honeycomb question

[nieuwhout83]

I am a semi-professional Dutch windsurfboard builder. My mainproducts are custombuild race- / slalom- / speedboards. I use carbon, kevlar and glass with epoxid resin in combination with pvc sandwichfoam by vacuümbagging. I want to make some real progress in weigth and stifness, so I'm testing kevlar honeycomb in stead of pvc foam sandwich. I am very aware of the importance of the bonding between the sandwich layer and the two glass /carbon / kevlar layers. Esp. with honeycomb this is difficult, because of the small contact. I'd like to have some good tips from people who are using this material for a longer period.

My questions are:
1. What gives the best bonding with kevlar honeycomb? Glass, carbon or kevlar?

2. Are there some tricks to get better bonding?
(I'm thinking about sanding the HC and rolling epoxy resin on the HC before putting on the cloth and filling the HC with pu foam on the critical aerias)

3. What are the expected problems in short and longer period of use?

Regards,
Gertjan Nieuwenkamp
info@iq-boards.com
www.iq-boards.com

[Herman]

Roll some thickened epoxy (not too thick) on the core. Then apply the top laminate. The thickened epoxy should be there to create small fillets on every cell wall.

Highly impacted cells can be potted, eg filled with thick epoxy.

The laminate you put on is not much of an influence for bonding strength, the right amount and right consistency of the epoxy for the fillets is...

Problems you can expect is a limited compression strength, with catastrophic failure (below the footstraps). Even foam is of limited use there. Unless a high density foam is used, of course.

[gonzo]

If extreme light weight is not necessary, a honeycomb filled with foam will give you much better adhesion. In some cases you can use a lighter laminate.

[nieuwhout83]

After doing some destructive research (this is what the technicians love to do; hurt the material with a big hammer and look when & where it goes wrong)I discovered a very big difference between honeycomb sandwich and pvc foam sandwich. The HC breaks first on the inner skin of the sandwich while the pvc breaks on the outside. My simple expanation is, that HC doesn't get any thinner under pressure 90 degrees to the surface as with impact forces. So all the impact forces will hurt the inside skin. A pvc foam sandwich on the other hand is far more soft, elastic, so most of the impact forces will be absorbed by elastic deforming of the core and mainly hurting the outer skin.

So my first conclusions are:
1. HC sandwich will collapse very easy on the inside, while pvc sandwich will break on the outside.

2. The danger is, that with honeycomb you can't see the damage, but also important; you can't repaire it that easy!

3. To build HC as light as possible you have to build the outer skin thinner because the innerskin must be stronger. But how thin can you go?

4. Because of the difference between the inner and outer skin I think on the inside you need ud carbon (forces in the direction of the fibers) while on the outer skin you better use carbon + kevlar. Here the forces are 90 degrees to the fibersdirection and kevlar is much more capable for that.

Any replies from the experienced users ???

[marshmat]

I haven't done much wet layup with honeycomb, but have built a couple of solar cars using prepreg carbon/Kevlar and Nomex honeycomb.

I don't consider this type of laminate to be suitable for applications with large point or impact loads (eg. a boat running aground). But it's excellent for distributed loads- such as an aircraft's wing.

I concur with Herman's advice- the core to skin bond is dependent on the thousands of epoxy fillets that form between the cell walls and the skins. Getting just the right consistency of thickened epoxy to create these fillets, without flooding the cells, is the trick.

If the skin thicknesses and core strength are appropriately chosen for the application, this "breaking on the inside first" doesn't happen. In your experiment, nieuwhout83, the compressive strength of the core you used is higher than necessary for the strength of the skins you used, thus the inner core-to-skin bond failed before the core itself.

Hexcel publishes a nice engineering guide, #7586, on how to estimate a panel's limits under various failure modes. It's free but a bit hard to find on their site, so I hope nobody will call "copyright violation" if I repost it here.