1

I have two, three-wire load sensors (https://cdn.sparkfun.com//assets/parts/4/5/9/5/10245-01.jpg) hooked up in a half wheatstone bridge (I think), with two 1K resistors. Using a multimeter, I am reading small voltage changes when applying pressure to either strain gauge, when measured on between the pins A+ and A- on the HX711 ADC amp, however my Arduino serial output shows all 0's when using a variety of code examples (for example, this one: http://www.dfrobot.com/wiki/index.php/Weight_Sensor_Module_V1#Sample_code).

My concern is that my circuit is not setup correctly as I am fairly novice, however everything I have researched about wheatstone bridges, and half wheatstone bridges seems fairly straightforward, and seems to match my circuit. Any guidance/advice on the attached breadboard circuit diagram would be extremely helpful.

enter image description here

  • The white wires in your diagram are not in the center as in the linked picture. Also, the E+/E-/A-/A+ labels seems off by one pin. If the whites are the center wire on your device, they are likely the middle, sensing terminal of the load cells, and should connect to the A+/A- of the HX711 See the last paragraph of arduino.stackexchange.com/a/18698/6628 for using the sensors's other gauges as resistors. Could be the precision of the resistors swamp the bridge. – Dave X Jan 19 '16 at 11:56
  • Perhaps you should move your greens from +5/GND to the E+/E- to let the HX711 excite your bridge rather than the power supply. – Dave X Jan 19 '16 at 11:57
  • I was directed in an Arduino forum to wire it up like this: imgur.com/OkEuWV4 however I still cannot get it to work. Going to move the greens around as well. – zqillini4 Jan 21 '16 at 21:23
  • What is the no-load voltage difference across A-/A+? I didn't dig into the code you pointed to, but it looks like it might be multiplying the value by 1992.0 and I'm not sure what offset it adds. If the no-load voltage is far from zero, it might overflow or underflow something and give you a constant value. I'd use the strain gauge's other halves as in the forum's diagram, instead of discrete resistors as in yours, since you need to balance things to near 0.1%. The white-black connections they show put the tension(or compression) gauges on opposite sides, difft than your connections. – Dave X Jan 21 '16 at 22:11
  • Also, check the resistances between the gauge terminals. The sense/A+/A- should be the same color wire, and should lie halfway between the resistance between the other two terminals. It is essentially the center tap of a ~2k potentiometer, and it should move about +/-0.1% or 2 ohms from the center at full load. – Dave X Jan 21 '16 at 22:16
2

In your diagram, the fixed 1KΩ resistors both tie to +5V at one set of ends, and their other ends tie to A+ and A-. The strain gauges are from A+ to ground and A- to ground.

In a typical scale, where several strain gauges are similarly loaded, that topology will produce about zero volts between A+ and A-. By contrast, the following topology (from: blog.l3l4.com) produces higher voltages for higher loading. (Diagram uses S+, S- rather than A+, A-)

(I realize your question shows some voltage changes when sensors are separately loaded, so this answer might not be relevant to your current problem.)

Wheatstone bridge on a SparkCore with HX711

  • 1
    The OP has the strain gauges both on the bottom side of the wheatstone bridge, where if similarly loaded would pull both E+ and E- in the same direction when loaded. What might be more of a problem, is if the values of the discrete resistors R1, R3 do not match the strain gauges to 0.1% perhaps making zero-load S+/S- voltage too large to work with. – Dave X Jan 19 '16 at 12:07

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.