1
\$\begingroup\$

I have a Wheatstone bridge but I don't know the values of the resistances in it.

My excitation voltage is 5 V. But instead of having 2.5 V on each output lines I get one line at 3.3 V and the other at 1.7 V

enter image description here

How can I do to compensate this situation and have a 2.5 V output on each line?

\$\endgroup\$
  • \$\begingroup\$ What's the application? \$\endgroup\$ – Eugene Sh. Jul 8 '16 at 15:19
  • \$\begingroup\$ Use it as strain gage and put the two output lines into an opamp (the ADS1115) \$\endgroup\$ – Stéphane Jul 8 '16 at 15:21
  • \$\begingroup\$ All of the resistors are a part of the strain gauge? In this case the only thing you can do is to calibrate your receiver (by adjusting the opamps gain/offset, for example). \$\endgroup\$ – Eugene Sh. Jul 8 '16 at 15:22
  • \$\begingroup\$ Add a parallel resistor between SP and EN or EP and SN. This will change the offset, but it will also affect the temperature response. Is the offset within specification? Has it been over-strained or could there be any damage/corrosion? \$\endgroup\$ – user57709 Jul 8 '16 at 15:24
  • \$\begingroup\$ Yes all the resistors are part of the strain gage, I have only access to the connector. The problem with a parallel resistor is that it change both lines proportionally. So for example -1V on both lines, but not on just one line. I can assure you that it is not damaged or anything. \$\endgroup\$ – Stéphane Jul 8 '16 at 15:30
3
\$\begingroup\$

Your bridge connection is most likely rotated by 90 degrees. No bridge would ever have this much offset. If you are sure it is not damaged, first disconnect any circuitry on the output pins and measure again. Then, check the pinouts and make sure they are properly connected. Best bet is to try rotating the strain gauge; that is, apply power to SP and SN and measure EP and EN. Many bridges are only symmetrical in one orientation, and the measure taps are not exactly halfway between power and ground.

\$\endgroup\$
  • \$\begingroup\$ A rotated bridge is exactly what I was thinking could be the problem. \$\endgroup\$ – Michael Karas Jul 8 '16 at 17:48
  • \$\begingroup\$ That was it. I check all the possibilities and succeed ! Thanks ! Seems out that the wire colors are not right.... \$\endgroup\$ – Stéphane Jul 10 '16 at 15:38
2
\$\begingroup\$

By definition, your bridge is unbalanced. you don't say where the bridge comes from, or what it is, or if all the elements are active, or which ones are simple resistors, but its pretty clear that on one side the resistors have about a 2:1 ratio, and on the other a 1:2 ratio (give or take).

Its possible that you've reversed your excitation with your outputs. Try rotating things 90 degrees and see if that helps.

Other that that, the way bridge unbalance is dealt with is to put a pot somewhere in the bridge, and tweak to balance.

\$\endgroup\$
  • \$\begingroup\$ +1 - Rotated bridge is very likely the issue here. \$\endgroup\$ – Michael Karas Jul 8 '16 at 17:48
  • 1
    \$\begingroup\$ Thanks ! It was in fact not well wired. Wire's colors issue... \$\endgroup\$ – Stéphane Jul 10 '16 at 15:38
0
\$\begingroup\$

How can I do to compensate this situation and have a 2.5 V output on each line?

It would be easier and more practical to use an instrumentation amplifer (AD620 et al) with a low gain setting and use the reference pin to null the offset to 0V (or some other mid-point value). Then you have a single ended nulled output that you can feed into the rest of your circuit/system: -

enter image description here

Pin 5 is the reference pin that can adjust the output offset of the amplifier but take heed, with this much offset, a near-unity gain configuration is going to be needed. Once it is nulled you can apply gain (if needed) using an op-amp after the AD620.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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