Carbon fiber is a particularly interesting material. When you refer to carbon fiber, you are usually referring to carbon fiber composite, which is a combination of about 50% carbon fibers, and 50% plastic (usually epoxy). This means it is essentially just reinforced plastic. Many people know about fiberglass, but not many people now about carbon fiber. They are very similar except for the fact that fiberglass is made from teensy tiny glass fibers and carbon fiber is made from even teensy tinier carbon fibers.
Carbon fiber is somewhat more expensive by weight than fiberglass, but it is also much stronger. Since you can use less for the same strength, the cost is very similar between carbon fiber and fiberglass. Both carbon fiber and fiberglass work by attaching to tons of molecules of plastic and holding it together very strongly. Plastic by itself can crack and snap and break, but with carbon fibers weaved through it, its strength is greatly increased. This is in part due to the immense strength of the fibers on their own, but it is also because there is such an enormous surface area for the fibers and the plastic to hold onto each other.
So now that we understand the 2 basic parts of carbon fiber, plastic and carbon fibers, we can look at how you manufacture parts with it. The most important concept for this is compression. Anything you make with CF must be compressed so that the plastic is evenly distributed throughout the fabric, there are no air bubbles where cracks can form under stress, and the ratio of carbon fibers to plastic is just right. Compression also makes things an even size and density. The most important part of this is the ratio of fibers to plastic. An ideal part should be just over 50% fibers. Any extra plastic adds to the weight of the part, but contributes almost nothing to the strength of the part.
There are many different ways to compress a part. Different methods are used for different parts. The simplest method and shape would be a flat sheet of carbon fiber. You would simply layer a few sheets of carbon fiber fabric, add in the plastic resin, then clamp it between two flat pieces of something else (aluminum, smooth wood, plastic, etc).
Unfortunately as soon as the shape gets more complicated than a flat sheet, the molding process gets much more complicated. Most methods rely on atmospheric pressure to press on the model through a plastic bag. Here are some methods that employ atmospheric pressure:
- Vacuum Infusion – This process gives the best part quality and can be used with most different part shapes being molded. You lay your dry carbon fiber onto a master mold, put a bag around the part and mold, then you apply a vacuum using a vacuum pump at one end of the part and you attach a container of plastic resin at the other end. The vacuum pump pulls out all the air in the carbon fiber fabric and the resin gets sucked in, replacing all the air. This results in a perfect part without any voids and with a perfect ratio of fabric to resin. The problem is that this process is very labor intensive and prone to failure because if there is a leak in the bag, the infusion is ruined. The bags get thrown away, so it is expensive. This method would be good for large one-off parts that need to be perfect regardless of cost. Here’s a video that helps you visualize what’s going on: http://www.youtube.com/watch?v=Moh9WQ2ZnTo
- Vacuum molding – This is somewhat similar to vacuum infusion. Instead of putting the fabric onto the mold dry, you add in the resin by hand and put the wet fabric onto the mold. Then you put the whole thing in a bag and apply a vacuum. In both this and vacuum infusion, you need a layer of peel ply between the bag and the part to keep it from sticking, but in vacuum molding you also need a layer of what’s called breather fabric which is meant to soak up any extra resin and prevent the resin from going into the vacuum pump. All of these extra materials add to the final cost of the part because they can’t be re-used. The parts aren’t quite as perfect as those made with vacuum infusion, but they are generally free from voids and well compressed. This process is also less prone to failure because if there are leaks in the bag, the part isn’t usually ruined.
- Bladder molding – Vacuum molding is ideal for making parts that are single halves of a shell. Think about a bottle cut in half. With vacuum molding you could make each half of the bottle, but if you wanted to make the whole thing as one piece then you would use bladder molding. Basically you have a two part mold of what you want to mold, then you put the wet carbon fiber over a “bladder” (usually a latex balloon) and you inflate the bladder so that the carbon fiber gets pushed to the outside of the mold. When the part is cured, you can just pull the bladder out of the part and take the mold off the part. The pressure from the bladder is what compresses the carbon fiber. This creates a very high quality part, similar to that created with vacuum molding. It is perfect when the outside surface of the part needs to be perfect and the inside doesn’t matter very much.
- Heat shrink tubing – I don’t like this method very much. If you’re making a straight simple tube then you can make a metal internal mold that can be removed after it is cured. For compression you simply put some heat shrink tubing over the mold and carbon fiber then you heat it up with a heat gun. This makes the heat shrink tubing shrink down over the part and compress it evenly… In theory. I’ve tried this a few times and I’ve found that the heat shrink tubing doesn’t compress evenly unless you heat it all up very hot. The problem is that the heat can damage the plastic resin. The heat shrink also doesn’t put a lot of pressure on it. The heat from the heat gun also makes the resin bubble quite a lot, so you get a ton of air bubbles in the final part, which is not good if you’re trying to make an airtight tube.
- Vacuum chamber bag molding – This is still an idea. All of the other methods are tried and true (not by me), but this method is just something I thought of a few days ago that I think would work well. This operates with the same principles as regular vacuum molding, but instead of connecting a vacuum pump directly to a bag, you put the wet carbon fiber on the mold, the mold in a bag, and the bag into a vacuum chamber. Then the air is pumped out of the chamber and the bag is sealed. When the chamber is allowed to go back to atmospheric pressure, the bag will contract on the part and compress the carbon fiber. This method would be better than vacuum molding because there is no chance of resin going into the vacuum pump, the vacuum pump does not need to be connected the whole time, and it would be significantly less work to make a ton of similar parts with this method. The eventual goal of this project is to manufacture and sell a trumpet, so I’m taking manufacturability into account very highly. This method has the potential to be really great, but it could also not be perfect. It’s possible that the epoxy will create bubbles while curing and since the bag is not connected to a constant source of vacuum, the bubbles will remain in the part. This is the only real problem I foresee with this method.
All of these methods need some kind of mold for the carbon fiber to shape onto. If it is a half part (like a bottle cut in half), then you can mill a mold out of plastic or aluminum. For bladder molding you would make a two part mold that is milled from plastic or aluminum. For heat shrink molding, you would use a tube as an internal mold or an aluminum bar that is machined on the lathe.
In a trumpet there is a very difficult situation. You need to make curved tubes that have very accurate internal surfaces. You can make curved tubes with the bladder molding method (see my post about this), but this doesn’t give a perfectly smooth and accurate internal surface. After making my first curved tube, I put out some extra mental effort to try and figure out a new method of making my curved tubes. I figured that if I could make some kind of internal mold that could be removed later, then I could make these parts using any of the vacuum methods. I’ve heard of people using a polystyrene foam (or similar) as a mold that can then be dissolved using acetone or another solvent. This might work ok, but I’m afraid the foam is too soft and wouldn’t give a good internal finish because of the pressure from the vacuum molding. So the next thought is to use something else that can be dissolved that is very hard. I’ve heard of people melting salt (yes, as in table salt) and pouring it into a mold and then using the salt as a mold. After curing, the salt could be dissolved out using water. This sounds awesome until you realize that salt melts at about 1500°F which is well above the melting point of Aluminum (~1200°F), so you would have to use a mold made of something more difficult to machine. Also, I don’t even know how I would melt salt. Maybe a blow torch… So basically this isn’t a viable solution for me. The next idea is to use a plastic that I’ve heard of that dissolves in water, PVA. I know you can use it in 3D printers as a support material. The problem is that 3D printers don’t give a perfect surface finish and waste a lot of time, so if I wanted to make PVA molds, I would have to get them injection molded. Injection molding is good if you’re making thousands of parts, but if you want to just make one, then it would cost thousands of dollars. So this is not a good solution for me.
So finally I thought of the idea of using wax as a mold material. You can machine wax or melt it at reasonable temperatures and pour it into a mold. Once the part has cured, you can simply melt the wax out. It even has the positive byproduct of sealing the inside of the part! The wax that I’ve chosen as the best candidate is carnauba wax. It has a melting point of about 180°F. You aren’t supposed to heat up the epoxy above 230°F once it has cured, so this leaves a nice safety margin. This wax is also the hardest natural wax available and it is frequently used in making things shiny, which is perfect for sealing the inside of these parts. The only thing that I foresee being a problem with this is that epoxy heats up slightly as it cures so it is possible that it would heat up enough to melt or soften the wax. I haven’t noticed the epoxy heating up much in the parts I make. It is really only a problem when you have a large volume of epoxy that isn’t spread over a large surface area. If it is a problem even with my thin tubes, I can always just mold using 1 or 2 layers of carbon fiber so that the epoxy is even thinner and then I can mold it a second time adding on extra layers. I also had the idea of using a water soluble wax that had a much higher melting temperature (above the safe epoxy temp) so then I could just dissolve out the wax core. Unfortunately I haven’t found any water soluble wax formulations with melting points higher than that of carnauba, so using it wouldn’t help.
So those are the basics of carbon fiber manufacturing. I got a little more detailed than I was planning to near the end, but I wanted to connect it to this project as much as possible.