# Simpler sandbag testing?

### Help Support HomeBuiltAirplanes.com:

#### akrotodd

##### Member
Howdy,

I am new. I've built a couple plans-built and custom jobs. I've got the aero degree. But I realized pretty early that I don't know much from school.

My current project is a very heavily loaded wing. In designing the spar and trying to finish the stress analysis, I find that simple Beam Theory of just the spar, 3-spar box theory, and many others I've beat up on aren't giving me valid answers. I've purchased a FEA package that has allowed me to at least easily model all the components of this wing, and get results that are valid. But how accurate the data is remains an unknown.

SO, I AM BUILDING THE SPAR AND I'M GOING TO BREAK IT. Actually, to save materials and time I'm building a simpler, shorter, but still representative beam that I'll model in FEA the same way, predict the deflections and stresses, and take huge amounts of data while I load it up. This, I hope, will make me feel good about the FEA results of the real thing.

So here's the biggest problem: the spar needs to support a BIG load, and to break it I'll need to stack a quarry-worth of sandbags on top, if they're each about 75 lb. There simply won't be room to make a big enough pile without them shifting or falling off! And making a "table" on top of the spar means making something so strong, it will change the strength of the spar itself.

I've thought of suspending a flexible (hinged?) table under the spar with straps, so I can adjsut the point loads at each strap as I go outboard along the spar. But what I'd really like to do is strap the spar to a large steel I-beam at many points outboard from the center, and jack up the middle with a force gage that tells me what the load is.

SO, THE REAL QUESTION: How could I make sure that the straps that attach the spar to the "base beam" are transferring an elliptical load? If I were to simplify to a uniform lift distribrution, I could envision a series of pulleys that all take the same rope, at both the spar and the base beam. Each pulley is bolted to the base beam, and at the spar they're each held near the spar with straps around the spar (no holes!) When I jack up the middle, the tension at the pulleys will be the same. But the lift distribution is not uniform--it is the modified elliptical distribution due to Schrenk's, and from a Purist's standpoint this bothers me to simplify.

What have other people done when sand isn't dense enough? Where's the local spent uranium shop?

-akrotodd

#### wassbiplane

##### Well-Known Member
Forgot who said it---but---

Aircraft design is 90% educated guesses, worked out to four decimal places."

Jerry

#### Birdmanzak

##### Well-Known Member
I've seen video (if you want to see it, it was "somewhere on the internet" - helpful, I know) of a sailplane wing being tested in a way similar to that you suggest. The load was applied by a crane through a series of steel beams arranged like a child's mobile. It gave an approximation of an elliptical distribution and you could get as close as you like by adding more beams.

Of course, it doesn't get you any further than the steel beam/pulley combination you mention:

The point is it worked well enough to certify the wing. It won't be exactly elliptical, but close enough.

#### Jeremy

##### Well-Known Member
The whiffle tree drawn above works well enough to be the usual means by which big public transport aircraft wings are tested.

One thing to watch is that you apply the loads in the right axes. The wing spar sees simultaneous loads vertically, forwards and in torsion, so the load applied needs to be offset across the chord as well as the span, and needs to also be inclined forwards.

The usual way of doing this with sandbags (or lead shot bags for high loads) is to load the bags towards the front of the wing (with the wing upside down) and also to incline the wing at or about the stall AoA.

If using a whiffle tree to apply vertical loads, then the point of load application on the wing needs to be set on the chord such that it produces the right torsion load. The wing still needs to be inclined to the stall AoA for the test.

Jeremy

#### akrotodd

##### Member
Well,

You guys' input is just about perfect. Kid's Mobile concept is a good one.

I can do that.

Thanks.

-Todd

#### wsimpso1

##### Super Moderator
Staff member
Log Member
Whiffle trees work well. You have to design them to distribute the load elliptically.

You will probably have a lot of weight in the whiffle tree, so you will not want to rely upon the hydraulic pressure gauge. Place a load gauge in one of the cables that is forecast to have low loads.

We expect pictures of the setup and then of the broken parts!

Billski