Help needed with charge sensitive amplifier circuit

[mohindar]

hello sir, we are designing radiation measurement readout front-end electronics, as part of our project ,in very 1st step to understand working of CSPA(charge amplifier) i had done little experiment by connecting photodiode with opa657 .here i have used general purpose ir photodiode and Rf=147k ohm, Cf= 1pf. in feedback with opa657. i am giving 5v pulse train having 50ns pulse width and 2Mhz repetition rate to IR LED which is placed very near to IR photodiode which simulated as radiation source and photodiode will work as detector. now problem is as per theory output of cspa have long decaying pulse having time constant equals to Rf.Cf , but in my experiment cspa output is same like input pulse having 70ns width and 350mV peak amplitude .can you please tell where i am making mistake? i don't understand what's wrong with this circuit. http://file:///C:/Users/Administrator/AppData/OICE_15_974FA576_32C1D314_8B1/msohtmlclip1/01/clip_image002.jpg

 

[davenn]

hi there
welcome to PF :)

can you please repost, better still upload your circuit image to the forum
as at the moment it isn't visible

regards
Dave

 

[mohindar]

here is circuit attached.

 

[davenn]

OK to make it easier to read ...

 

[mohindar]

can you please tell me where i am making mistake in my experiment ? why its output signal has same pulse width as input signal according to theory output pulse is decaying step signal with RfCf time constant.

 

[davenn]

OK
first major problem is that you are supplying it +- 12V when its maximum is +- 6V

 

[mohindar]

oppsss... that is mistake in circuit diagram...in actual circuit i have used +5v and -5v power supply..rest of component according to above post.

 

[mohindar]

thank you for your quick reply. my main doubt is that is it ok to use opa657 for this experiment? and for testing it with ir photodiode ?

 

[davenn]

OK no probs :)

you mite like to try 2 other things to come in line with what is shown in the datasheet

reverse the photodiode and connect its cathode to -V rather than common(GND) rail
place a ~ 50pF cap from the -in of the op=amp to common

I haven't played with op-amps for this particular purpose before, my stuff has all be low noise high gain for
seismic signal amplification before the ADC circuitry

Dave

 

[mohindar]

yes i have already done that reversing that photodiode , but output is same ! main problem is that Cf should discharge with Cf.Rf time constant but here it is discharging within 70ns but in circuit time constant is in terms of us.

 

[pretty_fly]

I had similar problems and couldnot undestand still. This made me think that the situation is about the properties of opamp simulation models. All my treatments about simulating especially such small signal stuff was unsuccefull. But you may find some solution when you look at from this view.

 

[LvW]

yes i have already done that reversing that photodiode , but output is same ! main problem is that Cf should discharge with Cf.Rf time constant but here it is discharging within 70ns but in circuit time constant is in terms of us.

Calculation: T=147ns , OK? And you are measuring (simulation) 50% of this ? So the problem is a factor of 2, right?

What do you mean with "in circuit time constant is in terms of us"?

 

[jim hardy]

i am giving 5v pulse train having 50ns pulse width and 2Mhz repetition rate

can you please tell me where i am making mistake in my experiment ? why its output signal has same pulse width as input signal according to theory output pulse is decaying step signal with RfCf time constant.

I gather you're trying to differentiate your input signal?
Recheck that theory...

Your circuit has DC gain, so it's a preamplifier not a differentiator.
so i'd expect output to be same shape as input with perhaps slower rise and fall times.

See fig 2 here:
http://freshlab.org/detektor/?lang=eng
first amplifier inside "amplifier & shaper" block is a differentiator.
It will produce 'decaying pulses' like you describe.

a quick google search turned up this page which looks at first glance to be rich with detail

http://iopscience.iop.org/0031-9155/55/7/003/pdf/pmb10_7_003.pdf

but I'm not well versed in such circuits.

good luck,

old jim