0
\$\begingroup\$

This is the first time I'm handling a professional load cell, and nobody else at my company has worked with them before.

I'm quite sure I've broken the load cell. I just checked the datasheet and see I should not have squeezed it. It's only rated for 300g. There is such a thing as "ultimate static overload" (250%), and I've definitely surpassed it. Furthermore, the datasheet says I should see an input impedance of ~415Ω and an output impedance of ~350Ω. I find it bizarre that they're not the same. Meanwhile, my LCR meter tells me the output is exactly 292.8Ω and the input is exactly 292.8Ω.

When I connect it to my HX711 setup, I get the same constant reading. I see in the datasheet it has a recommended excitation of 10VDC. All the load cell ADCs I can find have 2.7-5V in their datasheets. Does anyone know any ready-to-go Arduino-esque load cell ADC modules that can provide a steady >10VDC excitation voltage?

The only thing I know for sure is I'll be a lot more careful with load cells from now on, and I'll add overload protection to my test rig, but I'm confused about the 10V excitation and about the input impedance being higher than the output impedance. As I can't be sure my load cell is broken or not, I'll try to get a replacement as a sanity check. Feels bad.

Looking for advice. Thank you.

https://docs.rs-online.com/1a3c/0900766b8010f46a.pdf

\$\endgroup\$

1 Answer 1

Reset to default
2
\$\begingroup\$

There's no reason for the input impedance to equal the output impedance since there are apparently resistors in the supply leads, so for the bridge composed of 4 matched resistors R and supply lead resistors Rx the input impedance will be R+2Rx and the output impedance R.

enter image description here

From the numbers provided, the strain gauge elements are the standard 350Ω and there are 33Ω resistors in each excitation lead.

You can always provide less excitation, but the voltage output will obviously be proportionally lower so there is more potential for issues with thermal EMFs and amplifier drift. Modern auto-zero amplifiers have mostly eliminated the latter as a concern.

I suggest using a multimeter to measure the actual output voltages (excitation, +Output, -Output, difference between the latter) and see what they are. Also inspect closely to see if the beam elements themselves are actually bent or cracked.

\$\endgroup\$
5
  • \$\begingroup\$ Measured the output voltages. They do in fact change when I apply pressure. \$\endgroup\$
    – Popeye
    Commented Jul 3, 2023 at 16:31
  • \$\begingroup\$ So is there a huge (in relative terms) permanent offset? \$\endgroup\$
    – Spehro 'speff' Pefhany
    Commented Jul 3, 2023 at 16:36
  • \$\begingroup\$ Ok. I've measured the actual voltages of the load cell and see they are ~1.25V between green and white with no load applied. Generic, cheap, Amazon-bought load cells have ~0.00V with no load applied because they have similar input and output impedances. Meanwhile, on my load cell, it turns out the input and output impedances I measured were on the wrong wires. Following this schematic, I see I have exactly 350.0Ω between my white and red wires (outputs) and 411.0Ω between my green and black wires (inputs). Seems the issue I'm having is that my ADC only accepts -40mV to +40mV. \$\endgroup\$
    – Popeye
    Commented Jul 3, 2023 at 16:54
  • \$\begingroup\$ You should have a regulated excitation voltage between GREEN (+) and BLACK (-). The voltage will depend on whether you are using the regulator in the HX711 and, if so, the feedback resistors. Then INA+ to the HX711 goes to RED and INA- to the HX711 goes to WHITE. So the important voltages are that between GREEN and WHITE (excitation) and the voltage between RED and WHITE. If there was serious damage the voltage at either the RED or WHITE might be much different from the excitation voltage/2. \$\endgroup\$
    – Spehro 'speff' Pefhany
    Commented Jul 3, 2023 at 17:12
  • \$\begingroup\$ Output voltage can only be as stable as the excitation voltage so just slapping it across the 5V supply may not be ideal. \$\endgroup\$
    – Spehro 'speff' Pefhany
    Commented Jul 3, 2023 at 17:14

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.