[sdiy] Comparator high frequency hysteresis

[Rutger Vlek]

Hi Tony,

Thanks for the explanation! That is indeed a neat trick. In my case, I'm feeding the comparator output to an analog switch. Depending on the logic thresholds of the switch, it may be adding an additional bit of uncertainty when operating on a slow output transition of the comparator. But probably that's of a different order of magnitude. I'll give it a shot!

Rutger




On 28 jan 2016, at 10:46, Oakley Sound wrote:

> > I've read about Tony Allgood's 'transistor starving trick' in the Oakley Sound dual comparator 2 documentation.
> 
> Well 'starvation' seemed a good term for it.
> 
> Basically it's deliberately using a high resistance for the collector resistor on the output of the comparator. This reduces the speed at which the output can transition between low to high thanks to the combination of the now high output resistance and stray capacitance. Combined with the standard two resistor hysteresis network this helps in cutting down the possible problems of a noisy transition.
> 
> Normally one uses a low resistance on the output of a comparator like 1K. This gives mostly symmetrical fast rise and fall times and symmetrical hysteresis. But it also creates perturbations on the power supply rails which can be picked up elsewhere. Using a high resistance on the open collector output, I use 100K, reduces the rise time dramatically and reduces the impact of current pulses. Of course, you then need to tidy the output which is done in the Oakley Dual Comparator by a transistor switch and finally an op-amp for some level shifting and buffering.
> 
> But getting any comparator to be both precise (ie. low hysteresis) and to deal with very slow moving CVs is not trivial.
> 
> Tony
> 
> www.oakleysound.com
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy