Op-Amp Oscillating

[wuchy143]

Hello,

If you look at the attached schematic you will see an op-amp (U1A) in the bottom middle of the page. This is the area of concern. I'm trying to debug the issue with this PCB and I've landed to this amplifier. There are a couple issues:

1. I don't understand the op amp configuration. It looks like an integrator due to the cap in the feedback network but I'm not really sure because it doesn't function like one. I thought integrators operate such that depending on how long a voltage is at the input the amplifier increase or decrease its output(depending on polarity). When i look at the input terminals with a scope neither of these voltages vary at all. They both sit at 3V. Though, the output swings from 0V up to 4V.

2. The output when at 4V oscillates peak to peak by about 3/4V!!! Why is this? It should be a clean 4V.

 

[crutschow]

The resistor across the capacitor makes the circuit a low-pass filter not a integrator. The -3dB point of the 0.1uF cap and the 75k ohm resistor is 21.2Hz.

The oscillations could be due to C37 and R54 which appear to form a positive feedback path between the op amp output and the plus (+) input. Don't understand the purpose of those components.

 

[wuchy143]

The resistor across the capacitor makes the circuit a low-pass filter not a integrator. The -3dB point of the 0.1uF cap and the 75k ohm resistor is 21.2Hz.

The oscillations could be due to C37 and R54 which appear to form a positive feedback path between the op amp output and the plus (+) input. Don't understand the purpose of those components.

Thanks. That makes sense as the cap and resistor work together to form the low pass filter.

I didn't understand the function of those discretes either. Does anyone else know why they would be there? I think I may just rip those out one by one to see if there is a difference.....

 

[wuchy143]

I made some progress with this issue and have a quick update.

I removed C37 and R54 from the amplifier to rule them out in creating positive feedback. After doing that I still have oscillations.

I now suspect that the oscillations are further upstream in this signals path. Meaning, U5(DAC to the left of the amp) as the source for the problem. The oscillations are seen at the output of the DAC. I checked the power rails of this device and they are nice and stable.

I'm sort of lost as where to go from here with this issue. What would make a DAC oscillate like this? I guess it could be the software doing something weird but it's a perfect sine wave @2Hz.

I have the device setup with MPLAB in debugger mode and I"m trying to wrap myself around how the code works at the moment as I was not the programmer for this device. Stumped

One piece that I'd like to mention is when I pause the microcontroller in debug mode whatever D/A reading it was at it stays just like that on my scope. Nice and clean. No oscillations. So it appears that software may be doing this....

 

[crutschow]

I believe the SDI input is what controls the output of the DAC. So I would check there. I does sound like the D/A is just responding to the input.

 

[wuchy143]

I thought the same thing but nice clean digital signals on that line.(and all inputs to the D/A)

You're right I think the DAC is doing what's it's told. Time to dig into the code.

Thanks

 

[wuchy143]

I wanted to add that the sine wave oscillation I'm seeing on my output is at 2Hz.

 

[wuchy143]

Blah. Narrowed it down to the DC power supply. Used a battery and the oscillations went away entirely. doh!

 

[ericgibbs]

The resistor across the capacitor makes the circuit a low-pass filter not a integrator. The -3dB point of the 0.1uF cap and the 75k ohm resistor is 21.2Hz.

The oscillations could be due to C37 and R54 which appear to form a positive feedback path between the op amp output and the plus (+) input. Don't understand the purpose of those components.

hi Carl,
The capacitor and series resistor across that OPA input [-inp/+inp] is sometimes used to reduce common mode noise, in this case possibly ADC state switching. [C37 , R54]

 

[SneaKSz]

I can give you the example code for using the MCP DAC ( software SPI), if your interested. I pulled LDAC low so the DAC always updates its output with the new incoming data.


Kind regards

 

[Warpspeed]

Check that the isolated +5v supply is clean and at +5v.
Then check that the outputs from the clock and data opto isolators are switching between proper logic levels.

Sick clock or data bits going into the serial DAC will drive the output nuts, as might noise spikes on the 5v supply.

My money is either on a sick opto, or a faulty decoupling cap.
The battery may have changed the supply voltage just enough to get a marginal opto functioning.
It may not be noise on the supply, but a dying opto isolator.