Go Down

Topic: calibrating a straight-bar load cell (Read 346 times) previous topic - next topic

mz243100

I am currently utilizing a straight bar load cell with the hx711 amplifier for a project of mine. I was wondering if anyone knew the easiest way to calibrate a straight bar load cell? My load cell has a max load rating of 20kg. I am using it for grip strength measurements and have fashioned a grip. I can easily remove the grip of the straight bar if need be in order to calibrate. My current problem is I am really not sure what would work to calibrate the load cell. With a ring type load cell it is way easier as there is a single load point. But it is slightly different with the straight bar. Any help will be much appreciated.

jremington

#1
Jan 13, 2018, 05:27 pm Last Edit: Jan 13, 2018, 05:27 pm by jremington
Please post a picture of your cell, and explain how you intend to measure weights or forces with it.

mz243100

#2
Jan 13, 2018, 11:32 pm Last Edit: Jan 13, 2018, 11:46 pm by mz243100
I have attached an image of my cell, it is a straight bar type load cell. The max load is 20kg. Essentially, I am combining the load cell with the myoware muscle sensor. I wanted to investigate correlation between force applied and muscle potential. I am focused specifically on the muscles that control grip so I am essentially using the load cell to measure applied grip force. If need be, I can easily remove the grip if that is required in order to calibrate. Note: please ignore the numbers on the image of the load cell
Really appreciate the help

jremington

#3
Jan 14, 2018, 01:17 am Last Edit: Jan 14, 2018, 06:26 am by jremington
I wonder if calibration will be meaningful or useful. What did you have in mind?

I guess from the picture that the red bars are meant to be gripped by someone's hand, in which case different ways of gripping the setup might lead to large variations in the observed strain, even if the effort is the same. Have you already investigated this, and if so, what was the outcome?

If the above test leads to repeatable measurements, you can do comparisons without having to calibrate (which seems problematic in any case).

DangerToMyself

20kg max? (44lbs)

Don't let anyone who works with their hands try it. I can squeeze more than 5 times that (both hands combined).

Just putting that out there in case you're designing something to gather real world data for some sort of research. 20kg is no where near large enough, if you are.

mz243100

#5
Jan 16, 2018, 08:26 pm Last Edit: Jan 16, 2018, 10:23 pm by mz243100
Hey guys really sorry for the late reply. Have been a bit swamped with some other work. Yes, I definitely agree that 20kg is way too low, unfortunately,  messed up and ordered the wrong straight bar, I asked for a 50-100kg one.  I am more interested in looking at the correlation between stronger applied grip force and muscle potential, so it does not necessarily need to be the max grip force but just progressively stronger grip force between each level(if that makes sense). In regards to jremington's reply, the test is leading to repeatable measurements. To be honest I am not very interested in calibrating, but it is an assessed piece of work where they ask on how to calibrate it, it seems that calibrating it will get me marks. The criteria says that if the device is not calibrated, mention how you would calibrate it. If anyone has any idea on how I would hypothetically do that, I would appreciate it. Thanks alot for the help guys

TonyWilk

#6
Jan 16, 2018, 08:56 pm Last Edit: Jan 16, 2018, 08:56 pm by TonyWilk
If you mounted the sensor on a bar (of wood or metal) and arranged it like this:

Diagram:


Then you would be able to measure more than the 20Kg of just the sensor alone
and you could simply sit it on the edge of a bench and hang weights where the "GRIP" goes to do your calibration.


Yours,
  TonyWilk

jremington

You could calibrate the grip for a particular well defined configuration, like applying a known force at two points in the center of each hand grip bar. Consider drilling holes at those positions, insert a screw eye or bolt to attach a luggage scale, or apply weights.

If you have access to some professional device for measuring grip force, calibrate your device empirically by asking various people to grip both, applying the "same" force to both. With lots of trials and averaging, you can come up with a reasonable approximation to a scale factor.

Adafruit has a good tutorial on sensor calibration.

mz243100

Hey guys thanks for the quick replies. Jremington, thanks, I will take a look, my main problem is the deadline is coming up soon so am not sure If I will have time to acquire a professional device to measure grip force and calibrate using that method. I will check adafruit's website, appreciate the help. Tonywilk, thanks for the diagram,  I was wondering what do those black lines between both the load cell and the wooden/metal bar represent. Does that mean I should attach them both to each other, or should I screw the load cell directly into the metal/wooden bar in this situation, and the black lines represent screws?

TonyWilk

Tonywilk, thanks for the diagram,  I was wondering what do those black lines between both the load cell and the wooden/metal bar represent. Does that mean I should attach them both to each other, or should I screw the load cell directly into the metal/wooden bar in this situation, and the black lines represent screws?
Sorry, I meant the load cell should be bolted at both ends to the metal/wooden bar - so that it bends with the bar. The two bars are bolted to some block at the left hand end.

Yours,
  TonyWilk

mz243100

Thanks again Tonywilk, one more question, so does that mean that the straight bar is bolted to both bars?


TonyWilk

#11
Jan 17, 2018, 01:20 am Last Edit: Jan 17, 2018, 01:21 am by TonyWilk
Thanks again Tonywilk, one more question, so does that mean that the straight bar is bolted to both bars?
Is this any better....
Crappy scribbled Diagram:



Yours,
 TonyWilk

TomGeorge

Thanks again Tonywilk, one more question, so does that mean that the straight bar is bolted to both bars?


Hi,
A load-cell doesn't actually measure force, it measures the change in dimension of the cell, using resistive sensors.
So deflection of the loadcell can be calibrated to read force.

By fixing another bar alongside it, you are now measuring the deflection of load-cell and bar, this will need a greater force to deflect the load-cell compared to it when it is on its own.

When calibrating, I'm mot sure how accurate you want to be,  1 litre of water == 1kg.

Tom... :)
Everything runs on smoke, let the smoke out, it stops running....

TonyWilk

By fixing another bar alongside it, you are now measuring the deflection of load-cell and bar, this will need a greater force to deflect the load-cell compared to it when it is on its own.
Yeh, did that on purpose;  earlier the OP said (re: having a 20kg load cell):
Quote
Yes, I definitely agree that 20kg is way too low, unfortunately,  messed up and ordered the wrong straight bar, I asked for a 50-100kg one.
Yours,
 TonyWilk


TomGeorge

Yeh, did that on purpose;  earlier the OP said (re: having a 20kg load cell):
Yours,
 TonyWilk
Yup, just agreeing and explaining the method in the madness... :) :)


Tom... :)
Everything runs on smoke, let the smoke out, it stops running....

Go Up