Layer Adhesion Research
- Maciek Pajak (Deactivated)
Delamination is the most common source of failure, it happens between the outer carbon fiber skin and the Nomex core. The epoxy adhesive film we currently use doesn’t work as well as we would like since the epoxy we use may not be strong enough and the bonding area may be minimal since Nomex has a honeycomb shape, we can’t get as much epoxy on it as we would like. The following research will be on trying to find ways to fix this.
Shear-key insertion
Taken from the research paper: Improving delamination resistance capacity of sandwich composite columns with initial face/core debond.
This research paper was attempting to find new ways to combat the issue with delamination, they outlined how a popular method for doing this is by Z-pinning which is a very “expensive” method although I don’t know what they classify as “very expensive”. They were trying to find a more cost-efficient way to do this, it’s called shear-key insertion.
This method involves the “insertion of shear-keys in grooves within the PVC foam core”. Although this experiment was done using PVC foam as a core and not honeycomb, I believe it won’t be too different? (correct me if I’m wrong) The following are pictures of the actual inserts and what the sandwich layup will look like:
The shear keys can be made out of virtually any material and they can be made of any shape and size (this is all conceptual). They used pre-manufactured semi-circular shear-keys made out of glass-fibre and epoxy resin. The way they made them in this research paper was “woven roving strands were placed in a mold which was designed based on the shape and size of the foam core groove. If we were to pursue this option, it would be worth making this mold out of something that may be used many times such as aluminum, not a one-off. The shear-keys pictured above were the roving strands vacuum infused with epoxy resin and hardener followed by curing at room temperature. This method can also be combined with other methods to make ultra-strong sandwich layups!
Stitching
As you can possibly suspect, this method is done by literally stitching the sandwich panels together using Kevlar thread. I think this method is super cool and interesting.
There are many different types of stitching that people have used but the most popular stitch style is a “modified lock stitch in which the knots linking needle and bobbin threads are formed at one surface of the laminate to minimize in-plane fibre distortion”. It is pictured below:
Although this method is used on a foam core, I don’t see why it wouldn’t work for a honeycomb core since we just have to thread the yarn through the holes in the Nomex? One issue I saw while reading is that they say “in order to minimize possible damage to composite sandwich panels, through-the-thickness stitching was performed on the sandwich panels before the resin application” as we use prepeg, this may be an issue since this type of stitching was used on a wet layup/resin infusion? I don’t see why it wouldn’t exactly work on a prepreg but I may be wrong in thinking that.
They used Kevlar 129 to stitch the layers together which I imagine won’t be cheap since they say Kevlar 129 is specifically strong? so it may be more expensive? idk. I genuinely have no idea how much Kevlar yarn actually costs.
The following is a picture of the machine they used to sew the yarn in:
Another issue that arises is that We will have to most likely make our own machine? Since conventional sewing machines cannot pierce through the sandwich layup.
Debulking
In this article by explore composites, he gives a full rundown of what he does for prepreg carbon parts and how he believes it should be done. This guy has been doing this for around 20 years and I believe in his judgement, he has experience with larger yacht-scale projects and smaller DIY projects.
He mentions that debulking is “absolutely necessary to get successful pre-preg parts” which we are missing with our layups. Although it is time-consuming, it gives many benefits to the quality of our parts.
What is debulking you may ask? It just means vacuum bagging the material to compact it and remove air from between the layers. You simply remove the backer on the last layer of pre-preg that you’ve laid up in the mold and then put a vacuum-promoting flow media on top and then, you vacuum bag it for about 10-30 minutes. Also, note that debulking works in cooler environments so the pre-preg resin is less runny.
A useful tool that I believe we should invest in is a spike roller. It is commonly used to perforate material before debulking. You just roll the spikes over the pre-preg, that’s it. Below I have linked a roller that’s specifically used for this just so you can get an idea.
https://www.aircraftspruce.ca/catalog/cmpages/vb9060porcupineroller.php
When debulking, it is also imperative that you put some sort of mesh over the pre-preg, he says that his material of choice is an infusion flow-mesh, this helps spread the vacuum pressure. He says that “breather with perforated film instead of flow media works ok” he says that a downside is that it leaves little breather fuzz and sometimes. it can be more work.
The debulking bag is a bag that you’ll have to take on and off over and over again. The “setup” is pictured below.
Even though this method doesn’t directly impact the bonding of Nomex to carbon fibre, the author of this website explained in his video that it does help, although I don’t know exactly how much. He does a full layup with debulking in the following video for whoever wants to watch it, I believe it contains a lot of valuable practices that we may be able to emulate in our own layups.
https://www.youtube.com/watch?v=LQQKapu7QWA&feature=emb_title
Conclusion
As you see, there aren’t many novel options to help work against delamination. I did manage to find a couple of solutions though. It’s not even like big companies use some different ways, I found a PDF on formula 1 composite manufacturing and they even just used an adhesive film to stick the carbon fibre prepreg to nomex. Other than using one of the above three options or the solution we got from Iowa state, I don’t see there being any other solutions other than getting a stronger epoxy. The shear-key insertion is very interesting and seems like the actual shear-keys don't cost much but, we would have to cut the Nomex in possibly weird ways to achieve this. The kevlar sewing is a very, very interesting solution but, I don’t know how much Kevlar yarn costs (I imagine it is expensive) and unless we want to do it by hand manually, we will most likely have to build some sort of machine to do this. Debulking, although not a direct solution to delamination should possibly be integrated into our work to improve the quality of our fabricated parts.
For Reference
The following is a picture of what Iowa State came up with