Composites 101: Carbon Fiber, Fiberglass, etc.

Discussion in 'Some Assembly Required' started by HickOnACrick, Jul 14, 2011.

  1. HickOnACrick

    HickOnACrick Groovinator

    Joined:
    Dec 5, 2007
    Oddometer:
    2,881
    Location:
    captures.crunching.farewell
    Welcome aboard - please feel free to post any knowledge you gain building whatever part(s) you build.
    #41
  2. HickOnACrick

    HickOnACrick Groovinator

    Joined:
    Dec 5, 2007
    Oddometer:
    2,881
    Location:
    captures.crunching.farewell
    Vacuum bagging - kinda sounds like the latest West Hollywood fad, but for the DIY Compositeer, it is a method to creating a strong, lightweight, and cosmetically appealing composite part. More information and another video here.

    Some background information and terms:

    Vacuum pumps generate vacuum, plain and simple. They vary in price from about $100 dollars to hundreds of dollars for a non-industrial pump. In addition to the pump, one also needs the hose, brass fittings, and the connection for the bag. I bought all of the above (with the exception of the connector) at harbor freight for a fraction of the price, and have been satisfied so far. One can also use an air compressor to generate vacuum (Venturi principle), but I have no experience with this method.

    Peel Ply is a cloth-like substance that will not adhere to epoxy or polyester resin. It is used to separate the layer of CF and resin from the bleeder/breather cloth. It is disposable.

    Bleeder/breather cloth is a thick, spongy, fabric that is used to suck up and hold excess resin during the vacuum bagging process. It is also disposable.

    Sealant tape is a very tacky, double-sided, elastic tape used to tape the vacuum bag film to the mold. It is also disposed of after a single use. It is best to buy tape in the winter. In the summer, the tapes will melt together when shipped, especially when one buys multiple rolls. The best thing is to find a local supplier so you know your money isn't wasted on gummed-up tape.

    Vacuum bag film is a stretchy elastic film that actually contains the vacuum within a specified area. If you are careful, this can be re-used a few times. Be advised that different films are designed for different resins. An epoxy resin uses a different film than a polyester resin.

    Core mat is a marginally thick, very lightweight, polyester fabric that can be used to add thickness and stiffness to a composite part. It is sandwiched between layers of CF, Kevlar, or fiberglass as the piece is layered. I don't think it contributes to the overall strength of the composite part, but since most non-aeronautical parts don't need the optimal strength composites offer, one can use core mat to add stiffness and thickness, for a fraction of the cost. By using core mat, I can make parts with 4 layers of CF, that would normally take 8-10 layers to get the same thickness and stiffness. There are a variety of core materials than can be used depending on the function you need from your final part.

    Here is a project I did this afternoon:

    I began by applying wax to my mold:
    [​IMG]

    Using a HVLP air gun, I applied 3 layers of PVA. While that was drying, and cooling down next to the air-conditioner, I readied my supplies:

    1. 3-4 pairs of latex gloves
    2. 600 mL of polyester resin, separated into 200 mL aliquots, with MEKP pre-measured in a syringe
    3. 4 layers of CF. I used paper to create a pattern pf the mold, so I would not use excess CF.
    4. 2 layers of core mat, again cut according to the pattern.
    5. 1 layer of peel ply, cut to the pattern
    6. 3 layers of bleeder/breather fabric, cut to the pattern
    7. Sealant tape applied to the mold, with the paper backing still adhered to the up-side.
    8. 1 layer of vacuum bag film cut to the shape of the perimeter of the mold.
    9. Make sure the vacuum pump is working, the hoses are leak-free, and the connector is clean and ready to use.

    [​IMG]

    Some pitfalls to avoid:

    1. Cut the peel ply 2-3 inches wider than the perimeter of the CF
    2. Cut the bleeder/breather fabric the same size or smaller than the CF
    [​IMG]
    3. If you need to layer the core mat over complex shapes, it may need to be cut in a way to facilitate this, as it does not drape over curves easily.
    [​IMG]
    4. Triple or quadruple glove before you begin so as one set of gloves gets sticky, you just pull them off and your next pair is already on your hands.
    5. Pre-measure the resin and MEKP into small aliquots, especially when the ambient temperature is above 70-degrees. When it's hot, the pot-life of the resin/hardener is greatly diminished.

    The process is:
    1. after adding the MEKP to the resin, quickly apply a layer to the PVA-ed mold
    2. Add your first layer of CF - this is the cosmetically critical layer - try and make it as neat as possible, use a chip brush to push the resin into the CF - be quick about it, but this is the layer that demands a bit more time and attention
    3. Add the second layer of CF. Add more resin and dab on more resin with the chip brush. Quickly. Mix more resin/MEKP.
    4. Now I add a layer of core mat, with more resin, mix more resin/MEKP
    5. One more layer of CF and resin
    6. Another layer of core mat, with resin, mix more resin/MEKP
    7. Last layer of CF, with resin

    8. Cover with peel ply - you want all the CF covered with peel ply, as close to the sealant tape as possible. If you can avoid folds and wrinkles, the inside of your part will look nicer, but many times this is unavoidable.
    9. Add the layers of bleeder/breather fabric. The B/B fabric should not extend beyond the perimeter of the peel-ply otherwise it will stick to the CF and make a mess later. Another pearl of wisdom: when the vacuum connector is added, it should sit atop 4-6 layers of B/B fabric (just use small pieces), and their should be a pathway of B/B fabric all the way to the main part, so the air has a pathway through the B/B fabric, to the vacuum. Piling up layers of B/B under the vacuum connector will prevent excess resin from entering your vacuum hose and destroying your vacuum.

    In the photo below, directly under the vacuum connector are 6 layers of B/B fabric that are about 4 inches square. The remainder constitutes 2-3 layer of B/B fabric that were cut to the template of CF.

    [​IMG]

    10. Remove the backing from the sealant tape, cover the whole thing with vacuum bag film, attach your vacuum connector, and start your vacuum. You will likely have some leaks that need addressed around the sealant tape, and often around the connector as well. Use extra sealant tape to fix the leaks. Optimally, you want the vacuum applied before the resin begins to gel, so excess resin (and therefore excess weight) are sucked into the B/B fabric. Speed is of the essence.

    I keep the part under vacuum overnight, then let it cure for another day before removing from the mold.
    #42
  3. twodollardoug

    twodollardoug the calimesa kid

    Joined:
    Jan 15, 2009
    Oddometer:
    337
    Location:
    southern california (IE)
    great write up on vacuum bagging hick. it looks like i could almost tackle it:rofl. i my try the venturie instead of the pump. i have a huge 2 stage 5 hp 240v. compressor. and no dough for a pump.......but that is another day.

    im getting started on my plugs, but before i go any further......is it cool to use 2 plugs for a single mold?

    in case anybody is wondering what these are.....they are rain hats for dual weber idf 40/44 carburetors.
    [​IMG]
    #43
  4. Orangecicle

    Orangecicle On a "Quest" Supporter

    Joined:
    Apr 24, 2007
    Oddometer:
    5,175
    Location:
    West Des Moines, IA
    I have an odd question. Maybe someone here knows the answer. I made a mold yesterday from a painted part. The part was a buildup of various materials and then primed. The part was properly waxed and covered with PVA. I surrounded the part with standard non-hardening, oil-based modeling clay and then painted on orange tooling gelcoat.

    First, the gelcoat slides off of the clay. Is there a better solution?

    Second, the paint on my model part deteriorated during the molding process and stuck to the gelcoat. I was able to peel away the bad paint, but the gelcoat was not perfect because of this. I'm not sure what was reacting to cause the paint to do that. It's a chemical reaction causing the problem and not just related to heat.

    Thanks for any help.
    #44
  5. tmotten

    tmotten Lefthand ride Dutchy

    Joined:
    Jun 28, 2006
    Oddometer:
    2,705
    Location:
    Calgary
    I had the same thing. The plasticine I used for the flanging caused alligatoring on the gelcoat. I 'lined' the plasticine with playing cards after that which solved it. But I use 3M spray body filler to fill the minor imperfections on my plug and this also reacted (not everywhere though) with the gelcoat the same way the paint with Orangecicle did.

    I was hoping that PVA would create enough of a barrier but it sounds like it may not. Interested in the expert oppinions here. Dealing with product compatibility is turning into a real headache.
    #45
  6. HickOnACrick

    HickOnACrick Groovinator

    Joined:
    Dec 5, 2007
    Oddometer:
    2,881
    Location:
    captures.crunching.farewell
    After wasting many hours and dollars with gelcoat(s), I have thrown in the towel. I got fed up with the smell, the cost, the way it melts everything it touches, it's tendency to crack when drying, etc, etc, etc.

    I now just use West System Epoxy and fiberglass to make my molds. I crank out molds in a fraction of the time, and overall, I am saving money because I have much less waste. The epoxy doesn't melt everything it touches so it opens the door for a variety of media for making plugs.

    I LOVE the idea of playing cards...that is brilliant and will find its way into a future project for me.
    #46
  7. tmotten

    tmotten Lefthand ride Dutchy

    Joined:
    Jun 28, 2006
    Oddometer:
    2,705
    Location:
    Calgary
    No worries. Wish I could say it's mine, but found it suggested somewhere else on a random composites forum. Forums are everywhere. Doing flanging on complex curves sucks, so although it's difficult to cut the card in the exact curve it's close enough. The overlap between cards can serve as location dowels a bit as well.

    I've thought about ditching gelcoat as well, but was suggested not to because pinholes may be introduced. Are there no less aggressive maybe epoxy based gelcoats?
    #47
  8. HickOnACrick

    HickOnACrick Groovinator

    Joined:
    Dec 5, 2007
    Oddometer:
    2,881
    Location:
    captures.crunching.farewell
    I have made 2 molds from epoxy recently, and pinholes weren't an issue, but I did have a number of craters that I had to wet-sand out. After my recent successes with the epoxy methods I have outlined, and all the failures filling my dumpster from gelcoat debacles, I don't see myself ever using gelcoat again.
    #48
  9. tmotten

    tmotten Lefthand ride Dutchy

    Joined:
    Jun 28, 2006
    Oddometer:
    2,705
    Location:
    Calgary
    Fair enough. I'm undecided on it still for the faux tank I'm building. Although the more I think about it the more I think I'll stop using it as well. Got to sand the gelcoat when it reacts anyway, and I'm painting my jobs anyway.

    How big are those 'craters' in size and depth/ height?
    #49
  10. HickOnACrick

    HickOnACrick Groovinator

    Joined:
    Dec 5, 2007
    Oddometer:
    2,881
    Location:
    captures.crunching.farewell

    Craters were small, the size of pinheads. I couldn't see them, but could feel them. They wet-sanded out without any difficulty.
    #50
  11. Orangecicle

    Orangecicle On a "Quest" Supporter

    Joined:
    Apr 24, 2007
    Oddometer:
    5,175
    Location:
    West Des Moines, IA
    I was finally able to pull a decent CF part from my heat shield mold. I made this part in five layers of 2X2 twill. It's fine, but it does get a bit flexible when heated by the exhaust. Maybe seven layers next time:

    [​IMG]
    #51
  12. HickOnACrick

    HickOnACrick Groovinator

    Joined:
    Dec 5, 2007
    Oddometer:
    2,881
    Location:
    captures.crunching.farewell
    That came out really nice. How did you resolve the issues with your mold?
    #52
  13. Orangecicle

    Orangecicle On a "Quest" Supporter

    Joined:
    Apr 24, 2007
    Oddometer:
    5,175
    Location:
    West Des Moines, IA
    The mold isn't perfect, and the surface of the part shows it a little as well. I wetsanded the mold as far as I could (through the gel coat in a couple of places). Then I waxed the crap out of it to help fill the voids. I did three coats of PVA painted on with a brush before I started molding. In the end, the part is not absolutely perfect, but the imperfects aren't really that visible. I could also wetsand the part and coat the top of it with more epoxy if I wanted to. What I care most about is a part that actually performs. So far, this thing does the job of getting rid of all that heat on the leg. I've melted or burned numerous pairs of pants with the original metal heat shield, but that's not a concern anymore.
    #53
  14. tmotten

    tmotten Lefthand ride Dutchy

    Joined:
    Jun 28, 2006
    Oddometer:
    2,705
    Location:
    Calgary
    What did you lay up the mould onto? I'll have to do some insitu laying up on the bike in difficult areas, and am wondering about effective ways to keep the resin dripping into the small hard to get to areas around the motor.
    #54
  15. Orangecicle

    Orangecicle On a "Quest" Supporter

    Joined:
    Apr 24, 2007
    Oddometer:
    5,175
    Location:
    West Des Moines, IA
    I molded directly on my bike originally. In my mind you pretty much have to start by taping off areas of the bike that could be damaged by the epoxy and then just start working from there with whatever product works. I used tape and fiberglass to make my initial shapes. But, my initial heat shield was first molded with non-hardening modeling clay. That first heat shield didn't work out so great, so I used it as part of the mold for the second heat shield -- literally holding it in place with tape while the new fiberglass cured. After I had the basic part captured in fiberglass, it was just a matter of building up the shape as desired. I used bondo and other products to get the part as I wanted it to look on the outside, paying close attention to the edges. You don't want to build up the edges. After bondo, treat it like any car part -- primer, sand, primer, sand, primer, paint. Watch out using too much of any product in one application. I put too much bondo on to fill a gap at one point, and that warped the part.

    Once you have the part you want, you need to lay the part out on a board to create the mold. There are some really good examples of how to do that on You Tube. Just search for things like "carbon fiber mold" or "how to make a mold for carbon fiber."
    #55
  16. HickOnACrick

    HickOnACrick Groovinator

    Joined:
    Dec 5, 2007
    Oddometer:
    2,881
    Location:
    captures.crunching.farewell
    I know I sound like a broken record when it comes to Epoxy, but you could lay up some fiberglass with epoxy - say 1-2 sheets thick. Once it is dry, hold the smooth area over your difficult areas, then apply a heat gun. One of the properties of the epoxy is it softens when heated. While it's hot, use a gloved hand to mold it around. You might be able make an in situ mold in this way. If it doesn't work, you wouldn't be out that much time or money.
    #56
  17. twodollardoug

    twodollardoug the calimesa kid

    Joined:
    Jan 15, 2009
    Oddometer:
    337
    Location:
    southern california (IE)
    have you seen this high temp resin? suppose to be good to 400 degrees....has to be heat cured to 200 degrees tho. http://cgi.ebay.com/EPOXY-RESIN-HIG...247?pt=LH_DefaultDomain_0&hash=item483836807f

    or what about laying kevlar for your final layer.
    #57
  18. Orangecicle

    Orangecicle On a "Quest" Supporter

    Joined:
    Apr 24, 2007
    Oddometer:
    5,175
    Location:
    West Des Moines, IA
    I can hear it now.

    My wife: "Honey, what the #$%^&* are you cooking in the oven?"
    Me: "Uh. It's, uh. Well, carbon fiber."
    My wife: "Are you %$^&* kidding me?"

    :rofl

    I read that Kevlar has a lower temp capacity than carbon, so I went with carbon throughout. I did think that one solution might be to layer the inside of the part with heat resistant tape. I can say that I rode down the interstate on the way home tonight. I didn't wear gloves on the way home. The five-layer shield was only warm to the touch at the first light I came across. :eek1 The stock shield would have been melting my pants. The shield does heat up when sitting in traffic, but I think anything would. Still, it's not NEARLY as hot as stock.
    #58
  19. twodollardoug

    twodollardoug the calimesa kid

    Joined:
    Jan 15, 2009
    Oddometer:
    337
    Location:
    southern california (IE)
    right....i got an old electric oven for powder coating....so i guess i forgot not everbody has one.....but from the info about the resin, heat lamps can be used. i plan on using heat lamps myself. it only has to be cured at 200 degrees.

    i havnt read that about kevlar could you post up a link? it might keep us from using it in places we shouldnt....

    maybe over time your heat shield will harden up.... i powder coated mine. and it seams like the first couple heat cycles the powder coating softend a little. but know its rock hard.
    #59
  20. Orangecicle

    Orangecicle On a "Quest" Supporter

    Joined:
    Apr 24, 2007
    Oddometer:
    5,175
    Location:
    West Des Moines, IA
    http://www.intexa.com/downloads/hightemp.pdf

    This is what I looked at. It suggested that Kevlar was good up to a continuous operating temperature of 375 degrees Fahrenheit, and carbon fiber was good up to a continuous operating temp of 570 to 1000 degrees. I made my heat shield out of CF because of this article.

    "I found it on the Internet, so it must be true." :deal
    #60