Kai Klaas wrote:

"Ok, you measure the time of one period. I assume that you average signal before displaying? How do you do it?"

No, since I wont about 500ms stable displaying of the last result, I use another approach:

I count for minimal 500msec both, the periods and the time which was gone during this count of periods.
Otherwise there is the danger, that on high speed I get to less counts to achieve the wanted resolution.

So I get always a wide range and a high resolution on the minimum and also the maximum measuring value.

Orientation on the human sight gave the best solution for RPM or frequency measurement.


"I think of many audio applications..."

Yes, on some applications a PLL must be used.
I made also some experience with it. Fortunately I needed only an accuracy of only 5%. So the jitter was visible but tolerable.

If you search for jitter and PLL on Google you see, that both are married very strong. And often its a real challenge to divide both far enough for the wanted accuracy.

E.g. on one article I have seen there is a bug on the 74HC4046 which can cause big jitter under some circumstances and it was suggest to replace it with the Philips 74HC9046 where the bug was removed.


Peter