PXI-4351 Problems

[dilkgeet]

I'm not sure if this is a software problem or a PXI-4351 problem.  I have PXI-4351 boards installed in a chassis.  I am using the board to read the temperature from 16 thermistors.  The 25 uA current source flows through all thermistors as they are attached in series.  I have built a temperature reading software based on the library provided off the NI website.

I observe the following. 
1)  When the PXI-4351 cable is attached to the breakout board, I obtain correct readings.  When the temperature of the thermistors increases,  I read it immediately.

 2)  However, when the cable is disocnnected from the board, I still obtain reasonable readings, and these readings change over time as if the cable was connected to an actual board. 

3)  When the cable is connected to the board, and I break the circuit by removing one of the 16 thermistors, the temperature readings in Labview become unreasonable, as expected.

I do not understand why 2) above happens.  Is there a general setting on the PXI-4351that I have wrong? Thanks

[Hani R]

Hi dilkgeet,

When you disconnected the cable from your breakout board, the PXI-4351 was reading a floating signal (not grounded).  The voltage at the positive and negative terminals is at an unknown potential.  If you were to ground one of your terminals you may notice that the measurements no longer look like the thermistor readings. 

When you removed one thermistor from your circuit but left the cable connected, the signal was still grounded to the breakout board.  That is probably why you noticed the readings were as expected for an open circuit.  I hope this helps. 

[dilkgeet]

Thanks for the explanation.  However, is there a solution to this problem? We would like to know if the cable has become disconnected from the board so that we know we are not reading the true thermistor temperatures. Thanks

[J_Thomas]

Hi dilkgeet,

The only solution to the “problem” is to not disconnect the cable.  Not much can be done to get a correct reading again except to reconnect the cable. It may be possible to setup a detection method so that an operator could be alerted if the cable does become disconnected.  One thing you could try would be to use the DIO lines on your breakout board.  You could configure one digital line to output a 5 volt signal and use a second digital line to read it back.  In your program you could continuously check if the second digital line is high (reading 5 Volts) and terminate the program if the signal goes low (meaning the cable has become disconnected and the second line can no longer read the 5 Volt signal).  I haven’t had a chance to try this myself, but something of this nature could be used to detect a disconnected cable.