Method and integrated circuit for bandgap trimming

Information

  • Patent Grant
  • 6590372
  • Patent Number
    6,590,372
  • Date Filed
    Tuesday, February 19, 2002
    22 years ago
  • Date Issued
    Tuesday, July 8, 2003
    21 years ago
Abstract
An integrated circuit for generating a bandgap reference voltage (VBG) includes a first circuit and a second circuit. The first circuit includes an op-amp for equalizing emitter currents of a first bandgap transistor and a second bandgap transistor. The second circuit trims out error in at least one emitter current to achieve a desired frequency tolerance. The second circuit includes at least a single transistor digital to analog converter (DAC).
Description




This application relates to copending applications entitled “Method and Apparatus For Light to Frequency Conversion”, inventor William W. Wiles, Jr. (Attorney Docket 26783.18) and “Method and Integrated Circuit For Temperature Coefficient Compensation”, inventor William W. Wiles, Jr. (Attorney Docket 26783.20), filed concurrently herewith, assigned to the assignee of the present disclosure, and incorporated herein by reference.




BACKGROUND




The present invention relates generally to semiconductor devices, and more particularly, to a method and integrated circuit for bandgap trimming.




A number of optoelectronic systems applications require an accurate measurement of the intensity of a light beam or incident light, the measurement being performed over large ranges of input signal amplitude and wavelength. It is desired that the measurement accuracy be maintained over a wide range of environmental conditions. In addition, it is desired that the measurement be made in a minimal volume as dictated by packaging considerations and at a cost commensurate with consumer type systems.




Typical system requirements dictate that an incident light intensity needs to be converted to a digital form for use by a digital processor. To accomplish this, the light intensity is converted to an electrical form that can be digitized. For example, one technique applicable to sampled data processor systems includes converting the light intensity to a voltage that can be applied to an A/D Converter (ADC). This method however requires an additional analog block (i.e., the ADC), but can be used in those applications requiring a higher bandwidth. A second technique, applicable to very low bandwidth applications, includes directly converting the light intensity into a frequency that can be counted by a digital processor.




U.S. Pat. No. 5,850,195 entitled “MONOLITHIC LIGHT-TO-DIGITAL SIGNAL CONVERTER,” issued Dec. 15, 1998, discloses a converter having a switched capacitor oscillator in which the reference function is included in the oscillator circuit. The switched capacitor oscillator requires that calibration be accomplished by trimming the capacitors of the oscillator. However, the oscillator has a high level of parasitic capacitance which limits its high frequency performance. In addition, certain applications, including infrared incident intensity, for example, require a different temperature coefficient which cannot be implemented in an efficient manner in the oscillator circuit of the '195 converter.




It would be desirable to provide a method and integrated circuit for bandgap trimming for overcoming the problems in the art.




SUMMARY




According to one embodiment of the present disclosure, an integrated circuit for generating a bandgap reference voltage (VBG) includes a first circuit and a second circuit. The first circuit includes an op-amp for equalizing emitter currents of a first bandgap transistor and a second bandgap transistor. The second circuit trims out error in at least one emitter current to achieve a desired frequency tolerance. The second circuit includes at least a single transistor digital to analog converter (DAC). A method for generating a bandgap reference voltage (VBG) is also disclosed.











BRIEF DESCRIPTION OF THE DRAWINGS





FIG. 1

is an electrical block diagram of a light-to-frequency converter according to one embodiment of the present disclosure;





FIG. 2

is an electrical block diagram of a current controlled oscillator of the light to frequency converter of

FIG. 1

, according to another embodiment of the present disclosure;





FIG. 3

is an electrical schematic diagram of the one shot circuit of the current controlled oscillator of

FIG. 2

;





FIG. 4

is an electrical schematic diagram of the non-overlapping clock generator of

FIG. 2

;





FIG. 5

is an electrical block diagram of the programmable gain amplifier of

FIG. 1

, according to one embodiment of the present disclosure;





FIG. 6

is an electrical schematic diagram of the decoder of

FIG. 5

;





FIG. 7

is an electrical schematic diagram of the temperature coefficient generator of

FIG. 1

;





FIG. 8

is an electrical schematic diagram of the current proportional to absolute temperature (IPTAT) current generator of

FIG. 7

;





FIG. 9

is an electrical block diagram of the trimmable bandgap generator of

FIG. 1

, according to one embodiment of the present disclosure;





FIG. 10

is an electrical schematic diagram of the bandgap trim DAC of

FIG. 9

;





FIG. 11

is an electrical block diagram of a current controlled oscillator of the light to frequency converter of

FIG. 1

, according to another embodiment of the present disclosure;





FIG. 12

is an electrical schematic diagram of a photodiode with current buffer of

FIG. 11

, according to another embodiment of the present disclosure;





FIG. 13

is an electrical block diagram of a plurality of photodiodes with current buffer of

FIG. 11

, according to another embodiment of the present disclosure;





FIG. 14

is an electrical schematic diagram of the diode multiplexer of

FIG. 13

, according to another embodiment of the present disclosure; and





FIG. 15

is a block diagram view of an apparatus including a light-to-frequency converter, according to another embodiment of the present disclosure.











DETAILED DESCRIPTION




With reference to

FIG. 1

, a light-to-frequency converter according to one embodiment of the present disclosure is referred to, in general, by the reference numeral


10


. The light-to-frequency converter


10


includes a current controlled oscillator (Oscillator)


12


, a programmable amplifier (PGA)


14


, a temperature coefficient generator (TC_Gen)


16


, and a bandgap voltage generator (Reference)


18


. Briefly, the bandgap voltage generator


18


outputs a bandgap reference voltage (VBG). Temperature coefficient generator


16


outputs the bandgap reference voltage with temperature coefficient compensation (VBG_TC), in response to receiving VBG on its input. Programmable gain amplifier


14


outputs an oscillator reference voltage (VREF) in response to VBG_TC. Lastly, current controlled oscillator


12


includes at least one photodiode


20


. In response to VREF, a photodiode control current, and a feedback current, the current controlled oscillator


12


produces an output signal having a frequency (F


O


) proportional to an intensity of incident light upon the photodiode, as will be discussed further below.




In one embodiment, the current controlled oscillator


12


includes a switching capacitor configured to provide a feedback current. With respect to the switched capacitor feedback current, the bandgap reference voltage with temperature coefficient compensation (VBG_TC) modifies a temperature coefficient of the feedback current to match a temperature coefficient of the photodiode control current.




Referring now to

FIG. 2

, according to one embodiment, the current controlled oscillator


12


includes at least one photodiode


20


, an integrator


22


, a comparator


24


, a one shot


26


, a non-overlapping clock generator


28


, and a switched capacitor resistor configuration


30


. Photodiode


20


provides a source of controlling current (I


D


). The control current (I


D


) is directly proportional to an intensity of light


32


incident on the diode


20


. Integrator


22


provides for the summing of the control current (I


D


) against a feedback current (I


R


). Switched capacitor resistor configuration


30


generates the feedback current (I


R


) with the application of VREF to the switched capacitor resistor operating at a frequency of operation (F


O


).




The switched capacitor resistor configuration


30


includes the switching capacitor (CQ) identified by reference numeral


34


and MOS switches (MN


2


-MN


5


) identified by reference numerals


36


,


38


,


40


, and


42


, respectively. Switched capacitor resistor configuration


30


has a first input coupled to a first non-overlapping clock output


50


of non-overlapping clock generator


28


, a second input coupled to a second non-overlapping clock output


52


of the non-overlapping clock generator


28


, and a voltage reference input coupled to the oscillator reference voltage (VREF), identified by reference numeral


92


. In one embodiment, switching capacitor (CQ)


34


includes a single untrimmed capacitor.




Responsive to the first, second, and voltage reference inputs, the switched capacitor resistor configuration


30


provides feedback current (I


R


) operating at the frequency of operation (F


O


). The frequency of operation (F


O


) is a function of the control current (I


D


), the oscillator reference voltage (VREF), and a capacitance of the switching capacitor (CQ).




Under steady state conditions, feedback current (I


R


) is equal to (VREF)(F


O


)(CQ). This results in the following expression for the frequency of operation:








F




O




=I




D


/(


VREF


)(


CQ


)






The current controlled oscillator


12


further includes a one-shot


26


responsive to a trigger input at


44


for providing a one-shot clock output at


46


. Responsive to the one-shot clock output at input


48


, the non-overlapping clock generator


28


provides non-overlapping clock signals, as further discussed below, on the first and second non-overlapping clock outputs,


50


and


52


, respectively.




Integrator


22


includes an op-amp (OA


1


)


54


and a feedback capacitor (C


1


)


56


, the feedback capacitor coupled between an inverting input


58


of op-amp (OA


1


)


54


and the output


60


of op-amp (OA


1


)


54


. In the embodiment of

FIG. 2

, the non-inverting input


62


of op-amp (OA


1


)


54


is coupled to ground potential. Integrator


22


integrates a sum of the control current (I


D


) and the feedback current (I


R


) to provide an output voltage V


X3


at


60


.




Comparator


24


includes an op-amp (CA


1


) with an inverting input


64


, a non-inverting input


66


, and an output


68


. Responsive to the reference voltage (VREF) coupled to the inverting input


64


and the integrator output


60


coupled to the non-inverting input


66


, comparator


24


provides a pulse signal output on


68


having a frequency of oscillation (F


O


) representative of the intensity of incident light. The pulse signal output


68


is further coupled to the trigger input


44


of the one-shot


26


.




Referring briefly to

FIG. 3

, one-shot


28


includes a NOR latch


70


having an input corresponding to the trigger input


44


. The NOR latch


70


furthermore has a first output


72


coupled to a MOS switch


74


and a second output


46


coupled to the clock input


48


of the non-overlapping clock generator


28


(FIG.


2


). The switch


74


is configured to discharge a capacitor


76


coupled to a non-inverting input


78


of a comparator (CA


3


)


80


. The comparator


80


outputs a reset signal input


82


to the NOR latch


70


in response to a voltage of the capacitor (C


1


)


76


reaching the level of a voltage across a resistor (R


1


)


84


. Resistor (R


1


)


84


is coupled to the inverting input


86


of the comparator


80


. Lastly, current sources (I


1


)


88


and (I


2


)


90


provide current to capacitor (C


1


)


76


and resistor (R


1


)


84


, respectively.





FIG. 4

is an electrical schematic diagram of the non-overlapping clock generator (NOL)


28


of FIG.


2


. Generator


28


provides a first non-overlapping clock output at


50


and a second non-overlapping clock output at


52


in response to a clock input at


48


. Non-overlapping clock generator


28


can include any suitable circuit configuration known in the art.




Current controlled oscillator


12


of

FIG. 2

operates as follows. Assume that the outputs of comparator (CA


1


)


24


and the one shot


26


are both low and that the photodiode current (I


D


) is applied to the integrator


22


. As noted above, integrator


22


includes operational amplifier (OA


1


)


54


and capacitor (CI)


56


. Accordingly, this results in the voltage (V


X3


) at


60


ramping positive. When V


X3


reaches VREF, comparator (CA


1


)


24


switches high and sets the NOR latch


70


in the one shot


26


(FIG.


3


). The first output


72


of the latch releases switch (MN


1


)


74


, while the other output


46


is applied to the input


48


of the non-overlapping clock generator (NOL) circuit


28


(FIG.


4


).




With the one shot output


46


initially low, NOL output (PH


2


)


52


is also low, holding switches (MN


2


)


36


and (MN


3


)


38


off while output (PH


1


B)


50


is high. Output (PH


1


B)


50


high turns switches (MN


4


)


40


and (MN


5


)


42


on, thus shorting both sides of capacitor (CQ)


34


to ground. Shorting both sides of capacitor (CQ)


34


to ground sets the capacitor charge to zero (0). When one shot output


46


(NOL input


48


) goes high, output (PH


1


B)


50


goes low initially. Output (PH


1


B)


50


going low turns off switches (MN


4


)


40


and (MN


5


)


42


. This is followed by output (PH


2


)


52


transitioning high.




Output (PH


2


)


50


transitioning high turns switches (MN


2


)


36


and (MN


3


)


38


ON. The switching ON of switches (MN


2


)


36


and (MN


3


)


38


injects a packet of charge equal to (VREF)*(CQ) into the summing node


58


and a current is caused to flow through the integrator capacitor (CI)


56


, driving V


X3


at


60


negative by an amount equal to (VREF)(CQ)/(CI). Accordingly, this results in the output voltage V


X3


at


60


having a sawtooth waveform, as shown in FIG.


2


. As the charge transfer is completed, V


X3


begins to ramp positive, eventually reaching VREF and the cycle repeats itself at a frequency proportional to the input control current (I


D


).




If perfect components were available, the V


X3


waveform would be the theoretical sawtooth and the one shot


26


would not be needed. However, parasitic switch resistances and a finite op-amp (OA


1


) bandwidth introduce a minimum time requirement for settling which is of paramount importance for accurate charge transfer. Since the comparator (CA


1


)


24


pulse width will be smaller than this minimum time, the need for a one shot is introduced to guarantee enough time for settling.




The one shot circuit of

FIG. 3

operates as follows. When the input signal (TRIGGER) at


44


goes high, it sets the NOR latch


70


, releasing switch (MN


1


)


74


and generates the output clock signal CLOCK at


46


. The voltage at node (Vc)


78


ramps positive until it reaches the voltage at node (Vr)


86


, i.e., equal to (


12


)(R


1


). At that time, (T=R


1


*C


1


*I


2


/I


1


), the comparator output


82


switches high, resetting the NOR latch


70


, which removes the CLOCK signal at


46


and discharges the capacitor


76


through switch (MN


1


)


74


. The one shot circuit


26


is then ready for the next input trigger at


44


.




As discussed above, the charge packet size (VREF)*(CQ) determines the frequency of oscillation (F


O


) that will occur for a given incident light intensity. To minimize the parasitics in the oscillator circuit


12


(which maximizes the operational frequency), oscillator


12


utilizes a single untrimmed capacitor (CQ)


34


. The problem that this introduces, however, is that the capacitor


34


usually has a significant tolerance associated with it and this tolerance is typically wider than the calibrated limits desired. Accordingly, to offset the capacitor tolerance, it is necessary to vary the reference voltage in a manner that doesn't affect the temperature stability of the reference.




To accomplish varying the reference voltage in a manner that doesn't affect the temperature stability of the reference, the light to frequency converter


10


includes programmable gain amplifier (PGA)


14


inserted between the actual reference


18


and the oscillator input VREF at


92


(FIG.


1


). The programmable gain amplifier


14


includes a standard non-inverting potentiometric configuration using discrete gain switches controlled, for example, by a 3 to 8 decoder. The gain range and the number of gain settings are determined by the initial capacitor tolerance and the final frequency tolerance requirement for a particular light-to-frequency converter application. Accordingly, the gain range and number of gain settings may vary from those described and shown herein.




Referring now to

FIG. 5

, in further detail, the programmable gain amplifier


14


includes an op-amp


94


having a non-inverting input


96


, an inverting input


98


, and an output


100


. The bandgap reference voltage with temperature coefficient compensation (VBG_TC) couples to the non-inverting input


96


. The inverting input


98


couples to the output


100


via a potentiometric configuration, generally indicated by reference numeral


102


. The potentiometric configuration


102


is adapted for selectively switching, via switches (MN


2


-MN


9


) generally indicated by reference numeral


104


, one of zero, one, and more than one resistor of a plurality of serially coupled resistors (R


1


-R


8


), generally indicated by reference numeral


106


, between the inverting input


98


and the output


100


.




With reference still to

FIG. 5

, the potentiometric configuration


102


further includes an m-to-n decoder


108


. The m-to-n decoder


108


includes m inputs (A


0


-A


2


), generally indicated by reference numeral


110


, and n outputs (Q


0


-Q


7


), generally indicated by reference numeral


112


. The m inputs (A


0


-A


2


)


110


are configured by input signals (G


0


-G


2


) for selecting a desired one of the n outputs (Q


0


-Q


7


)


112


. The n outputs (Q


0


-Q


7


)


112


are coupled to respective switches (MN


2


-MN


9


)


104


and configured to selectively switch none, one, and more than one resistor of the plurality of serially coupled resistors (R


1


-R


8


)


106


between the inverting input


98


and the output


100


. In one embodiment, the m-to-n decoder


108


includes a 3-to-8 decoder, such as shown in FIG.


6


.




One of the practical aspects of light intensity to current (frequency) conversion involves the temperature coefficient versus the wavelength of the incident light intensity. In the visible spectrum, the coefficient is very flat, but it increases for wavelengths in the infrared region. For that reason, it is desirable to be able to modify the temperature coefficient of the switched capacitor current injection to match the temperature coefficient of the photodiode current. This can be accomplished by altering the bandgap voltage (VBG) to a voltage where the desired temperature coefficient resides, but to attain the proper range of temperature coefficients would require a very large voltage deviation. This would place additional range constraints on the PGA


14


, or it would have to be compensated for by adding trim capability to the charge packet capacitor (CQ)


34


. Either of these alternative would drive the complexity of the system up and reduce the higher frequency performance due to additional parasitic elements.




Accordingly, to add temperature coefficient modification capability without altering the PGA


14


or oscillator


12


, the temperature generation circuit of

FIG. 7

has been placed between the bandgap reference generator


18


and the PGA


16


, such that:








VBG









TC=VBG+


(


IPTAT−IREF


)*


RTC.








Referring now to

FIG. 7

, the temperature coefficient generator


16


includes a circuit responsive to the bandgap reference voltage (VBG) at input


114


for generating a temperature stable reference current IRI-F equal to (VBG/RREF) via op-amp (OA


1


)


116


, resistor (RREF)


118


, and cascode transistor (MN


4


)


120


. The temperature coefficient generator


16


applies the temperature stable reference current (VBG/RREF) through a current mirror


122


to a current summing junction


124


of an op-amp (OA


2


)


126


. The current mirror


116


includes MOS devices (MP


2


)


128


and (MP


3


)


130


. The temperature coefficient generator


16


sums the temperature stable reference current against a current proportional to absolute temperature (IPTAT) and converts the summed current, via the bandgap reference voltage applied to the non-inverting input


132


of op-amp (OA


2


)


126


and a feedback resistor (RTC)


134


coupled between the inverting input


124


and the output


136


of op-amp (OA


2


)


126


, into the bandgap reference voltage with temperature coefficient compensation (VBG_TC) on output


136


.




The temperature coefficient generator


16


of

FIG. 7

further includes a generator


138


for providing the current proportional to absolute temperature (IPTAT). With reference now to

FIG. 8

, the generator


138


for providing the current proportional to absolute temperature (IPTAT) includes a circuit for deriving the current proportional to absolute temperature from matched emitter currents flowing in first and second transistors (QP


2


)


140


and (QP


3


)


142


, respectively, and described by the expression:








IPTAT=A


1*(


kT/q


)*


ln


(


A


2


/A


3)*(1


/R


1),






where A


1


is a current mirror gain provided by first and second output CMOS devices (MN


10


)


144


and (MN


20


)


146


, k is Boltzman's constant, T is absolute temperature, q is electronic charge, A


2


is the emitter area of transistor QP


2


, A


3


is the emitter area of transistor QP


3


, and R


1


is the resistance coupled to the emitter of transistor QP


2


.




In one embodiment, the value of the current proportional to absolute temperature (IPTAT) is set to match a reference current defined by the bandgap reference voltage divided by a reference resistance (VBG/RREF) at a temperature equal to 27 degrees Celsius (T=27° C.) such that the bandgap reference voltage with temperature coefficient compensation (VBG_TC) equals the bandgap reference voltage (VBG) at that temperature. The intrinsic temperature coefficient of this current can then be described by taking the derivative of the IPTAT expression, yielding:









IPTAT/∂T=A


1*(


k/q


)*


ln


(


A


2


/A


3)*(1


/R


1),






which results in an output voltage temperature coefficient of:









VBG









TC/∂T=A


1*(


k/q


)*


ln


(


A


2


/A


3)*(


RTC/R


1).






Accordingly, the temperature coefficient described can be controlled by the selection of resistor (RTC)


134


. Controlling the temperature coefficient by selecting a value for resistor (RTC)


134


(

FIG. 7

) allows for operation at optimum temperature coefficient (TC) levels in the infrared spectrum without placing additional constraints on a remainder of the circuit.




Referring now to

FIG. 9

, the bandgap reference voltage generator


18


of

FIG. 1

provides for temperature stability and shall now be described in further detail. The circuit of

FIG. 9

generates the bandgap reference voltage (VBG) at output


148


, in part, with the use of a standard Bandgap configuration which produces an ideal output of:







VBG=VBE+ΔVBE


(


R


2


/R


3),




where:






Δ


VBE


=(


kT/q


)*


ln


(


A


2


/A


1).






Bandgap reference voltage generator


18


includes a circuit having an op-amp (OA


1


)


150


for equalizing emitter currents of a first transistor (QP


1


) and a second transistor (QP


2


), indicated by reference numerals,


152


and


154


, respectively. However, the op-amp and transistor configuration is prone to offset errors, in addition to normal bandgap tolerances. The errors will modulate the ΔVBE term such that the total differential voltage (V


X


) is the sum of ΔVBE and an error term V


OS


according to:








V




X


=(


kT/q


)*


ln


(


A


2


/A


1)+


V




OS


.






Accordingly, the circuit of

FIG. 9

further includes means for trimming out the above-mentioned error to achieve desired calibrated frequency tolerances and temperature coefficients. The error includes, for example, an offset error V


OS


. In one embodiment, the means for trimming out error includes a transistor digital to analog converter (DAC)


156


configured to alter an effective emitter area of the first transistor (QP


1


)


152


for nulling the offset error, according to:






Δ


VBE


=(


kT/q


)*


ln


((


A


2


/A


1)±Δ


A


1)+V


OS


.







FIG. 10

is an electrical schematic diagram of the bandgap trim (DAC)


156


of

FIG. 9

in greater detail. The DAC


156


includes at least two inputs


158


and an output


160


. DAC


156


is responsive to the at least two inputs


158


for providing a voltage on the output


160


as a function of the at least two inputs. More particularly, DAC


156


provides a variable impedance from the output


160


to ground as a function of the at least two inputs


158


. The variable impedance varies the current flowing in the first and second bandgap transistors (


152


and


154


, respectively), which in turn, varies the bandgap output voltage (VBG) at output


148


.




In another embodiment, the trimming circuit and technique, as used in the embodiments of

FIGS. 9 and 10

, can also be used to trim any errors in the PTAT current flowing at T=27° C. in the embodiment of the IPTAT generator of FIG.


8


. For example, in a circuit for equalizing emitter currents of a first transistor and a second transistor, a second circuit is provided to trim out error in at least one emitter current to achieve a desired frequency tolerance. The second circuit includes at least a single transistor digital to analog converter (DAC). In one instance, the error may include an offset error and the DAC is configured to alter the effective emitter area of one of the transistors for nulling the offset error. Still further, the DAC includes at least two inputs and an output, the DAC responsive to the at least two inputs for providing a variable impedance from the output to ground as a function of the at least two inputs. Accordingly, the variable impedance will vary the current flowing in the first and second transistors.




As shown in

FIG. 2

, the oscillator configuration includes photodiode tied directly to the integrator summing junction. While this biases the diode at its optimum level (Vb=0), it is not practical due to the large value of parasitic capacitance of the photodiode and its effect on the high frequency operation of the oscillator. To alleviate this problem, in another embodiment, the light to frequency converter


10


includes a current controlled oscillator


12


A as shown in FIG.


11


. The current controlled oscillator


12


A of

FIG. 11

is similar to that shown in

FIG. 2

with the following exceptions. Current controlled oscillator


12


A includes a buffered current controlled oscillator. The buffered current controlled oscillator


12


A includes a current buffer


162


coupled between the at least one photodiode and a summing junction for isolating the control current I


D


. The consequence of this buffering is that an AC ground reference (VBIAS) must be moved positive or the oscillator has to be run from split power to keep the buffer circuit from saturating.




The buffered current controlled oscillator


12


A of

FIG. 11

further includes one-shot


26


responsive to a trigger input for providing a one-shot clock output. Non-overlapping clock generator


28


, responsive to the one-shot clock output


46


, provides a first non-overlapping clock output


50


and a second non-overlapping clock output


52


. Switched capacitor resistor configuration


30


includes the switching capacitor (CQ)


34


, having a first input coupled to the first non-overlapping clock output


50


, a second input coupled to the second non-overlapping clock output


52


, a voltage reference input coupled to the oscillator reference voltage (VREF)


100


, and a voltage bias ground reference (VBIAS) at


164


. The switched capacitor resistor configuration


30


is responsive to the first, second, and voltage reference inputs and the voltage bias ground reference for providing the feedback current operating at a frequency of operation (F


O


) at input


58


of integrator


22


. The frequency of operation (F


O


) is a function of the control current, the oscillator reference voltage (VREF), and a capacitance of the switching capacitor (CQ).




Furthermore, in the embodiment of

FIG. 11

, the integrator


22


includes an op-amp


54


having the control and feedback currents coupled to an inverting input


58


of the op-amp and the voltage bias ground reference


164


coupled to a non-inverting input


62


of the op-amp. The integrator


22


integrates a sum of the control current I


D


and the feedback current I


R


to provide an output voltage (V


X3


) at


60


. Comparator


24


is responsive to the reference voltage VREF coupled to the inverting input


64


and the integrator output


60


coupled to the non-inverting input


66


for providing a pulse signal output at


68


having a frequency of oscillation representative of the intensity of incident light. The pulse signal output at


68


is further coupled to the trigger input


44


of the one-shot


26


.





FIG. 12

illustrates an electrical schematic diagram of the photodiode


20


with the current buffer


162


of

FIG. 11

for isolating the photodiode, according to one embodiment of the present disclosure. In this embodiment, current buffer


162


includes op-amp (OA


2


)


166


and switch (MN


2


)


168


coupled to photodiode


20


, as shown in FIG.


12


.





FIG. 13

illustrates an electrical block diagram of a plurality of photodiodes with current buffer


162


of

FIG. 11

, according to another embodiment of the present disclosure. The at least one photodiode


20


includes a plurality of photodiodes


20




a


coupled to the current buffer via a multiplexer


170


. The multiplexer


170


is configured to multiplex a desired one of the plurality of photodiodes


20




a


to the current buffer


162


under digital control. Accordingly, the multiplexer


170


enables different photodiode currents to be input to the oscillator


12




a


. That is, the multiplexer switches photodiodes D


1


-D


4


in or out of the circuit under digital control.

FIG. 14

is an electrical schematic diagram of the diode multiplexer


170


of FIG.


13


.




Referring now to

FIG. 15

, a block diagram view of an apparatus having a light intensity control function according to another embodiment shall be discussed. The apparatus having a light intensity control function is generally indicated by the reference numeral


172


. Apparatus


172


includes a monolithic light to frequency converter


10


, as shown and described with respect to FIG.


1


and subsequent Figures. The apparatus further includes a controller


174


, such as a microprocessor, micro controller, computer, control logic circuitry, or the like for controlling a device parameter of device


176


. In one embodiment, the controller


174


is responsive to the output signal of the light-to-frequency converter


10


, the output signal having a frequency (F


O


) proportional to the intensity of the incident light


32


, for controlling the device parameter according to the particular requirement of apparatus


172


and/or device


176


application. In another embodiment, controller


174


may include any suitable control device, circuit, or processor for performing a desired functionality, as discussed herein.




For example, controller


174


provides light intensity information on output


178


in response to the output signal of the light to frequency converter


10


. Output


178


provides light intensity information according to the requirements of a particular light detector application. In addition, controller


174


may provide suitable control signals on control output


178


to device


176


, according to the requirements of the particular light intensity control application.




In one embodiment, controller


174


may also contain circuitry for conditioning the output of the light-to-frequency converter


10


and translating the results of the conditioned output into alternate forms of light intensity information. For example, conditioning may include averaging, or other conditioning suitable for a particular light intensity measurement application.




According to yet another embodiment, controller


174


further performs a translation of the conditioned output that includes, for example, arithmetic processing and use of a table look-up. The translation may be also be accomplished with at least one of hard-wired logic and stored program logic, such as via a microprocessor.




Device


176


may include any device responsive to a light intensity control signal for producing a desired response as a function of the intensity of light. An illustrative device may include a rear view mirror of an automobile, the rear view mirror for being controlled in a desired manner in response to a measure of the intensity of light by the light to frequency converter


10


, for example.




Device


176


may also include a display, such as for a laptop computer or other device, for example. Controlling the backlighting of the display screen according to the intensity of incident light on the laptop computer can facilitate improved power management, extending a useful battery life between recharging periods of the same. Other electronic devices, such as hand held electronic devices, capable of providing backlighting with an artificial illumination, are also contemplated.




In the example of the immediately preceding paragraph, controller


174


may include a means for controlling a display parameter in response to the measure of light intensity of the incident light. The display parameter may include any parameter to be controlled in response to the measure of light intensity of the incident light. The display parameter may include, for example, a backlight control parameter of the display device. The backlight control parameter may include a first level for the intensity of incident light and a second level for the intensity of incident light. The first and second levels can correspond, for example, to providing backlighting and providing no backlighting, respectively. The first and second levels may also include additional levels of light intensities, with corresponding control actions.




Although only a few exemplary embodiments of this invention have been described in detail above, those skilled in the art will readily appreciate that many modifications are possible in the exemplary embodiments without materially departing from the novel teachings and advantages of this invention. For example, the functionality of the various embodiments as discussed herein can be provided on a single monolithic integrated circuit. Accordingly, all such modifications are intended to be included within the scope of this invention as defined in the following claims. In the claims, means-plus-function clauses are intended to cover the structures described herein as performing the recited function and not only structural equivalents, but also equivalent structures.



Claims
  • 1. An integrated circuit for generating a bandgap reference voltage (VBG), comprising:a first circuit having an op-amp for equalizing emitter currents of a first bandgap transistor and a second bandgap transistor; and a second circuit for trimming out error in at least one emitter current to achieve a desired frequency tolerance, said second circuit including at least a single transistor digital to analog converter (DAC).
  • 2. The integrated circuit of claim 1, wherein the error includes an offset error, and wherein the DAC is configured to alter the effective emitter area of one of the bandgap transistors for nulling the offset error.
  • 3. The integrated circuit of claim 2, wherein the DAC includes at least two inputs and an output, the DAC responsive to the at least two inputs for providing a variable impedance from the output to ground as a function of the at least two inputs which will vary the current flowing in the first and second bandgap transistors, which, in turn, will vary the bandgap reference voltage.
  • 4. A method for generating a bandgap reference voltage (VBG) comprising:equalizing emitter currents of a first bandgap transistor and a second bandgap transistor, wherein equalizing emitter currents includes using an op-amp; and trimming out error in at least one emitter current to achieve a desired frequency tolerance, wherein trimming out error includes utilizing digital to analog conversion.
  • 5. A method for generating a bandgap reference voltage (VBG) comprising:equalizing emitter currents of a first bandgap transistor and a second bandgap transistor; and trimming out error in at least one emitter current to achieve a desired frequency tolerance, wherein trimming out error includes utilizing digital to analog conversion, wherein the error includes an offset error, and wherein the trimming out error includes utilizing digital to analog conversion in a manner configured to alter an effective emitter area of the first bandgap transistor and null the offset error.
  • 6. The method of claim 5, wherein the digital to analog conversion includes the use of a transistor digital to analog converter (DAC), the DAC including at least two inputs and an output, the DAC responsive to the at least two inputs for providing a variable impedance from the output to ground as a function of the at least two inputs which will vary current flowing in the first and second bandgap transistors, which, in turn, will vary the bandgap reference voltage.
  • 7. An integrated circuit comprising:a bandgap reference voltage generator for generating a bandgap reference voltage (VBG), said bandgap reference voltage generator including a circuit having an op-amp for equalizing emitter currents of a first transistor and a second transistor; and means adapted to trim out an offset error between the emitter currents of the first and second transistors to achieve a desired frequency tolerance, wherein said offset error trimming means is configured to alter an effective emitter area of the first transistor for nulling the offset error.
  • 8. The integrated circuit of claim 7, wherein said trimming means includes a transistor digital to analog converter (DAC).
  • 9. The integrated circuit of claim 8, wherein the DAC includes at least two inputs and an output, the DAC responsive to the at least two inputs for providing a variable impedance from the output to ground as a function of the at least two inputs which will vary current flowing in the first and second transistors, which, in turn, will vary the bandgap reference voltage.
US Referenced Citations (2)
Number Name Date Kind
5850195 Berlien, Jr. et al. Dec 1998 A
6075354 Smith et al. Jun 2000 A