Acquiring a hi-speed square wave

[fuzzycontrolfreak]

Hello everyone,

I am acquiring a digital signal from a rotating shaft encoder. The square wave comes at a rate of 36,000 pps, at almost constant intervals. What sampling rules does this application follow. Shall I go after the Niquist Criterion, or a 36kS/s is enough to catch all the signals.

Thankyou very much for your help.

[gleichman]

QUOTE(fuzzycontrolfreak @ Feb 25 2008, 08:35 AM)

Hello everyone,

I am acquiring a digital signal from a rotating shaft encoder. The square wave comes at a rate of 36,000 pps, at almost constant intervals. What sampling rules does this application follow. Shall I go after the Niquist Criterion, or a 36kS/s is enough to catch all the signals.

Thankyou very much for your help.

That depends. What data are you trying to extract from your signal?

[ASTDan]

I would sugest using a counter timer input to measure encoder signals

Using a counter/timer channel will be more than adequate to capture your 36k/S signal

http://zone.ni.com/devzone/cda/tut/p/id/3321

Dan

[Tim_S]

QUOTE(fuzzycontrolfreak @ Feb 25 2008, 08:35 AM)

Hello everyone,

I am acquiring a digital signal from a rotating shaft encoder. The square wave comes at a rate of 36,000 pps, at almost constant intervals. What sampling rules does this application follow. Shall I go after the Niquist Criterion, or a 36kS/s is enough to catch all the signals.

Thankyou very much for your help.

If you know you're getting a 36 kHz signal and you want shape data, then you should be sampling at ten-times the expected maximum frequency (twice for frequency). The sugestion of the encoder is good if you want to be synchronous to the shaft (data collected in constant degrees apart rather than time). What you intend to do with the data after you've collected determines which method would be best.

Tim