Wingspar layup process

General Composites Discussions Engineering Talk
#1

Hi Everybody

I need to infuse a wingspar and started with a first approach. However that approach was not appropiate because the seam between inner shearweb and outer shearweb wasnt wrapped together. I enclose the setup in the first picture. Its a section view of the spar layup.

Spar Construction.jpg
Spar Construction.jpg960×720 48.8 KB

Accordingly I tried to arrange the infusion in a way that alowed the inner shearweb to be wrapped around the outer shearweb as shown in the second picture.
In this arragement the infusion is done out the mold. I have one worry. Could it be that the spar flanges are permeable after the first infusion and that i get a vacuum short circuit and consequently the flanges dont wet out correctly?
Do you see any other issues?
Would you bond the rohacell core and poltruded carbon rods with epoxy or methacrilate? Could you think of a weight efficient way to bond them during the infusion and not in a separate stage?

Spar Construction1.jpg
Spar Construction1.jpg960×720 50.5 KB

Thanks everybody

#2

Hi

Just to know if you had finally do this layup ? It was looking like a good idea to proceed. But I may try to do the infusion of the pultruded rod, with the 1rst layer of carbon on the bottom.
Or you propose to bond the rod outside…?

thanks

#3

Not yet. But i am thinking of laying at the mold wall compoflex.

The outstanding problem curently is how to guarantee that the epoxy will flow between the graphlite rods.

#4

Did you manage to get this process figured out yet?

I have not done it yet, but i need to approach a similar problem.

The method i was thinking was;

  1. Bond the pultruded rods together via manual glueing together with epoxy, and using clamps or vacuum pressure on a flat mold to hold them together and form the spar caps only.

2.Lay the shear web fabric into the mold.

  1. Lay the core (perforated foam core) and spar caps into the mold on top of the first shear web fabric and ensure a close fit with no large void areas between foam core and spar caps.

  2. Lay the second shear web fabric over the top, and infuse both shear webs and the bonding of core and spar caps in a single infusion…

How does that sound?

#5

Sounds good but i would say not optimal. The problem I see is that to allow for shear transfer from the caps to the shear web every rod should be wrapped and the wrapping fabric should reach to the shear web. Thats the difficult point i havent manage to solve. I doubt that those fabric sections between the rods get wetted out.
Apart from that i think on similar a process as you describe.

  1. glue the rods together in the mold corner manually in a first stage
  2. the u mold is a little wider than the spar height. The widening corresponds to the thickness of the compoflex under full vaccum.
  3. I place compoflex on the mold face. Lay the fabric and the glued caps over the compoflex place the sandwich over the fabric and ahain cover everything with the outer shearweb. Cover everything with an outside layer of compoflex. That way the sandwich core doesnt need to be perforated.
  4. Infuse.

I am pretty convinced this is the best method however the caps dont transfer the shear to the web ideally in this manner.

#6

In regards of the compoflex, it’s just a kind of mesh media as related to be acting
as a peelply too. Putting it inside will not give you a flat C shape face…

In my view, bonding the rod altogether could be the best fit. Perhaps with an Hysol epoxy. The result gonna need to be sand and scurf properly to made the proper bonding area with the flange material. But having the rod into a block /solid form, that mostly equal to a foam core itself. So it should be ok for the infusion. And to fold back the flange fiber over to cover back.

As without being pre-pack all together, i was thinking it’s gonna cause some issues.
Packing dry rod will try to go all around and moving… Or maybe just applying in some local area epoxy on the rod itself, that could be a way too. (like half inch every feet)
As the void in between those will be filled during the infusion. In both case, a sample cut with diamond blade will be the best to see.

On the core, I look further and perhaps a Divinymat could be slightly better than a grid score has having smaller channel and so no big hole too as being bond on the scrim. But haven’t tested yet…

#7

The compoflex against the mold shouldnt be a problem. The finish of compoflex is quite good as i have been told and there arent aestetic requirements. The c being not perfectly flat isnt structurally critical either.
An outstanding issue is if i will be able to hold the compoflex tight in the corners of the c shaped mold.
Rohacell is much better core and without groves as the infusion will run on both sides.

#8

If you want a flow media type of fabric, 300gsm Continuous Filament Mat works well and should conform to the mold provided you have 15mm or larger radius in the corners of the U shaped mold. My preference is to infuse from the center outwards…

I dont see a problem with half of the shear fabric on each side of the spar caps which give the full width of the cap for the shear transfer area. You can see many aircraft spars made with all of the shear fabric on the sides of the spar caps, and only a single layer wrapped over the top in an i-beam configuration.

#9

I’m not sure when you said :

I dont see a problem with half of the shear fabric on each side of the spar caps

Do you mean just the shear fabric facing the mold, like the 1frst picture on top. And not that fabric fold back into ?
And do you have example of glider that are made like this it could be interesting? As i don’t recall on those…

#10

Here like this;
http://www.continuo.com/marske/ARTICLES/Carbon%20rods/carbon.htm