Download pdf - AN1494 - Using MCP6491 Op Amps for Photodetection …ww1.microchip.com/downloads/en/AppNotes/01494A.pdf · · 2013-05-09Using MCP6491 Op Amps for Photodetection Applications

AN1494Using MCP6491 Op Amps for Photodetection Applications

INTRODUCTION

Low input bias operational amplifiers (op amps) areoften required for a wide range of photodetectionapplications, in order to reduce current error andimprove the accuracy of the output signal.

The typical photodetection applications are listedbelow:

• Smoke Detectors

• Flame Monitors

• Airport Security X-Ray Scanners

• Light Meters

• Brightness Controls

• Bar Code Scanners

• Pulse Oximeters

• Blood Particle Analyzers

• CT Scanners

• Automotive Headlight Dimmers

• Twilight Detectors

• Photographic Flash Controls

• Automatic Shutter Controls

• Optical Remote Controls

• Optical Communications, etc.

This application note discusses the features of Micro-chip’s MCP6491 low input bias current op amps [1], thecharacteristics of photodiodes, and the strengths of theactive photodiode current-to-voltage converter (i.e.photodiode amplifier), compared to the passive ver-sion. Next, the focus shifts to the design techniques ofphotodiode amplifier circuitry. Several key designpoints are discussed in order to improve the circuit’sperformance. Then, a practical application examplewith PSpice simulation results is provided, to help illus-trate the design techniques in depth. In addition, thenoise analysis of the photodiode amplifier and thedesign of a companion low pass filter are discussed.Finally, the PCB techniques that help reduce the cur-rent leakage are briefly introduced.

MCP6491 LOW INPUT BIAS CURRENT OP AMPS

Microchip’s MCP6491 family of op amps has low inputbias current (150 pA, typical at 125°C) and rail-to-railinput and output operation. The MCP6491 family isunity gain stable and has a gain bandwidth product of7.5 MHz (typical). These devices operate with a single-supply voltage as low as 2.4V, while only drawing530 μA/amplifier (typical) of quiescent current. Thesefeatures make the MCP6491 family of op amps wellsuited for photodiode amplifier, pH electrode amplifier,low leakage amplifier, and battery-powered signal con-ditioning applications, etc.

Features:

• Low Input Bias Current

- 1 pA (typical at 25°C)



• Low Quiescent Current:

- 530 µA/amplifier (typical)

• Low Input Offset Voltage:

- 1.5 mV (maximum)

• Rail-to-Rail Input and Output

• Supply Voltage Range: 2.4V to 5.5V

• Gain Bandwidth Product: 7.5 MHz (typical)

• Slew Rate: 6 V/µs (typical)

• Unity Gain Stable

• No Phase Reversal

• Small Packages

- Singles in SC70-5, SOT-23-5

• Extended Temperature Range

- -40°C to +125°C

Related Parts

• MCP6481: 4 MHz, Low Input Bias Current Op Amps [2]

• MCP6471: 2 MHz, Low Input Bias Current Op Amps [3]

Author: Yang ZhenMicrochip Technology Inc.

2013 Microchip Technology Inc. DS01494A-page 1

AN1494

PHOTODETECTION APPLICATIONS

There are many detectors which can be used for pho-todetection applications, such as photodiodes, photo-transistors, photoresistors, phototubes, photomultipliertubes, charge-coupled devices, etc.

In this application note, we will focus on the photodi-ode, as it is the most common photodetector andwidely used for the detection of intensity, position, colorand presence of light.

Photodiode

The photodiode is a type of photodetector capable ofconverting light to a small current which is proportionalto the level of illumination.

FEATURES

The photodiode’s features can be summarized asbelow:

• Wide spectral response

• Excellent linearity

• Low noise

• Excellent ruggedness and stability

• Small physical size

• Long lifetime

• Low cost

EQUIVALENT CIRCUIT

The equivalent circuit for a photodiode is shown below,in Figure 1.

FIGURE 1: A Photodiode Equivalent Circuit.

A photodiode can be represented by a current source(I), a junction shunt resistance (RJ), and a junctioncapacitance (CJ) in parallel with an ideal diode. Theseries resistance (RS) is connected with all other com-ponents in series. Dark current (ID) only exists underreverse bias conditions.

• Junction Shunt Resistance (RJ)

RJ represents the resistance of the zero-biased photo-diode junction. An ideal photodiode will have an infiniteRJ, but the actual value of RJ is typically on the order ofthousands of MΩ, which depends on the photodiodematerial, and decreases by a factor of 2 for every 10°Crise in temperature. The high value of RJ yields the lownoise current of the photodiode.

• Series Resistance (RS)

RS is the resistance of the wire bonds and contacts ofthe photodiode. An ideal photodiode should have noseries resistance, but the typical value is on the orderof tens of Ω, which is much smaller than RJ. The RS isused to determine the linearity of the photodiode underzero bias conditions. For most of applications, it can beignored.

• Junction capacitance (CJ)

CJ is directly proportional to the junction area andinversely proportional to the diode reverse bias voltage.For a small area diode at zero bias, the typical value ison the order of tens of pF.

• Dark Current (ID)

ID is the small leakage current that flows throughphotodiode under reverse bias conditions. It existseven when there is no illumination and approximatelydoubles for every 10°C rise in temperature. There is nodark current under zero bias conditions.

OPERATION MODES

There are two operation modes for the photodiode, thephotovoltaic mode and the photoconductive mode, asshown in Figure 2 and Figure 3. The two modes havetheir own strengths and drawbacks, and mode selec-tion is dependent on the target application.

• Photovoltaic Mode

This mode has zero voltage potential across the photo-diode. No dark current flows through the photodiode,the linearity and sensitivity are maximized, and thenoise level is relatively low (RJ’s thermal noise only),which make it well suited for precision applications.

FIGURE 2: Photovoltaic Mode.

CJRJ

RS << RJ

Light

I

I = current source

RJ = junction shunt resistanceCJ = junction capacitanceRS = series resistance

Ideal Diode

ID

ID = dark current

(photocurrent generated by the incident light)

Light

DS01494A-page 2 2013 Microchip Technology Inc.

AN1494

• Photoconductive Mode

This mode has a reverse bias voltage placed acrossthe photodiode. The reverse bias voltage reduces thediode junction capacitance and shortens the responsetime. Therefore, the photoconductive mode is suitablefor high speed applications (e.g., high speed digitalcommunications). The main drawbacks of this modeinclude dark current appearance, non-linearity, andhigh noise level (RJ’s thermal noise and ID’s shotnoise).

FIGURE 3: Photoconductive Mode.

Photodiode Current-to-Voltage Converter

This circuit is used to convert the photodiode’s smalloutput current to a measurable voltage. Typically, thereare two types of circuit implementations, which arepassive and active versions.

PASSIVE PHOTODIODE CURRENT-TO-VOLTAGE CONVERTER

The Passive Photodiode Current-to-Voltage Converteris implemented by only passive components, as shownin Figure 4. Its output resistance is roughly equal to thevalue of large resistor (RF) and the output voltage isequal to I*RF.

The large RF can cause loading effects for subsequentload resistance and capacitance, such as an inaccu-rate VOUT and a relatively long response time.

Moreover, the variation of photocurrent can cause thephotodiode’s biasing voltage to be unstable, which willchange the junction capacitance (CJ) and affect thefrequency response of photodiode.

FIGURE 4: Passive Photodiode Current-to-Voltage Converter.

ACTIVE PHOTODIODE CURRENT-TO-VOLTAGE CONVERTER

The active photodiode current-to-voltage converter isalso called a photodiode amplifier. Based on the photo-diode operating modes, two circuit implementations ofphotodiode amplifiers are shown in Figure 5 andFigure 6. For the strengths and drawbacks of eachimplementation, please refer to the section “OperationModes”.

Both implementations have a large resistor (RF) in thefeedback loop. The output resistance of the photodiodeamplifier is roughly equal to RF/AOL, where AOL is theopen loop gain of the op amp. Therefore, the outputresistance becomes very small and the loading effectscan be ignored.

For the photoconductive mode amplifier, the biasingvoltage is equal to VBIAS. For the photovoltaic modeamplifier, the biasing voltage is just zero. Both biasingvoltages do not change when the photocurrent varies,so the photodiode's frequency response will not beaffected.

These strengths of the photodiode amplifier make itwidely used in photodetection applications.

FIGURE 5: “Photovoltaic Mode” Photo-diode Amplifier.

FIGURE 6: “Photoconductive Mode” Photodiode Amplifier.

Light

VBIAS 0V

Light

RF

+

–

VOUT I RF=

I

Light

VOUT

VDD

RF

I –

+

MCP6491

VOUT I RF=

Light

VOUT

VDD

RF

–

+

MCP6491

VBIAS

I

VBIAS 0V VOUT

I RF=


AN1494

Photodiode Amplifier Key Design Points

Several key design points for the photodiode amplifierwill be analyzed next, in order to improve the circuit’sperformance.

OP AMP SELECTION

Selecting a suitable op amp for the photodiodeamplifier is critical. There are many DC and AC specsin an op amp data sheet, and the key op amp specs forthe photodiode amplifier are shown and discussedbelow.

Low Input Bias Current (IB)

The DC output voltage error due to IB is equal to IB*RF.IB increases with temperature rise, so the error will belarger at higher temperature. Usually, the voltage errorcan be reduced to IOS*RF by adding a compensationresistor RC with a value of RFǁRJ in series with the opamp non-inverting input.

However, at high temperatures, the value of RC isdifficult to determine because the value of RJsignificantly drops with temperature rise. In thiscondition, the value of RJ could be less than the valueof RF.

Moreover, RC will develop a noise voltage as the opamp input noise current flows through it. The RC alsogenerates a thermal noise voltage. Both noise voltageswill be amplified by the circuit's noise gain. Thus, theoutput noise level will increase.

IB also generates a voltage across RC at the op amp'snon-inverting input. This causes the same voltage atthe inverting input. Now, the biasing voltage is nolonger stable, which causes the photodiode's responseto become nonlinear.

Therefore, adding the compensation resistor RC toreduce the voltage error IB*RF is not an effectivemethod in general. The op amp should have IB lowenough to keep the voltage error within an acceptablerange of target applications.

Low Input Offset Voltage (VOS)

The DC output voltage error due to VOS is equal toVOS*(1 + RF/RJ) at room temperature (25°C), which isabout VOS because RF is much less than RJ and thegain is approximately 1 V/V. At high temperatures, theerror could be much larger because the value of RJsignificantly decreases and the gain can be higher than1 V/V. Moreover, the VOS drift could make the erroreven worse. Therefore, low VOS and low VOS drift willbe very helpful to reduce the output error at hightemperatures.

Common Mode Input Voltage Range

The common mode input voltage range needs to atleast include ground because the non-inverting input ofthe op amp is grounded.

Rail-to-Rail Output

The rail-to-rail output is helpful to maximize thedynamic output voltage range and improve the signal-to-noise ratio (SNR).

Wide Gain Bandwidth Product (GBWP) and High Slew Rate (SR)

The GBWP and SR should be large enough to meet therequirement of output step response time, which will bediscussed in more detail later.

Low Input Noise Current Density and Low Input Noise Voltage Density

When noise current flows through the photodiodeamplifier, resistor noise voltages will result. The op ampinput current noise density (2qI, where q is electroncharge, I is current) is determined by IB, so that lowerIB gives lower op amp input noise current density. Thelow input noise voltage density also plays a very impor-tant role for the output noise of the photodiode ampli-fier. It will be amplified by the noise gain so that theoutput noise level will be significantly affected. This willbe explained later in this section.

In conclusion, the Microchip’s MCP6491 op amp’s keyfeatures include low IB, low VOS, low VOS drift with tem-perature, rail-to-rail input/output, wide GBWP, high SR,low input noise current density and low input noise volt-age density, etc. These features make it well suited forthe photodiode amplifier.

FEEDBACK RESISTOR

The value of the feedback resistor (RF) should be setas large as possible to give a high transimpedance gainto the photocurrent. Usually, this gain should be highenough to use most of the op amp’s output voltageswing when the photocurrent is at its maximum value.For precision applications, a large resistor with tight tol-erance and a low temperature coefficient should beselected.

It is possible to add more gain with subsequent stages,however, the noise performance will not be as good asusing a large RF in one stage, which can easily improvethe SNR.

For a given bandwidth f, the thermal noise voltage ofRF is given by 4kTRFf, where k is Boltzmann’sconstant (1.38 x 10-13J/K), T is absolute temperature(K), RF is feedback resistance (Ω).

The output signal is given by VSIGNAL = I*RF and theSNR = 20*log(VSIGNAL/VNOISE). When RF is doubled,the resistor thermal noise voltage is increased by 2and the output signal voltage is increased by 2. Thus,the SNR is increased by 3 dB.


AN1494

FEEDBACK CAPACITOR

The photodiode amplifier does not always behave asdesired. The gain peaking and step output ringing aretypical phenomena, which could happen in frequencyand time domains (refer to Figure 7 and Figure 8).

Moreover, the noise gain peaking results in very highoutput noise levels (Figure 9), which may severelydegrade the integrity of the output signal.

FIGURE 7: Signal Gain (i.e. Tran-simpedance Gain) Peaking.

FIGURE 8: Output Ringing.

FIGURE 9: Noise Gain Peaking.

These phenomena make the photodiode amplifierunstable. A small capacitor (CF) can be added in thefeedback loop to eliminate the gain peaking, step out-put ringing and noise gain peaking issues (refer toFigure 10).

FIGURE 10: Photodiode Amplifier Using Feedback Capacitor.

In the next section, we will discuss the stability of thephotodiode amplifier, explain why the amplifier will bestable after adding a feedback capacitor, and learn howto determine the value of the feedback capacitor to getthe optimum output response.

peda

nce

Gai

n (V

/A)

100k

100M

10k

1M

10M

1.00 10.00 100.00 1,000.00 10,000.00 100,000.00 1,000,000.00 10,000,000.0

Tran

sim

p

Frequency (Hz)1 10 100 1k 10k 100k 1M 10M

100

10k

1k

2

3

4

5

6

utpu

t Vol

tage

(V)

0

1

2

Step

Ou

Time (ms)0 0.5 1.0 1.5 2.0 2.5 3.0

20

40

60

80

100

Noi

se G

ain

(dB

)

-20

0

20N


VOUT

VDD

RF

–

+

MCP6491

CF

CJRJIdeal Diode

I

VOUT I RF=


AN1494

AMPLIFIER STABILITY ANALYSIS

Figure 11 shows the noise gain bode plot of the photo-diode amplifier in log-log scale. It is important to clarifythe difference between the noise gain and the signalgain because the system stability is dependent on thecharacteristics of noise gain, not signal gain.

The noise gain is the gain seen by a testing voltagesource in series with the op amp non-inverting input,which is equal to the signal gain when the signal isapplied to the op amp non-inverting input.

FIGURE 11: Noise Gain Bode Plot.

The stability of the system is determined by the netslope between the noise gain (GN) and the open loopgain (AOL) at the frequency where they cross over.

• For an unstable photodiode amplifier, the net slope between GN and AOL is equal to +40 dB/decade as shown in Figure 11, where the dotted line of GN intercepts the curve of AOL. The dashed line shows the extended GN curve without adding CF.

• For a stable photodiode amplifier, the net slope between GN and AOL is equal to +20 dB/decade as shown in Figure 11, where the solid line of GN intercepts the curve of AOL. The solid line shows the GN curve with adding CF.

The explanation on the noise gain Bode plot is shownbelow:

• When f is less than f1:

- GN is equal to 1 + RF/RJ, which is roughly equal to 1 V/V or 0 dB when RF << RJ.

- The zero of GN is located in f1.

• When f is between f1 and f2:

- GN increases by +20 dB/dec.

- The pole of GN is located in f2, which is equal to 1/(2*RF*CF). This is also the signal gain bandwidth.

• When f is between f2 and f3:

- GN is equal to 1 + (CJ + COP)/CF.

- The crossover frequency of AOL and GN is located in f3, which is equal to GBWP/GN.

• When f is larger than f3:

- GN is determined and limited by AOL, which decreases by -20 dB/dec.

The value of CF affects the location of f2, whichdetermines the signal gain bandwidth and the phasemargin of the photodiode amplifier.

When CF becomes larger, the phase margin will beincreased, which makes the system more stable withless gain peaking, step overshoot and noise gainpeaking. However, this also will result in smaller signalgain bandwidth and longer output response time.Table 1 below shows the percent overshoot as a resultof different phase margins.

TABLE 1:

f (Hz)

(dB)

AOL

GN

f1 f2 f3

-20 dB/dec

+20 dB/dec

without adding CF (Unstable)

with adding CF (Stable)

f11

2 RJ RF CJ COP C+ +F

--------------------------------------------------------------------------------=

f21

2 RF CF-----------------------------=

f3GBWP

GN----------------- GBWP

1CJ C+

OP

CF-----------------------+

---------------------------------= =

RJ = junction shunt resistance

Where

RF = resistance of feedback resistor

CJ = junction capacitance

COP = op amp input capacitance = CCM + CDM

CF = feedback capacitance

CCM = op amp common mode input capacitance

CDM = op amp differential mode input capacitance

GBWP = op amp gain bandwidth product

f2 = signal gain bandwidth

f3 = noise gain bandwidth

f1 = the location of GN’s first zero

GN = noise gain

AOL = op amp open loop gain

(i.e. transimpedance gain bandwidth)

Phase Margin (°) Overshoot (%)

45 25

55 13.3

65 4.7

75 0.008


AN1494

For most photodetection applications, the optimumvalue of CF is typically considered when the phasemargin is 65°, which gives a negligible gain peakingand 4.7% overshoot at output, while keeping reason-able signal gain bandwidth and response time.

The value of CF at 65° phase margin is approximatelyshown in Equation 1 where RF << RJ is assumed.

EQUATION 1:

In Figure 11, the maximum signal gain bandwidth isachieved at 45° phase margin when f2 is equal to f3,and the corresponding value of CF will be half of theone shown in Equation 1.

If we consider the effect of RF’s parasitic capacitance,CF will be the value of the one shown in Equation 1minus RF’s parasitic capacitance.

Normally, the parasitic capacitance is less than 0.1 pFfor a surface mount resistor due to its small size. Thus,the effect of the parasitic capacitance can be ignored.

APPLICATION EXAMPLE

Here we provide an example to illustrate the circuit’sperformance improvement in frequency and timedomains after the feedback capacitor is added.

In Figure 12, the photodiode’s RJ = 2000 MΩ at 25°C,CJ = 100 pF, MCP6491 op amp’s VDD = 5.5V,RF = 10 MΩ, and assume VOUT switches between 2Vand 4V for the two alternating illumination levels.

FIGURE 12: Photodiode Amplifier Circuit Example.

MCP6491 op amp’s typical GBWP is 7.5 MHz and itsinput capacitance is COP = CCM + CDM = 12 pF.

To make the photodiode amplifier stable, a feedbackcapacitor CF is needed. Based on Equation 1, thevalue of CF is 1 pF when the amplifier’s phase marginis 65°.

At room temperature (25°C), the DC voltage error atoutput due to IB and VOS of MCP6491 is given byIB*RF + VOS = 1 pA*10 MΩ + 1.5 mV = 1.51 mV.

The graphs in Figure 13 — Figure 17 show the relatedoutput response plots with and without adding CF.

FIGURE 13: Signal Gain vs. Frequency.

FIGURE 14: Step Output Response.

CF 2CJ C+

OP

2 RF GBWP ----------------------------------------

VOUT

VDD

10 MΩ

–

+

MCP6491

CF

100 2000pFMΩ

I

VOUT I RF=

Note: These plots are PSpice simulation resultsby using MCP6491 op amp Spice macromodel, which is free on the Microchip website at www.microchip.com. The model isintended to be an initial design tool. Benchtesting is a very important part of anydesign and cannot be replaced withsimulations.

1.00E+05

1.00E+06

1.00E+07

1.00E+08

1.00E+09

peda

nce

Gai

n (V

/A)

100k

100M

1M

10M

1G

with CF

without CF

1.00E+02

1.00E+03

1.00E+04

Tran

sim

p


10k

1k

100

2

3

4

5

6

Out

put V

olta

ge (V

)

with CF

without CF

0

1

2

Step

O

Time (ms)0 0.5 1.0 1.5 2.0 2.5 3.0


AN1494

FIGURE 15: Noise Gain vs. Frequency.

FIGURE 16: Total Output RMS Noise Voltage Density vs. Frequency.

FIGURE 17: Total Output RMS Noise Voltage vs. Frequency.

Although the added CF eliminates a lot of output noise,we still need to further reduce the noise in order toimprove the SNR and achieve better signal integrity.Now we will focus on the noise analysis of the photodi-ode amplifier.

Photodiode Amplifier Noise Analysis

Figure 18 shows the noise model of the photodiodeamplifier.

FIGURE 18: Noise Model.

Two ways to quickly estimate total output root-mean-square (RMS) noise are provided:

• Hand Calculation

• PSpice Simulation

NOISE ESTIMATED BY HAND CALCULATION

The resistor voltage noise density is given byVN = 4kTR and is spectrally flat. For a 1 kΩ resistor,the VN is 4 nV/Hz.

The typical input noise voltage density and input noisecurrent density of MCP6491 are 19 nV/Hz and0.6 fA/Hz, respectively. The input noise voltagedensity vs. frequency plot can be found in theMCP6491 data sheet. The 1/f noise is dominant in thelower frequencies while the thermal noise is dominantin the higher frequencies.

The total output RMS noise is calculated by the squareroot of the sum of the squared values of the individualoutput noise contributors. Each output noise contribu-tor is calculated by integrating its squared output noisedensity over the equivalent noise bandwidth in asquare root. The output noise density is calculated bymultiplying its input noise density by an appropriategain. Note that the worst output noise contributor willdominate the total output RMS noise.

20

40

60

80

100N

oise

Gai

n (d

B)

with CF

without CF

-20

0

20N


F

put R

MS

Noi

se V

olta

ge

ensi

ty (n

V/H

z)

with CF

without CF

1k

10k

100k

Tota

l Out

p De

Frequency (Hz)0.1 1 10 100 1k 10k 100k 1M 10M

F

100

10

3

4

5

6

7

ut R

MS

Noi

se V

olta

ge

(mV)

without CF

0

1

2

Tota

l Out

pu


with CF

VOUT

VDD–

+

MCP6491

CF

+–

+–

+–

RF

RJ CJ

VNVN_RJ

VN_RF

IN-

VN_RJ = RJ’s noise voltage density

Where

VN_RF = RF’s noise voltage density

IN+

VN = op amp input noise voltage densityIN-, IN+ = op amp input noise current density


AN1494

Table 2 shows the input noise density of each noisesource, the corresponding output noise density and theequivalent noise bandwidth.

For a single pole system, the equivalent noise band-width is equal to the -3 dB bandwidth multiplied by 1.57.Because there is no resistor in series with the op amp'snon-inverting input, IN+ does not contribute to outputnoise.

TABLE 2:

Note 1: The noise gain bandwidth is given byGBWP/GN.

2: The signal gain bandwidth is given by 1/(2*RF*CF).

For the circuit shown in Figure 12, the noise gain band-width is (7.5 MHz)/(113 V/V) = 66 kHz and its equiva-lent noise bandwidth is 66 kHz*1.57 = 104 kHz. Thesignal gain bandwidth is 16 kHz and its equivalentnoise bandwidth is 16 kHz*1.57 = 25 kHz.

The GN is dependent on frequency; it is 1 V/V at lowerfrequencies and gradually becomes higher with amaximum of 113 V/V at higher frequencies. Instead ofintegrating GN over frequency, we simply use 113 V/Vas the noise gain over the equivalent noise bandwidthfor quick noise estimation.

Thus, the output noise from each contributor can beestimated, according to Table 2, and the results areshown in Table 3.

TABLE 3:

The total output RMS noise is equal to 695 µV, which isthe square root of the sum of the individual squaredoutput noise values.

Notice that the op amp’s input noise voltage density(VN) needs to be multiplied by noise gain GN to get thecorresponding output noise density, and the noise gainbandwidth is much larger than the signal gain band-width. This makes VN dominate the total output RMSnoise voltage.

NOISE ESTIMATED BY PSPICE SIMULATION

Figure 19 shows the MCP6491 op amp input noisevoltage density spectrum simulation plot by using theMCP6491 op amp Spice macro model in PSpice, whichmatches the noise density spectrum plot of MCP6491data sheet well.

FIGURE 19: MCP6491 Op Amp Input Noise Voltage Density vs. Frequency.

Figure 20 shows the total output RMS noise voltagedensity spectrum.


Input Noise Density

Output Noise Density

Equivalent Noise Bandwidth

VN VN*GN 1.57*Noise Gain Bandwidth

IN- IN-*RF 1.57*Signal Gain Bandwidth

VN_RJ VN_RJ*(RF/RJ) 1.57*Signal Gain Bandwidth

VN_RF VN_RF 1.57*Signal Gain Bandwidth

Input Noise Density

Output Noise Voltage Density

(nV/Hz)

Individual Output Noise

Voltage (RMS in µV)

VN VN*GN = 19*113 692

IN- IN-*RF = 6 1

VN_RJ VN_RJ*(RF/RJ) = 28 4.4

VN_RF VN_RF = 400 63

oise

Vol

tage

Den

sity

(nV/

Hz)

100

1k

Inpu

t No


10pu

t RM

S N

oise

Vol

tage

D

ensi

ty (n

V/H

z)

100

1k

10k

Tota

l Out D


100

10


AN1494

Figure 21 shows the total output RMS noise voltagespectrum. Within 10 MHz, the total output RMS noisevoltage is 650 µV.

The noise estimated by hand calculation (695 µV) issimilar to the one simulated by PSpice.

For a 4V output voltage signal, the SNR is equal to20*log(VSIGNAL/VNOISE) = 20*log(4V/650µV) = 76 dB.


NOISE FILTERING

In Figure 22, a single pole RC low pass filter can followthe photodiode amplifier to eliminate the noise beyondthe signal gain bandwidth.

FIGURE 22: Noise Filtering.

In Equation 2, the low pass filter’s cut-off frequency (fc)is set to be equal to the maximum allowed signal gain,which gives the minimum rising time (tR) of the outputstep.

For a fixed RF, tR can be further reduced by choosingan op amp with higher GBWP. The higher GBWPmakes the value of CF smaller based on Equation 1,and thus makes fc larger.

EQUATION 2:

As shown in Figure 12, RF = 10 MΩ, CF = 1 pF, thus fCis 16 kHz and tR is 22 µs based on Equation 2.

In Figure 23, the step output responses are shown forthe low pass filters with different fC. Notice that the filterwith lower fC yields longer tR.

FIGURE 23: Step Output Response vs. Low Pass Filter’s fC.

The low pass filter also serves as an anti-aliasing filterfor the subsequent analog-to-digital converter (ADC).The ADC’s sampling rate should be at least two timesof the low pass filter’s fc.

Note: In PSpice probe, the trace expression“SQRT(S(V(ONOISE)*V(ONOISE)))” canbe used to integrate output noise voltagedensity over bandwidth.

700

age

500

600

se V

olta

400

500

S N

ois

V)

300

put R

M (μV

100

200

al O

utp

0

100

Tota

0


VOUT

VDD

RF

I –

+

MCP6491

CF

R

C

fC1

2 RF CF-----------------------------=

tR0.35fC

----------

tR = 10% to 90% rising time (s)

Where

fC = cut-off frequency of low pass filter

2

3

4

Out

put V

olta

ge (V

)

fC = 16kHz

fC = 1.6kHz

fC = 318Hz

0

1

Step

O

Time (ms)0 2 4 6 8 10 12


AN1494

We chose R = 100 kΩ and C = 0.1 nF to make the lowpass filter with fc = 16 kHz. The noise generated by thefilter itself is negligible.

In Figure 24 and Figure 25, the related output RMSnoise spectrum plots with and without filtering areshown, which are PSpice simulation results.



In Figure 25, the total output RMS noise voltage is205 µV within 10 MHz.

For a 4V output voltage signal, the SNR is equal to20*log(VSIGNAL/VNOISE) = 20*log(4V/205µV) = 86 dB,which is 10 dB higher than the SNR without filtering.

PCB Surface Leakage

In photodetection applications, PCB surface leakageeffects need to be considered. Surface leakage iscaused by humidity, dust or other contamination on theboard. Under low-humidity conditions, a typicalresistance between nearby traces is 1012. A 5Vdifference would cause 5 pA of current to flow, which isgreater than the MCP6491 family’s bias current at+25°C (1 pA, typical).

There are several ways to reduce surface leakagesuch as cleaning, coating and guard rings.

Cleaning with isopropyl alcohol helps removeresidues, and coating isolates the surface frommoisture, dust, etc.

The more reliable and permanent solution to reducesurface leakage is using guard rings. As shown inFigure 26, the guard ring drawn in the dotted line is alow impedance conductive trace and it surrounds thesensitive inverting input pin area. The guard ring isbiased at the same voltage as the sensitive invertinginput pin so that there is no leakage current betweenitself and the guarded sensitive pin. In a photodiodeamplifier circuit, the guard ring is directly connected tothe op amp’s grounded non-inverting input pin. Thus,the guard ring blocks the leakage current which wouldflow into the sensitive pin, and sinks it to ground.Moreover, to minimize coupling effects, the circuitconnections within the guard ring should be kept asshort as possible.

For more information on PCB layout techniques,please refer to Microchip’s AN1258 (“Op AmpPrecision Design: PCB Layout Techniques”).

FIGURE 26: Guard Ring Technique.

t Noi

se V

olta

ge D

ensi

ty

(nV/

Hz)

No low pass filter

100

1k

10k

Tota

l Out

put


Adding low pass filter with fC = 16 kHz

100

10

300

400

500

600

700

put R

MS

Noi

se V

olta

ge(μ

V)

No low pass filter

Adding low pass filter with fC = 16 kHz

0

100

200

Tota

l Out


VOUT

VDD

RF

I –

+

MCP6491

CF


AN1494

SUMMARY

This application note reviews the features of Micro-chip’s MCP6491 low input bias current op amps [1], thecharacteristics and operation modes of photodiodes,then it focuses on designing photodiode amplifier cir-cuitry, and several key design points are discussed inorder to improve the circuit’s performance. The noiseanalysis of a photodiode amplifier and the design tech-nique of a low pass filter are also discussed. Finally, thePCB techniques that help reduce the current leakageare briefly introduced as well.

REFERENCES

[1] MCP6491 Data Sheet, “7.5 MHz Low Input BiasCurrent Op Amps”, Microchip Technology Inc.,DS22321, 2012

[2] MCP6481 Data Sheet, “4 MHz Low Input Bias Cur-rent Op Amps”, Microchip Technology Inc.,DS22322, 2012

[3] MCP6471 Data Sheet, “2 MHz Low Input Bias Cur-rent Op Amps”, Microchip Technology Inc.,DS22324, 2012


Note the following details of the code protection feature on Microchip devices:

• Microchip products meet the specification contained in their particular Microchip Data Sheet.

• Microchip believes that its family of products is one of the most secure families of its kind on the market today, when used in the intended manner and under normal conditions.

• There are dishonest and possibly illegal methods used to breach the code protection feature. All of these methods, to our knowledge, require using the Microchip products in a manner outside the operating specifications contained in Microchip’s Data Sheets. Most likely, the person doing so is engaged in theft of intellectual property.

• Microchip is willing to work with the customer who is concerned about the integrity of their code.

• Neither Microchip nor any other semiconductor manufacturer can guarantee the security of their code. Code protection does not mean that we are guaranteeing the product as “unbreakable.”

Code protection is constantly evolving. We at Microchip are committed to continuously improving the code protection features of ourproducts. Attempts to break Microchip’s code protection feature may be a violation of the Digital Millennium Copyright Act. If such actsallow unauthorized access to your software or other copyrighted work, you may have a right to sue for relief under that Act.

Information contained in this publication regarding deviceapplications and the like is provided only for your convenienceand may be superseded by updates. It is your responsibility toensure that your application meets with your specifications.MICROCHIP MAKES NO REPRESENTATIONS ORWARRANTIES OF ANY KIND WHETHER EXPRESS ORIMPLIED, WRITTEN OR ORAL, STATUTORY OROTHERWISE, RELATED TO THE INFORMATION,INCLUDING BUT NOT LIMITED TO ITS CONDITION,QUALITY, PERFORMANCE, MERCHANTABILITY ORFITNESS FOR PURPOSE. Microchip disclaims all liabilityarising from this information and its use. Use of Microchipdevices in life support and/or safety applications is entirely atthe buyer’s risk, and the buyer agrees to defend, indemnify andhold harmless Microchip from any and all damages, claims,suits, or expenses resulting from such use. No licenses areconveyed, implicitly or otherwise, under any Microchipintellectual property rights.

2013 Microchip Technology Inc.

QUALITY MANAGEMENT SYSTEM CERTIFIED BY DNV

== ISO/TS 16949 ==

Trademarks

The Microchip name and logo, the Microchip logo, dsPIC, FlashFlex, KEELOQ, KEELOQ logo, MPLAB, PIC, PICmicro, PICSTART, PIC32 logo, rfPIC, SST, SST Logo, SuperFlash and UNI/O are registered trademarks of Microchip Technology Incorporated in the U.S.A. and other countries.

FilterLab, Hampshire, HI-TECH C, Linear Active Thermistor, MTP, SEEVAL and The Embedded Control Solutions Company are registered trademarks of Microchip Technology Incorporated in the U.S.A.

Silicon Storage Technology is a registered trademark of Microchip Technology Inc. in other countries.

Analog-for-the-Digital Age, Application Maestro, BodyCom, chipKIT, chipKIT logo, CodeGuard, dsPICDEM, dsPICDEM.net, dsPICworks, dsSPEAK, ECAN, ECONOMONITOR, FanSense, HI-TIDE, In-Circuit Serial Programming, ICSP, Mindi, MiWi, MPASM, MPF, MPLAB Certified logo, MPLIB, MPLINK, mTouch, Omniscient Code Generation, PICC, PICC-18, PICDEM, PICDEM.net, PICkit, PICtail, REAL ICE, rfLAB, Select Mode, SQI, Serial Quad I/O, Total Endurance, TSHARC, UniWinDriver, WiperLock, ZENA and Z-Scale are trademarks of Microchip Technology Incorporated in the U.S.A. and other countries.

SQTP is a service mark of Microchip Technology Incorporated in the U.S.A.

GestIC and ULPP are registered trademarks of Microchip Technology Germany II GmbH & Co. & KG, a subsidiary of Microchip Technology Inc., in other countries.

All other trademarks mentioned herein are property of their respective companies.

© 2013, Microchip Technology Incorporated, Printed in the U.S.A., All Rights Reserved.

Printed on recycled paper.

ISBN: 978-1-62076-987-4

Microchip received ISO/TS-16949:2009 certification for its worldwide

DS01494A-page 13

headquarters, design and wafer fabrication facilities in Chandler and Tempe, Arizona; Gresham, Oregon and design centers in California and India. The Company’s quality system processes and procedures are for its PIC® MCUs and dsPIC® DSCs, KEELOQ® code hopping devices, Serial EEPROMs, microperipherals, nonvolatile memory and analog products. In addition, Microchip’s quality system for the design and manufacture of development systems is ISO 9001:2000 certified.


AMERICASCorporate Office2355 West Chandler Blvd.Chandler, AZ 85224-6199Tel: 480-792-7200 Fax: 480-792-7277Technical Support: http://www.microchip.com/supportWeb Address: www.microchip.com

AtlantaDuluth, GA Tel: 678-957-9614 Fax: 678-957-1455

BostonWestborough, MA Tel: 774-760-0087 Fax: 774-760-0088

ChicagoItasca, IL Tel: 630-285-0071 Fax: 630-285-0075

ClevelandIndependence, OH Tel: 216-447-0464 Fax: 216-447-0643

DallasAddison, TX Tel: 972-818-7423 Fax: 972-818-2924

DetroitFarmington Hills, MI Tel: 248-538-2250Fax: 248-538-2260

IndianapolisNoblesville, IN Tel: 317-773-8323Fax: 317-773-5453

Los AngelesMission Viejo, CA Tel: 949-462-9523 Fax: 949-462-9608

Santa ClaraSanta Clara, CA Tel: 408-961-6444Fax: 408-961-6445

TorontoMississauga, Ontario, CanadaTel: 905-673-0699 Fax: 905-673-6509

ASIA/PACIFICAsia Pacific OfficeSuites 3707-14, 37th FloorTower 6, The GatewayHarbour City, KowloonHong KongTel: 852-2401-1200Fax: 852-2401-3431

Australia - SydneyTel: 61-2-9868-6733Fax: 61-2-9868-6755

China - BeijingTel: 86-10-8569-7000 Fax: 86-10-8528-2104

China - ChengduTel: 86-28-8665-5511Fax: 86-28-8665-7889

China - ChongqingTel: 86-23-8980-9588Fax: 86-23-8980-9500

China - HangzhouTel: 86-571-2819-3187 Fax: 86-571-2819-3189

China - Hong Kong SARTel: 852-2943-5100 Fax: 852-2401-3431

China - NanjingTel: 86-25-8473-2460Fax: 86-25-8473-2470

China - QingdaoTel: 86-532-8502-7355Fax: 86-532-8502-7205

China - ShanghaiTel: 86-21-5407-5533 Fax: 86-21-5407-5066

China - ShenyangTel: 86-24-2334-2829Fax: 86-24-2334-2393

China - ShenzhenTel: 86-755-8864-2200 Fax: 86-755-8203-1760

China - WuhanTel: 86-27-5980-5300Fax: 86-27-5980-5118

China - XianTel: 86-29-8833-7252Fax: 86-29-8833-7256

China - XiamenTel: 86-592-2388138 Fax: 86-592-2388130

China - ZhuhaiTel: 86-756-3210040 Fax: 86-756-3210049

ASIA/PACIFICIndia - BangaloreTel: 91-80-3090-4444 Fax: 91-80-3090-4123

India - New DelhiTel: 91-11-4160-8631Fax: 91-11-4160-8632

India - PuneTel: 91-20-2566-1512Fax: 91-20-2566-1513

Japan - OsakaTel: 81-6-6152-7160 Fax: 81-6-6152-9310

Japan - TokyoTel: 81-3-6880- 3770 Fax: 81-3-6880-3771

Korea - DaeguTel: 82-53-744-4301Fax: 82-53-744-4302

Korea - SeoulTel: 82-2-554-7200Fax: 82-2-558-5932 or 82-2-558-5934

Malaysia - Kuala LumpurTel: 60-3-6201-9857Fax: 60-3-6201-9859

Malaysia - PenangTel: 60-4-227-8870Fax: 60-4-227-4068

Philippines - ManilaTel: 63-2-634-9065Fax: 63-2-634-9069

SingaporeTel: 65-6334-8870Fax: 65-6334-8850

Taiwan - Hsin ChuTel: 886-3-5778-366Fax: 886-3-5770-955

Taiwan - KaohsiungTel: 886-7-213-7828Fax: 886-7-330-9305

Taiwan - TaipeiTel: 886-2-2508-8600 Fax: 886-2-2508-0102

Thailand - BangkokTel: 66-2-694-1351Fax: 66-2-694-1350

EUROPEAustria - WelsTel: 43-7242-2244-39Fax: 43-7242-2244-393Denmark - CopenhagenTel: 45-4450-2828 Fax: 45-4485-2829

France - ParisTel: 33-1-69-53-63-20 Fax: 33-1-69-30-90-79

Germany - MunichTel: 49-89-627-144-0 Fax: 49-89-627-144-44

Italy - Milan Tel: 39-0331-742611 Fax: 39-0331-466781

Netherlands - DrunenTel: 31-416-690399 Fax: 31-416-690340

Spain - MadridTel: 34-91-708-08-90Fax: 34-91-708-08-91

UK - WokinghamTel: 44-118-921-5869Fax: 44-118-921-5820

Worldwide Sales and Service

11/29/12

http://support.microchip.com

http://www.microchip.com