Strain gauge for Autothrottle disengage

[upbeat sundial]

I am trying to figure out how to use a BF350 strain gauge to detect manually induced deflection of a throttle lever and automatically disengage the auto-throttle. This, instead of installing switches.

The throttle lever can be 3D printed following how strain gauge bars are made. I think that a 3D printed bar would offer much less resistance, so easier to measure deflection.

There are HX711 modules used in scale applications (which I also got a couple of those) but they use what seems like an SPI type connection (data and clock), which would require creating a custom device for Mobiflight.

I am leaning towards trying to read the change in resistance as a change in voltage, very much like a potentiometer that can be used directly by Mobiflight.

All sources I find point to using a Wheatstone bridge resistor arrangement. I also think that connecting an op-amp chip will be necessary to raise the voltage to Arduino measurable levels.

In my view, precision is not important in this application. I am just trying to detect a deflection on the throttle lever strain gauge beyond a reasonable threshold that can be used to trigger the disengage event.

I cannot find any documentation on how to do this. Everything I find is geared towards using the HX711 module and not directly measuring the voltage with the arduino ADC.

Any comments or help would be appreciated.

This is what I have bought from Amazon
https://www.amazon.com/DAOKI-BF350-3AA-High-Precision-Pressure-Resistance/dp/B07X87CJD8/ref=pd_ci_mcx_mh_mcx_views_0?pd_rd_w=vJWO1&content-id=amzn1.sym.8b590b55-908d-4829-9f90-4c8752768e8b%3Aamzn1.symc.40e6a10e-cbc4-4fa5-81e3-4435ff64d03b&pf_rd_p=8b590b55-908d-4829-9f90-4c8752768e8b&pf_rd_r=DQF00Z1X293BJFVYNXWF&pd_rd_wg=GnaWB&pd_rd_r=2c4f59db-618b-4b55-8c24-dc58efc4410c&pd_rd_i=B07X87CJD8

the throttle lever could be 3D printed like the strain gauge bars used for scale application

Then install the strain gauges to the throttle lever

I think that in this way, minute deflections of the throttle lever can be detected and obtain almost immediate response to disengage the auto-throttle.
On the other hand, using limit switches to detect the manual force trying to move the lever, will require deflections in either direction of several degrees, creating an undesired dead zone around the current throttle position.

[upbeat sundial]

Addittional searching found this...

https://edg.uchicago.edu/tutorials/load_cell/

[upbeat sundial]

Would anyone know if this would achieve what I am trying to do?

[royal crystal]

a force detector that, when a force is applied to the trothle lever, deactivates AT

Perhaps an easier way to do this, considering that the AT motor is not strong enough to prevent the lever from moving, would be to move the lever a little to detect the difference between the real position and the position where it should be and if it is out of a range "for example 5%" it is deactivated.

[upbeat sundial]

Hi, thanks for replying. I am not sure I can move the lever by hand against the nema17 motor, without damaging the 3d printed gears.
I don't use a clutch.
In my control scheme, the stepper motor and potentiometer are kept in sync all the time. I think if I let the potentiometer slip, they will become out of sync. I do not use a pid type scheme.
I will need to think about this.

The control scheme works beautifully, but requires deactivating AT manually by pressing a button to transition between auto and manual.
I thought of this way just to automate that button press without changing the control scheme.

[compact pebble]

You're thinking following the pattern from the uchicago tutorial above, but replacing the load cell they used with your own wheatstone bridge using the strain gages from Amazon above?

...and you'd put the home-brew load cell on the front (or rear) face of the throttle shaft?

(I'm not familiar with the wheatstone bridge thing, but I get the idea. It's not necessary that all four legs of the bridge are sensors, right? Three of them can just be fixed resistors?)

[upbeat sundial]

What you say is not out of the question. I think I can 3D print the lever with the holes to make a weak load cell bar that can hold a couple of strain gauges.
Those strain gauges would form two sides of the wheat stone bridge.
I think I need to connect maybe a 100x op amplifier to get the small difference in resistance manifested as a large enough voltage that the arduino ADC can measure.

[compact pebble]

They're using an INA125 in their circuit as the amp, but it seems like a INA128 would work just as well.

[upbeat sundial]

I don't want to use the HX711 module

[compact pebble]

Right, I understand why.

MF can use an analog input? --and an action can triggered when the input crosses some set value?

[upbeat sundial]

Exactly what I have in mind

I would like to just push the lever say with one pound force and have the AT disengage automatically.

[compact pebble]

Well, this seems doable to me. Obviously want to build a prototype, but that doesn't seem like an excessive amount of work.

And the approach is kind of elegant (if it works :-))

[upbeat sundial]

I saw someone attempt at installing a load cell bar into their throttle lever, but I felt it was too bulky.

I think the load cell can be integrated into the design

[compact pebble]

It sure seems like a worthy experiment.

[upbeat sundial]

For this to work, the triggering force must be much less than the stepper holding torque

[compact pebble]

Right. And less than the force needed to strip the gears 🙂

[upbeat sundial]

Yes, definitely

I still need to buy the op Amp. Wasn't sure which to buy.

Never done anything like this