回复： Clarification and Confirmation About the Practical Implications of Common Mode Voltage on Differential Measurements

[chassan]

Hello folks,

I have read multiple documents on the NI site regarding differential vs. common mode measurements and I think I understand it, but I'm looking for confirmation through two examples (please).

Both examples assume a NI daq card with +/-10V inputs and a max working voltage of +/-11V (e.g. PCI-6052 with programmable gain set to 0.5). 

Example 1:  Assume a DC voltage divider with two resistors and a differential input configuration on the daq card.  First resistor = 4-Ohms connected to 100VDC, second resistor = 96-Ohms connected to ground (0VDC).  This produces a 4VDC drop across the first resistor.  However, I believe that the differential voltage of 4VDC cannot be connected directly to the daq card because the absolute (i.e. common mode?) voltage is actually from 96VDC to 100VDC which is outside the "working range" of +/-11V for the card.  Is this conclusion correct?

Example 2: Assume a differential measurement between two AC waveforms and a differential input configuration on the daq card.  First signal = 25v@0deg, second signal = 10v@180deg.  From some of the literature on the NI website, I have gathered that the common mode voltage is the mean of the vector representation of the input signals.  In this case it would be (25v@0deg + 10v@180deg)/2 = 7.5v@0deg.  I believe that this signal cannot be directly connected to the daq card because one of the signals (25v@0deg) is outside the working range of the card (even though the resultant "mean of the vectors" is within the working range of the daq card.  Is this conclusion correct?

Thanks in advance for any assistance,

chassan   

[Aaron_W. · Accepted solution]

Chassan,

You are correct with both of your examples. The best way to consider this is that even though you are taking a differential measurement the max voltage that you can subject a single channel to is the +/- 11V for the 6052E. Hope this clears things up a little bit. 

[MF]

So let me get this straight, you should never use a differential measurement with a DAQ board if either of the 2 inputs to the differential are over 11V whether AC or DC? This means voltage dividers do not work in the differential mode. 

I have a 26V AC signal that I need to measure. I am using a PCI-6221 DAQ and I have a voltage divider with a 39 kOhm and a 13 kOhm resistors for an expected voltage of 6.5 V AC. What is the mode to measure the voltage in, RSE or NRSE?Thanks so much for all you help.

Michael

[MF]

Hello Again,

I will answer my own post above. I found out that it is possible to use a voltage divider if you are working with a voltage over 11V. The whole post made me realize that there is a correct and an incorrect way to connect the voltage divider reduced signal. For example (mine), I reduced a 26VAC signal down to 5.2VAC and 20.8VAC signals using a simple resistor voltage divider One way is to hook it so that the voltage supply hot lead is connected to the 5.2VAC signal and the other lead is connected to the 20.8VAC signal. 

With that you have 26 VAC -> 1.5 kOhm -> 20.8 VAC -> 6 kOhm -> 0 VAC     - as you can see the 26VAC and the 20.8VAC go to the DAQ and even though the measured signal is only 5.2VAC, the 2 end points are greater than 11VAC and the DAQ cannot handle it. 

Reversing the order and connecting the hot lead to the 20.8 VAC you get:

26VAC -> 6 kOhm -> 5.2 VAC -> 1.5 kOhm -> 0 VAC  - now the DAQ receives 5.2 VAC and 0 VAC for the differential measurement and life is good.

Thanks NI for your support.

Michael

[xcuresme]

But I still have one question: if the ground of voltage soure is isolated from the earth ground, the ground of DAQ, then can it work?

[johnsold]

In that case you will need an isolated input module because you cannot determine the instantaneous voltage between the DAQ input and the DAQ ground.  If your signal is AC only, another option is to use a transformer with suitable isolation rating between primary and secondary windings.

Lynn

[xcuresme]

Hi, Lynn.

I think I am a little bit confused about "you will need an isolated input module because you cannot determine the instantaneous voltage between the DAQ input and the DAQ ground."

Let's look at example 1. As shown in figure1, if I isolate the ground of the DC voltage source by using an isolated AC/DC power supply and connect the DAQ and the signal following the NI suggestion about differential measurement of a floating signal, then I think the voltage between the DAQ input and the DAQ ground is specific, that is ±V1/2. 

Is that right? Can it work?

Thanks a lot!

xcuresme

[johnsold]

I just posted to your other thread.

Similarly to what I said in that thread, you could ground the junction of the 4 and 96 ohm resistors and use a single-ended input.

Will you be making both voltage and current measurements?

Please tell us the whole story and keep it all in one thread.

Lynn

[xcuresme]

I have posted a new thread here.(http://forums.ni.com/t5/Multifunction-DAQ/Can-floating-a-signal-be-a-proper-way-to-solve-the-problem...

Waiting for your suggestion.

Thank you!