Comparator Ringing

[kieran.french]

Hey,
I was wondering if anyone out there can help me with a problem im having with 'ringing' on a comparator output. As the output changes from logic level '0' to logic level '1' there is a short burst of oscillations on the output. This is really undesirable as the comparator output is used as an input to a quadrature waveform generation circuit that ultimately inputs encoded signals to a digital tachometer. The ringing on the comparator output causes spikes to occur on the velocity profile of the digital tach.
At the moment I am using positive feedback on a LM311N comparator. The positive feedback has reduced the ringing considerably but it is still affecting the velocity profile on the digital tachometer.
If anyone has any idea how to get rid of the comparator ringing entirely I would be extremely grateful as this is all part of a final year undergrad electronics degree project and is due for assessment in 4 weeks.
Thanks,
Kieran.

 

[Styx]

The multiple ringing will most likely be donw to multiple crossings of the V+ and V- signals

you have stated that you are using +ve feedback and that has improved things.

there is a chance that your source signals are quite noisy (are they?)

If that is the case think ablut putting some caps to ground to clean them up a bit

Also a 22pF cap between the +ve and -ve terminals can aid (or hinder) a comparators performance

 

[Roff]

Hey,
I was wondering if anyone out there can help me with a problem im having with 'ringing' on a comparator output. As the output changes from logic level '0' to logic level '1' there is a short burst of oscillations on the output. This is really undesirable as the comparator output is used as an input to a quadrature waveform generation circuit that ultimately inputs encoded signals to a digital tachometer. The ringing on the comparator output causes spikes to occur on the velocity profile of the digital tach.
At the moment I am using positive feedback on a LM311N comparator. The positive feedback has reduced the ringing considerably but it is still affecting the velocity profile on the digital tachometer.
If anyone has any idea how to get rid of the comparator ringing entirely I would be extremely grateful as this is all part of a final year undergrad electronics degree project and is due for assessment in 4 weeks.
Thanks,
Kieran.

Good power supply decoupling, a good ground plane, and a little hysteresis will generally clean up a comparator output unless you have a lot of nonmonotonicity (noise or ringing) on the input. The hysteresis needs to exceed the amplitude of the noise to be effective.
Edit:
I guess I just restated what Styx said.

 

[eblc1388]

Hi kieran.french,

In addition to what Ron and Styx had suggested, please makes sure you are not causing the ringing simply by measuring it with a probe with high capacitance.

Isolate the comparator output so that it is not going to anywhere, and use a 100K series resistor on the scope probe tip to see whether the ringing is still there.

 

[kieran.french]

thanks guys, the ringing is there because of multiple crossings of positive to negative voltage, i guess ill just use hysteresis and hopefully that will sort it out as i cannot afford any ringing on the output whatsoever.
Thanks again

 

[Roff]

thanks guys, the ringing is there because of multiple crossings of positive to negative voltage, i guess ill just use hysteresis and hopefully that will sort it out as i cannot afford any ringing on the output whatsoever.
Thanks again

Do you know why the input transition is nonmonotonic? That is the best place to tackle the problem.

 

[Styx]

Since you have identified the ringing if due to the comparator doing it's job and not its fault, have you thought about using a low-pass filter on those signals

it will produce a slight delay in the forward path but it will clean it up

 

[eblc1388]

Could the output of the comparator changing state affects the supply voltage and thus also changes the comparator reference input if this is derived from the same supply using two resistors? Try bypassing the reference voltage with a capacitor and see if thing improves.

 

[kieran.french]

hey,
I have bypassed the supply voltages using 0.1uF capacitors, it makes a slight improvement but still not quite good enough. I have tried adding additional hysteresis but again to no avail. Iv attached a scope shot of the ringing. The top waveform is the comparator outptut, and the bottom two waveforms are a set of quadrature waveforms for which the comparator output is used to generate.