Center Section Spar Layup

[hdmx]

Hello everyone,

I got a question regarding the composite construction of the center section spar. According to the plans you fabricate the box and the lid. The "large box" has a U-shaped cross section. It is laminated with glass inside, the glas is then trimmed after curing. The fat red line represents the inside layup - see first attached picture. Then the laminated and cured CS4 foam panel (forward foam) is bonded to the U-shaped box with micro slurry. You nice see it in Ary's well documented build log (great read by the way):

- http://www.aryjglantz.com/2015/01/spar-interior-layup.html

- http://www.aryjglantz.com/2015/01/spar-lid-cs4.html

So my questions are the following:

1) Why is the inside layup trimmed which results in a simple and weak tee joint? Wouldn't it be better to overlap the inside glass layup (see second picture) and peel ply it so it the CS4 will not get bonded to CS2/CS3 foam but directly to the inside glass layup with a much larger surface? It doesn't add any noticable weight, it doesn't add any extra steps, it's not more work - what do I miss here? Even if overlapping has negative side effects, why not at least use a floxed corner for a better connection between glas on CS4 and glas on the U-box?

2) According to the plans, CS4 is bonded with wet micro to the U-box. Why not flox, which is supposed to be structurally stronger?

Maybe there are very simple and obvious reasons that I might have overlooked. Thanks for any insights

[Aclouston]

I'm not an aeronautical engineer, but if you may, I'll venture an answer. You're interrogation is implying two things: 

1) you have the impression that a foam to foam bond is a weak bond in the sense that we should try to avoid it.

2) you have the impression that a single glass to foam bond is weaker than a glass to glass bond.

On (1), I'll comment that most higly stressed parts of the plane have foam to foam bonds in both direction. Look at wings and canards. In those cases, the foam assembly is encapsulated in layers of glass from all sides. If this was extremely weak, the designer would never accepted to bond pieces together if it created a risk of separation. He would either have called for a one piece foam for the wing or have builder's glass one or bothe pieces before bonding them together.  I will agree however that we do not a whole lot of foam bonded perpendicularely. I want to insist that I am not saying here that the foam carries the load. I'm simply stating that bonded foam encapsulated in glass is part of the design everywhere on the plane.

Now, the center spar carries the load mainly in torsion and flexion. Internally - layer by layer - that means that foam and glass want to shear off from each other. If you look at the location of the spar caps, which takes a lot of sweeping flex and some limited vertical flex load, they have been positioned by the designer at the aft end of the spar. This means the designer has calculated that the back of the spar is more critical ( in flex anyway) than the front. It makes sense since the aft part of the spar is bolted to the wing. The shear in the front is obviously less than at the back. If it was more the spar cap would have been positionned at the front or extended to cover the entire top section. 

Looking in torsion, ideally we want a round shape to distribute evenly the load. When not practicle for the application, the designer will however try to homogenize the section, to be basically the same all around hence the foam to foam in all corners. A junoir designer - not knowing any better or designing without weight limits -  would have made cs1 and cs4 the same thickness as cs2,cs3 as they would carry more torsionnal load and also more flex load not because there is more foam but because the distance between the outer and inner glass of both fwd and aft wall would be greater so more stable (less likely to encounter buckling during flex). 

I'll add that the further you are from the point of application of a force, the less strong you have to be. This is the simple lever effect. Here the bolting joint is where almost  everything the wing feels is transmitted to the spar. So again it makes sense that the fwd section is "simpler".

On (2), consider the shear web of the canard. It is basically a foam - the leading edge microed to the multi-layer shear web. Then everything is surrounded with glass. A pure glass to foam of a highly stressed part.

Also, don't forget that the whole strake is installed over the taper of that fwd section giving it added support.

With all that said, your suggestion is, I think a good one, especially if it is accepted that glass to glass bond is generally significally stronger than any other cases. Obvoiusly Mr. Rutan has decided to go that way because he was cconfident in that all the loads were being taken care of.

Sorry I'm french speaking in my native tongue and I used to many words trying to express my comprehension of why this is. I hope it helped... or is maybe making you ask more questions!

Regards

Allan

 

 

 

 

 

[Aclouston]

I must add that you have a very good sense of observation. I am at my third read of the manual and will read it again with simillar eyes as you just have. Are you a composite buff?

[hdmx]

Thank you very much for your reply. I think your answer boils down to:

vor 25 Minuten schrieb Aclouston:

Obvoiusly Mr. Rutan has decided to go that way because he was cconfident in that all the loads were being taken care of

Basically the center section spar is strong enough as is to accommodate all forces given in the original design. Adding more complexity would not make the spar better. The aft section will take most of it and the front is not as stressed so altering the design might not even have a significant impact for the original design. I fully agree with your conclusion here.

I plan to incorporate a retractable landing gear which will be bolted onto the front face of the center section spar. This changes the loads for this part. I thought about possible load paths - the forces would need to be transferred to the thick spar caps so this glas-to-foam connection will be stressed significantly in comparison to the original design. Altering the connection probably isn't enough to take the landing loads but I feel like this change from glas-to-foam to glas-to-glas makes sense in addition to introducing some thick layups that will transfer the loads to the aft spar caps.

So, condensing it down to two sentences: would you think that this change makes the spar stronger but was not done by Rutan because it's simply not needed? Or can you think of any reason why this change might even have some negative side effects and shall be avoided?

By the way I have not really worked with composites so far so this is new territory for me

[Kent Ashton]

2 hours ago, hdmx said:

I got a question regarding the composite construction of the center section spar.

Not that I know-it-all but I have been fooling around with canard airplanes for 30+ years and never heard of a centerspar failure or delamination.   so I would say, "don't overthink the plans".  There are a few weak areas in the Long-EZ plans I can think of that could be addressed:  Main gear mounts (airplanes are flown heavier than Rutan anticipated), engine mount angles (for heavier engines), nose extended (for heavier engines).  Other than those few things and a few I may be forgetting, there is no need to deviate much from the plans.  When people do that, they generally just take longer to build and it does not make the airplane measurably safer.

Maybe you have seen the pics of Rutan testing a canard to failure.  It was a donated canard that was judged to be too dry and still went beyond the G limits.  I can dig it up the picture if you're interested.

[hdmx]

vor 5 Minuten schrieb Kent Ashton:

and never heard of a centerspar failure or delamination

absolutely, however I would introduce new loads by mounting a retractable landing gear on the forward face of the spar. The original spar was designed to take loads introduced on the aft face and it does a good job doing exactly that - it was never intended to take loads introduced in the front face

[Kent Ashton]

If you are using Infinity retracts, I imagine JD could advise you on how he reinforces the center spar.   https://infinityaerospace.com/     He was sick for a while.  Not sure what his status is these days.  Make yourself happy but IMO, retracts improve the
"gee whiz" factore but they don't add very much to the top speed and make it less likely you will complete your project.  A few years ago, there was a problem with brazed (?) welds cracking on the struts.  Maybe those are fixed. 

[zolotiyeruki]

Why is the foam bonded to the glass with micro instead of flox?  Because (at least the way I've heard it), between fiberglass, micro, flox, and foam, the foam is the weakest link.  Replacing micro with flox would add weight (albeit not a ton) but the structure would be no stronger.  In other words, whether you bond it with micro or bond it with flox, the foam will fail before anything else.

[Voidhawk9]

5 hours ago, Kent Ashton said:

retracts improve the "gee whiz" factor but they don't add very much to the top speed

Assuming all else is equal, which is often not the case for birds with upgrades of this sort. They become much more significant if speeds are pushed up a bit by other means.