Help me plan out the process to build an FSAE body

General Composites Discussions Composites Talk
#1

Hey guys,

So I am pretty new to composites but after an internship and some experience I am the composites lead for my university’s FSAE team and my goal is to build the body among other components. I have been doing a lot of research and have basically put together all the steps to building our body, but was hoping you guys could give me any advice or let me know if I am doing something wrong. Our plan is to use 2" thick insulation foam to build a plug of the body, build a tooling part off that of fiberglass, and then use that to layup a carbon fiber/core body. Here is our procedure:

  1. Design the body in Solidworks with surfaces (currently in progress).
  2. Once we have the design, cut out cross sections of plywood with a waterjet or CNC router to use as guides to cut foam. (Any advice on how many of these we would need, one for every 2 inches of foam, 6 inches of foam, etc?)
  3. Purchase 2" thick insulation foam from a building supply store and hot wire cut the foam into the shapes using the wood templates.
  4. Use contact cement to bond the foam pieces together and then sand/hot wire the plug to the final shape.
  5. Attach the shape to a flat piece of foam to act as a flange.
  6. This is where I am unsure the next step, do we then directly apply tooling gelcoat to the foam and then layup fiberglass, or is there something else we put on the foam first before applying the gelcoat? PVA?
  7. Layup the fiberglass tool out of 1.5oz mat. (I was thinking roughly 6 plies, would we need to add reinforcement?)
  8. Remove the tool and trim the outside surface.
  9. Release the tool (should we sand it at all first?)
  10. Layup our body part on the tool and then vacuum infuse it.)

We want two of the molds to be used for two parts each, would that change how we build the molds at all? Is there anything I am missing here?

Any recommendations on tooling gelcoats?

Thanks for any input!

#2

For step 6, no - tooling gel needs to be applied to a smooth surface so the foam will need to be covered with something that will stop the foam soaking it up. Most likely this will be iso-pon/body filler for you. So cover the foam with body filler and then use something like the gelcoat. I wouldnt use the tooling gel however but would use this instead for example:
http://www.mbfg.co.uk/mould-preparation/duratec_904.html

Oh and be prepared for lots and lots of sanding with the body/filler!

#3

Sanding blows… if possible machine the pattern. Foam is fairly cheap but you can’t use polyesters with it unless you coat it and sanding the whole surface smooth is lots of labor. You could also use that EPS foam, cut the shape and then hot coat it with epoxy. Do some test for sure… make a few small pieces with some shape to it, go through all the steps and see what works and what doesn’t. I could write a bit more but, gotta run…

#4

I’ve made a number of one-five part molds for boat parts in a similar method. As Sammy mentioned insulation foam + vinyl or polyester is a no-no, this includes any bondo or car filler. Polystyrene foam is melted by the styrene in the resins. If you have access to a CNC machine I would recommend skipping the plug and going straight to a female mold.

  1. Design your part in Solidworks, then make a female mold model from this, making sure to pffset for the layers of fiberglass you will laminate on the machined foam.

  2. Slice the mold into 2" sections (measure the actual foam and slice to that thickness, might not be exactly 2") and add alignment holes. I like the 2" pink Foamular XPS foam they sell at home depot for $35 for a 4x8 sheet as it’s cheap and machines and sands much nicer than the white EPS but. Machining it in 2" thick pieces will mean a little more time putting the mold together but will save a ton of machine time.

  3. Once the pieces are machined a glued together there will be some sanding involved but it shouldn’t be too much (really depends on the machining parameters), just enough to remove machining ridges.

  4. Then hot coat with epoxy mixed with a small amount of micro balloons to seal and give a quick sand to smooth out any imperfections.

  5. At this point you can switch to poly or vinylester but I would just stay with epoxy. laminate a few layers of 3-4oz cloth over the foam. It you want to bag these layers on be aware that you can crush the foam with the vacuum if your bag is attached to the surface your foam is sitting on. If you make a large flange on the mold and bag on the mold itself not so much of an issue. If you stay with epoxy chop strand won’t be as good as the binder is designed to be broken down by the styrene. If you have a lot of complex curves look at a 4 harness satin weave. US composites has 3oz fiberglass 4HS that will conform to complex shapes nicely

  6. Top coat with your choice of gel coat (some polyester gelcoats wont stick to epoxy very well), epoxy primer or epoxy and graphite and sand until your happy, then you should probably sand some more. It can be helpful to make some custom sanding tools depending on your mold shape. They can be made quickly with 5 minute epoxy and fiber glass to match the curvature of your mold.

  7. Seal and mold release. Lots of options here, just make sure you release it well as the mold will be fine for a couple of parts but won’t hold up to a difficult release.

8.infuse

  1. Remove part. Once again a little care here is good as molds built this way aren’t the most robust but will be good for a small run.
#5

Thanks for the reply! I have axis to a 3axis CNC mill so this is certainly feasible. If I understand correctly, we would machine the female mold in foam, then slowly coat it starting with epoxy mixed with micro balloons, do I wait for that to cure? Then I basically reinforce the mold with fiberglass to be able to pull vacuum on it? And then finally a layer of gelcoat that will stay on the mold? After these three coats the surface will be rough because it is going to air cure so that’s what you mean by sanding?

Does the gel coat cure and then get sanded down and that is my tool surface for the infusion?

Also, it sounds like you say to machine each ~2" piece of foam first, then connect them together. Is there any problem with bonding them all and then just milling one big piece of foam?

Thanks for the input!

#6

For machining the pieces my suggestion about 2" slices is more about saving machine time than anything else. If you have access to unlimited machine time not so important.

I let the first coat of epoxy and micro balloons fully cure and then give a light sand. Sounds like you get the idea regarding the other steps

#7

If you can machine the entire piece at once it’s great. A 3 axis may or may not make this a pain, depending on geometry.

IF you just work in a flat 2d world, the cutting is a simple through cut. You can also machine in a registration hole that will fit a piece of Pvc pipe through it so they all line up nicely. Bonding them together is simple, just use the correct glue. I’ve use a water activated glue that is made for foam, works great, fills gaps. If you use an epoxy or some hard glue, harder than the foam, you’ll have a tougher time sanding down the glue and not griding the foam. Use similiar hardness glues is what I’m saying.

IF you machine it that way, there shouldn’t be too much finish work. As kilojoule said, make up a paste of epoxy for the hot coat, put it on as nice as possible. I’d probably add talk to it to make it easier to sand and come out smoother. THen you can either go with that, or I’ve used Durakotes red gel coat sprayed for the final coat. This one should go on nice and even and when cured will be a pretty nice gloss. After A quick buffing maybe going through two or three grits of polishing compount, you’ll have a nice gloss finish that’s pretty durable. Enough so that you could use this as the mold or as the master to make a mold. Though a mold of this sort is low temp, and fairly fragile.

#8

Great Point!

#9

Check out this Project thread. It outlines everything you are trying to do.
http://www.instructables.com/id/[Video]-Large-Scale-Mold-Making---Part-I/

#10

Hey Kilojoule, which University do you attend?

I am on the CSU, Chico team and we finished our nosecone a few months ago (AL sheet side panels) and went with a similar process that you have explained in your original post. If you haven’t started already, prepare for a lot of bondo dust… everywhere.

We did not have access to a large CNC router or water jet for our foam, so we did all of our shaping by hand which took a long time. Make sure your surfaces are as smooth as possible because everything shows up on your part and is a pain in the ass to fix later. Sand, polish and wax both the plug and the tool to a mirror finish before you apply resin and use some kind of mold release.

We used 3 layers of 3k 2x2 Twill Fiberglass and it seemed to be stiff enough everywhere except on long flat sections of the nosecone. On the flexible sections, we used a 1/8" particle board core and layed up another layer of glass on top of it and it stiffened it up a lot.

Here is a link to a picture of our car a few months back (https://drive.google.com/file/d/0B8fgGxXY5aJyS3NYa1lhMElwRVowSEc2Zy1kVEVveHBsaXMw/edit?usp=sharing). It definitely isn’t perfect and needed a lot of work after we removed it from the tool, but it works and looks pretty badass for a first year car.

We will be attending the Lincoln competition this year. Maybe we will see you there?

Happy sanding,
Josh

#11

Kilojoule,

I thought I posted a long reply to this (putting off writing a composites paper) but apparently it didn’t get posted.

Anyway, I am a FSAE member at CSU, Chico and we finished our nosecone a few months ago. It was our first time working with composites on such a large scale but it turned out nicely.

I looked over your process in the original post and it’s similar to what we did, with the changes that were suggested in some of the replys from other users. One word of advice is to just make sure you sand, polish and wax your surfaces before you lay up the tool and part. It will save you headaches in the future. Here is a picture of our car about a month ago with the nosecone and body panels on (https://drive.google.com/file/d/0B8fgGxXY5aJyS3NYa1lhMElwRVowSEc2Zy1kVEVveHBsaXMw/edit?usp=sharing)

Post pictures on how the panels are coming along, we’d love to see what the other cars are looking like.

Happy sanding,
Josh Miranda

P.S.: We are attending Lincoln this year, hope to see you there!

#12

jmiranda8, I go to Washington State University, thanks for the advice I will definitely post pictures!

We have adjusted our process somewhat substantially, and I have one question I was hoping someone could answer. We are now gonig to make a plug, build a tool off it, and then infuse off the tool for our part. For the tool we were gonig to use fiberglass and it seems like everyone does, but we’re in a weird situation where we have a lot of carbon fiber and not a lot of fiberglass. Would there be any problems using a 2x2twill 6k carbon fiber to build the tools with?

Anyways, our new process is going to be as follows (although we’ll see how it goes haha):

  1. Buy 2" thick insulation foam from the hardware store and then cut it in to rectangles we then glue with contact cement together to make boxes with.
  2. CNC the foam boxes to the male mold pattern (we have a small mill so we will have to do it in sections, then glue and biscuit joint them together.)
  3. Sand down the foam post CNC and then apply a layer of Duratec Grey Surfacing Primer, and sand that down really well, probably up to ~1200 - 2000 grit.
  4. Release the Duratec surface.
  5. Apply a coat of PTM&W PT1105 to the duratec surface, and when this is in the fingerprint stage, meaning touching it leaves a fingerprint but none of it comes off, laminate it with USComposites 635 medium epoxy and fiberglass or carbon fiber.
  6. Add support to the tool.
  7. Sand the tool surface to ~1200 - 2000 grit and then seal and release it.
  8. Infuse the part with Carbon Fiber, core, and PTM&W PT5712 on the tool, and then cure at 120F overnight.
  9. Demold the part and post cure.
  10. Sand the final part to a desirable surface finish/paint it.

Questions:
a. Can we laminate the tool with carbon fiber instead of fiberglass? We would need to buy fiberglass to do it, but we have lots of carbon fiber.
b. People have recommended polishing the surfaces, but that sounds like it would involve buying buffing pads and sanding compound which could be like ~$100. Is sanding it to ~1200 - 2000 grit not enough?
c. We are going to seal and release parts with Frekote FMS and 770NC, do we need to seal the plug with Duratec or just release that, and then seal and release the tool?
d. One of my biggest questions, how should we reinforce the tool? I was thinking we could cut plywood pieces the shape of the female tool and then either laminate them on while we are laminating the tool, or laminate them on after the tool has cured. Would that be sufficient? Is it over kill? What do you guys recommend?

#13

Be sure to seal the foam with epoxy before spreading with duratec as it’s vinyl ester and will melt the foam

#14

a) probably yes. I use carbon all the time for infusions. IF it’s unidirectional, it’s a little different, if it’s fabric then it should be the same.

b) Buffing is easier than sanding. You just need a buffer, which I"m sure you can get for little or borrow. THen you need a few pads, which are like $20 maybe and the buffing compounds. It’s pretty fast and gives very gratifying results. Again, do a test panel to see what your fresh cured part looks like and then buff it to see how it looks after just a buffing.

c) So the plug. You make the foam shape, seal it; either hot coat it wish epoxy or use an appropriate sealer. Spray with grey duratec and sand. I’ve used the orange duratec at this stage as a final coat which polishes to a beautiful shine. Then release the plug with freekote, apply surface coat, make the mold using fiber, epoxy and core. Pull mold, fix any issues of voids etc. Polish with buffing compound. Seal and release. Parts are not sealed or released, the mold is.

D) you should always reinforce the mold! You can use core while building it. Sometimes I"ll laminate in pieces of cardboard tube cut in half, or pieces of foam, to add dimension and strength. The mold should be thicker than the part. Also you can cut piece of ply wood, foam, etc, as you mention to add to the back side so you have a flat bottom so the tool will sit nicely while you laminate. This backup structure add strength as well. The backup structure may be optional if you think it’s too much, or would be too much extra depending on time, materials available. If you’r only going to make a few parts, then you could probably just leave out the structure but still add core to increase rigidity. Especially for the post curing.

Also if you’re planning to post cure the part in the mold, be sure to post cure the mold to a higher temp than the cures will be done at. Leave the tool on the plug if you can while you post cure it.

#15

are you actively pursuing sponsors? Precision board is a good material to use for a direct cut mold for very limited use and loose tolerances. They do a lot of sponsorship programs for college projects and are familiar with FSAE.

#16

check out these guys Harbor Freight, they sell buffers for cheap. Not the best quality but plenty good for a light duty job like yours.
http://www.harborfreight.com/power-tools/polishers/sanderpolisher-7-variable-speed-60626.html