Embedding Metal Components into Composite Part

[grantdunlop]

Dear All,

I am working on a composite part design where there will be some additional components bonded into it during the lamination process. The components are small/thin attachment plates (approx. 1.5mm thick) which primarily have to resist tensile loads (and perhaps some lateral and torsional loads). The plates will be positioned centrally within the lamination and have an exposed fastening point at one end. The overall part is to be produced from epoxy pre-preg of carbon-kevlar composite in a closed steel tool.

I have done a little bit of research and it seems like titanium would be the preferred material due to its mechanical properties, corrosion resistance (the part will be used in a corrosive environment) and general compatibility with carbon composites. Is this a reasonable assumption or are there other options? I am aware of the issues with using aluminium with carbon composites - corrosion, CTE, etc... and the need to isolate the components through layers of glass.

I would be interested to know whether anyone is aware of any possible plastic materials which may be able to be embedded in the lamination and withstand the heat and pressure during the curing process. A similar assembly used a 10% glass reinforced nylon 6 component which was then mechanically attached to the composite part and we would like to do away with the mechanical fasteners by embedding it into the composite part.

Our composites manufacturer has experience with embedding aluminium components but none with titanium. Any information or suggestions regarding part design, lay-up, surface preparation and general manufacturability would be greatly received.

Regards,

Grant Dunlop

 

[wes616]

Are you considering this metal, or the plastic component, due to production volume? If your volumes are not very high, why not consider an all composite assembly. It is possible to replace the metal bit with some type of composite layup that will achieve the same result with less headache. It seams you could locally build up a quasi-isotropic laminate to, or if this is a shaped piece, possibly build this component ahead of time and secondarily bond (or even co cure) it in place.

Regarding co curing metals in a laminate. Abrade, abrade, abrade. Ensure that the surface is cleaned just prior to placing it within the layup. Speak with your prepreger - they should be able to give you much more tailored advice. It is important to use a film adhesive that will have good metal/composite properties. Do some test panels to ensure that you are not getting disbonds across the surface of your metal bit.

Even after all this, you may need to consider a secondary bond anyway. With a quality surface preparation and the right epoxy system, you may be able to eliminate the fasteners without much of a change to the process you have in place currently.



Wes C.
------------------------------
No trees were killed in the sending of this message, but a large number of electrons were terribly inconvenienced.

 

[grantdunlop]

Wes,

Thank you very much for the reply and your advice.

Unfortunately the parts will be produced at quantities which make it uneconomical to manufacture the embedded plates as composite components as you guessed.

Some one has suggested that we might be able to use something like G10/Garolite to produce the embedded plates. I will have to look further into this.

I will speak with our manufacturer to see what they recommend in terms of surface prep and general lamination specs.

Thanks,

Grant

 

[RPstress]

Ti is a pretty good choice for metal in carbon composite. However, it does have a bigger coefficient of thermal expansion than carbon/thermoset. For a typical carbon layup you'll probably wind up with a CTE of about 2 or 3 microstrain/K. Ti is about 9. This can cause warpage problems on cool down from cure.

Surface prep of Ti is a slightly black art. A hydroflouric acid etch followed by what's essentially an anodise can work. Simple mechanical abrasion can indeed be effective, but most people that I know of do a chemical treatment to optimise adhesion. It depends a bit on the use. Will there be relatively high stresses in the Ti-carbon bond, and what are the consequences of failure? If it's lightly loaded you may well get away with thorough cleaning and abrasion.

G10 seems to be a common glass laminating material with the usual relatively low mechanical properties. If it'll do the job you could use it, but it will have a CTE rather higher than Ti. On the other hand glass is a lot less stiff than Ti, so warpage problems might be about the same by the time the glass is thick enough to do an equivalent mechanical job to Ti.

 

[PanelGuy]

Good advice around. I definitely recommend toalking to your pre-preg supplier. Most of them have the lab and the ability to solve the problem given a little time.

As well, check with your supplier or better yet, if you can find a titanium coater, they have to bond to it all the time, just the other side And yes, abrade, but do it right before you bond it. Priming it is even better as it preserves that surface where you want it.

As well, I will go for the crude method usually overlooked...Is there any reason not to surface mount? Adhesive technologies support it well and (gasp)you could bolt through to a plate if needed.

It is maybe not as sexy as comolded parts, but it can certainly get the job done and make the load well distributed. Also, if the adhesive you use is flexible (2040 has 400% elongation) it will absorb any CTE issues.

 

[grantdunlop]

Dear RPstress & PanelGuy,

Thank you both for your replies and suggestions.

I am hopeful that our pre-pregger will be able to handle any warpage issues from the curing process. They are a little unsure about the surface prep required so I will pass on your suggestions. I had found some information from Henkel describes a similar process to what you mention.

I don't think we will be using the G10 option as it doesn't sound like it has many advantages over Ti or other metals.

Unfortunately the size and appearance requirements of this part will not allow a surface mounted mechanical fastener solution although we may look at adhesives. We had used a G909 one-part epoxy adhesive for bonding glass reinforced nylon components to a carbon composite part, with good results.

Thanks,

Grant

 

[RPstress]

3M also recommend Ti surface prep using Turco 5578 in their AF 163 film adhesive spec. This is easily found on their website.

And yes, you should almost certainly prime.

Beware: there is usually a time limit on how long after prep you can bond and the prepped surface needs protecting.

NB: Usually a film adhesive would be used even if the carbon is not pre-cured.

 

[rodneybennet]

I definitely agree with RP Stress:

Titanium can be incorporated in carbon - epoxy composite, but indeed surface prep of Titanium may be essential in case (small) gas and/or water/moisture permeation must be low, or in case a sustainable service life is required in a humid environment.

Perhaps - and indeed - thorough cleaning and abresion will be sufficient if ones uses (i) sufficient volume of (ii) a rather polar epoxy (100% water uptake larger than 1 weight percent relative to case epoxy at 20 degrees Celsius and 1 bar atmospheric pressure).

I should test the specific composite configuration if the composite will be applied in a humid and stressed enviroment.

More information regarding permeation and composite configuration can be found here:

http://www.composite-agency.com/materials-forum.htm

and some tables for gases and liquids in their "library" section.

 

[grantdunlop]

Dear RPStress & Rodney,

Thank you very much for your replies and suggestions.

I will pass on this information to our supplier and discuss with them the lamination type, insert materials and surface prep. options with them prior to producing some samples.

Thanks,

Grant Dunlop