PXI-6225 loading when measuring DC Voltage Ripple

[SpineTester]

When measuring DC voltage ripple with NI PXI-6255 board, it is causing a DC drops of ~70mV when LabView executes "start task" in PXI-6255 LabView Driver. No loading is applied to that measuring DC voltage node and it appears that this PXI-6255 is internally loading and causes a DC drops and resultes in higher ripple. I can understand this output if I apply some sort of load in this node. When task is clear, this DC level goes back to where it support to. Based on the 6255 specifiation, it has the internal resistance of 10G Ohms. Differential measurement method was used in my LabView code.

My question is why causing a DC drops whenever LabView executes "start task" with no load applied to this measuring DC output node? Am I missing something? Your input is greatly appreciated.

Test system: LabView 2009 Service Pack 1, Window Vista Business (32-bit), NI PXI-6255 (M Series)

[D-Cubed]

Hi SpineTester,

Can you please provide more information about your system? I specifically would like to know the following:

1. What are you using to measure this 70mV drop when the start task executes?
2. How do you know the voltage you read in LabVIEW is incorrect?
3. Why are you performing a differential measurement?

With more information, I think we will be able to work this problem out.

Regards,

[zqhector]

I'm seeing something similar.

On the measuring side, I have a USB-6255 connected to a SCB-68.

On the circuit side, I have:

1. A battery

2. An analog multiplexer (connected to the battery).

When I measure the output of the analog multiplexer with an oscilloscope, I looks fine and clear. The same as if I was measuring directly at the battery.

Then, when I connect the analog input (differential mode, of the SCB-68:USB6255) to the output of the analog multiplexer, a lot of noise gets added to the voltage at that point.

If I measure directly at the battery, the voltage looks clean. The, if connect the analog input to the battery, the signal stays the same: clean.

I don't understand what is the root cause of this.

Can you guys please help me out?

Thank you,

Hector

[SpineTester]

My setup is very simple using a differential measurement; targeted voltages to be measured were connected to PXI-6225 thru cable (connector 0) from my hardware board. I was measuring the total of 4 different DC voltages and they all reference to common AGND simultaneously.

Originally, I setup the sampling rate as 250Ks/s for all 4 channels (wrong!) and that causes DC-drop (30 to 70mV range - my original issue) depending on source voltage driving to these DC voltages. The issue was, I think, I cannot use 250Ks/s as I was measuring 4 channels simultaneously and the sampling rate should have been 250Ks/s divided by 4. Then I started using ~60Ks/s as my sampling rate and this DC-drop was still causing. Remember this only happened when LabView executed start task. I was then starting to decrease the sampling rate until I didn't see this DC-drop and right around 10Ks/s; I no longer see this loading effect! Bottom line is the higher my sampling rate, the higher the DC-drop (delta from original voltage).

I still don't understand why this is causing based on the sampling rate. PXI-6225 claims 10G ohms parallel with 100pF cap as an analog input resistance. So using 60Ks/s for all 4 channels should not violate over-sampling rule. It seems to me that using higher sampling rate causes loading effect on these measuring DC nodes. By the way, I was measuring unloaded voltage for all 4 channels and determining the ripples.

As always, your input is greatly appreciated!

[D-Cubed]

Hi SpineTester,

Have you ensured that for the measurement everything has been grounded properly?  A great document which goes over this can be located here: NI Developer Zone: Field Wiring and Noise Considerations for Analog Signals.  Please let me know if this is helpful.

Regards,