strain gauge loadcell. 3

[CezarAndrei]

i want to build a load cell that's using one or more strain gauges connected to a wheatstone bridge. so i was thinking to a beam (12mm X 24mm, 80mm long), fixed to one end and free on the "load" side. the strain gauge will be glued to the bottom of the beam (on the surface of 12X80) (would be a compression cell)

than for a maximum elongation, i'll drill a hole of 18.5mm in diameter on the surface of 24X80 with the center above the gauge,.

any idea of what is the best method to estimate how much the elongation would be based on the force applied at the free end of the beam?

Thank you very much
(sorry for my english)

 

[rb1957]

roughly, I = 12mm*(24^3-18.5^3)/12 (thx for using 12mm width !) = 7500mm4 bending moment = P*L (you don't say weher the s/gauge is mounted with respect to the loaded end, and bending stress = (PL)*12/7500, and strain = stress/E (you don't say what material)

good luck

 

[CezarAndrei]

<<(thx for using 12mm width !) >>

you're funny, thanks man

cezar

 

[MikeHalloran]

The most common form of load cell (a "binocular") comprises a cantilever beam with three transverse holes drilled in it, two quite large and one smaller, so the flanges of the beam form two parallel flexures, with four strain gages mounted on the outside faces opposite the big holes where the strain is greatest.

If they think you're going into the load cell business, the strain gage manufacturers will be happy to bury you in technical information about using strain gages and building load cells.

If your time is worth anything at all, and this is not just a learning experience, and you want to end up with a load cell that you can trust, it's much much much cheaper to just buy it.



Mike Halloran
Pembroke Pines, FL, USA

 

[CezarAndrei]

thanks for the info.
the answer is yeah, it's a learning experience as well as a building something i will use.

the company i'm working for is in rubber industry (automotive). we have to overmold a metal washer (insert) with rubber for an air conditioning application, using a primer that will create a bond inbetween the rubber and metal.
the primer we utilize is very very sensitive to the relative humidity during the drying time (drying of the primer). also the bond is affected by the process parameters.
i was designing a fixture to be able to quantify what a "good bond" is. right now all we can say is: ooohh, that's a good bond, oops, that's not very good and so.
i asked about $5,000 for a load cell (compression, 1kN), it wasn't approved.
Now I want to build it myself and after running about 1,000 parts (destructive test) I'd have an idea of what a bond is. i need a way to verify the consistency in time, maybe with a spring, i did not think about it now.
so, as you can see I don't want to know perfectly what the maximum penetration force is as an absolute value. i'll use just the voltage as a reference. also for the test i will use an old drilling press in order to have the same speed for all the parts tested.
(btw i had to work a lot with office guys to let me bring the press on the office area to get stable environmental conditions. lol)
after that i'll be running a DOE and figure out where we actually are

ohh, it's so hard to get what u want sometimes, but i never give up.

thank you too

cezar andrei
guelph, ontario

 

[GregLocock]

I think you'll learn a lot, including arcane things like what the fatigue properties of load cell cement are.

The cheapest load cell is often a pressure gauge on a hydraulic ram, or a dead weight. Are neither of those applicable?



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[MikeHalloran]

Figure six months to learn how to build a load cell that gives a stable repeatable reading. You will build a lot of bad load cells and collect a lot of irreproducible data along the way.

I'm with Greg on this: Pressure gage and cylinder, spring and dial indicator, dead weight, etc.

There's also the political issue to consider. If you ask to buy something, are refused, then go out and build the same thing, you might be perceived as insubordinate, at least in some outfits. If you build something else, you have a more defensible position.




Mike Halloran
Pembroke Pines, FL, USA

 

[rb1957]

just like greg above, a simple test fixture with a dead load would probably tell you what you want to know, and be cheaper. maybe make a sacrifical test piece with each batch, or maybe a special test piece (not a production piece, but a test piece with the production bond and suitable for mounting in a rig).

guelph, eh
i'm in toronto

 

[Philrock]

If you're serious about learning about load cells and strain gauges, it would be well worthwhile to take one of the top-notch training courses offered by Vishay Measurements Group. They offer hands-on workshops, seminars, a wealth of technical information, and excellent support.

 

[CezarAndrei]

Thanks Philrock.
Very good info from Vishay's guys

 

[CezarAndrei]

Hi guys,

I'll go in parallel, pressure gage and load cell. It's a nice exercise from the repeatability standpoint.
Thank you again.

Any idea what kind of steel would be better (for the load cell)?

Thanks again

ps. hi rb1957

cezar

 

[Philrock]

In order to get the highest possible signal, you want the cell to have the maximum possible strain without yielding. This means high stress and low modulus of elasticity. For this reason, 7075 aluminum is popular for load cells. It has the low modulus of aluminum and a fairly high yield strength. If you prefer steel, you might consider 4140 heat treated to Rc 33-38 (many distributors stock it in this condition). This has a high yield strength but is still machinable.

 

[GregLocock]

Titanium would be a tad better then. tiAL6V4 or whatever it is at T6, given the cost of the exercise, would probably be worthwhile. easy to machine, 1100 N/mm^2 yield, 2/3 the stiffness of steel.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[CezarAndrei]

thanks again
i knew i can count on you guys

cezar
guelph, on