Multifunction DAQ

cancel
Showing results for 
Search instead for 
Did you mean: 

Inaccurate peak to peak voltage

I am using an usb 6358 DAQ.  In the process of troubleshooting something in the field we were applying known signals to the AI of the DAQ.  For DC signals the DAQ reported accurate values (both through the data collection VI and through MAX).  However when we applied a sine wave we got twice the peak to peak voltage.  For example if we applied a 1 Vpp signal (which should be -0.5 V to 0.5) we would see 2 Vpp via the DAQ (again both through the data collection VI and through MAX).  We tried a range of frequencies (1 Hz to several kHz) and got the same result.  I am sure there is a simple explanation but it escapes me right now.

 

Thanks.

0 Kudos
Message 1 of 12
(3,865 Views)

Just some quick thoughts without knowing yours system.

 

Is there a scale being applied?

Is the peak-to-peak measurement incorrect (assuming coding was involved to calculate it)?

Is there a buffer that is applying a gain in the circuit?


GCentral
There are only two ways to tell somebody thanks: Kudos and Marked Solutions
Unofficial Forum Rules and Guidelines
"Not that we are sufficient in ourselves to claim anything as coming from us, but our sufficiency is from God" - 2 Corinthians 3:5
0 Kudos
Message 2 of 12
(3,862 Views)

I am viewing the signal so I am not calculating the peak to peak, just reading it off of a graph.  I have no gain settings for my DAQ.  I have tried changing the range (which affects the internal gain and resolution) but this does not affect the reading.  I am using a differential measurement (no other choice for this particular DAQ) and plugging directly into the function generator.

 

Another piece of the puzzle:  The function generator (a modern digital model) reports that it is outputing a 1 Vpp signal and shows a nice little graph with a sine wave oscillating b/t -0.5 V and 0.5 V (as you would expect).  We checked this on an o-scope and it was indeed within this range.  When I read the Vrms directly from the o-scope using a multimeter, I get a Vrms of 0.7071 which matches a Vpk of 1 V which would then imply the signal really is 2 Vpp. 

0 Kudos
Message 3 of 12
(3,860 Views)

Is the AI GND connected to the system ground?  Even though you are using differential measurements, you still need to have grounds common.  I've seen floating grounds do some really weird things.


GCentral
There are only two ways to tell somebody thanks: Kudos and Marked Solutions
Unofficial Forum Rules and Guidelines
"Not that we are sufficient in ourselves to claim anything as coming from us, but our sufficiency is from God" - 2 Corinthians 3:5
0 Kudos
Message 4 of 12
(3,856 Views)

I am putting the signal into AI+ and the function generator ground to AI-. 

0 Kudos
Message 5 of 12
(3,847 Views)

What is this source? What does it expect to be terminated with? Sounds like an impedance mismatch where the source might be expecting 50 ohms.

Message 6 of 12
(3,845 Views)

I've been wondering about that.  I connected the function generator directly to the o-scope via a 50-ohm BNC cable and set the o-scope channel to 50-ohms.  The output on the function generator is labeled as 50 ohms. 

0 Kudos
Message 7 of 12
(3,842 Views)

There you go, then. The DAQ card and DMM are high impedance. The setting of the source is double when you don't terminate in 50 ohms. If you were to change the scope and use a regular probe in high impedance mode, you would see exactly the same thing as the DAQ card.

0 Kudos
Message 8 of 12
(3,839 Views)

yep, slap a 50Ω across the terminals.  I've run into that problem more times than I care to admit.

0 Kudos
Message 9 of 12
(3,836 Views)

Thanks for the feedback.  I don't quite understand why the low impedance would result in double the voltage.  We're in the middle of tesing now but I will give this solution a try.  Thanks to everyone for their input.  I'll give you guys a tenative "Kudos" until I can verify!

0 Kudos
Message 10 of 12
(3,831 Views)