Improving aluminum flame spray adhesion on composite? 1

[KirbyWan]

Howdy all,

We have had issues with maintaining a good bond between the flame spray aluminum coating used as lightning strike protection on RB211 inlet cowls. The inlet cowl outer panels are carbon fiber and the process is to apply a base coat of an epoxy adhesive, I'm assuming for electrical isolation to prevent galvanic corrosion, then before the adhesive is cured, flame spray the part, then apply a sealing coat of the same epoxy before going to paint. We have had issues where the flame spray just flakes off of the epoxy base coat, showing it has poor adhesion. We don't do the flame spray here so we send it out to be done and I'm trying to learn as much as i can to find and solve the problem.

My questions are:

How much cureing should the epoxy get before begining the flame spray process?
Are there adhesive formulations that work better/worse for this application?
Are there other factors that help/hurt flame spray adhesion?

Thanks for your help.

-Kirby


Kirby Wilkerson

Remember, first define the problem, then solve it.

 

[Compositepro]

Has this process ever worked well? I've never heard of this being done and have serious doubts that it is a good idea. However, it's one of those things where you don't know for sure until you try it. Molten aluminum droplets hitting uncured epoxy will likely volatilize some of the epoxy, which will not help adhesion. The chemistry of the cure is also probably altered at the aluminum surface.

 

[berkshire]

Is there any chance that you can substitute Thorstrand ( aluminium coated fiberglass.) as that first layer?
B.E.

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[KirbyWan]

Compositepro,

The manual sections that refer to this process are not all unanimous, but one indicates a 6 hour partial cure before flame spraying.

We're trying to follow the process in the RR manual, which is now a Bombardier manual which references the RR materials, processes and other specs. I would think a change of the order you suggest would count as major and need testing to validate how well it protects against lightning. If It was my choice I would go with expanded copper wire mesh adhesively bonded to the CFRP skin. Eliminates galvanic corrosion, provides an excellent conductive path around the periphery of the inlet cowl. But we try and stay as close to the manual as possible.


-Kirby

Kirby Wilkerson

Remember, first define the problem, then solve it.

 

[IRstuff]

Does a mesh work well? We use a copper foil on one of our products, and supposedly, lightning strikes will still blow holes through the foil.

TTFN
faq731-376

 

[KirbyWan]

Well since lightning raises the temperature of the ionized air around to 30 000°C so there really is no absolute protection. It will melt holes in aluminum extrusions where it goes into and out them. The point is not to protect where the lightning hits, but give it a path so it only damages where it enters and exits the plane. At least that's my understanding.

-Kirby

Kirby Wilkerson

Remember, first define the problem, then solve it.

 

[Compositepro]

I've worked on developing lightning strike products. All light-weight solutions will see the metal vaporized right at the point of strike for about a 2" diameter. There will be surface damage to the composite at this point but no penetration. For about 6" around the strike point the mesh will have charred the surface somewhat due to the current density heating the metal mesh. Aluminum vaporizes to a larger diameter than copper or nickel due to the low melting point. The diameter of visible heat damage is highest with nickle because of its lower conductivity.
Aluminum skins are thick enough that the electrical and thermal conductivity usually results in barely visible surface marks due to lighting strike.

 

[blakmax]

Where is the failure occuring? Between the adhesive and the surface, or between the adhesive and the aluminium spray material? If it is failing at the surface of the part, then you have an issue with the preparation of the carbon part before applying the adhesive. Are you applying it to a surface which has had a peel ply removed? If so what type of peel ply?

Regards

Blakmax

 

[berkshire]

Kirby,
There are patents on flame spraying aluminum onto composite substrates here is one, Publication date: 2008-12-18
Patent application number: 20080311374
Check this one out and see if it pertains to the process spec you are using now.
It mentions using glass micro spheres in the resin as an adhesion promoter.
B.E.

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[MikeHalloran]

I'm guessing that the timing of the flame spray application relative to cure state of the epoxy is critical to the process, and I'm also guessing that the original process includes some undocumented time, like a trip across a city, or filling out tonnes of paperwork.

I.e., my WAG is that the original process applied the flame spray just after the epoxy reached B-stage, so the molten aluminum was impacting, not on wet runny resin, but on sticky almost-cured resin, such that the first flame sprayed globules penetrated for most of their diameter into sticky resin, and the second pass of flame sprayed globules bonded to those mostly embedded globules.

Maybe you can social-engineer your way to speaking with some of the original crew who figured out how to do it.


Mike Halloran
Pembroke Pines, FL, USA

 

[KirbyWan]

BlackMax, The failure is adhesion failure between the flame spray and the base adhesive coat. We are useing a peel ply to protect the surface and peeling just before flame spraying, I've gotten the suggestion to scuff sand and solvent wipe after this step and this may help.

Berkshire, I hadn't thought of doing a patent search, that's a genius idea, I'll check out your suggestion.

MikeHalloran, I think your right about the epoxy being in the b-stage. We've prepared a set of test coupons to see how the cure stage affects adhesion, with and without the peel ply (and I may have them scuff sand and solvent wipe half of the peel play area for a comparison.

Thanks to all for your suggestions.

-Kirby

Kirby Wilkerson

Remember, first define the problem, then solve it.