--- In "j.chitte" <> wrote:
>
>
>
> --- In "hans.schillstrom" <hans@> wrote:
> >
> >
> >
> >
> >
> >
> > --- In "j.chitte" <j.chitte@> wrote:
> > >
> > >
> > >
> > >
> > >
> > > --- In "mccarl1978" <mccarl1978@> wrote:
> > > >
> > > >
> > > > >
> > > > > The one big downside is that it seems to be set (non adjustable) at
> > > > > +/-10V and I need 0-5V
> > > > >
> > > > > The readme says it's set in the code but at a first reading I can't
> > > > > see where. Anyone looked at this?
> > > > >
> > > >
> > > > The lines that you are looking for are lines 168 & 188 of adc_hwtimer.c
> > > >
> > > > Check adc_hwtimer.h for the defines that you can use for different
> > > > ranges, etc.
> > > >
> > > > Cheers
> > > > Phil
> > > >
> > >
> > > Thanks , it was a great help knowing that's all I needed to look at.
> > >
> > > I now have the platform and /dev drivers and my userland C code to work
> > > with.
> > >
> > > Am I correct in thinking that the primary limit on sampling freq is the
> > > timer resolution? It seems it can't do much better than 10kS/s on single
> > > channel readings.
> > >
> > > What would be the best strategy for approaching the 100k limit of the MAX
> > > chip? I know I won't get there but I'm interested in the limits of the
> > > board.
> > >
> > > Oh, and thanks for a nice clean example of how to write a kernel driver.
> > > ;)
> > >
> >
> > The timer resolution on the EP93xx is normally 508kHz for Timer1-3, and
> > that's not the limit in this case.
> >
> > My experience from the board is that the jitter will be 1-15ms
> > The kernel will turn off interrupts for several ms, (up to 15ms on a 7250
> > board with 2.6.29 kernel)
> > The requirements in my case was a sampling rate of 2.5 kHz with jitter less
> > than 200us. I guess that you can figure it out the limitations of the
> > kernel.
> >
> > I implemented Xenomai (a Real time Linux patch) to the 7250 board with
> > dynamic timer ticks. It works great, with an interrupt latency of ~20-120us
> > Right now I'm running Max197 at 5kHz with RMS calc on 4 channels and it
> > gives a CPU load of 1.5%.
> >
> > Regards
> > Hans
> >
>
> Thanks for the info. It's interesting to see what others have managed to get
> out of the board.
>
> MAX197.pdf says:
> 6us conversion time , 100ksps sample rate.
>
>
> TS manual states:
> After the write cycle to location 0x10F0_0000, the MAX197 completes the A/D
> conversion in about 12 mS
>
> Three orders of magnitude slower that the chip spec. (Did they mean 12us ?!)
>
> Your results on jitter suggest it can not sample reliably faster then 8ksps
> (120us) , a Nyquist freq of 4kHz even with a RT layer.
>
> That's rather disappointing.
>
That's the way Linux works...
If you need high speed sampling use DMA and a suitable DAC.
EP93xx and MAX197 can't do it for you.
There is a bunch of "intelligent" PC-104 boards with ADC:s on the market, or
simpler boards with buffers so readings could be done in batch mode.
------------------------------------
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/ts-7000/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/ts-7000/join
(Yahoo! ID required)
<*> To change settings via email:
<*> To unsubscribe from this group, send an email to:
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|