OPA657 Op-Amp oscillating for unknown reasons

Thread Starter

pargregory

Joined Apr 14, 2021
1
Greetings,
I am working on a Lidar system for a senior project and am running into issues I am unable to solve with my circuit. The project requires the use of an amplification and comparator circuit in order to take the detector output up to a level that will trigger a TDC controlled by a MSP430. The detector output voltages are around 2 mV and in order to trigger the TDC we require a >3.3 V signal, thus the ampflifier and comparator. The system requires high speed components in order to maintain the integrity of the time of flight calculations so I chose the OPA657 decomponsated op-amp for its greater performance, and planned on adding external compensation to account for this.

I used Texas Instruments simulating software to simulate the design(included in the link below) and saw that I got the amplification and timing I wanted. I saw no signs of oscillation at this point. When moving to actually building the circuit however measuring the output of the amplifier with an oscilloscope revealed low frequency oscillations. I have tried active bandpass filters, low & high pass filters, DC block at the input, and used suggested resistor/capacitor values to reduce oscillations as seen in the datasheets. The circuit is currently mounted on a demo-board designed for this kind of op-amp, so I am at a loss as to the source of these oscillations. Even with a basic non-inverting setup, there were still oscillations.

If anyone has any insight into this problem or anything I can try, help would be very much appreciated. I am including links at the bottom to the basic non-inverting setup, and can try to answer any questions to the best of my ability or post anything you need to see. Thanks.

Circuit Setup(non-inverting):
 

Papabravo

Joined Feb 24, 2006
21,225
It looks like you have a gain of 50 and you have an AC source coupled directly into the amplifier. Is that correct? If so a couple of millivolts on the input will send the output close to the rails. I see no reactive elements in the schematic so it seems un likely that oscillation would occur naturally, but there is obviously something we cannot see from your diagram The image is horrible and I can't see clearly what is going on. Is there any way you can provide a better image, with more contrast and better readability?
 

Bordodynov

Joined May 20, 2015
3,180
What is the source of the signal, its impedance? How long are the power wires going to the amplifier circuitry? What is the load on the amplifier - load impedance?
 
Last edited:

du00000001

Joined Nov 10, 2020
117
At high gains, even the slightest supply ripple and/or ground bounce can lead to "oscillations". (Might look like oscillations for you. In reality it's just the amplified disturbances.)
Thinking about it: are these oscillations in the 50/60 Hz range? (Or 100/120 Hz?)
 

DickCappels

Joined Aug 21, 2008
10,180
Who knows...at 1.6 GHz the smallest layout problem could make it oscillate. One thing to try is to use a reduced gain-bandwidth amplifier and see whether it performs ok oscillation wise. That will tell you something. In the mean time, people here have asked quite a few relevant questions.
 

RPLaJeunesse

Joined Jul 29, 2018
254
It could be oscillating at a far higher frequency than you can detect. Use of a 5K feedback resistor seems rather high for a 1.6GHz amplifier, and the demo board is rather generic so not RF optimized. Just for fun put a 10pF chip cap across the 5K feedback resistor and see what happens.
 

Bordodynov

Joined May 20, 2015
3,180
It could be oscillating at a far higher frequency than you can detect. Use of a 5K feedback resistor seems rather high for a 1.6GHz amplifier, and the demo board is rather generic so not RF optimized. Just for fun put a 10pF chip cap across the 5K feedback resistor and see what happens.
This is a bad idea. With such a capacitor the circuit gain at high frequencies will be close to one. And for this amplifier a stable gain of 7 is guaranteed! There will be oscillation even if there was none! For such high frequency circuits the simulation should be set with the input source impedance! Otherwise you may get nonsense.OPA657_tran.png
 

RPLaJeunesse

Joined Jul 29, 2018
254
This is a bad idea. With such a capacitor the circuit gain at high frequencies will be close to one. And for this amplifier a stable gain of 7 is guaranteed! There will be oscillation even if there was none! For such high frequency circuits the simulation should be set with the input source impedance! Otherwise you may get nonsense.View attachment 235713
Good point, I missed that part about the amp. So maybe put 1K in series with the added cap to roll off the HF gain and still keep it above 7? It seems all the datasheet examples expect either 50 Ohms or added capacitance at the non-inverting input. If the non-inverting input doesn't have either of those, how would you expect the amp to behave?
 
Top