Linux UL - error in differential voltage with USB-1608G

[matt_sgf]

Hi

I'm using a USB-1608G and I connected a voltage device in the differential set-up. I installed the Linux driver.

I adapted the sample code 'AIn.c' to work for differential voltages by setting:

    inputMode=AI_DIFFERENTIAL;
    range=BIP1VOLTS;

However, the voltage reading is low. If I apply 1.0V I read around 0.998V. This uses the ulAIn function.

If I plug the device in to a Windows laptop and read the voltage using DAQami I get 1.0V.

Also, I adapted the sample code 'AInScan.c', which uses the ulAInScan function, to read differential voltages and this gave approx 1.0V.

Is there a problem with ulAIn voltage reading? Perhaps I am missing a required setting? Why does the ulAInScan give a different (and correct) number?

Thanks

Matt

[DAQman]

Hello,

I did the same here, switching Ain.c to AI_DIFFERNTIAL and BIP1VOLTS. I used a voltage calibrator set to 1.0000 volts for my test voltage. If I let the program run for a few seconds, the measurement toggles between 0.999 and 1.000. I got slightly better results by modifying the program to use lowChan = 0 and highChan = 0 and attaching my test signal to channel 0. I didn't see any appreciable difference using the DAQami or the AInScan example. 

Best regards,

John

[matt_sgf]

Hi

I've just tried again. I made the edits to the example, see attached. I applied 1.0V and the value reads low still.

I am using channel 0 with a 100k resistor between the low and the ground as instructed.

I also edited the Python code example with the same result. I'm using OpenSuse 15.4.

Can you suggest anything I might be missing?

Thanks

AIn_dif.c

[DAQman]

Please try the attached C example and tell me what you get for the average value.

 

AIn_Average.c

[matt_sgf]

Thanks for the code.

I consistently get +0.9988 @ 1.0V

I get +0.6988 @ 0.7V
and +0.4987 @ 0.5V
 

[DAQman]

I'm not able to duplicate the lower measurement using ULDaq 1.2.1. I run the AIn_Average.C program, which returns a consistent 0.9998v. I also have a 6.5-digit multimeter on my voltage source output, which reads 0.99997v. 

DAQami doesn't have an average feature.  Instead, I wrote a Windows C program similar to the AIn_Average.c and it returns a consistent 0.9959v. The lower value may have to do with the counts to voltage conversion function using a float type instead of the double used by ULDaq. 

 

[DAQman]

I figured out why I was getting inconsistent results; the low-side input reference resistor connection was loose. I'm using a 47k ohm resistor from the low side input to the ground primarily because my voltage calibrator is isolated from the system ground. Now I get good results using AIn, AInScan, Linux, or Windows.  I get a 200 reading average of 0.9999v consistently. I suspect that when you switch from your Windows computer to the Linux box, your system grounding changes ever so slightly.