On Wed, Feb 18, 2004 at 03:29:19AM -0800, Pieter Hoeben enlightened us thusly
> Hi,
>
> I am working on an application where either
> a microprocessor or electronics must measure
> two times. The data comes in a 100 MHz clocked
> A/D. Two pulses are detected, looks like (if it
> looks garbled, use font courier new 8):
>
> | /\
> | / \
> | / \ /\
> |____/ \___/ \__
> ---------------------
> The time from 0 to peak 1 and peak 2 are needed.
> The voltages can differ, so the processor must
> look where two peaks are. The time between the
> peaks can be something like 100 nsec to 5 usec.
> The problem is that the processor only has
> 50 usec to measure the two peaks (data will
> probably come through DMA), analyse them and
> send them somewhere (probably flash or to another
> processor through a fifo). We cannot afford
> a expensive environment.
>
> My question:
> What microprocessor/controller would you recommend?
> Do you think a 'normal' controller would do, or
> would it be an ARM or DSP? Which compilers are
> (low cost) available for those processors?
>
> I can also try to do it in electronics, that would
> be no problem for me, but software would be more
> intelligent and better in finding the peaks. And
> easier to change...
The minimum time between 0, peak1 and peak2 is the vital information you
didn't give us. Gathering that it will all happen in 50uSec, peak
detection would be difficult. Certainly timing form 0 to peak would be
difficult.
Can you use the voltage value for peak 1? If there is a constant filter,
that might work. Another lazy approach if the pulses are symmetrical
would be to feed them into a digital 'delay line', and pick out the
midpoint. Here's what I mean
> | /\
> | / \
> | / \ /\
> |____/ \___/ \__ becomes the following pulse train
000000000111110000011000
You'd probably sample a bit faster than that, but you get the idea.
Either way, your software wouldn't have to break out a sweat.
--
With best Regards,
Declan Moriarty.
--
Author: Declan Moriarty
INET: [EMAIL PROTECTED]
Fat City Hosting, San Diego, California -- http://www.fatcity.com
---------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB CHIPDIR-L
(or the name of mailing list you want to be removed from). You may
also send the HELP command for other information (like subscribing).
