of Fig. 5 ;

0029Fig. 8 illustrates various voltage waveforms vs. time for a specific implementation of the method of Fig. 5;

0030Fig. 9 is a flow diagram illustrating a method for setting a floating gate using the differential dual floating gate circuit of the present invention; Fig. 10 illustrates various voltage waveforms vs. time for a specific implementation of the method of Fig. 9;

0031Fig. 11 illustrates various voltage waveforms vs. time for a specific implementation of the method of Fig. 9; and

0032Fig. 12 illustrates various voltage waveforms vs. time for a specific implementation of the method of Fig. 9.

0033DETAILED DESCRIPTION OF THE INVENTION Fig. 3 is a circuit diagram of a differential single floating gate circuit 30 according to the present invention for accurately setting a floating gate to an analog voltage during a high voltage set mode or set cycle. Fig. 4A is a circuit diagram of a differential dual floating gate circuit 40 according to another embodiment of the present invention. Circuit 40 is also used to accurately set a floating gate to an analog voltage during a high voltage set mode. Once the analog voltage level is set, both circuit 30 and circuit 40 can then be configured during a read mode as a precise voltage comparator circuit with a built-in voltage reference or a precise voltage reference circuit. Circuit 30 and circuit 40 are preferably implemented as an integrated circuit manufactured using industry standard CMOS processing techniques. Since the sequence used during the set mode is similar for both circuits, circuit 30 and the method for programming a floating gate using circuit 30 will be described first.

0034Circuit 30 comprises a floating gate fgø at a node 2 that, at the conclusion of a set mode, is set to a voltage that is a function of, and preferably is equal to an input set voltage Vsetø received at an input terminal 300 coupled to a node 1. This set mode may be instituted at the factory to cause floating gate fgø to be set to a desired voltage. Alternatively, a later user of circuit 30 can cause circuit 30 to enter a set mode wherever the user wishes to update the voltage on fgø as a function of the Vsetø voltage input by the user during this later, or in the field, set mode operation. Circuit 30 further comprises a circuit 310 that includes: a programming tunnel device TP0 formed between floating gate fgø and a programming electrode Ep0, at a node 3; an erase tunnel device Te0 formed between floating gate fgø and an erase electrode Ee0, at a node 4; and a steering capacitor Cl coupled between floating gate fgø and a node 5. Preferably, programming electrode Ep0 receives a negative voltage during the set mode, and erase electrode Ee0 receives a positive voltage during the set mode. Moreover, Tp0 and Te0 are Fowler-Nordheim tunnel devices that are reasonably well matched by layout. The bottom plate of steering capacitor Cl is coupled to a predetermined voltage during the set mode that is preferably ground gl . Steering capacitor Cl is used to provide a stable ground reference for floating gate fg0.

0035Setting fgø to a specific charge level during the set mode, which corresponds to a specific voltage at node 2, is achieved by taking Ep0 negative and Ee0 positive, such that the voltage at node 4 minus the voltage at node 3 is two tunnel voltages or approximately 22V. An alternative is to take Ep0 negative and Ee0 positive such that approximately 5nA of current flows from node 4 to node 3. In either case, both tunnel devices are conducting, i.e., the tunnel devices are in "dual conduction." By operating in dual conduction, the voltage on the floating gate fgø can stabilize at a DC voltage level for as long a time as needed for Circuit 30 to settle to a very precise and accurate level. Operating two Fowler-Nordham tunneling devices in dual conduction is key to making it possible to set the floating gate fgø voltage very accurately using either on-chip circuitry or test equipment off-chip. In dual conduction, the tunnel devices, Te0 and Tp0, which are reasonably well matched as a result of their chip layout, will modify the charge level on the floating gate fgø by allowing electrons to tunnel onto and off of floating gate fgø so as to divide the voltage between nodes 4 and 3 in half. Thus, the floating gate voltage, i.e., the voltage at node 2, will be Vfgø = Vnode3 + (Vnode3-Vnode3)/2, which is halfway between the voltage at node 4 and the voltage at node 3. Under these conditions, the dual conduction current can typically charge or discharge node 2, which typically has less than 5 pF capacitance, in less than 1 mSec. As this occurs, the floating gate voltage "tracks" directly with the voltage at nodes 3 and 4 and settles to a DC voltage that is halfway between those two voltages in a few mSec. Accordingly, Vfgø can be set to a positive or a negative voltage or zero volts depending upon the voltages at electrodes Ee0 and Ep0. For example, if the tunnel voltage is approximately 1 IV for the erase and program tunnel devices Te0 and Tp0, and the voltage at electrode Ee0 is set to about +16V and the voltage at electrode Ep0 is about -6V, then Vfgø will settle at about +5 V, which is the midpoint between the two voltages. If the voltage at electrode Ee0 is set to about +1 IV and the voltage at electrode Ep0 is about -1 IV, then Vfgø will go to about 0V. If the voltage at electrode Ee0 is set to about +6V and the voltage at electrode Ep0 is about -16V, then Vfgø will go to about -5V.

0036Note that, in a preferred embodiment, a specific voltage is not generated at node 3 during the set mode. The voltage used to control the charge level on floating gate fgø is the voltage at node 4. A current source Ip0, which is preferably implemented as a charge pump, provides the necessary voltage compliance to generate a negative voltage sufficient to generate the voltage difference required to produce dual conduction tunneling in tunnel devices Te0 and Tp0. Circuit 30 further includes a circuit 320 that compares Vfgø, the voltage on the floating gate fgø, with the voltage at node 1 and generates an output voltage Vout, at a node 6, that is a function of the difference between Vsetø and the voltage at node 1. Circuit 320 preferably includes a differential amplifier (or differential stage) 322 that is preferably configured to have an inverting input coupled to floating gate fgø, a non-inverting input coupled to node 1, and an output at a node 7. Circuit 320 preferably further includes a gain stage 324 with an input coupled to node 7 and an output terminal 326, at node 6. The differential stage compares the voltages received at its inputs and amplifies that difference, typically by a factor of 50 to 100. The gain stage then further amplifies that difference by another factor of 50 to 100. Moreover at the conclusion of the set mode, circuit 320 ideally settles to a steady state condition such that Vfgø = Vsetø.

0037Referring again to Fig. 3, the differential stage 322 preferably includes enhancement mode transistors TI, T2, T3 and T4. Transistors TI and T2 are preferably NMOS transistors that are reasonably well matched by layout, and transistors T3 and T4 are preferably PMOS transistors that are reasonably well matched by layout. The sources of NMOS transistors TI and T2 are coupled together at a node 8. The drain of NMOS transistor TI is coupled to a node 9, and its gate is floating gate fgø. The drain of NMOS transistor T2 is coupled to node 7, and its gate is coupled to node 1. PMOS transistor T3 is coupled common drain, common gate, to node 9, with its source coupled to node 10. The gate of PMOS transistor T4 is coupled to node 9. Its drain is coupled to node 7, and its source is coupled to node 10. A voltage supply Vcc, typically 3 to 5 volts, is coupled to node 10, and a current source It0 is coupled between node 8 and ground gl to cause transistors TI, T2, T3 and T4 to operate in either the prethreshold or linear region during the set mode. Current source It0 can be implemented using any number of conventional circuits.

0038One benefit provided by differential stage 322 is that temperature and stress effects track in transistors TI - T4 because the temperature coefficient Tc of these transistors is approximately the same. That is, any variation in the temperature of the integrated circuit chip on which a floating gate circuit according to the present invention is implemented will have the same effect on transistors TI - T4, such that differential stage 322 is in a balanced condition essentially independent of temperature. Similarly, mechanical and thermal stress effects are also common-mode and so their effects are also greatly reduced.

0039The gain stage 324 preferably includes a PMOS pull-up transistor T5 biased by Vcc, and includes a current source pull-down load Ig0. The source of transistor T5 is coupled to node 10. Its gate is coupled to the differential stage PMOS pull-up T4 at node 7, and its drain is coupled to node 6. Current source pull-down load Ig0 is coupled between node 6 and ground gl . The gain stage 324 also preferably includes a compensation capacitor C2 coupled between nodes 6 and 7. Current source pull-down load Ig0 is preferably an active load using an NMOS current mirror or a depletion device. Using an active current source with relatively high output resistance, the gain stage 324 can provide a voltage gain of about 100. The output swing of the gain stage 324 is nearly full rail from ground to Vcc. Stability and response of this circuit can be easily adjusted for various processes using compensation capacitor C2. In this configuration, transistor T5 provides good current sourcing capacity, but current sinking is limited to the current in the current source pull-down Ig0. Therefore, the current in Ig0 should be greater than the pull-up current required by the load on Vout so that the gain stage 324 is capable of adequately controlling Vout, at node 6, by sinking all of the current that flows to node 6.

0040Circuit 320 further operates in the following manner during the set mode. When biased by Vcc and current source It0, TI senses Vfgø relative to input set voltage Vsetø (300), which is sensed by transistor T2, and the amplified difference appears as Vout at node 6. If Vfgø is initially less than Vsetø, T2 is turned on more than TI , and the current flow through T2 (and through T4 since they are connected in series) is initially greater than the current flow through TI (and correspondingly T3). The gate of the pull up transistor T3 is tied to the drain of T3 and also to the gate of pullup transistor T4, which makes the current in T4 a mirror of the current in T3. When more current flows through T4 than T3, the voltage, V7, on node 7 drops below the voltage, V9, on node 9. The lower voltage on node 7 causes the current through T5 to increase which pulls Vout high. The voltage gain of the differential stage 322 is typically about 80 and the voltage gain of the output stage 324 is about 100, giving an overall gain from Vsetø to Vout of about 8000. A negative feedback path or loop from Vout to the inverting input fgø is necessary for the differential circuit 320 to settle at the point where the voltage on fgø is equal to Vsetø. During the set mode, this feedback path is provided by tunnel devices TF0, Te0 and transistors T6 and T7, as described in the next section. When Vout goes high, the negative feedback path pulls Vfgø higher. As Vfgø rises, the current in TI increases until it matches the current in T2. At this point, the differential circuit 320 settles to a steady state condition where the currents in transistors TI, T2, T3, and T4 match, and Vfgø = Vsetø.

0041Those skilled in the art will realize that circuit 320 can be implemented using PMOS transistors for TI and T2 and NMOS transistors for T3 and T4. For this implementation, the gain stage 324 comprises an NMOS pull-down transistor T5 coupled to a current source pull-up load Ig0. Circuit 30 also includes a feedback loop coupled between nodes 6 and

00422. During the set mode, this feedback loop causes the voltage differential between tunnel electrodes Ee0 and Ep0 to be modified by modifying the voltage at node 4 as a function of the output voltage at node 6. The feedback loop preferably comprises a level shift circuit that is preferably a tunnel device TF0 formed between node 6 and a node 11 and a transistor T7, preferably an NMOS transistor, coupled common gate, common drain to a node 12, with its source coupled to node 11. Also included in the feedback loop is a transistor T6, preferably an NMOS transistor, having its gate coupled to node 12, its source coupled to node 4, and thereby to erase tunnel device Te0, and its drain coupled to a node 13. As earlier indicated, the maximum output of the gain stage is approximately Vcc. However, this is not high enough to drive Vefb at node 12 directly, because Vefb typically needs to go to about 14 to 19 volts, which is well above the usual 3 to 5 volt Vcc supply level. The level shift circuit TF0 and T7 shifts the relatively low output voltage at node 6 (Vout) up to the desired 14 to 19 volt range. Preferably, TF0 and Te0 are reasonably well matched by layout and transistors T6 and T7 are reasonably well matched by layout. Under these conditions, when the same tunnel current flows through both TF0 and Te0, the level shift tracks the erase tunnel voltage as measured by the voltage drop from node 4 to node 2, which drives the gate of transistor TI (fgø) to the same voltage as the voltage on the gate of transistor T2 (Vsetø) when circuit 320 settles. This adds to the improved setting accuracy of the circuit.

0043One advantage of having the level shift track the erase tunnel voltage is that, as the voltage necessary to create tunneling changes, due to charge trapping in the dielectric as more and more set cycles are performed, output voltage Vout continues to follow the input set voltage Vsetø and operate in the same voltage range. Another advantage is that when the output voltage Vout is not quite equal to the input set voltage Vsetø, the error introduced by the finite gain of circuit 320 is very small.' For example, if circuit 320 has a gain of 10,000 and Vout is 1 volt lower than Vsetø and Vfgø when circuit 30 settles, Vfgø will have an error of 1V/10,000, or only O.lmV. Circuit 30 also preferably includes current sources 12 and Ip0, and a capacitor Cp0. Current source 12 is coupled between node 12 and a high voltage supply HV+ at node 13 for establishing Nefb at the beginning of the set mode and for providing tunnel current through TF0. Current source 12 can be implemented using any number of conventional methods. However, current source 12 is preferably a current regulator that is biased by HN+, such as a current mirror comprising P- Channel devices that operate in the prethreshold region. In this manner, current source 12 will automatically go to whatever positive voltage needed at node 12 to establish the tunnel current through tunnel device TF0. Current source 12 preferably generates a current that is about the same as Ip0. This means the current through tunnel device TF0 is about the same as the current through tunnel devices Te0 and Tp0.

0044Current source Ip0 is coupled between node 3 and ground gl . Current source Ip0 is preferably a P-Channel charge pump that is used as a negative current source to pump a controlled tunnel current out of programming tunnel device Tp0. As mentioned above, since Ip0 is a current source, it functions to automatically goes to whatever negative voltage at node 3 that is needed to establish the tunnel current at the desired level. Current source Ip0 has sufficient voltage compliance to provide this negative voltage. Moreover, once the current through the tunnel devices is established, the voltage across the tunnel devices is also well defined by their Fowler- Nordheim characteristics. Therefore, current source Ipø produces Np, the voltage at node 3, by controlling the current through tunnel device Tp0. Using a current source Ipø is the preferred way to assure that tunnel devices Te0 and Tp0 are operating at a current level that is high enough to allow dual conduction and to allow the feedback circuit to work, but low enough to avoid excessive current flow which damages the tunnel devices. Capacitor Cp0 controls the discharge of current through the tunnel devices when, as explained in more detail below, Ipø is shut down at the conclusion of the set mode.

0045Those skilled in the art will realize that Np can also be produced using a fixed voltage supply that is about 24 to 30 volts below Nefb. However, this topology should be used with caution because the current in Fowler-Νordheim tunnel devices varies exponentially with the applied voltage. In particular, very high current will flow through the tunnel devices if the voltage differential is too high, and extremely low current may flow if the voltage differential is too low. Very high currents will damage or "wear out" the tunnel devices due to rapid charge trapping in the dielectric, and if the tunnel current is too low, the feedback circuit will not be able to tunnel charge onto or off of fgø, and thus will not be able to control the voltage on fgø. Moreover, it is also possible to connect Nefb to a current source and connect Np to the feedback circuit such that Np controls the voltage on fgø. However, this would require the feedback circuit to produce a controlled negative voltage, which is more difficult to integrate in a standard CMOS process.

0046Fig. 5 is a flow diagram illustrating a method 50 for setting a floating gate that may be implemented during a set mode, for instance, by circuit 30 of Fig. 3. Figs. 6-8 illustrate voltage waveforms for Vout, Vp, Vefb, Vfgø and Vsetø, for the specific implementation of method 50 discussed below relative to those figures. Each of the four waveforms shown in Figs. 6-8 are the same, only the voltage axes of some of these waveforms are modified to illustrate specific details. In the circuit implementation illustrated in Figs. 6-8: Vsetø = 4.00V; Vcc = +5V, HV+ is about 22V, Ipø is about 6nA, 12 is about 6nA, It0 is about 5nA; and Ig0 is about 20nA. At step 51, circuit 30 is powered up at the beginning of the set mode, which is illustrated in Figs. 6-8 as time t₀, and at some point thereafter receives input set voltage Vsetø. Figs. 6-8 further illustrate Vsetø being held at a constant voltage of 4.00V. In addition Vcc is set to +5V, HV+ is ramped up to a high positive voltage of about +22V, which turns on 12, and current source Ipø is turned on to enable this current source to begin generating its corresponding current. Thereafter, according to the preferred implementation of the remaining steps 52-56 of method 50, circuit 30 can set Vfgø to within about 0.5 mV of Vsetø in about 30 mSec, as illustrated in Figs. 6-8. At step 52, circuit 30 causes tunnel devices Te0 and Tp0 to operate in a dual conduction mode under the control of the voltage differential between the erase and programming electrodes Ee0 and Ep0, respectively, for modifying the charge level on floating gate fgø. Dual conduction occurs when tunnel current flows through both Te0 and Tp0. Tunnel current flows through Te0 and Tp0 when the voltage differential between the erase and programming electrodes is at least two tunnel voltages or approximately 22V as discussed earlier.

0047Preferably, circuit 30 causes dual conduction in the following manner. Current source 12 pulls node 12, Vefb, up relatively quickly to about +18V. Vefb (node 12) turns on transistor T6, which pulls Ve0 (node 4) to one Vt below Vefb. Charge pump Ipø gradually charges capacitor Cp0 and ramps Vp (node 3) down to a negative voltage of about -1 IV in about 2 mSec. Once Vp ramps down to the point where the difference between Ve0 and Vp is at least two tunnel voltages, tunnel current flows through both tunnel devices Te0 and Tp0, under the control of Ipø, and Vfgø is controlled directly by Vefb. 12 continues to pull up Vefb until Vefb reaches Vout + ITV + IVt, where ITV is the tunnel voltage across tunnel device TF0, and IVt is the threshold voltage of transistor T7. When at least one tunnel voltage exists across TF0 tunnel current flows through TF0, and TF0 and T7 act as level shift devices such that Vefb is controlled directly by Vout. At step 53, circuit 30 compares Vfgø with Vsetø and generates an output voltage Vout that is a function of the difference between Vfgø and Vsetø. Circuit 30 then, at step 55, causes the voltage differential between Vefb and Vp to be modified as a function of Vout, by modifying Vefb, and circuit 30 repeats steps 52 through 55 until circuit 30 settles to a steady state condition, at step 54, where Vfgø is approximately equal to Vsetø. At this point circuit 30 is powered down, at step 56. As a result of method 50, fgø is set to a charge level that will remain essentially the same over time. The voltage waveforms of Figs. 6-8 illustrate how circuit 30 functions during steps 52 through 55. Dual conduction occurs after about 0.5 mSec, which is illustrated as time t_t in Figs. 6-8. Prior to time ti, Vout=0V, Vefb is pulled up by 12, and Vfgø is not controlled by Vefb. However, once tunnel current is flowing through Te0, Tp0 and TF0 at time t] : the differential stage senses that Vfgø is not equal to Vsetø; Vout is a function of the difference between Vfgø and Vsetø; Vefb follows Vout; and Vfgø follows Vefb. For about the next 2.5 mSec, which is illustrated as time ti to time t₂ in Figs. 6-8, Vfgø oscillates above and below Vsetø as Vefb moves up and down as a function of the negative feedback loop.

0048At the beginning of this oscillation period at time ti, it can be seen in Fig. 6 that Vfgø is below Vsetø. Thus, transistor TI is OFF and transistor T2 is ON, which pulls down node 7. This turns on transistor T5, which quickly pulls up Vout from zero volts, also illustrated in Fig. 6. Since tunnel current is flowing through TF0, TF0 and T7 act as level shifters such that Vefb pulls up ITV and IVt above Vout. Vefb then pulls up Vfgø through tunnel device Te0. Since Vp is continuing to ramp down to a predetermined negative voltage, Vfgø is pulled greater than Vsetø after about 1 mSec. At that point, the differential stage 322 senses that Vfgø is greater than Vsetø, and the gain stage 324 amplifies that difference, quickly pulling Vout low, which pulls Vefb low and pulls Vfgø back down low. When Vfgø is approximately equal to Vsetø, circuit 320 ceases to oscillate except for some noise coupled to circuit 320 from the charge pump Ipø, as best shown in Figs. 7 and 8 beginning at time t₂.

0049Beginning at time ti, current source Ig0 in the gain stage 324 produces a current that is much larger than that generated by current source 12. Therefore, the gain stage 324 is able control Vout by sinking all the current from 12 that flows through T7 and TF0 to Vout. In addition, the compensation capacitor C2 in the gain stage 324 is made large enough to assure the feedback loop is stable and settles in less than about 1 mSec. The level shift in Vefb caused by the Vt across T7 approximately matches the voltage drop in T6. The level shift in Vefb caused by the tunnel voltage across TF0 approximately matches the voltage drop across tunnel device Te0, so that when the differential and gain stages settle, Vfgø and Vout are about the same. This can be seen in Fig. 8 where Vout settles to within about 30mV of Vfgø, beginning at time t₂. This 30mV difference is generated by noise coupled to fgø from the Ipø current source. Specifically, negative charge pump Ipø, which pumps charge from the program tunnel device Tp0, produces noise on Vp. This noise is coupled to floating gate fgø through program tunnel device capacitance Cp0. The noise on Vp cannot be seen in the Vp waveform in Fig. 8 because the voltage axis is shown in volts, whereas the voltage axis for the Vfgø vs. Vsetø waveform is shown in millivolts.

0050Referring again to Fig. 5, once circuit 30 settles at step 54 such that Vfgø is approximately Vsetø, circuit 30 is powered down at step 56. Powering down circuit 30 ramps Vefb and Vp toward ground as seen beginning at t₃ in Figs. 7 and 8. Step 56 may be performed by simply concurrently shutting off the charge pump Ipø and HV+, and thereby current source 12, at time t₃. However, this may significantly impact Vfgø once Vefb and Vp have ramped back to 0V. As explained above, noise from Ipø limits the accuracy of setting Vfgø equal to Vsetø when the negative charge pump that generates Vp is ON. This means Vfgø may not be equal to Vsetø at the beginning of the ramping of Vefb and Vp to ground. If Vfgø is not equal to Vsetø when this ramp down begins, then Vfgø will not equal Vsetø after Vp and Vefb reach 0V. Moreover, during the ramp down, the current that continues to flow through tunnel devices Te0 and Tp0 is typically not the same. This further affects the final charge level on floating gate fgø.

0051To overcome this limitation and thereby maintain the same charge level on floating gate fgø during the ramping of Vefb and Vp to ground, the current in the erase and program tunnel devices must be the same during this time. In order to maintain the same current in both tunnel devices, the voltage across each of the tunnel devices must be the same, which means Vefb must ramp down to 0V at the same rate as Vp ramps up to 0V. Also the tunnel device characteristics must be well matched. Accordingly, circuit 30 should be powered down, at step 56, in the following preferred manner. Once circuit 320 and the feedback circuit have stabilized for a time and it is clear that further accuracy to setting Vfgø is limited primarily by the charge pump noise, shown beginning at t₂, Ipø is shut off at t₃ to eliminate the pump noise. However, HV+, and thereby current source 12, are left on such that the feedback circuit is still active and continues to control Vefb. At the point when the negative charge pump is shut off, tunnel current continues to flow through Te0 and Tp0 as Cp0 discharges, which pulls up Vp back towards 0V. This tunnel current and the capacitance Cp0 determine the ramp rate on Vp. As Vp ramps up, the voltage on floating gate fgø is capacitively coupled upwards. Circuit 320 senses Vfgø moving upwards and ramps Vefb down towards 0V through the feedback circuit. As Vefb ramps down and Vp ramps up, the tunnel current in tunnel devices Te0 and Tp0 decreases rapidly due to the steep slope of their Fowler-Nordheim tunnel device characteristics. Since feedback response time depends directly on the current in the erase tunnel device, the feedback circuit response slows down as Vefb ramps down. As the tunnel current decreases, both the ramp rate and feedback response times slow down and Vfgø gradually moves closer to Vsetø. For instance, Fig. 8 shows that Vfgø has converged to within about 0.5 mV of Vsetø for a set mode time of 30 mSec, and Vfgø may be set even more accurately by allowing a ramp down time of greater than 30 mV. After Vfgø is allowed to converge on Vsetø for an amount of time determined by the level of accuracy desired, the HV+ voltage supply and thereby the 12 current source can be shut off, for instance at t₄, without affecting the charge on fgø. Moreover, Vcc may be shut off. In other words, once Vfgø is detected as being within a predetermined threshold level of Vsetø, a steady state condition has been reached and power to circuit 30 can be shut off without affecting the value of Vfgø. It is important that the response of the feedback circuit is slow enough to assure Vfgø is always slightly above Vsetø so circuit 320 and the feedback circuit continue to ramp Vefb down. If Vfgø goes below Vsetø and the feedback switches the direction Vefb is ramping, the feedback system will start to oscillate very slowly and Vfgø will diverge from Vsetø instead of converge towards Vsetø. After Vefb and Vp have ramped a few volts towards 0V and Vfgø is very close to Vsetø, Vefb and Vp can be ramped to 0V quickly, as illustrated at time t in Fig. 6, by shutting off ^' HV+ because the current in Te0 and Tp0 is so low it no longer affects the charge on the floating gate fgø. Cp0 must be carefully set to assure that as Vp rises to 0V, the feedback path through the differential stage 322, gain stage 324, TF0 level shift and Te0 devices to floating gate fgø is able to ramp down Vefb and move Vfgø closer and closer to Vsetø. If Cp0 is too small: Vp rises very quickly; the delay through the feedback path causes Vefb to ramp down too slowly; and Vfgø will rise above Vsetø instead of converging towards Vsetø. If Cp0 is too large, the response of the feedback path is too fast and Vefb is ramped down too much, such that Vfgø may undershoot which causes the circuit to oscillate slowly. If circuit 320 is allowed to oscillate, Vfgø will tend to diverge instead of converge towards Vsetø. Accordingly, Cp0 is designed such that the feedback response time is slightly slower than the discharge rate of Cp0. Preferably Cp0 should be set at about 2.4pf

0052At the end of the set mode, at time t₄, floating gate fgø will then continue to indefinitely store the charge level programmed on floating gate fgø during the set mode, subject to possible charge loss, e.g., due to detrapping of electrons or dielectric relaxation over time, without any external power being supplied to circuit 30. In addition, although in the example illustrated above Vfgø was set to be equal to Vsetø, those of ordinary skill in the art will realize that in another embodiment of the present invention, circuit 30 can be configured such that Vfgø is set to a voltage that is some other predetermined value of Vsetø.

0053With the above understanding of the differential floating gate circuit 30 of Fig. 3 and of the method 50 of setting floating gate fgø illustrated by the flow diagram in Fig. 5, we now turn to the differential dual floating gate circuit 40 of Fig. 4A. Circuit 40 preferably comprises a reference floating gate fgr at a node 15 and a second floating gate fgl at a node 14. At the conclusion of a set mode, both floating gates fgr and fgl are programmed, respectively, to charge levels such that the difference in charge level between fgr and fgl is a function of an input set voltage capacitively coupled to fgr during the set mode. Thereafter, during a read mode, circuit 40 may be configured as a voltage reference circuit such that an output reference voltage is generated that is a function of the input set voltage and is preferably equal to the input set voltage. The set mode may be instituted at the factory to cause fgr and fgl to be set to their respective desired charge levels, and thereby, to cause circuit 40 to generate a desired output reference voltage whenever circuit 40 is later caused to enter its read mode. Alternatively, a later user of circuit 40 can cause circuit 40 to enter a set mode whenever the user wishes, to thereby update the difference in charge levels between fgr and fgl as a function of the Vsetø voltage input and thus to update the output reference voltage generated by circuit 40 during subsequent read mode.

0054The sequence used to program floating gates fgr and fgl in circuit 40 is similar to the sequence used to set the charge level on floating gate fgø in circuit 30 of Fig. 3. One major difference between the previously described single floating gate circuit 30 and the dual floating gate circuit 40 is that the gate of transistor T2 in Fig. 3 is replaced by a floating gate, fgl, in Fig. 4 A, that cannot be connected directly to an external voltage. In order to set the voltage on fgl, a voltage Vx is coupled at a node 27 to the gate of a transistor T15 in circuit 40, such that Vfgl is set to Vx- IVt- ITV, where 1 Vt is the threshold voltage of transistor T15 and ITV is the tunnel voltage of an erase tunnel device Tel.

0055In a preferred embodiment, Vx is generated by a second floating gate voltage reference circuit, e.g., circuit 30. Fig. 4B is a combined schematic and block diagram illustrating this embodiment. Circuits 30 and 40 in Fig. 4B are identical to the circuits illustrated, respectively, in Figs. 3 and 4A. In the embodiment shown in Fig. 4B, a high voltage set cycle is performed on both the single floating gate differential circuit 30 and the dual floating gate differential reference circuit 40 at the same time. During the set mode, circuit 30 generates the voltage at node 12 such that floating gate fgø is set as described earlier, wherein Vsetø for circuit 30 is an internally or externally supplied predetermined voltage, such as +4v. Floating gate fgl is therefore set to a voltage that is a predetermined function of the voltage on floating gate fgø, and is preferably set to be approximately equal to Vfgø assuming the tunnel devices in both differential circuits, i.e., circuits 30 and 40, are reasonably well matched. The voltage set on floating gate fgl is then used to set the voltage on floating gate fgr, such that Vfgr is a predetermined function of Vfgl, and preferably approximately equal to Vfgl, as described in greater detail below.

0056Circuit 40 further comprises a circuit 410 that includes: a programming tunnel device Tpr formed between floating gate fgr and a programming electrode Epr, at a node 16; an erase tunnel device Ter formed between floating gate fgr and an erase electrode Eer, at a node 17; and a steering capacitor Cfgr coupled between floating gate fgr and a node 18. Circuit 40 also comprises a circuit 420 that includes: a programming tunnel device Tpl formed between floating gate fgl and a programming electrode Epl, at node 16, and an erase tunnel device Tel formed between floating gate fgl and an erase electrode Eel, at a node 28. Preferably, programming electrodes Epr and Epl receive a negative voltage during the set mode, and erase electrodes Eer and Eel receive a positive voltage during the set mode. Moreover, tunnel devices Tpr, T l, Ter and Tel are preferably Fowler-Nordheim tunnel devices that are reasonably well matched as a result of their chip layout, and these tunnel devices are ideally reasonably well matched with tunnel devices Tp0 and Te0 of circuit 30. Also included in circuit 40 is a steering capacitor Cfgl coupled between floating gate fgl and a node 32. The bottom plate of steering capacitor Cfgl is coupled to a predetermined voltage during the set mode that is preferably ground gl. Steering capacitor Cfgl is used to provide a stable ground reference for floating gate fgl . Circuit 40 also includes a transistor TI 5 that has its drain coupled to a high voltage supply HV+, at a node 26, its source coupled to node 28, and its gate coupled to node 27.

0057Setting a voltage on floating gate fgr during the set mode is achieved by taking electrode Epr negative and electrode Eer positive such that the voltage at node 17 minus the voltage at node 16 is two tunnel voltages or approximately 22V. The dual conduction current at 22V is typically approximately one to two nanoamps. An alternative is to create a sufficient voltage differential across electrode Epr and electrode Eer to generate a current flow of approximately 5nA from node 16 to node 17. In either case, both tunnel devices are conducting, i.e., the tunnel devices are in "dual conduction." By operating in dual conduction, the voltage on the floating gate fgr can stabilize at a DC voltage level for as long a time as needed to enable circuit 40 to end the set mode process in a controlled fashion such that the voltage on floating gate fgr settles to a very precise and accurate level. Operating in dual conduction with feedback through at least one of the tunnel devices is key to making it possible to set the floating gate fgr voltage very accurately. In dual conduction, the tunnel devices Ter and Tpr, which are reasonably well matched by layout, will modify the charge level on floating gate fgr by allowing electrons to tunnel onto and off of floating gate fgr so as to divide the voltage between nodes 17 and 16 in half. Thus, the floating gate voltage, i.e., the voltage at node 15, will be Vfgr = Vnodelό + (Vnodel7-Vnodel6)/2, which is half way between the voltage at node 17 and the voltage at node 16. Under these conditions, the dual conduction current can typically charge or discharge node 15, which typically has less than 1.0 pF capacitance, in less than 1 mSec. As this occurs, the floating gate voltage "tracks" directly with the voltage at nodes 16 and 17 and settles to a DC voltage that is halfway between those two voltages in a few mSec. Accordingly, Vfgr can be set to a positive or negative voltage or zero volts depending upon the value of the voltages existing at electrodes Eer and Epr. For example, if the tunnel voltage is approximately 1 IV for the erase and program tunnel devices Ter and Tpr, and the voltage at electrode Eer is set to about +16V and the voltage at electrode Epr is set to about -6V, then Vfgr will settle at about +5V, which is the midpoint between the two voltages. If the voltage at Eer is set to about +1 IV and the voltage at Epr is set to about -11 V, then Vfgr will go to about 0 V. If the voltage at Eer is set to about +6V and the voltage at Epr is set to about -16V, then Vfgr will go to about -5V.

0058As stated earlier, circuit 40 programs both floating gates fgr and fgl during the set mode. Correspondingly, tunnel devices Tpl and Tel similarly operate in dual conduction to modify the charge level on floating gate fgl by allowing electrons to tunnel onto and off of floating gate fgl so as to divide the voltage between nodes 28 and 16 in half. In addition, if circuit 30 is used during the set mode to generate the voltage Vx at node 27 in circuit 40, ideally, the tunnel currents in both circuits 30 and 40 are reasonably well matched, and transistors T13, T14, T15 are reasonably well matched, such that when circuits 30 and 40 settle, Vfgr = Vfgl = Vfgø. Although this condition is preferable, circuit 40 will set Vfgr = Vfgl even where floating gate fgl is not set exactly equal to floating gate fgø, since floating gates fgl and fgø are not in the same differential circuit.

0059Circuit 40 further includes a circuit 430 that compares Vfgr, the voltage on floating gate fgr to Vfgl, the voltage on floating gate fgl, and that generates an output voltage Vout, at node 19, that is a function of the difference between the voltages on floating gates fgr and fgl . Circuit 430 preferably includes a differential amplifier (or differential stage) 432 that is preferably configured to have a non-inverting input coupled to floating gate fgl and an inverting input coupled to floating gate fgr. Circuit 430 further includes a gain stage 434 with an input coupled to node 20 and an output terminal 436, at node 19. The differential stage 432 compares the voltages received at its inputs and amplifies that difference, typically by a factor of 50 to 100. The gain stage 434 then further amplifies that difference by another factor of 50 to 100. Moreover, at the conclusion of the set mode, Circuit 430 ideally settles to a steady state condition, such that Vfgr = Vfgl = Vout. Referring again to Fig. 4B, the differential stage 432 preferably includes enhancement mode transistors T8, T9, T10 and TI 1. Transistors T8 and T9 are preferably NMOS transistors that are reasonably well matched by layout, and transistors T10 and TI 1 are preferably PMOS transistors that are reasonably well matched by layout. The sources of NMOS transistors T8 and T9 are coupled together at a node 21. The drain of NMOS transistor T8 is coupled to a node 22, and its gate is floating gate fgr. The drain of NMOS transistor T9 is coupled to a node 20, and its gate is floating gate fgl. PMOS transistor T10 is coupled common drain, common gate, to node 22, with its source coupled to a node 23. The gate of PMOS transistor TI 1 is coupled to at node 22. Its drain is coupled to node 20, and its source is coupled to node 23. A voltage supply Vcc, typically 3 to 5 volts, is coupled to node 23, and a current source Itr is coupled between node 21 and ground gl to cause transistors T8, T9, T10 and TI 1 to operate in either the pretheshold or linear region during the set mode. Current source Itr can be generated using any number of conventional circuits. The gain stage 434 preferably includes a PMOS pullup transistor T12 biased by Vcc and a current source pull-down load Igr. The source of transistor T12 is coupled to node 23. Its gate is coupled to the differential stage pull-up transistor TI 1 at node 20, and its drain is coupled to node 19. Current source pull-down load Igr is coupled between node 19 and ground gl. The gain stage 434 also preferably includes a compensation capacitor C3 coupled between nodes 19 and 20. Current source pulldown load Igr is preferably an active load using an NMOS current mirror or a depletion device. Using an active current source with relatively high output resistance, the gain stage 434 can provide a voltage gain of about 100. The output swing of the gain stage 434 is nearly full rail from ground to Vcc. Stability and response of this circuit can be easily adjusted for various processes using compensation capacitor C3. In this configuration, transistor T12 provides good current sourcing capability, but current sinking is limited to the current in the current source pull-down Igr. Therefore, the current in Igr should be greater than the pull-up current required by the load on Vout so that the gain stage 434 is capable of adequately controlling Vout by sinking all of the current that flows to Vout.

0060Circuit 430 further operates in the following manner. When biased by Vcc and current source Itr, T8 senses Vfgr relative to Vfgl , which is sensed by transistor T9, and the amplified difference appears as Vout at node 19. If Vfgr is initially less than Vfgl, T9 is turned on more than T8, and the current flow through T9 (and through TI 1 since they are connected in series) is initially greater than the current flow through T8 (and correspondingly T10). The gate of the pullup transistor T10 is tied to the drain of TI 0 and also to the gate of pullup transistor Ti l, which makes the current in TI 1 a mirror of the current in transistor T10. When more current flows through TI 1 than T10, the voltage, V20, on node 20 drops below the voltage V22, on node 22. The lower voltage on node 20 causes the current through transistor T12 to increase, which pulls Vout high. The voltage gain of the differential stage 432 is typically about 80 and the voltage gain of the gain stage 434 is typically about 100 giving an overall gain from Vfgl to Vout of about 8000. A negative feedback path from Vout to the inverting input fgr is necessary for circuit 430 to settle at the point where the voltage on fgr is equal to the voltage on fgl . During the set mode, this feedback path is provided by tunnel devices TF1 and Ter and transistors T13 and T14 as described in the next section. When Vout goes high, the negative feedback path pulls Vfgr higher. As Vfgr rises, the current in transistor T8 increases until it matches the current in transistor T9. At this point the differential circuit 430 settles at the point where the currents in transistors T8, T9, T10 and TI 1 match and Vfgr = Vfgl .

0061Those skilled in the art will realize that circuit 430 can be implemented using PMOS transistors for T8 and T9 and NMOS transistors for T10 and TI 1. For this implementation, the gain stage 434 preferably comprises an NMOS pull-down transistor T12 coupled to a current source pull-up load Igr.

0062Circuit 40 also includes a feedback loop coupled between nodes 19 and 15. During the set mode, this feedback loop causes the voltage differential between tunnel electrodes Eer and Epr to be modified by modifying the voltage at node 17 as a function of the voltage at node 19. The feedback loop preferably comprises a level shift circuit, preferably a tunnel device TFl formed between node 19 and a node 24, and a transistor T14, preferably an NMOS transistor, coupled common gate, common drain at a node 25, with its source coupled to node 24. Also included in the feedback loop is a transistor T13, preferably an NMOS transistor, having its gate coupled to node 25, its source coupled to node 17, and thereby to erase tunnel device Ter, and its drain coupled to node 26.

0063As earlier indicated, the maximum output of the gain stage 434 is approximately Vcc. However, this is not high enough to drive the voltage at node 25 (Vefb) directly, because Vefb typically needs to go to about 14 to 19 volts, which is well above the usual 3 to 5 volt Vcc supply level. The level shift circuit TFl and T14 shifts the low output voltage at node 19 (Vout) up to the desired 14 to 19 volt range. Preferably, TFl and Ter are reasonably well matched by layout and T13 and T14 are reasonably well matched by layout. Under these conditions, when the same tunnel current flows through both TFl and Ter, the level shift tracks the erase tunnel voltage as measured by the voltage drop from node 17 to node 15 which drives the gate of transistor T8 (fgr) to the same voltage as the voltage on the gate of transistor T9 (fgl)when circuit 430 settles. This adds to the improved setting accuracy of the circuit.

0064One advantage of having the level shift track the erase tunnel voltage is that, as the voltage necessary to create tunneling changes, due to charge trapping in the dielectric as more and more set cycles are performed, the circuit 430 output, Vout, continues to follow Vfgl and operate in the same voltage range. Another advantage is that when the output voltage Vout is not equal to Vfgr, the error introduced by the finite gain of circuit 430 is very small. For example, if circuit 430 has a gain of 10,000 and Vout is 1 volt lower than Vfgl minus Vfgr when circuit 40 settles, Vfgl minus Vfgr will have an error of 1V/10,000, or only O.lmV.

0065Circuit 40 also preferably includes current sources I2r and Ipr, and a capacitor Cpr. Current source I2r is coupled between node 25 and HV+ at node 26 for establishing Vefb at the beginning of the set mode and for providing tunnel current through TFl . Current source I2r can be implemented using any number of conventional circuits. However, current source I2r is preferably a current regulator that is biased by HV+, such as a current mirror comprising P-Channel devices that operate in the prethreshold region. In this manner, current source I2r will automatically go to whatever positive voltage is needed at node 25 to establish the tunnel current through tunnel device TFl . Moreover, current source I2r preferably generates a current that is about half that of current source Ipr, so that the current through tunnel device TFl is about the same as the current through tunnel devices Ter, Tpr, Tel, and Tpl.

0066Current source Ipr is coupled between node 16 and ground gl . Current source Ipr is preferably a P-Channel charge pump that is used as a negative current source to pump a controlled tunnel current out of programming tunnel devices Tpr and Tpl . Since Ipr is a current source, it automatically goes to whatever negative voltage at node 16 that is needed to establish the tunnel current at the desired level, assuming the current source has sufficient voltage compliance. Moreover, once the current through the tunnel devices is established, the voltage across the tunnel devices is also well defined by their Fowler-Nordheim characteristics. Therefore, current source Ipr produces Vpl, the voltage at node 16, by controlling the current through tunnel devices Tpr and Tpl. Using a current source Ipr is the preferred way to assure that tunnel devices Ter, Tel, Tpr and Tpl are operating at a current level that is high enough to allow dual conduction and to allow the feedback circuit to work, but low enough to avoid excessive current flow which damages the tunnel devices. Capacitor Cpr, controls the rate of discharge of current through the tunnel devices when, as explained in more detail below, current source Ipr is shut down at the conclusion of the set mode. Moreover, when circuit 30 is used to generate the voltage Vx at node 27 in circuit 40 during the set mode, to achieve the ideal condition of setting Vfgr = Vfgl = Vfgø, preferably current sources I2r and 12 (of Fig. 3) are reasonably well matched, current source Ipr is about twice as large as current source Ipø (of Fig. 3), and capacitors Cpr and Cp0 (of Fig. 3) are reasonably well matched. In addition, HV+ is the same in circuit 30 and in circuit 40. Those skilled in the art will realize that Vpl can also be produced using a fixed voltage supply that is about 24 to 30 volts below the voltage at nodes 17 and 28. However, this topology should be used with caution because the current in Fowler-Nordheim tunnel devices varies exponentially with the applied voltage. In particular, very high current will flow through the tunnel devices if the voltage differential is too high, and extremely low current may flow if the voltage differential is too low. Very high currents will damage or "wear out" the tunnel devices due to rapid charge trapping in the dielectric, and if the tunnel current is too low, the feedback circuit will not be able to tunnel charge onto or off of fgr, and thus will not be able to control the voltage on fgr. Moreover, it is also possible to connect erase electrode Eer to a current source and connect programming electrode Epr to the feedback circuit such that Vpl controls the voltage on fgr. However, this would require the feedback circuit to produce a controlled negative voltage, which is more difficult to integrate in a standard CMOS process.

0067Finally, circuit 40 also preferably includes a circuit 440. Circuit 440 preferably comprises a switch S4 that is preferably a MOS transistor that is coupled between nodes 18 and 19 and a MOS transistor switch S5 coupled between node 18 and an input voltage terminal 450. In the set mode, switch S4 is OFF, and switch S5 is ON such that the input set voltage Vset can be coupled to the bottom plate of steering capacitor Cfgr.

0068Coupling input voltage Vset to terminal 450 during the set mode enables circuit 40 to program a charge level difference between floating gates fgr and fgl that is a predetermined function of Vset. Thereafter during a subsequent read mode, circuit 40 generates a reference voltage that is a predetermined function of Vset, and is preferably equal to Vset. Specifically, during the set mode, the voltage programmed across capacitor Cfgl is the same as that programmed on floating gate fgl, since Cfgl is preferably coupled to ground during the set mode. Whereas, the voltage programmed across capacitor Cfgr is Vfgr (which is ideally equal to Vfgl) minus Vset. Thereafter, when power and Vset are removed at the conclusion of the set mode, node 18 goes to zero volts and Vfgl remains the same, but Vfgr is equal to the voltage across Cfgr, which is equal to (Vfgl -Vset). Thus, a difference in charge level exists between floating gates fgr and fgl that is equal to the difference in charge remaining on capacitors Cfgl and Cfgr at the conclusion of the set mode. This difference in charge level between fgr and fgl, which is a predetermined function of Vset, is what causes a reference voltage to be generated at node 19 during a read mode for circuit 40 that is a predetermined function of Vset, and is preferably equal to Vset. To produce a voltage reference output equal to Vset, S5 is turned off and S4 is turned on, which connects Vset to node 18, which is coupled to fgl through Cfgr. Vout settles at the voltage where Vfgr = Vfgl , which occurs when node 18 = Vset.

0069Fig. 9 is a flow diagram illustrating a method 90 for setting a floating gate that may be implemented during a set mode, for instance, by circuits 30 and 40 of Fig. 4B. Figs. 10-12 illustrate voltage waveforms for Vout, Vpl, Vefb (circuit 40), Vfgr and Vfgl, for the specific implementation of method 90 discussed below relative to those figures. Each of the four waveforms shown in Figs. 10-12 are the same, only the voltage axes of some of these waveforms are modified to illustrate specific details. Preferably, Vfgl is set to 4 volts, such that Vfgl = Vfgr = 4V at the conclusion of the set mode. However, Vfgl may be set to any voltage in order to set Vfgr during the set mode. In the following example, Vfgl is set to 4V during the set mode. In the circuit implementation illustrated in Figs. 10-12: Vin = 4.00V, Vcc = +5V, HV+ is about 22N, Ipø, 12 and I2r are each about 6nA, Ipr is about 12nA, It0 and Itr are each about 5nA; and Ig0 and Igr are each about 20nA.

0070At step 91, circuits 30 and 40 are powered up at the beginning of the set mode, which is illustrated in Figs. 6-8 and Figs. 10-12 as time t₀. Circuit 30 at some point thereafter receives an input set voltage, e.g., Vsetø, and the Vx signal from circuit 30 is received at node 27 into the gate of transistor T15 in circuit 40. In addition Vcc is set to +5V, HV+ is ramped up to a high positive voltage of about +22V, which turns on current sources 12 and I2r. Finally, charge pumps Ipø and Ipr are turned on to enable these current sources to begin generating their corresponding currents. Thereafter, according to the preferred implementation of the remaining steps 92-96 of method 90, circuit 40 can set Vfgr to within about 0.5 mV of Vfgl in about 30 mSec, as illustrated in Figs. 10-12.

0071At step 92, circuit 40 causes tunnel devices Ter, Tpr, Tel and Tpl to operate in a dual conduction mode under the control of the voltage differential between the corresponding floating gate erase and program electrodes for modifying the charge level on floating gates fgr and fgl . Dual conduction occurs when tunnel current flows through these four tunnel devices. Tunnel current flows through both Ter and Tpr when the voltage differential (Vefb-Vpl) is at least two tunnel voltages or approximately 22V as discussed earlier, and tunnel current flows through Tel and Tpl when the voltage differential (Vx-Vpl) is at least two tunnel voltages. Preferably, circuit 40 causes dual conduction in the following manner.

0072Current sources 12 and I2r are turned on and start to pull up Vx (node 12) and Vefb (node 25) respectively. For example, Vefb ramps up to about 18 volts in less than 0.5 mSec. The negative current sources Ipø and Ipr are turned on and pull Vp (node 3) and Vpl (node 16) negative. Respectively, in this instance, charge pump Ipø gradually ramps Vp down to about -1 IV voltage in about 2 mSec, and charge pump Ipr gradually ramps Vpl down to about -1 IV voltage in about 2 mSec. Current source Ipø controls the tunnel current that flows through tunneling devices Tp0 and Te0 in circuit 30, and current source Ipr controls the tunnel current that flows through tunneling devices Ter, Tpr, Tel and Tpl in circuit 40. Circuit 30 produces a Vx signal controlled by feedback from circuit

0073320 as described earlier. Vx (node 27) turns on transistor TI 5, which pulls up Vel (node 28) to one Vt below Vefb. When Vpl ramps down to the point where the difference between Vpl and Vel is 2 tunnel voltages, tunnel current flows through tunneling devices Tel and Tpl. Once tunnel current is flowing in Tel and Tpl, the voltage on floating gate fgl (node 14) is controlled directly by Vx and to first order tracks the voltage on floating gate fgø in circuit 30 for the rest of the set mode. Circuit 40 produces a Vefb signal controlled by feedback from circuit 430 in a manner analogous to circuit 30. Vefb (node 25) turns on transistor T13, which pulls up Ver (node 17) to one Vt below Vefb. When Vpl (node 16) ramps down to the point where the difference between Vpl and Ver is 2 tunnel voltages, tunnel current flows through tunneling devices Ter and Tpr, and the voltage on fgr (node 15) is controlled directly by Vefb. I2r continues to pull up Vefb until Vefb reaches Vout + ITV + IVt, where ITV is the tunnel voltage across tunnel device TFl and IVt is the threshold voltage of transistor T14. When at least one tunnel voltage exists across TFl, tunnel current flows through TFl, and TFl and transistor T14 act as level shift devices such that Vefb is controlled directly by Vout (node 19). At step 93, circuit 40 compares Vfgr with Vfgl and generates an output voltage Vout that is a function of the difference between Vfgr and Vfgl . Circuit 40 then, at step 95, causes the voltage differential between Vefb and Vpl to be modified as a function of Vout, and circuit 40 repeats steps 92 through 95 until circuit 40 settles to a steady state condition, at step 94 where Vfgr is approximately equal to Vfgl . At this point circuit 40 is powered down, at step 96. As a result of method 90, floating gates fgr and fgl are each set to a charge level that will remain essentially the same over time.

0074The voltage waveforms of Figs. 10-12 illustrate how circuit 40 functions during steps 92 through 95. Dual conduction of tunnel devices Tel and Tpl occurs after about 0.5 mSec, as best seen in Fig. 10. Prior to this time, Vfgl is zero volts. However, once tunnel current is flowing through tunnel devices Tel and Tpl, Vfgl is controlled by and oscillates with Vx from circuit 30, and Vfgl tracks Vfgø. Dual conduction of tunnel devices Ter and Tpr, on the other hand, occurs slightly later at about 1.5 mSec, which is illustrated as ti in Figs. 10-12. Prior to time t] Vout = 0 V, Vefb is pulled-up by I2r and is ramping toward about 18 V, and Vfgr is not controlled by Vefb. Once tunnel current is flowing through tunnel devices Ter, Tpr, and TFl at time ti: circuit 430 senses that Vfgr is not equal to Vfgl ; Vout is a function of the difference between Vfgr and Vfgl; Vefb follows Vout; and Vfgr follows Vefb. For about the next 2.0 mSec which is illustrated as time ti to time t₂ in Figs. 11 and 12, Vfgr oscillates as Vefb moves up and down as a function of the negative feedback loop. Thereafter, the negative feedback loop causes the differential and gain stages 432 and 434, respectively, to settle to a steady state condition, where circuit 430 ceases to oscillate except for about 30mV of noise coupled to circuit 430 from the charge pump Ipr as best shown in Figs. 11 and 12 beginning at time t₂.

0075Beginning at time ti, current source Igr in the gain stage 434 produces a current that is much larger than that generated by current source I2r. Therefore, the gain stage 434 is able to control Nout by sinking all the current from current source I2r that flows through T14 and TFl to Nout. In addition, the compensation capacitor C3 in the gain stage 434 is made large enough to assure that the feedback loop is stable and settles in less than about 1 mSec. The level shift in Nefb caused by the Nt across transistor T14 approximately matches the voltage drop in T13. The level shift in Nefb caused by the tunnel voltage across tunnel device TFl approximately matches the voltage drop across tunnel device Ter, so that when the differential and gain stages settle, Nfgr, Vfgl and Vout are about the same. This can be seen in Fig. 12 where Vout settles to about 3.7V beginning at time t₂, reflecting about 30 mV of noise coupled to floating gates fgr and fgl from current source Ipr.

0076Referring again to Fig. 9, once circuit 40 settles at step 94 such that Vfgr is approximately equal to Vfgl, circuit 40 is powered down at step 96. Powering down circuit 40 ramps down the voltages at the erase and programming electrodes toward ground, as seen beginning at time t₃ in Figs. 10-12. Step 96 may be performed by simply concurrently shutting off all of the current and voltage sources in circuits 30 and 40 at time t₃. However, this may significantly impact Vfgr once Vefb and Vpl have ramped back to 0V. As explained above, noise from charge pump Ipr limits the accuracy of setting Vfgr equal to Vfgl when the negative charge pump that generates Vpl is ON. This means Vfgr may not be equal to Vfgl at the beginning of the ramping of Vefb and Vpl to ground. If Vfgr is not equal to Vfgl when this ramp down begins, then Vfgr will not equal Vfgl after Vpl and Vefb reach 0V. Moreover, during the ramp down, the current that continues to flow through tunnel devices Tel and Tpl and through Ter and Tpr is typically not the same. This further affects the final charge level on floating gates fgr and fgl . To overcome this limitation and thereby maintain the same charge level on floating gates fgr and fgl during the ramping of Vefb and Vpl to ground, the current in the erase and program tunnel devices must be the same during this time. In order to maintain the same current in these tunnel devices, the voltage across each of the tunnel devices must be the same, which means Vefb and Vx must ramp down to 0V at the same rate as Vpl ramps up to 0V. Also the tunnel device characteristics must be reasonably well matched.

0077Accordingly circuit 40 should be powered down, at step 96, in the following preferred manner. Once circuits 320 and 430 and the feedback circuits in both circuits 30 and 40 have stabilized for a time and it is clear that further accuracy to setting Vfgø, Vfgr and Vfgl is limited primarily by the charge pump noise, shown beginning at t , Ipø and Ipr are shut off at t₃ to eliminate the pump noise. However, HV+, and thereby current sources 12 and I2r, are left on such that the feedback circuit in circuit 30 is still active and continues to control Vx, and the feedback circuit in circuit 40 is still active and continues to control Vefb. At the point when the negative charge pumps are shut off, tunnel current continues to flow through tunnel devices Te0 and Tp0 as capacitor Cp0 discharges, which pulls up Vp back towards 0V. This tunnel current and the capacitance due to Cp0 determine the ramp rate on Vp. Similarly, tunnel current continues to flow through tunnel devices Ter, Tel, Tpr and Tpl as capacitor Cpr discharges, which pulls up Vpl back towards 0V. This tunnel current and the capacitance due to Cpr determine the ramp rate on Vpl . Feedback in circuit 30 drives Vx such that Vfgø is set as described previously. To first order, Vfgl tracks Vfgø, assuming Vp and Vpl track each other reasonably closely. Similarly to what occurs in circuit 30, in circuit 40 as Vpl ramps up, the voltage on floating gate fgr is capacitively coupled upwards. Circuit 430 senses Vfgl moving upwards and ramps Vefb down toward 0V through the feedback circuit. As Vefb ramps down and Vpl ramps up toward 0V, the tunnel current in tunnel devices Ter and Tpr decrease rapidly due to the steep slope of their Fowler- Nordheim tunnel device characteristics. Since feedback response time depends directly on the current in the erase tunnel device, the feedback circuit response slows down as Vefb ramps down toward ground. As the tunnel current decreases, both the ramp rate and feedback response times slow down and Vfgr gradually moves closer to Vfgl. For instance, Fig. 12 shows that Vfgr has converged to within about 0.5 mV of Vfgl for a set mode time of 30 mSec, and Vfgr may be set even more accurately with respect to Vfgl by allowing a ramp down time of greater than 30mV. After Vfgr is allowed to converge on Vfgl for an amount of time determined by the level of accuracy desired, the HN+ voltage supply, and thereby the I2r current source, can be shut off, for instance at time t , without affecting the charge on floating gates fgr and fgl . Moreover, Vcc may be shut off.

0078It is important that the response of the feedback circuit is slow enough to assure Vfgr is always slightly above Vfgl so circuit 430 and the feedback circuit continue to ramp Vefb down. If Vfgr goes below Vfgl and the feedback switches the direction Vefb is ramping, the feedback system will start to oscillate very slowly and Vfgr will diverge from Vfgl instead of converge towards Vfgl . After Vefb and Vpl have ramped a few volts toward ground and Vfgr is very close to Vfgl, Vefb and Vpl can be ramped to 0V quickly, as illustrated at time t₄ in Fig. 10, by shutting off HV+, because the current in tunnel devices Ter and Tpr is so low it no longer affects the charge on the floating gate fgr. Capacitor Cpr must be carefully set to assure that as Vpl rises toward ground, the feedback path through the differential stage 432, gain stage 434, TFl level shift and Ter devices to floating gate fgr is able to ramp down Vefb and move Vfgr closer and closer to Vfgl. If capacitor Cpr is too small, Vpl rises very quickly, the delay through the feedback path causes Vefb to ramp down too slowly, and Vfgr will rise above Vfgl instead of converging towards Vfgl . If Cpr is too large, the response of the feedback path is too fast and Vefb is ramped down too much, such that Vfgr may undershoot which causes the circuit to oscillate slowly. If circuit 430 is allowed to oscillate, Vfgr will tend to diverge instead of converge towards Vfgl . Accordingly, Cpr is designed such that the feedback response time is slightly slower than the discharge rate of Cpr. Preferably Cpr should be set at about 2.4pf.

0079At the end of the set mode, at time t₄, floating gates fgr and fgl will continue to indefinitely store the charge level programmed on them during the set mode, subject to possible charge loss, e.g., due to detrapping of electrons or dielectric relaxation over time, without any external power being supplied to circuit 40. In addition, although in the example illustrated above Vfgr was set to be approximately equal to Vfgl, those of ordinary skill in the art will realize that in another embodiment of the present invention, circuit 40 can be configured such that Vfgr is set a voltage that is some other function of Vfgl. As stated above, once floating gate fgø is set during the set mode, circuit 30 may be configured during a read mode as a voltage reference circuit or as a comparator circuit with a built-in voltage reference. Likewise, once floating gates fgl and fgr are set during the set mode, circuit 40 may be configured during a read mode as a voltage reference circuit or a comparator circuit with a built-in voltage reference. When circuit 40 is configured as a voltage reference, it provides a more accurate reference voltage at node 19 over that provided by circuit 30 when circuit 30 is configured as a voltage reference. This is because when high voltages are ramped down in circuit 40, any offsets coupled through the tunnel devices to the corresponding floating gates fgr and fgl are common mode and do not change the voltage difference between the two floating gates and thus does not change the reference voltage at node 19.

0080The differential dual floating gate voltage reference circuit and method for programming described in the text above was chosen as being illustrative of the best mode of the present invention. All embodiments of the present invention described above are illustrative of the principles of the invention and are not intended to limit the invention to the particular embodiments described. Accordingly, while the preferred embodiment of the invention has been illustrated and described, it will be appreciated that various changes can be made therein without departing from the spirit and scope of the invention as claimed.