Q about noise in time interval measurement averging
- From: "colin" <colin.rowe1@xxxxxxxxxxxxxxxxxx>
- Date: Fri, 06 Apr 2007 18:42:02 GMT
Hi,
I have a PIC measuring a time interval to 25ns resolution,
the interval is totaly asynchronous to the PIC clock.
So say if I average over 25000 measurements that will give me a limit of 1ps
resolution.
Im trying to do a system noise analysis and im wondering how to work this
out,
I cant seem to recall how the noise reduces with increasing samples,
is it 1/samples ? this would be too good to be true,
or 1/sqrt(samples) this seems rather low.
there is also a lot of noise in the signal as it is from an optical encoder,
seems mostly mechanical, I need to reduce this by averaging over a long time
too.
Im trying to work out the optimum rate of pulses per revolution to use.
thanks
Colin =^.^=
.
- Follow-Ups:
- Re: Q about noise in time interval measurement averging
- From: bill . sloman
- Re: Q about noise in time interval measurement averging
- Prev by Date: Re: Yo! RF dudes!
- Next by Date: Re: OT: An Inconvenient "Truth"
- Previous by thread: need simple sample and hold chip
- Next by thread: Re: Q about noise in time interval measurement averging
- Index(es):
Relevant Pages
|
