0
\$\begingroup\$

I am trying to measure mechanical deformations using a full Wheatstone Bridge. In order to do that, I connected the bridge to an instrumental amplifier (INA125PA).

The amplifier is powered with a dual supply of +-15 V, provided using a symmetric power supply (ELC AL991S). The Wheatstone bridge (350 Ohm gauges) is supplied at 10V using the reference output voltage provided via the amplifier. The following schematic represent the whole electronic circuit.

Electric circuit

However, one the amplifier is turned on, there is a 50 Hz noise of around 4 mV of amplitude, which interferes with the measured signal (a reflecting deformation wave of low frequency). The measurement was done using a Picoscope 4824A (Documentation) Measured noise on the amplifier's output

I wish to get rid of this noise. I suspect a problem of ground connection, creating a loop eventually, but I lack the knowledge to identify it.

Do you have any lead I could follow?

Edit: I've been asked to provide a picture of the circuit layout. Here it goes: Circuit layout I am curently using a Breadboard, which was usefull for the time being due to the flexibility it offers. I plan on soldering everything. I can confirm that the scope is not aliasing.

The wheatstone bridge is wired using same length wires soldered to the strain gauges. Regarding the connections in between the different elements, I use either BNC cables or banana plugs (for the power supply) (not visible on the previous picture)

\$\endgroup\$
5
  • 4
    \$\begingroup\$ Likely culprit is circuit board layout. If breadboarded then that's the main culprit. \$\endgroup\$
    – Andy aka
    Aug 11, 2023 at 12:17
  • \$\begingroup\$ Would you recommend soldering everything to a pcb ? \$\endgroup\$ Aug 11, 2023 at 12:28
  • 3
    \$\begingroup\$ Zoom your time scale in big time to make sure the sampling scope isn't aliasing. \$\endgroup\$ Aug 11, 2023 at 12:28
  • 2
    \$\begingroup\$ For this kind of problem, you can power your electronic device with a battery to determine whether the problem is related to radiation or conduction of the 50Hz. \$\endgroup\$
    – Vincent
    Aug 11, 2023 at 19:34
  • \$\begingroup\$ Can you show an actual picture of the layout and how the wheastone bridge is wired up? So we can identify problematic loops, for example. \$\endgroup\$
    – feynman
    Aug 12, 2023 at 13:53

1 Answer 1

1
\$\begingroup\$

Hard to tell without the actual physical layout of the PCB and the wheatstone bridge itself. It could also be just a measurement issue after all.

What you can (and should) do is identifying and minimizing all loops that begin at the positive and end at the negative input of the amplifier. Those loops are the ones most sensitive to induced voltages due to low frequency magnetic fields. You also need to minimize the loop of the wheatstone brigde itself.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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