Strain Gauges

[drawoh]

I am writing up a test plan to use strain gauges. I have no experience whatsoever with them. I want to take multiple readings for a series of static loads. I have been reading up on how to install the gauges themselves. There are some good links here in Eng-Tips.

I need multiple readings. If I have a controller that reads one channel only, can I apply a load, then plug each gauge into the controller, one at a time? I am okay with reading a display and manually recording the results, but automatically dumping into a computer would be nifty.

Is there reader/controller hardware and software out there that anyone recommends? I have found stuff in the Omega catalogue, and I am reading through some Vishay literature.

--
JHG

 

[rb1957]

you're looking for a Data Acquisition System. many out there. manually multiplexing and recording results sounds a bit "agricultural" though it would be cheap, also loses traceability (who's to say you plugged in the right wire ? wrote down the right number ?)

another day in paradise, or is paradise one day closer ?

 

[drawoh]

rb1957,

I label the cables A, B, C, etc, then I label my spreadsheet columns A, B, C, etc.

--
JHG

 

[rb1957]

and you connect cable C and write in column B and transpose some numbers ...

Is time important (ie manually multiplexing will take time) ? How critical are the loads (you wouldn't want to subject the part to an ultimate load for a long time whilst you connect cables and scribble down numbers) ?
Is traceability important ? (short of keeping an electronic copy of the data, you could take lots of pictures)
Is money important ? (a DAS will set you back some; you can rent these things)

another day in paradise, or is paradise one day closer ?

 

[gruntguru]

AFAIK you would need to record unloaded readings at each gauge before loading, then record them all again in the loaded condition. Of course you would not be able to "balance" each bridge if you use this method, so all gauges would need to be roughly the same resistance in the "unloaded" state. Sounds dodgy to me.

I don't know how many gauges you have but if you must get by with a single channel controller, it would be better to connect to one gauge, take all the readings you need from that gauge (unloaded (balance the bridge) plus each loaded condition) then repeat for each gauge.

je suis charlie

 

[3DDave]

The test engineers I've worked with have emphasized the care that is needed to attach the gage to begin with. I mention this because you are specifically using strain gages and not pre-configured load cells. Practice on some test bars with tensile loading to make sure you are doing it correctly before committing to the actual test.

 

[dhengr]

Drawoh:
Yes, you should learn something about strain gauges, so you understand what they are doing and so you can help direct their application, location and orientation and the type of gauge to apply, etc. You want to know that you can communicate with them; what you need and how they can achieve that. But otherwise, you would probably do well to hire someone who has strain gauging experience and the right equipment. They know how to apply them and wire them properly, they can zero them out, account for temp. changes, etc. and generally help you, and record the data, while you run and manage the complete testing program. The most gauges I’ve ever used on one structure, for one test program, was about 60 gauges, with about 20 progressive stepped loadings for several different loading conditions. I would not have wanted to do that testing program without their help, I’d still be applying the gauges years later, and never have gotten the testing program done with all the people watching our every move.

 

[rb1957]

agree with the previous posts ... if you don't do s/g work regularly I think it's better to get some help. There should be plenty of help out there (universities for example), it'll be somewhat expensive but it'll get the job done.

I think you can manually multiplex ... there was a post about balancing the different channels ... fair enough observation, but from my experience QED.

Do you Need strain gauges ? can you use "stress-coat" (now called "StressKote" to separate from water treatments) ... a surface "paint" that'll give you some indication of the stress.

another day in paradise, or is paradise one day closer ?

 

[drawoh]

3DDave,

I have watched some videos of people installing strain gauges. I can see why this is a delicate procedure. There is at least one person in-house here who has used them, and I am bugging him for info.

If I (carefully) attach a strain gauge to a beam and apply a known force and bending moment, will that calibrate all of my (carefully installed) strain gauges, or just that particular installation? Relative readings actually work for me, but it would be nice to have an estimate of actual strains.

My next problem is figuring out where to put the strain gauges.

--
JHG

 

[rb1957]

strain gauges give you actual strains. there are several factors ... temperature compensation. the biggest problem is often installing the gauges on unloaded structure, if you want true strains. A way around this is to apply known loads and use the deltas to get a stress output.

are you looking at a linear part (like a strut) or a 2D part (like a web, skin panel). Using axial gauges on 2D parts in suspect unless you can carefully determine the gauge orientation from loads.

another day in paradise, or is paradise one day closer ?

 

[drawoh]

rb1957,

I want to test a fairly elaborate structure. I will put strain gauges on the structure close to where I will apply force. I will put some other gauges on some equipment I would rather did not get strained. I will place some gauges in intermediate areas if at all possible. For a set of arbitrary forces, I will take readings and compare them.

There is minimal straight tension and compression on this system. Right now, I am running an FEA process to see where the strains would be on a bare frame. I sort of already know this, but the FEA make a nice sanity check.

--
JHG

 

[rb1957]

forgive the stupid question, but can't you isolate the equipment from the structural strains ?

s/g near loading points ? ... don't you trust your load control ? but it is a way yo validate the FEM.

another day in paradise, or is paradise one day closer ?

 

[3DDave]

Drawoh,

A practice run won't calibrate them - they come with a calibration certificate about how much deformation produces how much change. What it will do is provide feedback that you are installing them in such a way that they are working correctly. Strain gages are detecting changes in the millionths of an inch range; pretty easy to mess them up.

 

[MikeHalloran]

I used to work at a place that made refrigerant scales, among other things.
They had an incredibly detailed and finicky process for installing and calibrating strain gages and protecting them from the environment once installed.

You would probably be better off buying load cells to build into your equipment or associated test fixtures if at all possible. Load cells have well-defined transfer functions between force and strain, so you don't have to worry about odd strain behaviors unique to every different location on your product.



Mike Halloran
Pembroke Pines, FL, USA

 

[drawoh]

rb1957,

The design is fairly complete, and there is no way to structurally isolate critical components. There is no way to implement load cells. I am not looking at contact forces. I am concerned about internal components being strained excessively by external loads. I am hoping I can demonstrate that the internal strain is around 1% of that experienced by the frame next to the force input. Absolute numbers would be better, but that probably will do.

--
JHG

 

[rb1957]

well the good thing with a frame (as i understand it) is that at least it isn't strained when you install the s/gs (picture putting strain gauges on a wing of a plane sitting on the tarmac).

your equipment mounts onto the frame ... could you replace the equipment with some test piece that would allow you to infer the loads that the equipment would see ? could you use photoelasticity ? (I haven't since school ...).

I think s/gs will get you there, but it'll take time and/or money.

I assume your equipment is screwed into the frame (ie hard mounted). could you mount using jubilee/hose clips ?

another day in paradise, or is paradise one day closer ?

 

[3DDave]

Imagine putting strain gages on an airplane wing while the plane is not on the tarmac!