0
\$\begingroup\$

I have a load-cell setup with some noise that I would like to filter out. I have a vague understanding of low-pass RC filters.

I have a strain gauge with 350 ohms of resistance (no load). How can I create an RC circuit to attempt to filter out some higher frequency signals? Is it as simple as using capacitor to match the resistance of the load cell for a given cut off frequency? Or will I need external resistors?

If external resistors are required, will this not change the output voltage (and thus the signal I'm intent on measuring?)

\$\endgroup\$
0
\$\begingroup\$

This is a broad question. The short answer is, it depends on what range of frequencies you want to filter out, and you will need external resistors. The long answer is, it depends on the sensitivity of your strain gage, how precise and fast you need your output to be, and how much time you want to devote to understanding the answer to this question.

As for changing the output voltage, yes and no; yes, you would likely change your output voltage if you were to modify the circuit below, but no, because it's not applicable here. (TL;DR- if you don't want a more step-by-step answer, just click the link to the UC Santa Barbara laboratory exercise at the bottom.)

I've inserted a schematic below of the simplest possible low-pass filter with what you have described.

schematic

simulate this circuit – Schematic created using CircuitLab

This circuit is relatively stupid, and will probably not do what you need, but it's a good starting point. The cutoff frequency for a low-pass filter, as described by $$ f_c = \frac{1}{2 \pi R C}$$

can be easily calculated as 455 Hz, or roughly 2 ms. Another (more simplistic) way to think about it is that anything faster than 2 ms will not be detectable.

Ah, if only electrical engineering were so simple as just a resistor and capacitor in parallel.

One popular (i.e., precise) implementation of a strain gage is using a Wheatstone bridge, which I've found via an online PDF of a laboratory exercise: Strain Gage Sensors Laboratory Exercise - UC Santa Barbara

This lab was written by people far more experienced than I, so I'll let them take over from me here.

I hope I've given you some of the tools necessary for acquiring more of your own answers. Please let me know if this has helped, or if you have other questions!

\$\endgroup\$
  • \$\begingroup\$ Regarding the huge generalization "anything faster than 2 milliSec will not be detectable"----that is very misleading. Briefer events (noise spikes) will be attenuated as 1/event-duration, but are still detectable. 2uS events will be reduced to 0.1% of their unfiltered energy level, but are still detectable by a sensitive ADC. \$\endgroup\$ – analogsystemsrf Nov 17 '18 at 3:11
0
\$\begingroup\$

Is it as simple as using capacitor to match the resistance of the load cell for a given cut off frequency? Or will I need external resistors?

It can certainly be as simple as this; when |Xc| = R (350 ohms) you have the cut-off frequency: -

$$F_C = \dfrac{1}{2\pi RC}$$

If you have a bridge of strain gauges then you might need three capacitors to optimally filter your bridge output; one from each output line to 0 volts and 1 between the two outputs: -

enter image description here

\$\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.