Jump to content

Fibreglass, Carbon Fibre, Paper Mache - The composites chat thread


Recommended Posts


I've been interested in learning some techniques to produce some parts with composite materials.
I had a look at some existing threads but there were just scattered snippets of stuff here and there... So here we are! 
If nothing else, this will serve as my soapbox for blathering about various experiments and results I've had trying different techniques. 
But I'd love to hear about other peoples projects and experiences and so on.
So dont be shy, tell us all about some cool stuff that you've made, or otherwise varying degrees of success... good or bad. 

  • Like 5
Link to comment
Share on other sites

  • Replies 306
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Okay soooo here's something shitty I made, the first time I tried anything with Fibreglass.

A long time ago I had quad throttles on my engine, and I wanted to build an airbox for them. 

I spoke with a friend who worked with fibreglass a lot, and he suggested building a frame to the basic shape. Fill it with expanding foam. 
Then take the frame off, cut and sand the foam to shape.
Then cover with the brown packing tape stuff, and just wet layup fibreglass over the top...

Then pull the foam out. 

So, what I found here was... Fuck. This was so much work and the results were yuck hahaha. 
I didnt have very good bonding between layers, my inside layer of the airbox had a lot of spikey bits which I was concerned may break off. And it took a whole lot of sanding and finishing to get it looking okay-ish. 

It did the job but it was a giant ball buster for a low quality result. 


I ended up having a big paddy and smashing it to pieces with a hammer. Which was the most enjoyable part of this process. haha

  • Like 6
Link to comment
Share on other sites

So fast forward a few years to now-ish. 

And I've always wanted to do something more with fibreglass etc, but I didnt want to go through that same shitty process with making a mold. Which was 99% of the work.

Around about the same time, start reading up about 3d printers, and thinking this could be a great way for not only making an inside mold - but an outside one too so you can compress a shape rather than just blobbing layers together and praying. 

So after a million hours getting printer working nicely, printed some molds for a test piece. 




Compared to my historical attempt with an airbox this couldnt be considered anything apart from hideously successful! 

I used a straight out of printer mold, some carnuba wax as a release agent, (way too much) 3 or 4 layers of wet layup carbon then put a heavy bucket on top of the mold while it cured. 

It trimmed up nicely and had a cool "dry" look to the carbon, I think because I used so much wax that it made all of the epoxy not stick to that outer layer. But it turned out bloody strong. 

  • Like 7
Link to comment
Share on other sites



I'm currently in the process of turning these into science. Pultruded flat bar.


One batch has no filler, the rest have filler ranging from 5 phr (parts per hundred resin) to 30 phr and different particle sizes. 250 short beam tests over the last week and a half have resulted in a pile of somewhat meaningless data.

Here's what they look like up close (optical microscope, really poor image compared to with SEM). Getting onto the SEM tomorrow to analyse damage evolution which should be quite fun.

You can see the process of polishing the samples has dislodged a lot of the filler particles in this image, they are from a population with a median particle diameter of 2 micron.


One of the issues with using large (median bigger than about 6 micron) particles as a filler is that they don't really fit into the gaps between the fibers (which in this product are about 23 micron diameter) when there is a high fiber content and you start to get micro-crimping of the glass. This reduces the tensile strength of the product significantly.

When it comes to the short beam (mainly interlaminar shear) strength of the composites the fillers also play a significant role in the performance of the parts. The presence of fillers appears to aid in crack arresting, but they also seem to increase the ease with which cracks initiate. Up to a loading of about 20 phr you will retain a significant amount of the short beam strength, above this it seems to drop off significantly (increase in crack initiation ease seems to be outgrowing crack arresting ability).


Another thing is that with fillers there seems to be a critical filler loading rate at which you begin to get significant voids in the product. This loading rate depends on the particle size distributions. I think that the reason for this is that when you have enough particles in a small space they will try to close-pack around each other instead of being surrounded by resin. When the fillers are close packed (calcium carbonate in these samples) the filler has a bulk density of about 1 g/cm^3, while calcium carbonate has a density of approximately 2.7 g/cm^3 in solid block form. So basically, the particles tend to agglomerate and exclude the resin from within.

This can be observed when you compare density of the product to filler content (without close-packing and exclusion you would expect the density to increase with increasing filler loading). For the filler with a median particle diameter of 2 micron there is a peak in density at about 15 phr.



Short beam strength and tensile strength follow the same trend as density. Basically, adding a little filler to your resin will help reduce voids and provide crack barriers (increasing strength and density). Adding too much filler will result in the formation of voids, making it easier for cracks to initiate. You *may* be able to get around this to a degree by milling the filler particles into the resin and using a high pressure forming/curing process but you will still experience a decrease in properties with increasing filler content.


This is with a glass content of about 60% by volume.


Another method for improving interlaminar shear strength is to just remove some fibers, giving the matrix more volume - allowing it to deform and absorb damage better.


A good measure for the quality of the product is its stiffness.  Test results seem to indicate a strong correlation between initial stiffness (force/deformation in linear portion of response) and ultimate short beam strength. Tensile stiffness doesn't seem to be affected by the filler (I will have to analyse the data though to confirm this) as tensile properties are mainly dominated by the fibers. However, using the incorrect filler WILL drastically affect the tensile strength of the product.





Other things to note about fillers is that the size of the particles also has an effect on the process robustness when manufacturing stuff with pultrusion or extrusion. Larger particles cause fiber crimping as described above, this can have both positive and negative effects on processing. The positive effects are that you get a lot more scraping occurring within the die, which removes any build up (scale) and will prevent the product from having a crappy scaly surface or even locking up in the die. You will also be able to get a nicer looking product due to being able to remove more glass from the product (resin is nice and shiny). Negatives are that you probably won't be able to put as much glass into the product (so less strength) and that colours of the product will be affected more significantly (if you are using pigments).

You get the same sort of issues if you use a filler with really small particles. For example when using a filler with a supposed median particle diameter of 1 micron, the particles tend to agglomerate into large clusters with median diameters of about 100 micron. This causes very erratic product properties and the strength varies significantly throughout the product, causing unpredictable failure loads and modes.


A median particle diameter of about 2 micron seems to be rather benign in that it doesn't really cause many negative effects. However you don't get the die scraping effect with the 2 micron filler, so the process loses a lot of robustness - you can only operate within a narrow band of glass content, you'll either scale the product or lock the die. Mixing some 2 micron filler with 5 micron filler seems to give an optimal combination of robustness and product quality.

UV stability also seems to be controlled by filler type and loading rates, but a discussion of this will take up a whole load of space and time to write.

In short: You can add a little bit of filler to your products to save money on resin without compromising (possibly even improving) properties. Add too much and your product will turn to shit. Don't use huge filler particles.

/Random gibberish that is not really of any use to anyone.

Bonus graph:





  • Like 7
Link to comment
Share on other sites

So next thing to try here was something a bit bigger. One problem is that with my small mold I was using a lot of infill plastic. So it took ages to print and used up a lot. 
Also, I was running out of plastic.
So my next iteration was just to print with thicker outer walls and a hollow center. 

I pretty quickly learned that the mold was too flimsy and the result was rubbish because it couldnt squash the layers together.

Things learned: 

Mold rigidity is important 
Squishing things together is important 
Still too much wax as mold release

So I thought about filling it with something, tried plaster of paris. 
Great idea which made the mold massively more rigid and non compressible. 
But, the weight of the plaster bulged my mold and so it was now unusable. Pooz. But it was now strong enough that I could stand on it. Cool.

Sooooo after I got some more plastic, printed another mold with some rectangular infill patterns. So making some cells that I could fill with plaster and these would also hold the shape.


I used carbon again, wet layup of each layer, a bit less wax, and still a bucket on top to squish them together. 

The surface finished looked a bit crappy as there were air bubbles. Also with much less wax, it retained more epoxy on outer layer and it inherited a lot of the detail from the mold (aka printing lines) 

But with some work it turned out nice. Glued it to the hinge part of my glovebox lid and now I've got a new glovebox lid. (since painted satin black)


Things learned: 
Mold needs to be smooth if you want a nice finish
Needs to squeeze it better to get air bubbles out
Could still use even less wax
Mold is strong and 100% reusable with no damage
Dimensionally accurate and holds shape awesomely

  • Like 3
Link to comment
Share on other sites

Now with a good mold (But no carbon left) I can experiment with different techniques to see what gives a better looking result. 

I sanded the print lines out of the mold, and wiped off most of the wax. 

This time used fibreglass and clamped it all together with what I had available.


This squished a shitload of the epoxy out which was awesome, and my mold is rigid enough that it could handle it. 

Also before the mold was completely cured, pulled the part out. Trimmed edges with scissors. Then put back in mold and clamped back on to cure to final shape. 


Turned out great! 


Things learned: 
Smooth Mold = way better looking part 
Clamping made an awesome difference 
Need more clamps 
Trimming part while still "soft" is a lot easier than needing to use bandsaw later

  • Like 2
Link to comment
Share on other sites

2 minutes ago, ajg193 said:

Have you considered getting a vacuum pump and some bags?


If you do that then you won't need to put any weights on it as you will have 1 bar of clamping pressure all around. It will also help with sucking out the air bubbles.

I'm starting to lean towards this! 

Definitely keen to give it a go.

Link to comment
Share on other sites

2 hours ago, The Bronze said:

This thread seems like a good place for this link: https://www.maxbetteridgedesign.com/building-the-flat-white

Really love how this looks and quite interesting build thread including some shots of his mould making for granite fibre moulding.



Damn that was so cool! 

I love the slices of the MDF stuff that he's used to build the shapes from CAD. That is an exceptionally awesome idea.

  • Like 1
Link to comment
Share on other sites

1 minute ago, h4nd said:

I'm slightly surprised the CaFi got painted satin black. That's extra weight, yeah?

Also, titanium fasteners go well with composites? Could one leave fibres unepoxied in a band to act as a hinge?

Yes, but it's still half the weight of a standard glovebox lid :lol:

Titanium fasteners are cool buuuttttt one of my other "Things" about my car is I want to be as servicable as possible, and part of that is reducing toolset needed to work on it.
I've got no cap screws for this reason (except where unavoidable) So I can do pretty much everything with 10,12,14,17,19,21mm sockets or spanners apart from a few necessary exceptions like headbolts and so on. (Which if I'm needing to do something with headbolts, I'm in deep trouble already!)

  • Like 1
Link to comment
Share on other sites

3 hours ago, NickJ said:

I second vacuum infusion, far better results, getting a solid vacuum and even infusion can be infuriating, but when it works, :):):) 

Between my skin sensitising to the resin and hating dust, i'm not sure why I like working with it.

Do you have any pics of some of the things you've done? 

Link to comment
Share on other sites

30 minutes ago, Roman said:

Do you have any pics of some of the things you've done? 

Would love to share what we do in pictures, sadly i'd lose my job! 

We design/build alot of composite parts for the cryogenic industry, while they're often simple to look at, some of the materials used have interesting properties as to not tear apart from differential contraction and what not at 4K

  • Like 3
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Create New...