Cored Pre-preg Parts in One Cook

This post is about a technique that lets you speed up the construction of basic cored pre-preg parts by potting the core edges in thickened epoxy. While not the best idea for highly loaded or critical parts, it does let you cut one of the big corners that makes pre-preg parts expensive and time consuming – the price for this – thickened epoxy!

The key idea here is that the whole part gets cooked at once, but only after the core “chocking” material (ideally epoxy that needs post-curing filled with low density filler) had b-staged and is hard to the touch. You are using the cast-in-place core to apply pressure to the outside skin, so it will never be as good as a multiple cook part – but it sure is faster and primary bonds are nice too.

First Example: A Cored “Sawhorse”

Here’s an example of a test part I made a long time ago. It was a prototype bench for tuning snowboards and I wanted to core the top but also wanted it to be as quick and cheap to build as possible. Weight was important but not critical. This is what the core fit looked like before the second skin went on:

Nomex honeycomb potted around the edges with thickened epoxy so the part is cured in one cook.

The skin in the mold is not cured – it is just debulked in there well and glue film is debulked into the outside skin under where the core will go. Glue film is just pre-preg resin in a sheet – it is added when bonding core so there is some resin to soak into the core material and form a nice bond.

The honeycomb core was loosely fit and then I mixed up a batch of slow epoxy and thickened it with glass microspheres and about 10%-20% silica. Ideally you could use a resin system that requires a post cure so that there would be some primary bonding going on with the pre-preg – so they would co-cure. This isn’t necessary though – you just want to use a system that doesn’t produce amine blush. The filled resin was squirted into the radii under the core with a cake bag – you could do with a fillet stick too but it would be messier. Then I pushed the core (the corners had been radiused so it didn’t hang up in the corners) down into the fill and it shot up into the honeycomb cells. Then I used the cake bag again to squirt thickened epoxy around the edges to make a fillet. I used a lot!

Once the core was manually pressed in and the fillets were cut neatly, I placed a piece of peel ply (bias-cut no doubt!) then perforated release film and breather and bagged the core in using my debulk bag. This left the nice radii of cured filled resin, ready for the second skin. You can see the peel ply surface on the above picture – that’s why the uncured pre-preg doesn’t look shiny.

carbon sawhorse for tuning snowboards made in one cook with Nomex core - and cool aluminum leg inserts.

Here is the second skin laminate before cooking. The aluminum leg socket formers/mandrels were part of the experiment – they were wrapped in pre-preg material and bedded into the uncured outer skin before the inner skin was laminated over them. Worked pretty well! The part on the right shows the cured and trimmed part – the core didn’t print or leave any voids on the mold surface because it was essentially “potted” in place against the debulked skin.

Second Example: Foam Cored Stairs for a Yacht

SAN (Corecell) foam is carefully fit before the part is laid-up.  Once the first skin is debulked and film adhesive is applied...

Here is a more complicated cored part with Corecell foam core instead of honeycomb. One the left it the core fitting stage, which was done before the skins were laminated in the mold. The idea was that the skin thickness (about 0.04″/1mm) would offset the foam uniformly and that if the joints were loose, there would be room for the filled resin to pot the foam against the debulked inner skin. Here the inside and outside radii as well as the perimeter flange made fitting carefully a necessity. One the right – the pre-preg glue film is debulked onto the outer uncured skin.

the foam is potted down in thickened epoxy and then vacuum bagged (gently) so that the wet-filler is compressed and perfectly fit against the pre-debulked skin.

Here is the step where the core is pressed into the wet filled resin so that it forms a chock-fit – essentially a cast-in-place addition to the foam. One the perimeter of the steps there is a rebate for secondary bonding and this is why there is a wide skim of filled resin around the edge. During the cook, the foam will soften and “thermoform” over any excess glue – which won’t be a big deal because we are going to vacuum bag the foam down with the debulk bag as soon as it is all fit.

more filled epoxy applied to the uncured pre-preg to fill all the small gaps between the foam and the skin.

Above is a closer look at the process. This would probably get you taken aside for a stern talking-to at Boeing or SpaceX but in this case it saves time and makes a good solid part with only a small weight penalty. The core is all perforated to bleed air and resin when it is cooked – and the excess adhesive mix bleeds out and into the bag stack while the core itself is bagged. So a good deal of this white filled material will never make it into the final part. Here is the core the next day after the bag, peel ply, perforated film and breather was removed:

after the wet-epoxy has gelled, the excess is cleaned up.

We went around and detailed the radii by either scraping away high spots or filling carefully with more filled epoxy. The foam here is neatly potted into the outside skin all around and all the pre-preg will get cooked in one final cook at the end. Next we started laminating the second skin. The first layer of the second skin was “combo-laminated” with glue film on a debulking table so that we didn’t have to handle glue-film and then place pre-preg plies onto the film itself – a surefire mess! After cooking, the part was demolded and looked like this:

Finished prepreg carbon stairs for a large yacht with taping rebates around the perimeter and a bonding flange to meet the finished hull.  Once cook!

It came out great! There were no voids or nasty radii and the rebate are around the edge was tight. This would have been even easier with honeycomb core but it works ok with foam – you just have the make sure the core fits loose and there’s room for the excess filled resin to squeeze out. There’s not a ton of detail here but it should help you if you have never used this method before. If you have any questions, just ask!