Difference between notebook and desktop offset measurements

[NancyH]

Hello,

We are having an issue with measuring an offset in a analog input channel. 

On our instrument we have AISENSE and AOGND connected to the backplane of our instrument which is grounded to the electrical outlet.  We read the offset in an analog input channel which we have grounded to the electrical outlet.  Therefore, we expect it to read 0 V, which it does when we use a PCI 6024E (configured NRSE) in a desktop computer but when we use the PCMCIA 6062E (configured NRSE) in a notebook we do see a an offset.  The offset is measureable but is of a different magnitude when the notebook is powered by battery or AC adapter (not grounded).  Our guess is that there is a grounding issue.

Any ideas?  Should we connect the AIGND on the board to the chasis of our instrument? 

Thanks.

[Chris_K_]

Hi NancyH 

Thank you for using our discussion forums. 

I have some resources for you to look at.  First off here is the Field Wiring and Noise Considerations for Analog Signals .  This explains all the proper connections for each of the input configurations.

First off I will suggest if at all possible, switching to a Differential Input configuration.  This lowers the number of input channels that you will have available, however it will be the best way to reduce the voltage offset that you are seeing in your measurements.  If that is not possible and you must use NRSE I would suggest looking at “Table 1: Analog Input Connections”, and depending on the type of signal you are using configure your connections like the bottom two diagrams associated with NRSE measurements. These configurations should help reduce the offset that you are seeing.

If you are still having this offset issue after properly connecting your input signal, can you let me know the level of offset that you are seeing, what type of instrument you are measuring the signal from, and the results of the steps I have outlined for you above.

Chris_K

[NancyH]

Thanks for the information.   We are wiring to the board per the NRSE, grounded signal source recommendations and we are aware of the lack of true grounding in the laptop. 

Our instrument is a collection of pressure, displacement and thermistor sensors which are all grounded to earth ground via the the backplane of our instrument and a 3 prong AC plug.  AISENSE is grounded to the instrument ground.  The problem is in the thermistor measurements.  Our software program reads AICH2 (thermistor) voltage (typically 70 mV) and then reads a grounded signal (to our instrument ground) on AICH4.  The AICH4 voltage is then subtracted from the AICH2 voltage.  The idea is that any offset in AICH2 is equal to offset in AICH4.  The problem is that the signal on AICH4 is 0V with a desktop and varies on a laptop as much as 5 mV. 

This substraction procedure pre-dates me and may have been used to correct for problems when the instrument was configured for RSE (bad configuration).  Wouldn't NRSE configuration eliminate the need to do this substraction?

Thanks. 

[Chris_K_]

Hi NancyH,

Instead of using this “subtraction method”, I would suggest setting up a differential reading for your thermistor measurements.  The differential configuration basically does the subtraction that you described automatically and in hardware.  You can configure the different measurement configuration in the same task so this can easily be accomplished.  Note that you may need to move the actual inputs for your device as the differential input pair may be different from what you are currently using.

Chris_K