05-31-2009 12:00 PM
I’ve seen some funky behavior with a SCXI-1100, and was wondering if anyone knew why it behaved the way it did. It takes time for the signal to level out, why? For example, when MAX (Measurement and Automation Explorer) is run in an acquisition mode of 100 samples at 1kHz, the signal value takes time to ramp up (see attached image). In this example, it took about 140 ms to level out. That means at this voltage level, I have to wait this long for a good signal, killing my acquisition rate.
There’s another twist to this: it doesn’t happen the second time. If run again right way, you see a constant signal (except for noise). On the other hand, if you switch to another task in MAX and run it, and then switch back to the first task and run it, then it ramps up again. Why?
06-01-2009 12:26 PM
Hi Agile,
Do you see this behavior only for acquisition of this particular signal? If you try acquiring a known constant signal (eg a power supply), do you see the same behavior? In looking at the image you attached, it looks like you are reading the signal off of a Viatran500 pressure sensor, yes? Do you see the same behavior for the other DAQmx tasks you are working with?
06-01-2009 02:25 PM
What are your jumper settings for W2,3,4? You may have 4Hz filtering enabled.
See table 2-4 in the manual:
http://www.ni.com/pdf/manuals/320637c.pdf
-AK2DM
06-01-2009 06:19 PM
Thank you David.
Yes, I see it only for acquisition of the signals on the 1100. We are also using an 1102B for thermocouples, and don't experience the same ... it looks like an RC curve.
I don't see how applying a straight DC voltage would make a difference. The voltage coming into it is DC anyway and originally comes from a DC power supply through the pressure transducer, so if I understand it correctly, the pressure transducer is attenuating the 10VDC signal.
The DAQmx tasks involved are all analog input:
06-01-2009 06:32 PM
AnalogKid2DigitalMan wrote:What are your jumper settings for W2,3,4? You may have 4Hz filtering enabled.
See table 2-4 in the manual:
http://www.ni.com/pdf/manuals/320637c.pdf
-AK2DM
Thank you AK2DM,
I bet this is it. The response is like an RC time constant from a low pass filter. I don't have access to the hardware right now, but I'll ask someone to take a look. I'm pretty sure this is the solution, but I won't mark it as such until I hear back from them. It might be a while if they've started their testing.
Thanks again to everyone.
06-02-2009 09:38 AM
Agile:
Yes, I immediately thought it looked like an RC time constant too. Hope it is simple as that. You may also want them to check and document all the other jumper settings too for future reference.
-AK2DM