0
\$\begingroup\$

Here is the link for two different measurement systems:

http://zone.ni.com/reference/en-XX/help/370466V-01/measfunds/refsingleended/

In RSE case, the AIGND is the point where all channels' grounds are connected. This AIGND is directly connected to the measurement system ground. I was having some offset issues and found out the offset comes from AIGND and measurement system ground's potential differences.

In NRSE case, the AISENSE is the point where all channels' grounds are connected. But this AISENSE is not connected to the measurement system ground.

The company I contacted told me to try NRSE case to reduce the offset.

I have two questions here:

1-)Why in RSE case there is potential difference between AIGND and measurement system ground?(Why this wouldn't be an issue in NRSE case?)

2-)Why would one use RSE if NRSE would introduce less offset?(Does NRSE has some other disadvantages?)

\$\endgroup\$
6
  • \$\begingroup\$ Companies should be taxed for every acronym they introduce. \$\endgroup\$
    – Sredni Vashtar
    Commented Oct 30, 2015 at 10:53
  • \$\begingroup\$ I've noticed that you have asked many, many questions and done the right thing in "accepting" quite a few of them but there are still about 50% of the questions unaccepted. I've done a bit of a scan thru them and plenty seem to give perfectly good answers. Maybe you should consider that today is payback time for quite a few folk!!! \$\endgroup\$
    – Andy aka
    Commented Oct 30, 2015 at 11:01
  • \$\begingroup\$ Sorry I think my previous questions were not put clearly. \$\endgroup\$
    – user16307
    Commented Oct 30, 2015 at 11:02
  • \$\begingroup\$ That doesn't exempt you from abiding by the unwritten rules and giving back a little something for all the hard work put in by many people in taking the time to make very reasonable answers. \$\endgroup\$
    – Andy aka
    Commented Oct 30, 2015 at 11:04
  • \$\begingroup\$ Sure Im not happy either, just confused really. This is my last try for this question. \$\endgroup\$
    – user16307
    Commented Oct 30, 2015 at 11:09

2 Answers 2

Reset to default
1
\$\begingroup\$

Why in RSE case there is potential difference between AIGND and measurement system ground?(Why this wouldn't be an issue in NRSE case?)

Seperate grounds will be at different potentials - this is a fact of life unless you are making a measurement to a device up-close to the acquisition system. Earth currents are the usual culprit in an industrial environment - current flows thru earth and creates a volt drop. This produces a measurement error.

If you have an NRSE system you can remotely measure the ground reference point directly and because no current flows down this extra wire there is no measurement error caused by earth currents.

Why would one use RSE if NRSE would introduce less offset?(Does NRSE has some other disadvantages?)

NRSE can only work over a limited range i.e. it can only cope with a few volts of offset between measurement ground and acquisition ground.

Just to show you that a little bit of time put into both questions and answers (as well as trying to learn from answers) is its own reward, here's a picture and a simple (hopefully) explanation: -

enter image description here

There are two sensors labelled V1 and V2. They share a common ground connection back to the measurement device. They also share a common 5V excitation voltage (power supply if you wish). Sensor 1 takes 1mA whilst sensor 2 takes no current. The cable return back to the measurement device has 0.1 ohms in the 0V return wire. This will inevitably drop 100 uV due to the 1mA flowing. Even though sensor V2 is still producing 2V with respect to it's local 0V, the measurement device measures 2V + 100uV.

Now if the measurement 0V were seperated from power 0V then the measurement system would measure exactly 2V.

\$\endgroup\$
5
  • \$\begingroup\$ Here is my confusion: 1- ) There is no earth loop currents to loop because the inputs are isolated. So "Seperate grounds will be at different potentials " argument is still valid? 2- ) I still could not understand what you mean by "NRSE can only work over a limited range i.e. it can only cope with a few volts of offset between measurement ground and acquisition ground." It doesn't make a picture on my mind. Im wondering why? \$\endgroup\$
    – user16307
    Commented Oct 30, 2015 at 11:26
  • \$\begingroup\$ If the inputs are truly isolated then no problem but you have to consider what current flows down that line from all the sensors you might have - this can give rise to an error and crosstalk. The limited range of NRSE is due to the limited range that the input wire has with respect to its amplifier's local ground. You can't expect the floating AIN reference to be taken outside the power rails of the amplifier - it's not a true differential measurement device. \$\endgroup\$
    – Andy aka
    Commented Oct 30, 2015 at 11:31
  • \$\begingroup\$ I still didn't get it, it is not clear for me. I couldn't find any drawing would help me to get it. But since no one else could answer and you are complaining about accepting answers. I accept your answer even though it is still something mystery for me. Thanks for the answers. \$\endgroup\$
    – user16307
    Commented Oct 30, 2015 at 15:31
  • \$\begingroup\$ If sensors share a common return wire and that common return wire has non-zero impedance then current from one sensor will wiggle the sensor end of that wire about and in turn the other sensor's hot lead will have the same wiggle - this would be measured as 0V and a sensor output with a wiggle i.e. the wiggle is an error due to crosstalk. Does that make any more sense. I'm not complaining BTW. \$\endgroup\$
    – Andy aka
    Commented Oct 30, 2015 at 15:36
  • \$\begingroup\$ could you make a little drawing? i would be very glad. which currents? i cannot picture in my mind. \$\endgroup\$
    – user16307
    Commented Oct 30, 2015 at 15:39
1
\$\begingroup\$

NRSE is often called 'pseudo differential'.

If you measure with respect to ground, and due to currents flowing between chassis, the ground 'here' is a few mV different from the ground 'there', you will see errors imposed on your measurement.

Having a non ground referenced measurement avoids that difference, without the cost of two wires per channel.

\$\endgroup\$
1
  • \$\begingroup\$ But could you tell me what could be the disadvantage of NRSE over RSE? If so, why? \$\endgroup\$
    – user16307
    Commented Oct 30, 2015 at 11:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.