>
>
> That looks like a really nice chip. I was limiting myself to PDIP packages
> because they're easier for DIYers (like myself) to solder, but I've
> certainly been tempted by the newer chips that aren't available as PDIP.
Yes, thought I would supply the boards with the chip soldered on .... in
bulk shoudn't cost much ..
>
>
> Do you have a strategy for sampling the full voltage range with only the
> chip and input protection? I have been focusing exclusively on rise-time
> algorithms as I expected nearly all strikes to generate more voltage than
> the ADC could handle, but I am a software guy, not a hardware guy.
>
>
Since the piezo's are very high output impedence devices, though that I
could probably get away with just a resistor divider network on each input,
if the series resistor is high enough, the internal
pin protection should be enough .... but need to check this
more throughly...
>
>> I haven't done the measurements yet, either, so I can't answer the
> question, but related somewhat: can you do interrupt driven code so that
> you get signaled when the ADC senses the beginning of the strike? My hope
> was that I could get interrupted when the ADC sensed the voltage rise above
> zero by some small margin, and then get interrupted again when the voltage
> hit some predetermined value and calculate the velocity by the length of
> time between those interrupts...
>
>
Can't quite do the interrupts that easily without extra hardware support
(i.e routing each input through a diode to pin which can interrupt on a
rising/falling edge ..)
What I thought of doing was get the A2D to sample each input continuously in
time with a sample clock going quite fast (1ms or even 500us, I can
sample/convert all channels in 100us), at the end of each sample window I
would get interrupted and look at each channel for an edge, once I have an
edge I would start a counter for that channel, then on subsequent interrupts
check for falling below the threshold and report the count as a velocity
measurement ... or something like that ... I will see what I can get out of
the hardware.
Another possibility instead of doing it time based ... is once you have an
edge, to accumulate the A2D channel until falling below the threshold ...
that should give you a velocity as well ... might not be as good though so
I'll try the time based method 1st.
> Also, I think you only mailed me, but I think you meant to mail the list...
>
Oops ... just hit reply without looking.
Cheers,
Bernie
>
>
>
>
------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
Hydrogen-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/hydrogen-devel
