New Carbon Fibre Access Hatch

So, I’m replacing my mild steel access hatch with a carbon fibre hatch, and here are some pictures. I’ve also done a detailed cost analysis, and included 20% for wastage/accuracy as well.

It’s a 5 layer part and the cost for materials (on the open market) is £27.89 inc vat, and it doesn’t include my time.

Total weight saved – 500g.

Layer Specification Cost
1 – Carbon Fibre 2/2 Twill 200g £3.20
2 – eGlass 2/2 Twill 200g £0.54
3 – closed-cell foam ribs 3mm £0.10
4 – Carbon Reinforcement 2/2 Twill 12k 450g £2.98
5 – Soric Hard Point 3mm £1.33
6 – eGlass 2/2 Twill 200g £0.54
7 – Carbon Fibre 2/2 Twill 200g £3.20
8 – Peel Ply £1.26
9 – Infusion Spiral £0.25
10 – Infusion Mesh £0.72
11 – resin Tube £1.19
12 – Gum Tape £1.32
13 – Resin and Hardner 300g £6.63
14 – Labour 6 hours

Here are the inevitable pictures:


This is the part laid up before infusion





Vacuum bagged ready for infusion – not quite full vac yet.





full vac, ready to go





300g of resin with a 50/50 mix of fast and slow hardener.





This is the part out of the mound (well, a piece of double glazing window – gives a nice smooth finish).




clampsclamped to the original footwell access hatch for drilling. Note the bolts to keep it all in place when I moved the wood around. You need to drill through into another medium else you may chip it as the drill goes through.




wrapped in gaffer tape and being drilled. It was necessary to be sure the wood or the clamps or the metal face didn’t scratch the part. Carbon fibre manufacturing does seem to involve a lot of disposable materials, and gaffer tape is just one thing of many.



finished part, with ribs for extra pleasure or stiffness. This is the side you’re looking at here is the peel-ply side, so it’s rough in finish.



IMG_0034_2Finished cosmetic side. Pity this side is the one that goes down into the foot-well.



Lessons Learned

  1. I should have used washers when bolting the metal side to the part for cross-drilling. The nuts dug into the carbon a little, leaving some slight indentations which you can see in the picture above. No big deal though because the areas that have been damaged will be covered over by a gasket when the part goes in the car.
  2. I would have been better off making the ribs by cutting a cross shape out of closed-cell foam rather than 4 individual ribs. It would have been much easier to align it all and I could have made a larger center point which may reduce stress.
  3. It took me 6 hours to make this part – I could have saved an hour with getting the bagging right (I have a better technique now) and I didn’t make it from templates because I don’t see how I need to repeat it. If I need to remake it for any reason, I could get it down to 3 hours I reckon.

Mould Making Course write-up

So, a week or so ago, I attended the Carbon Copies Mould Making Course, and I thought I’d put in a write-up because I was so impressed. I decided to go on the course because I was soon going beyond the flat-panel CF work I was doing on this fury and thought … £180 of tuition will take me way further than multiple days of bumbling around in the workshop and getting it wrong.
The course is organised by Warren (owner and self-admitted carbon-fibre and composite addict). You get there at 9 and brace yourself for a full-on day of mould-making, part-taking, tooling, techniques, tips of the trade and general how-to-do-it, gaining advantage of all of his ex-aerospace years of carbon-fibre work.

The workshop is in an old cloth mill in Crescent St, Todmorden which caries a certain irony. The training room upstairs is massive, and on the walls are lots of parts Warren’s made, and lots of moulds on benches for you to see just how (for instance) you would make a cycle wing. You get a tour of the techniques when you want. Lunch and infinite cups of tea and coffee are included and not skimped on. This is an important point because Warren was willing to go on as long as I was and I needed to leave at 7pm 🙂 On the way you have a right giggle and plenty of time for reflection and questions. In fact, Warren encourages interruptions and wants you to be totally sure you’ve got something before he moves on and also sure that alternative approaches (which he encourages) are discussed.

The plan for the day is to make a mould and take it home to use it. As such he asks you to bring a pattern (or for our US cousins: a buck) that you want to make a mould from and you can do that during the day. You actually get something concrete in your hands at the finish.

The course doesn’t just go into this though, because you’d not get the mould made and set in the time for the course, so he does it a little backwards to squeeze the most out. After a general discussion, much cooing at the lovely things hanging on the wall and another cup of tea, you make a mould from one of Warren’s existing NACA ducts. This means you can do the “how to finish the mould” bit on an existing pattern before you then spend more time making your pattern.

We used the Easy Composites epoxy mould making kit for this because it’s quick and Warren reckons it’s the best for the job. I should point out that neither he nor I are associated with these guys. They just supply good stuff to their own formulas. Warren also goes into how to make moulds using polyester as well, and that crops up later when you’re starting out making yours.

Warren took us through the mould-release process, and then guided us as we mixed, kneaded, painted and dabbed the stuff on in the right order (and you’re encouraged to use plenty-lots) and let it set. Out it pops and there you have it, a mould. We went through the issues we may get, and the slight imperfections we may create, and lots of time was spent discussing mould-repair and maintenance and how to put the imperfections right. This was a good morning spent.

Here is the mould from which I made a negative mould.





Afternoon was spent discussing the polyester resin system, barriers and finish, etc so that you can make a mould with the cheap stuff and still take an epoxy part from it. He also discussed the merits of using kingspan bonded together with bog for forming because it’s lots cheaper than the modelling board that is used for cnc machining. Barriers discussed were high-build primer, 2k clear coat and gel-coat.

I bonded my bits together, and Warren had a lot to say about my mixing technique but I got there in the end. After bonding multiple layers (I was making a dash-pod for the new dash I need to make), I carved them into rough shape with a knife, and then we got to sanding the right shape in place. Once done, layers of “whatever cloth scraps you have lying around” are used to cover the shape, which is set with polyester resin before bog is applied and sanded smooth. Even with bog application there was advice about how to apply for the least sanding effort. I will not be a good cake icing person.

By the time the part is sanded to the right shape, the day is about over, but that’s the point where you can apply the techniques from the morning at home yourself to finish off.

All in, a great day, a good long course, value for money and a great laugh. It’s a treat to see just how much Warren knows, and that he shares it all and really wants you to leave with a mould and pattern in your hands that will be genuinely useful for you.

Definitely worth it.

First Carbon Fibre part

8So, this is my first carbon-fibre part. It is a resin-infusion aramid/foam/glass/carbon part designed to replace a aluminium floor panel.  Unlike the aluminium part which was riveted in, the composite part will be bonded in for good. This will reduce weight and increase rigidity in the chassis. All in all, I’m really pleased with the result, and the pictures below will show how I made it, and how strong it is.

The layup of the part is:

What Specification
Aramid 2/2 Twill Weave 300g
Closed Cell Foam 10mm
E-Glass 2/2 Twill 200g
Carbon Fibre 2/2 Twill 12k 450g
Epoxy Resin Normal, Slow

As I said in the previous post, there was a journey finding the right medium to lay the part out on, and found a double-glazing window worked a treat.


Now that the parts are laid on the glass and vacuum bagged, I infused 200g of resin through. Sure enough, this wasn’t enough, so I stopped the infusion process, added another 100g of resin and completed the process. The infusion process slowed down towards the end, and the far-right corner was the last to fill. Just as the corner filled, I had the input clamp tight, so the corner filled from resin already in the part. I don’t think I had too much resin in there.

3One fully infused, vacuum bagged part, sitting in the sun curing. I was lucky in having to work away from home for three days when the UK weather was the best its been for a while so the part got a proper baking. For slow resin, 24 hours is a minimum before demoulding, and 36 is better. I got 72, with three hot blasts of the mid-day sun.


4Here’s the part popped from the mound. The easy-lease compound (4 applications, as per the instructions) worked a treat and the part popped off the glass without a fight. You can see the aramid (yellow), which is the underside part of the car, facing the road surface. Aramid gives brilliant impact protection. It’s also incredibly strong, and a little more giving than carbon-fibre on its own.

5Here’s the side of the part that faces into the footwell. You can see the infusion spiral and resin bung still on the part because I haven’t removed the peel-ply yet that this lot is all stuck to.




Lo and behold it came to pass that the part was peeled and trimmed.




9Here’s the finished part, sat on two tins of soup, with my not inconsiderable self balancing on top of it. Note the lack of deflection.



The original aluminium part weighed 960g, and this part weighs 524g. That’s a 55% saving in weight.

Happy with this.


This wasn’t the easiest of starts, and I had some good technical support from Matt at Easy Composites and lots and lots of moral support, texts, phone calls, tea and training from Warren at Carbon Copies.

Glass is porus

So, I tried laying up my first part at the weekend, and got nowhere fast. lots of slow, unidentifiable leaks. I tried again and again, and couldn’t get my bagging film to seal on the glass.

In these times, if you’re lucky enough to have a mentor (Warren, Carbon Copies Ltd), that’s when you ring them. His first suggestion is that laminated, toughened glass can be porous. So, I liberated a mirror and did a small lay-up on that. Then I started struggling with the final seam on the bag. There’s always a bit of slack, and it ends up as a crease (in which vacuum escapes). So, he suggested not going into a corner (which was my default behaviour, not design), but to finish on the seam nearest you in the middle, and to put a bit of extra gum-tape on the bagging film where you know it will meet the gum tape you’ve laid down. Ta-daa – where the seam appears, you already have tape, and can treat it just like a pleat.

So, I pulled full vacuum, and left it overnight. it’s still there today. I suspect I have one more leak in my system to eradicate. I patched it last night, and removed the patch today. if the vacuum drops then I know this is an issue, and have a solution to fettle it.

I feel quite pleased – Warren did say to me “well, you’re at the point when people give up, plough through”. That I did.