Showing results for 
Search instead for 
Did you mean: 

Connecting board-level DPT to NI 9237

Go to solution



I have several NI 9237 modules which are used in a multi-unit pressure monitoring system. I was given this system and told to replace the existing unreliable transducers, which have unacceptable drift. I am trying to use TE Connectivity 13A-002G pressure transducers. These are board-level transducers and have connections according to this diagram:




I've only ever used packaged DPTs with 4 factory leads, so I'm not sure what to do with the gain set pins.The data sheet I linked shows an application schematic with two op-amps connected with negative feedback, but only one of the resistors is labeled. The NI 9237 can provide internal excitation - does that fulfill the role of these op-amps? If so, how should I connect the leads to the NI module which has the following internal structure:




I would greatly appreciate any input on how to configure these for use with the 9237 module. I've already got one in the shop and mounted on a prototyping board, ready to solder more connections or components as needed.




0 Kudos
Message 1 of 4
Accepted by topic author grabercars

Hi SG,


why don't you read the next page of the NI9237 manual?


It shows figure 2:

Your bridge sensor has 4 wires to connect to the bridge, the NI9237 has 6 wires to connect to the bridge…


I don't see a reason to use this GainSet resistor with two external OpAmps just to reach a certain output range. The NI9237 does that amplification on its own…

Best regards,

using LV2020 on Win8.1+cRIO
Message 2 of 4

(Edited after posting because I found more sources of information)




Thank you for your helpful response. I've hooked up the bridge and it functions as expected.


I will have about 20 sensors total in a long line. The wires to these sensors will range from 10-100ft in length. The longest wires (braided copper, 20AWG) will therefore have a resistance on the order of one Ohm. The resistance of the bridge is on the order of kOhms. Therefore, gain error for this full-bridge is on the order of ~(2*1Ohm)/1000Ohm = 0.002. Does gain error change the linear voltage response as a function of pressure, or does it result in a constant offset to the measured pressure? According to this page, which has the same content as the NI 9237 datasheet but with more explanation, it's not a constant, but a linear factor. i.e. measured values will be decreased by 0.2%.


The absolute accuracy of these sensors is not important to me, but I am interested in precision and linearity of each sensor in isolation. In the case that I do elect to use remote sense wires, and for my own future reference, I'm curious about the details of implementation. Practically, I'd want to use wires of the same size and probably in the same bundle as the excitation leads. How can remote sensing function if the remote sense wires have resistance equivalent to the excitation wire resistance? From the NI datasheet, "If you connect the remote sense signals directly to the bridge resistors, the NI 9237 senses the actual bridge voltage". With the resistance of the wiring being an unknown variable, how is this possible?


So, to reiterate, I've got two questions:

  1. Does gain error change the voltage vs. pressure response by a constant or by a linear factor?
  2. How does remote sense work if the resistance of remote sense wires is unknown, and is there any point in using remote sense when R_bridge>>R_leads?


Thanks for your time,


0 Kudos
Message 3 of 4
Accepted by topic author grabercars



sensing voltage requires (nearly) no current flow, so the resistance of the wires will have (nearly) no influence...


Or the other way around: the input resistance of voltmeters is in the >100MOhm range, several magnitudes higher than the resistance of the wires...


Wikipedia also explains such stuff, books about electric measurements too.

Best regards,

using LV2020 on Win8.1+cRIO
Message 4 of 4