Strain Gauging Instrumentation Help

[Burner2k]

Hello,
We are planning to strain gauge instrumenting our birds during ground static tests and flights. I am looking for some references & pointers on how to accomplish the same.

I need help in the following:
1. Determine suitable locations on Wings/Fuselage from Finite Element Models. If I am placing strain gauges on skins, do I place them both on the outside of skin & inside at the same location. Questions like these.
2. Data Extraction & Post-processing
3. Correlating the output with FE results.

If anyone knows of any good open source references, it would be of great help.

We are in preliminary stages of planning the above...so possible that we may have not thought it through completely and thus have many questions or issues for which we have no answers currently.

 

[rb1957]

oh are you in for a world of hurt and learning ! I can't be of much detail help, we have a consultant who does that (when we can't avoid it). Find someone local who does it for a living. Don't need OML and IML gauges on skins, IML should be good enough. Aways use rosettes on skins, use axials on axial members (like stringers). is it a wing or fuselage ? A huge issue (in my limited experience/knowledge) is going to be removing the on ground stress (that the strain gauge feels when installed. OEMs will put their static test article into the rig and tare it out, so the airplane is essentially unstressed.

Good Luck !

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[SWComposites]

Yep, there is lots of ways to get instrumentation wrong, and make your life completely miserable trying to explain away unpredictable results. And a lot is learned by experience. I don’t know of any publicly available info. You could hire someone like me as a consultant.

First, answer this question: what is the exact purpose for the strain gages? Is this a research project or a certification program?

A few of my “rules”:
- only put gages where there is significant strain at Ultimate, like 2000 mstrain or greater.
- only put gages in areas where there is minimal strain gradients.
- if a skin or web is buckled, you must have back-back gages.
- avoid putting gages in buckling areas unless you are very certain you can accurately predict the nonlinear post-buckling response (and you have done it before with good correlation).
- if you need to validate a load distribution, you need gages on every stiffener centerline (and a back-back one on the skin) across the width.
- unless you are certain the the strain is uniaxial, use rosettes.

 

[Burner2k]

@SWComposites & RB1957,
Our "bird" is a small UAV & we're located outside of US & EU . There are no certification requirements for our product. Our intentions of putting strain gauges are purely for research. Thanks for the few pointers you've shared.

My intuitions were correct...this ain't gonna be an easy task. I will keep on researching more but I thought there would be publicly available papers put out by NASA on how to conduct strain gauge instrumentation.

 

[rb1957]

my 2c ... use accelerometers ... 6 axes at a "CG", 3 axes at the four corners. This "will" tell you the airplane's response to inputs, your controls, external (unknown) gusts.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[3DDave]

The more problematic is that thermal effects are picked up by strain gauges. Any temperature drift during the testing will cause problems particularly for flight test; but you haven't mentioned how the loads are applied. Add in sunshine heating one side and shade on the other and it will look like your strain data is garbage compared to stress/load inputs.

Has your FEA been used with variations in material properties and variations in section properties/skin thickness?

Small UAV? From 10 cm to 10 meters?

 

[rb1957]

accelerometers would valid your CFD ... airplane manoeuvre loads.

I would validate the FEM with a static test article. MMMMMuch easier, with controlled loads, and can test to failure.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[Ng2020]

Fpr further reading try the AGARD flight test instrumentation report series.

 

[SWComposites]

typing "wing box static test instrumentation" into Google brings up lots of potentially useful docs

 

[Burner2k]

Hi all,
Firstly, thanks for the many replies. After some research & consultation with people who have had some experience with aerospace testing, we will be proceeding with instrumenting our ground test articles.

A few questions:
1. Our wings are composite fibers. Any special considerations that must be taken for composite vs metallic structures?
2. When we are correlating strain gauge outputs with FE results (strains), should we go with strains in the outermost OML ply?

 

[SWComposites]

re:
1) if you are using fabric composite materials, you want a strain gage size at least the size of a couple of weave unit cells, and definitely not smaller than a single unit cell. and when they install the strain gages, do not sand into the carbon fibers when preparing for bonding.
2) you should correlate to outer surface strains, which if there is minimal bending thru the laminate thickness is essentially the outer ply strain, but if the laminate is thin and there is significant bending, such as when post-buckled, there will be a significant difference between the surface strain and outer ply midplane strain.

 

[WKTaylor]

DANG... no one seems aware of 'texts' or 'Spec(s)' or guides or handbooks, etc for aircraft strain-gauging [-gaging? gauge, gage?... drives me crazy] for static/dynamic testing a major structure... sigh... was hoping to add to my knowledge base.

This field of live structures testing seems to be a closely held 'secret'... part 'technical' and part 'art'... learned by 'looking over shoulders' and by 'doing'.

Regards, Wil Taylor
o Trust - But Verify!
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation, Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", HBA forum]
o Only fools and charlatans know everything and understand everything." -Anton Chekhov

 

[rb1957]

yes, an OML gauge should be closest to the outer ply stress.

I would use principal stress from a rosette, since your s/g axes may not be aligned to your stress axes. Match the FEM principal and infer the component stresses. With ground test articles, you have a lot more control over the loads applied. Exercise each stress component separately.

Is your composite "black aluminium" or heavily tailored for specific directions ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[Stress_Eng]

Just a couple of aspects, have you determined how to calibrate each strain gauge response (use of A/C test rig), and what is to be the A/C reference state, i.e. 1g level flight or standing on ground?

 

[Burner2k]

Thanks all for the replies. Since we are starting from scratch (or near scratch), many of the points mention here have to be learnt. Please keep on bring in relevant pointers from your experience. I may not have the answer but it will help us on giving some thought towards accommodating the challenges.

@Stress_Eng, we have not determined that part yet. Initially we will be getting experience through instrumenting ground test articles before proceeding to flight worthy birds. So, probably our calibration have to be done on A/C test rig.

 

[rb1957]

you need to make a static test rig. typical method these days is to have a "flying" specimen ... so you apply balancing loads and reactions; This sounds quite risky, so to protect the specimen I'd constrain a point with stiff springs with load cells to check these loadpaths. A problem in balancing the specimen is how the direction of the loads changes due to deflection of the specimen. You may be able to find generic information about static testing, OEMs will usually publish "sanitised" presentations. Good Luck !

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.