Theoretical Question: Op amps induce dc noise. A friend once suggested that I should put a ac filter after each op amp in my circuit. I find that doing that greatly attenuates the signal. So I just use a ac filter just before the ADC. Is there merit to my friend's suggestion?

 

Theoretical Question: Op amps induce dc noise. A friend once suggested that I should put a ac filter after each op amp in my circuit. I find that doing that greatly attenuates the signal. So I just use a ac filter just before the ADC. Is there merit to my friend's suggestion?

There's no such thing as DC noise!
Generally, you AC couple between stages to remove any DC offset.
Choosing the right values of coupling capacitor would eliminate any signal attenuation.
What are you trying to measure with your ADC?

 

What is "DC noise" since by definition, DC has no noise?
If you mean op amps have AC noise on their outputs with a DC input, then yes, that is true.

You would typically use a low-pass filter before the ADC input with a corner frequency at or just above the highest signal frequency of interest.

If the filters after each amp attenuated the signal, then their corner frequencies were too low.
You can't just use some arbitrary filter to get the desired results.

 

What is "DC noise" since by definition, DC has no noise?
If you mean op amps have AC noise on their outputs with a DC input, then yes, that is true.

You would typically use a low-pass filter before the ADC input with a corner frequency at or just above the highest signal frequency of interest.

If the filters after each amp attenuated the signal, then their corner frequencies were too low.
You can't just use some arbitrary filter to get the desired results.

Yeah, I mean DC input. You are right, I get it now, the attenuation is due to low cutoff of the AC coupling ("coupling", I didn't remember that term earlier hence called it "filter"). So just before the ADC I do the coupling in whatever circuit I build (@Ian0 I am just asking a general question as I do this with every circuit). Is that sufficient? Any benefit to coupling after each op amp?

 

I think by "DC noise", you mean that opamps have input offset voltages that get amplified and appear at the output.

Whether you need to remove them after each stage depends on a few things. First, what is it you are trying to measure? More specifically, what is the range of frequencies in your signal of interest?

If the amplified offset doesn't cause saturation problems, then there is no compelling need to remove it at each step. Instead, you can let it accumulate until just before it becomes a problem and then remove it. Remember, every component in the circuit adds some amount of noise.

But, unless your signal contains pretty low frequencies that you want to keep, you shouldn't have much problem choosing appropriate coupling capacitors to avoid attenuating the signal you want to keep. If you ARE attenuating it, then you are also likely distorting it quite a bit, too.

 

I think by "DC noise", you mean that opamps have input offset voltages that get amplified and appear at the output.

Whether you need to remove them after each stage depends on a few things. First, what is it you are trying to measure? More specifically, what is the range of frequencies in your signal of interest?

If the amplified offset doesn't cause saturation problems, then there is no compelling need to remove it at each step. Instead, you can let it accumulate until just before it becomes a problem and then remove it. Remember, every component in the circuit adds some amount of noise.

But, unless your signal contains pretty low frequencies that you want to keep, you shouldn't have much problem choosing appropriate coupling capacitors to avoid attenuating the signal you want to keep. If you ARE attenuating it, then you are also likely distorting it quite a bit, too.

Thank you. Under what frequency would you say a frequency is low?

 

hi John,
Post an amplifier specification and if possible diagram.
E

 

hi John,
Post an amplifier specification and if possible diagram.
E

Depending on the application, I've got in my inventory LM324s, LM358s, OP07DPs, , MCP6004s, MCP6002s. All give some some DC offset. Sorry no diagrams, I am asking a very theoretical and general purpose question here to add to my circuitry SOPs.

 

Thank you. Under what frequency would you say a frequency is low?

It entirely depends upon what frequency range you are trying to measure, and you have yet to state that critical piece of information.

Incidentally, a coupling capacitor is a high-pass filter (attenuates frequencies below its corner frequency)

 

The lousy old LM324 and LM358 are extremely noisy and also produce crossover distortion.
An audio opamp (OPA134 single, OPA2134 dual and OPA4134 quad) are made for hifi audio have low noise that cannot be heard and is difficult to measure.

 

Thank you. Under what frequency would you say a frequency is low?

Low and high are both vague and subjective terms.

If you wanted to digitize human heartbeat which is around 1Hz, it would be difficult to remove DC offset using a high pass filter. It would be better to use a summing amplifier to subtract the DC offset.