Quantcast

Load Test Question

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,645
Location
NJ
But I didn't see (in calculations) the moment of air load.
In any case, a 3d element model with an excellent grid of elements is necessary.
Posts 26-28 have the moment calculations...... we used the area of the wing for the distributed load, I already had that calculated for my loading test, and Tdfks had the wing platform dimensions and obtained the areas and distribution that way. We were mostly in agreement on that as regard the numbers so posting of the calculations was not necessary. Going through the details of my loading spread sheet or his spread sheet you would be able to see the distribution.

FEA is a specialty all in itself, it is great when the output matches the other calculations but I do not have the necessary expertise to use it to any great extent such as my confidence level in the out put approaches that of a simple MY/I calculation. For example looking at my deflection outputs with different runs and only slightly different loading I have outputs of 3" or 7" ....if I hand calculate the deflection based on cuts along the span to get the necessary Moment of Inertia's, I would have a higher level of confidence on my result.
 

tdfsks

Well-Known Member
Joined
Aug 29, 2005
Messages
93
Best I have are at post 1.....the bare aluminum and reflections made it hard to see anything through the camera lens.
No problem, I just thought I would ask.

I didn't get a chance to do much with this today. Should have more time tomorrow.

I think that 0.020" angle is going to need some stiffening for starters. What rivets have you used ? Are they solids ? What dia ? Are you able to remove the inboard leading edge skin ?
 

tdfsks

Well-Known Member
Joined
Aug 29, 2005
Messages
93
But I didn't see (in calculations) the moment of air load.
There was a recent thread on the design of composite wing spars. See some of my posts in that thread for the method used to calculate the bending moment.

With regards to the use of FEA for this problem ..... it is a very difficult to model structures like this when you have a lot of buckling in thin sheet and no amount of stiffening is going to eliminate all the buckling. I have 35+ years experience in doing advanced FEA analysis, a lot of it for aerospace structures, and even with this experience, it would be a challenge to get the right answers for this structure. Also you are going to need software that is a bit more capable than the free stuff. See my earlier comments in this thread on the required modelling approach. I think we can solve this without FEA .....
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,010
Location
Saline Michigan
And much of what we are doing does not really need FEA. Hand calcs work pretty well, even when calculating critical loads or stresses for buckling, as has been demonstrated above. Another thing to know about FEA is that it is not usually some magic. It is doing many of the same calcs we would do by hand, but it does it many times and many places way faster than we can even hope to do them.

On other point - when you are building sheet metal structures, your options are usually thicker sheet or more layers or a combination of changed layer count and thickness. There is no avoiding the fact that the structure will change in discreet steps and tailoring to meet the loads will rarely land on exactly the needed structure. With substantial steps from one iteration to the next, one will be inadequate and the next will be somewhat overbuilt. That is OK - the structure will be a little over what it needed to be, but now it deflects less everywhere, and is less likely to have "surprises".

And if the weight offends you too much, there are always other combinations of thickness and position used in places that might also satisfy the strength and stiffness needs. Iterate, search the design space, and find the lightest solution that works.

Does this change when you go composites? Not much. We tend to standardize on a precious few lamina that we know well the behaviour of, and then it is a matter of "do I use another ply in the caps or another two plies in the web?".
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
13,582
Location
Port Townsend WA
Might need .020" at root. I recall that Lazair upgraded to .020" and the strutted Lazair wing is much lower bending moment.
 

plncraze

Well-Known Member
HBA Supporter
Joined
May 11, 2006
Messages
1,871
If you have a set of Jeanie's Teenie plans look at how the angle was put in the flange of the spar web. The outer corner of the angle was rounded to fit into the inner corners of the upper and lower spar web. My fantasy is that with some hole finders and a sander you could add material where necessary without significant disassembly.
 

christos

Well-Known Member
Joined
Nov 9, 2018
Messages
52
Location
Greece
I apologize, i didn't see it yet, i saw only the stress calculations without it.

Off course, engineers designed and built aircraft a lot of years before the first PC.

Finite element analysis is worthwhile in flutter analysis, off course in a light aircraft case, the static aeroelastic analysis should be fine.
I am working on a spreadsheet for static aeroelastic analysis if you are interested in.

Finite element analysis is another one solution, but every scientific solution is correct.
It is your own to go with FEA or by hand calculations. In some cases, FEA is far away faster and it is the main reason why we choose it. In other cases, it simply doesn't worth the effort.
You can choose FEA also and go with hand calculations. There a lot of engineers whose run FEA in mathematical Softwares, like Matlab.

Nowadays there are a lot of open sources software able to solve the most difficult model, like OpenFOAM, but it is more difficult to run than commercial software like ANSYS.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,645
Location
NJ
I think that 0.020" angle is going to need some stiffening for starters. What rivets have you used ? Are they solids ? What dia ? Are you able to remove the inboard leading edge skin ?
The .020 Angle is sandwiched between the .025 web and the flat cap strips. The rivets are Solid Aluminum AD rivets 3/32....the aft portion of the angle attached to the Leading Edge are solid AD 3/32 rivets.

I could cut out and drill out the rivets to remove the inboard panels, but that would not be my first choice....

If we are now talking about a fix I will be able to install stiffeners without removal of the skin and I can even get a Z stiffener at the aft .036 flange.....I believe the caps and spar are strong enough, the flange and LE are not .........the .02 angle right now is the only tie in to the leading edge from the spar.....and the area of that radius portion of the .020 angle is only .0028...not very much. With a 3/4 pitch in the caps and 3/8 min pitch for the 3/32 rivet a piece of angle or channel is a possible addition to the caps in order to better tie in to the LE.

If the 28K Z stiffener is a good number (see spread sheet) and the 1" and 2" span .016 plate calculation between stiffeners with a 24k and 18k buckling number are good numbers....I think I see a path. I just am not sure if the say 2" between stiffeners is correct or if that calculation should be 21" distance span-wise between the ribs. The FEA is consistent with the 2" numbers.
 
Last edited:

tdfsks

Well-Known Member
Joined
Aug 29, 2005
Messages
93
Yes .... removing the skin would be the last resort. I was just getting a feel for what is possible. Good thing you have used solid rivets ... at least they are easily removed (unlike say stainless steel blind rivets).

I need to do some more calcs but, do you think you can fit an extra angle between the rear face of the spar cap and the horizontal flange of the 0.020" angle ? The angle might be 0.032" or even 0.04" with an appropriate bend radii. Can you get the existing rivet through the skin and horizontal leg of the 0.02" angle through the new angle allowing for its bend radii ? See sketch below. This is not something I would do in a certified aircraft because it prevents future inspection of the spar cap .... but I do not see a lot of other option here since we are looking for an easy fix.

Also on another issue, have you calculated the spanwise distribution of pitching moment along the wing ?
What is your maximum design dive speed (VD) ?

IMG_8669.jpg

Also, is the following extract from your Autocad file (same as my sketch above) what you have actually built ?

Spar Cap Arrangement.png

Or do you have the spar cap on the front of the web with the stiffeners on the rear face ? Looking at your photos it looks like the cap is not on the rear face of the web ??
 
Last edited:

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,645
Location
NJ
Looking at your photos it looks like the cap is not on the rear face of the web ??
Can you get the existing rivet through the skin and horizontal leg of the 0.02" angle through the new angle allowing for its bend radii ? See sketch below.
Except for the flange being 1" instead of 1.25 as shown in my sketch yes.... for the second 4' length of caps. The caps and stiffeners swap sides of the web and are bolted together at the splice points.

I can rivet through the caps to the angle and also get rivets into the aft flange, I will need to make layouts but I have a basic question about that.....The attachment of a spar is always much less mass than the caps of the spar......I can get Z stiffeners on both sides of the spar at a 1" spacing....do I need to attach to the caps of the spar.

shown root top cap, 2ud rib and bolted splice, 4th rib and bolted splice.

I did not calculate any pitching moments.....I copied the plan-form (platform) of the Carbon Dragon and assumed the aerodynamics were good.

The Spar web is at about 27% cord.

IMG_1172.JPGIMG_1173.JPGIMG_1174.JPG
 

Attachments

Last edited:

tdfsks

Well-Known Member
Joined
Aug 29, 2005
Messages
93
OK ... so are you saying that the caps are in 4 ft lengths with overlaps where they are bolted and that each successive 4 ft length alternates to the front and back of the spar web ? Like this ...... where the blue is the web, the red the caps and the green the bolts .....

Spar Caps.png
Also when you did the static load test, where did you hang the loads from ? Just the ribs ? Were the ropes hard up against the rearward facing flange on the spar on all ribs like I can see in the photo above ?

The published flight envelope for the Carbon Dragon is as follows:

Carbon Dragon Flight Envelope 2.png

So it looks like the design VNE was 70 mph although other data I have suggest that this was actually set at 65 mph in the operating limitations.

Are you intending your wing to have the same operating envelope as the Carbon Dragon ?

Interestingly they seem to have only designed for a 20 fps gust which seems a bit low. However, it is stated that no alleviation was applied which is conservative. I am guessing that only Irv Culver could explain why they did that but he ain't around any longer so this will likely remain a mystery. Alleviation reduces the gust loads by using transient aerodynamic theory (Kussner functions etc) that allows time dependence of the lift build up, as the aircraft flies through the gust, to be considered. Essentially what it means is that the aircraft will fly through the gust before the lift builds up to the full value that corresponds to the angle of attack change caused by the gust, thereby reducing the peak gust loads.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,645
Location
NJ
OK ... so are you saying that the caps are in 4 ft lengths with overlaps where they are bolted and that each successive 4 ft length alternates to the front and back of the spar web ? Like this ...... where the blue is the web, the red the caps and the green the bolts .....
yes that is correct, and stiffeners on the opposite side of the caps so as to allowing stiffeners not to need filler under the caps to save weight.
Ropes were as shown for the spar test, and at the 1/3 location for the aileron test. The plan is to also stack sand bags on top a little forward/aft the spar but I did not get that far.

I used FAA Glider Criteria , tail loading, aileron loading, and tested to the proper lb/sq ft.
9 psf or 12 psf based on speed in the charts, Page 25-27

 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,645
Location
NJ
How do we know that little piece of angle is strong enough? Is it the calculated crippling strength ? Is it shear flow per inch/cross section area? This has to do with weather to attach any stiffeners to the caps or how thick the attachment to the caps have to be

1594033633468.png
 

tdfsks

Well-Known Member
Joined
Aug 29, 2005
Messages
93
We need to calculate the shear flow in the D Box for combined wing shear force and torque. I have that half done ... but it needs more work.

This style of spar (sheet metal C section with bars riveted to the webs) is quite common ... as are failures ... most of the failures result from the buckled spar caps.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,645
Location
NJ
Cap thickness including web thickness, angle attach thickness, up to 4 th rib ws. 72.375 is .3575..... caps are 7075-T6 .....rest of the items are 2024-T3
revised section properties.

1594042724928.png

1594042789744.png
 
Last edited:

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,645
Location
NJ
I have never done a shear flow calculation so be gentile here.
Shown is a simplified layout for the calculation.....seems low so I wounder if it is valid.
 

Attachments

tdfsks

Well-Known Member
Joined
Aug 29, 2005
Messages
93
Thanks for the links above to your building thread ... I hadn't seen that before. I just spent some time looking through it and now have a much clearer understanding of how the spar is built.

I will have a look at your shear flow calculations ..... although the spreadsheet does not seem to open for me at the moment.

Before doing shear flow calcs for the LE D box I first need to calculate the torque loading on the box. That is todays project. What are the flap deflections ? The same as the Carbon Dragon ?
 

tdfsks

Well-Known Member
Joined
Aug 29, 2005
Messages
93
The spreadsheet opens for me now.

Your approach to the analysis of shear is correct for the web of a beam however I need to check the numbers because something does not look right. For a beam like this with a thin web, and most of its mass in the caps, the shear stress given by VQ/It should be pretty close to the average shear stress in the web (i.e. 700/(11.67 x 0.025) = 2399 psi. Your spreadsheet is giving just 350 psi. The error is in your calculation for A and hence Q is wrong. A should be the area outboard of the section where you are calculating the shear (in this case the neutral axis) and Q should be the first moment of this area about the neutral axis. The A must include the contribution from the cap. You A is too small. See if you can fix it ... otherwise I will have a go later.

For a D Box that is loaded with both shear and torsion you need to use the following approach:

Single Cell, 2 Flange D Box Equations.jpeg
 
Top