Digital Multimeters (DMMs) and Precision DC Sources

cancel
Showing results for 
Search instead for 
Did you mean: 

NI PXIe-4140 Max Load Capacitance

Solved!
Go to solution

Hi All,

 

Is there a spec on the Maximum Load capacitance for the PXIe-4140? I've looked at the spec sheet and cannot see anything in there. Any help appreciated.

 

Thanks,

 

Nick

0 Kudos
Message 1 of 4
(3,465 Views)

There is no such spec.  The PXIe-4141 is a better choice for handling arbitrary capacitive loads, but it may not be necessary depending on what you're trying to do.  In general, heavy capacitive loads will compromise stability more the lower the selected current range.  Do you have a capacitance and current range in mind?

0 Kudos
Message 2 of 4
(3,402 Views)

Hi Chris,

 

Thanks for your response. Sorry, Spec was probably the wrong choice of word. I'm looking to use my existing 4140 to power 4 DUTs being tested in parallel inside a thermal chamber. Because of the length of connection wires involved I think I'll need quite high value decoupling caps (220uf) x 4 in parallel. The DUT has two modes, high and low power mode. In HP the current will be in the order of 20mA/DUT and in LP it will be in the order of 60uA/DUT. I thought there may be stability issues hence my question if there was a maximum load capacitance beyond which stability would not be guaranteed.

 

Thanks,

 

Nick

0 Kudos
Message 3 of 4
(3,390 Views)
Solution
Accepted by topic author nickstorton

That extra detail helps a lot.  Theoretically, that's too much capacitance for good stability.  Practically, though, it could work.  Are you planning on using remote sense or local sense?  If you're using local sense, all that connecting wire will add enough resistance to make the load more resistive than capacitive at the loop crossover frequency.  Ironically, adding even more capacitance at the DUT could make that even more true.

 

If you're using remote sense, then the resistance will be limited to the ESR of the capacitors you use, which might make it more problematic.  There's probably some load bypass network that could be made to work fine, though.

 

Another irony is that in situations like this, "Fast" compensation could be more stable than "Slow," because its higher loop bandwidth hits unity gain at a frequency where the load is more resistive, whereas it's capacitive at lower frequencies.

Message 4 of 4
(3,364 Views)