05-31-2023 03:17 PM
I am using a PCIe Series X DAQ card with LabVIEW to measure voltages around 12V in magnitude. Since the input voltage range of my card is GND+-10V I have been feeding in a 5V supply to the GND of the card and then adding it to the measured voltages to get the correct value. Recently I have upgraded to DAQmx 2023Q3 and LabVIEW 2023Q1 and it seems like this solution stopped working. Right now all the voltages that should be around 12V are clipping at 10V irregardless of what I feed into GND. Does anybody know if I can go back to the old way of how things worked? If it was one of the software updates that broke it which one should I revert to?
05-31-2023 03:27 PM
That doesn't sound like "software" to me, but rather like you've shorted your "offset voltage" (2 V?) to ground someplace. If you updated to a new version of LabVIEW, you probably got new Drivers and a new version of MAX. Check to see that the A/D converters don't have a software setting somewhere connecting them to Ground, for example ...
Do you have an ordinary (very analog) voltmeter? Can you see if you see a potential between the A/D "ground" and what you think is a reasonable "ground" for your equipment? I'll wager a dime it is close to 0 V.
Bob Schor
05-31-2023 04:41 PM
Hi Bob,
Thank you for the answer, so my set up simplified is as follows:
ai1 --> power supply negative
ai2 --> 12V analogue signal
aiGND --> power supply positive (5V)
My expected output is -5V on ai1 and 7V on ai2. Since the update the output I get is -5V on ai1 and 10V (clipped) on ai2.
Hope that might give some additional insight into the issue.
05-31-2023 07:53 PM
Hmm. Let's say you have a signal you want to measure that varies from 0 to +12 V relative to "system ground" (whatever that is). You have a measuring device (your DAQ card) that measures from 0 to +10 V, relative to the same "system ground" (let's further assume the two grounds are connected together, and you are recording differentially, not single-ended).
Suppose you solder a 10 Mohm resistor to a 2 Mohm resistor, connect the free end of the 10 M to the negative input of the A/D converter, the junction of the two resistors to the positive input of the A/D converter, and the free end of the 2 M resistor to your 12 V signal. You measure voltages from 0 to 10 V when your "signal" varies from 0 to 12 V, so you need to multiply your voltages by 1.2. There is probably a way to set up DAQmx to do this multiplication for you, so it "reports" 0 - 12V.
Note -- consult an actual Electrical Engineer about such a Voltage Divider circuit. There may be more to be considered than my simple model suggests.
Bob Schor
06-01-2023 07:08 AM
@Bob_Schor wrote:Suppose you solder a 10 Mohm resistor to a 2 Mohm resistor, connect the free end of the 10 M to the negative input of the A/D converter, the junction of the two resistors to the positive input of the A/D converter, and the free end of the 2 M resistor to your 12 V signal. You measure voltages from 0 to 10 V when your "signal" varies from 0 to 12 V, so you need to multiply your voltages by 1.2. There is probably a way to set up DAQmx to do this multiplication for you, so it "reports" 0 - 12V.
You will have a problem if using more than one channel due to the high impedance of the divider. You will want a simple Voltage Follower op-amp circuit after the divider to give that low impedance for the DAQ.
Also, DAQmx has Scales as a feature. You just create the scale and then apply that scale to the channel.
06-01-2023 09:24 AM
Don't try weird ground raising tricks, you are asking for trouble as it can blow the front end of your DAQ and worst case create unsafe conditions for the user.
What you are looking for is called Signal Conditioning and here's a link to a pretty good paper on the subject.
06-04-2023 01:38 PM
I have started a new thread where I tried this approach instead of ground raising, but it comes with its own set of problems: changing input voltage range with resistor divider bridge Can you help me determining what the exact problem is?
06-05-2023 09:41 AM
Dear crossrulz,
thank you for your answer. Above I have linked a post that shows how I have built up the divider. Could you please elaborate a bit on what the problem is and why I would need an extra amp if I want to measure multiple channels?