51pf vs 50pf vs 47pf

[ranchak]

I've noticed that some projects use a 51pf cap and some a 50pf. Would it hurt if one would substitute either value to use what you had on hand? What if you used a 47pf ?

[ildar]

Almost no audible difference. Given the variations in components due to tolerance a 50pf cap with, say a 10% tolerance can be anywhere from 45pf to 55pf.

[MikeH]

I would say it you'd never notice the difference between a 51pf cap and a 47pf cap.  I usually just use the closest value I have on hand, unless it's way off.  Depending on what type of cap it is, the tolerance could put it off by several pf's.  Some caps have a tolerance of 20%.  That means a 50pf cap with a 20% tolerance could actually be as low as 40pf, or as high as 60pf.

I that right?  Anyone want to back me up?

Edit:  Ooops, looks like someone already did.

[Mark Hammer]

Depends where.  If it is simply a cap in the feedback loop of an op-amp to keep a high-gain circuit from oscillating at high frequencies, often VERY large variations in value are quite acceptable.  I often see things like a 220k feedback resistor and a 10pf cap in parallel  That provides a rolloff above 72.3khz.  For guitar circuits, you can easily live with a rolloff down to 10khz for "clean" circuits, and often less than that for distortion.  So in that instance, finding out that you have some 33pf caps but no 1-pf left is no hardship whatsoever.  Stick the 33pf in there and you still have almost 22khz of smooth sailing.

Other times, every pf makes a difference.  For instance, when a cap less than 100pf is used to set the clock range for the delay chip on a chorus or flanger pedal, small differences in cap value can yield different "colours" of chorus or flange.  They won't impede functioning one single bit, but the delay range may be a little higher or lower than what you find satisfying and to your tastes.

So, the question to really ask is: how much will changes in cap value of this much impact on functioning in this part of the circuit?