TIDA-01187: OPA2356

Hello Shyam,

  Yes, you are correct the issue is the AC coupling capacitor. The reason why is because the OPA2356 is used in single-supply configuration, if it was used in split-supply configuration it would have worked. DC coupling is not required for differential to single ended conversion path in the usual sense, but since OPA2356 is used in 3.3V single supply, it will need a DC voltage to bias the voltage since it will saturate and not function properly if the common-mode is at 0V. 

   Another method is to keep AC coupling capacitor, then add a voltage bias at +IN pin of the amplifier, but above would be the best action to take for an easy fix. 

Thank you,
Sima 

Thanks for your response

My Vcm is 1V

I remove the 1p Cap inbetween the output terminals of THS and below is the output voltage

At -Ve Terminal

At +Ve Terminal

If i see the output on +ve terminal is double the -ve terminal, i am not sure of reasons

i dont see any output at the output of OPA

I tried removing coupling caps and shorted the pads, the output is nothing,

i tried the coupling caps to remain 0.1uF, at that setting the output comes only for optical pulse width of 200nS and less and if PW is increased the output goes OFF.

for above setting the output is only a 20nS pulse.

Kindly throw some light on where exactly i am going wrong

where as in simulation removing coupling caps also makes output to disappear

i strictly trusted simulation and put 49.9ohm in series with 0.1uF , but in reality output is nothing

so the coupling caps may not be the issue and the issue is something else.

Hello Shyam,

  Thank you for the additional details, yes it looks like something else is off due to the double output.

  Above scope shots are from VOUT+ and VOUT - after removing C50 from the schematic. Can you also provide the scope shots from OUT- and OUT+ nodes?

Thank you,
Sima

i dont have a scope shot now, but i can say, the output is as per simulation only

on OUT+ the swing is in mV and on OUT- the swing is almost 1V i think, the input is from TIA OPA857

i dont think there is a problem there.

do you want any other scope shots ?

Hello Shyam,

  Is below the correct configuration that you have set up on your board?

   The FDA is setting the output common-mode voltage at around 0.94V. This output voltage will become negative at the output of the amplifier. And if add back in the AC coupling capacitors, it will be as shown below which is a negative input to the amplifier which will violate both input and output of the amplifier.

   Unfortunately, the only solution to fix this is to use split supply on your difference amplifier, for input source above this will be 3V to -2V supply. Or add a voltage bias to VIN+ pin of both the amplifiers. 

   I know this is not possible for you, and apologize for the inconvenience this has caused. I will do a redesign and we can probably eliminate the FDA and use a single channel difference amplifier from the OPA857 to the comparator. I will be working on a redesign, and will update you by Monday. Let me know if this redesign direction is feasible for you.

Thank you,
Sima

Why cant i tie directly THS output to my comparator, where Vref i can put arround 1V+ ?

Hello Shyam, 

  Unfortunately, that will not work as well, because similar to setting VOCM pin to + value, it will just be inverted at the dual amplifiers, which will result in 0V difference:

  The only option that would work is to add a voltage reference here which will not be possible on your board since it is most likely tied to your ground plane:

  Sorry, I could not do the redesign yet, I will be working on it and send you an update by tomorrow.

Thank you,
Sima

I have set the Ref voltage of comparator TLV3601 to 1.2V and taken the ouput from +ve output of FDA THS and it worked, there is some damping that is observed, but i was able to manage.

Hello Shyam,

  Oh, I see what you mean I misunderstood and thought the 1V would be tied back to the THS output. Yes, that would work for now, but the issue with this method would be distortion performance will be not good due to one output being used.

  Would you like me to help with the redesign if you are able to re-spin the board?

Thank you,
Sima

Yes i would definitely need a glitch free FDA to single ended output to comparator to go ahead, for now i will manage with this work arround, kindly guide in redesign

Hello Shyam,

  I am really glad you found a work around for your testing. We will work on a redesign and have it reviewed, and I will get back to you for an update on our end.

Thank you,
Sima 

when i tried to tie the output of THS directly to comparator, due to gain peaking at my requirements i could not rely on it, but if there is a way to reduce the gain peaking i can avoid the multiple pulses which occur due to lower thresholds, hope you understand what i meant, as there will be sub harmonic peaks, the comparator will generate new pulse next the pulse i have

My PW is 1uS-20uS , but rise time is important to be maintained at 1nS

My PRF is 1mS-20mS

but its not necessary to maintain the PW at my Rx, so i switched back to the transformer approach by unmounting the OPA2356 circutry and mouted back the transformer

what i did not understand is as  i reduce the reference voltage of comparator TLV3601 the pulse width also decreases at transformer output, why this happens ?

below is the screen shot of the transformer output, which goes to comparator, the input given at Photo diode is 1uS PW and 20mS PRF

Here the Blue is transformer output and pink is the comparator output

what is happening now is the comparator output is not stable at falling edge, even though there is change in input to the comparator i am seeing a spurious pulses at output of comprator like below

This sometimes is more, currently my reference is 800mV, the comparator should not through a second rising pulse again as per my design

kindly help in understanding why this issue is happening, is it due to comparator output circuit? which goes to RC circuit for Loss of signal.

or is it because of comparator being fast ? is there any pin equivalent low speed comparator ? i am not able to understand this kind of performance, i feel this is purely due to comprator because you can see the input to comprator is not having any change in signal.