I got some troubles while conducting a low-level capacitance measurement using PXI-4072.
Our device is a pipe whose length is 40cm, internal diameter is 50cm, and external diameter is 60cm. We enwrap a layer of copperplate outside the pipe to get a pole for our measurement loop. The varnished wire inserting inside the pipe is the other pole.
We use a LCR meter in our lab to measure the capacitance of the pipe, and the value is about 10pF but when we use PXI-4072, the value is instable. Sometimes it is UNDRN, and sometimens it is a number. In general, we need to touch the negative to get a number. If we do not, the reading is UNDRN.
We think that the ground is not necessary for capacitance measurement, but we do not know why sometimes the negative should be touched to get a reading. At times we connect a wire which is about 3.2m long to the negative, and the reading is normal. Sometimes we smoothly conduct the measurement on the forenoon, but at night the reading returns UNDRN. The above phenomenas is odd.
Now we need a solution for our experiment.
Thank you for using the forums. In order for me to understand what is going on with your application, would it be possible to post a schematic of your system? It would be very helpful too, if you could post a screenshot of the UNDRN and where you are receiving it. I would like to be able to more fully understand your application before I can make a recommendation.
Some photos are added to the attachment to illustrate my application.
Thank you for taking those photo's for me they were very helpful. In regards to your application, is it possible to try taking a the capacitance of a different capacitor and using different ranges. Also do you happen to have another 4072 that you could test this out on. This will let us know if there is something wrong with the card or if just a bad measurement is being taken.
According the Digital Multimeter Help, if you search NaN the UNDRNG occurs when the measurement is below 5% except when using the 300 pF range, in which case there shouldn't be an UNDRNG. So everything should be right in your test setup.
Please let me know if those suggestions help.
Not sure if you got your capacitance issue under control. I looked at the pictures you submitted and it appears that the DMM leads are not fully seated. I might be wrong, but if those are NI probes in 3.jpg, they are not fully seated and thus the stimulus signal from the DMM is never making it to your DUT. Let me know if that does the trick; if not, we can dive deeper. Have a wonderful everything!
You might also try using a 1uF capacitor to see if you receive an overrange condition.
Can you give us a hint what resolution and accuracy you need for the cap measurement?
Would it help to measure at a certain frequency?
How often do you need to measure?
What is the other card I see in your PXI rack?
Looking at your pictures, I'm in doubt that for 10pF you use the right cabling (don't know the 4072)
OK, a short look into this paper explains a little bit more...
You definitly need to use a better cabling 🙂 (some hints are given in the paper)
Did you do a open/short compensation before the measurement?
The DMM measures with 1kHz, 10pF is -i*16MOhm@1kHz
Your tube is a great antenna, so grounding the outside will defintly help.(That's what you do by touching it 😉 )
3m cable is a lot of additional wire capapacity. Of course the open/short compensation will cancel it, but roughly 100pF/m -> 300pF to 10pF isn't that nice (make shure to stay in the 300pF range). Active guarded cabling might be a solution (PXI-4022?? )
My next try would be using a BNC coax to banana cable to your tube, shield outside to the tube and the Lo connected to ground at the PXI , the connector at the tube is used for open/short compensation.
If the open cable already gets too close to 300pF* look for cable with a lower capacitive load or use an active guard .
*) I don't know if the 300pF rage is before or after the compensation... something NI guy might better answer.
A few days ago, an engineer from NI came to our lab to check up our mearnment system. He was in doubt that the chassis, PXI-1042Q, may has problems. Changing the chassis is his suggestion, and now we are waiting for the mail of a new chassis.
Had you tried to acheive any overrange condition by changing the capacitor?