Jochen Kuepper wrote:
>
> David Schleef writes:
>
> > Hmmm... bad idea. Sound cards have a few properties that make them
> > good sound cards, but bad data acquisition cards.
>
> > - input is generally heavily filtered, with unknown filters, but
> > generally cut out anything below 20 Hz, around 50 or 60 Hz (depending
> > on your local line frequency), and above 10 kHz. The <20 Hz
> > filter is nasty, since it makes DC measurements impossible.
>
> Reminds me. I do have a ComputerBoards PCI-DAS1602-16 board. Currently
> I use it as a plain single channel D/A converter :-( to generate a
> voltage ramp.)
>
> The voltage of this ramp (or simply a constant output) shows very
> prominent 50 Hz -- though not all time ! Some days it is not visible
> on a 1 mV scale (the lowest setting of my bread-and-butter scope) and
> as such is good enough (for me). Most days I do have problems with my
> application due to the 50 Hz though (because is is big, like 10 mV
> p-t-p, or such). The shape isn�t exactly sinus-like, it�s almost a
> (bad) rectangular wave.
>
> (... I am pretty sure this is not a software problem, sorry for the
> off-topic posting :-)
>
> Are others out here seeing similar problems with D/A conversion ? What
> are you doing to avoid/nihilitate/minimize such problems ?
>
> I assume I simply see the line-frequency somewhat modulated by the
> card. The converters itself are hosted in a small Faraday-box (at
> least I hope it is one) on the card. Still, are such noises induced by
> e-smog inside the PC, through the PCI bus, or both ?
>
> I appreciate any hints,
> Jochen
I think you are right. Since you are in Europe, 50 hz are coming from
the power lines, both through the ADA card and trough the sensors. It is
also quite common in our labs.
Fixing comes by avoiding ground loops. Simply said but difficult to
solve. So
try to connect all groundings with a common wire, do not trust the
grounding of your plugs.
On the sensor side it can be of help to power them with battery packs,
thus avoiding any galvanic connection to the electric power net.
Powering them using DC-DC converters, or very well isolated DC supplies
can be an alternative solution, but it can be costlier.
Further improvements are brought by using diffrential signal. Most ADA
board have such a capabilty. It can be a problem if you have many imput,
as in that way you halve the input afforded by the card and must use
two.
My modest experience says that SB16 is not a so bad AD if used above 30
hz, and 16 bits at 44 Khz are enough. Since SB16 compatibles are
available at a cost 70 times less than a reasonable 16 input board, with
the advantage of as many AD and DA converters, sound cards can be an
alternative for > than 20 hz applications. The only problem is that you
have not so many slot in your PC.
We have used them under RTAI. Unfortunatly we have programmed them
directly in the real time functions using them and cannot provide a
driver. If anybody is interested I have available the DOS program from
which we took the coding of SB16. We derived our code by a simple cut
and paste. In my opinion sound cards programming is more easily learned
from the many program made available by DOS hackers than from Linux
drivers. That's what we did. The only problem is that DOS hacking can be
not up to date.
Ciao, Paolo
-- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
---
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
