Looking through the TRM for the PSoc 5 and the Switched Capacitance op amps, I notice that there exist either a 400 femtofarad or an 800 femtofarad cap on the input, and a selectable feedback resistor ( I believe 20k =<Rsub_f=< 900k). Is it possible to get these put together to make a differentiator? We obviously have the option to make an integrator/TIA, but it would be nice to be able to use the differentiator at the very least for edge detection. Can this be done? If not, could we get this option for Creator 2.1x purty please Cypress?

Differentiators are inherently unstable, so you have to build a modified one. This reference show possibilities

http://www.ti.com/lit/an/sboa092a/sboa092a.pdf

Note, this would take external components, and an internal OpAmp.

You could also use A/D and DFB to effect a differentiator. Tons of references

on web for approach.

Regards, Dana.

Thanks Dana, as usual, spot on answer. I guess I should have known that a classic differentiator is unstable (impedance of input element goes to zero as omega goes to infinity). External components it is...sigh. All: the good stuff is on page 38 of the document Dana links to.

 Yeah, I know you can use an ADC and some computation to differentiate (based on Taylor polynomial central or backward diiferencing) but you won't find any fast edges that way...good for noise immunity, bad if you want any decent performance (it would make even 1970's RTL logic look blazing fast).

RTL, cowabunga buffalo Bob.

I have some RTL and DTL laying around, sometimes they made great linear amps,

well maybe not that  great, sometimes great logic. Blazing fast Tr, Tf in the uS area.

Great memories.

Regards, Dana.