Calibrating pressure transducers with T7

2 posts / 0 new

Tue, 09/27/2016 - 16:48

Nick L

Hello,

I've purchased two PX302-500GV pressure transducers from Omega. I've got them hooked up through a LJTick-VRef-41 on a T7 board to get power. The PX302 output is 100mV +/- 1mV (10 mV/V). It is rated for 0-500 PSIG. The LJTick excites the transducers with 4.096 V.

In Kipling I have the input set to -.01 to .01 V Range and the negative channel set to the other analog input that the transducer is using. I followed instructions on an archived post for most of this & I think it should be correct.

I have the voltages scaled by multiplying by 500 psi & dividing by .01 V (max output) and dividing again by 4.096 (excite voltage). So basically P=Vout*500/(.01*4.096).

When taking readings on various locations in a refrigeration system, I get reasonable results, but not quite correct. I am comparing my results to past testing data know to be accurate. At equilibrium, both gauges are supposed to read ~90 psi, and my calculation has them at around 70-75 psi. At another point, the high side of the system is supposed to be around 270-280 psi and the low side around 30-40psi, but I'm reading 180psi and 15psi. These values are never constant, so I can't calibrate using known values. I'm wondering if there is a reason my offset calculation might not be accurate? Or if I just need to alter the offset until I get data that agrees with previous data (this seems like an imprecise way for an expensive sensor to work)? Any help is very much appreciated.

Thanks!

Wed, 09/28/2016 - 10:49

LabJack Support

Your sensor is a raw bridge

Your sensor is a raw bridge type with output rated for 10 mV/V, so with the 4.096V excitation you expect 40.96 mV to correspond to 500 PSI.

Note that since you expect voltages greater than 10mV you do not want to use the +/-10mV range on the T7. I suggest the +/-100mV range. If using Kipling to configure the range make sure you save the new power-up settings before you exit Kipling.

The previous forum topic you likely saw, which is the place to start when working with bridge circuits, is topic 4376 from our old forum. See post #13 in particular:

http://forums.labjack.com/index.php?showtopic=4376&p=23917

See the troubleshooting step 2b and start with these 2 tips. 1. Use a DMM to check Vexc (should be about 4.096V) and Vsignal (at 90 PSI the signal should be 0.04096 * 90 / 500 = 0.0074 V). 2. Replace your signals with GND jumpers to make sure the noise levels agree with what is documented in Appendix A-3-1 and then compare to the noise levels with your actual signal connected.