Carbon Fiber Q&A

Interviewed Andre from Multimatic. Here's a paraphrased version of our conversation:

Q: Best method for joining pieces together? [I told him we were considering clamping hubs as well as using an adhesive]
A: Adhesive will definitely give the strongest connection, at the cost of being more permanent. With clamping features you always run a risk of slipping and loosening. Also keep in mind for using adhesive: you can't just apply an arbitrary amount of adhesive and squish it together. If you squish it too tightly, all the adhesive will squeeze out and there will be none left on the bond surfaces. For this reason, when applying the adhesive you need to have spacers with thickness equal to the required adhesive thickness.

Q: How feasible is it to make custom parts?
A:  You definitely get a big advantage if you have the ability to make custom parts. You can go crazy on weight savings if you optimize your custom parts so that they're only heavy in the required places. But custom parts are just so much effort to make. You've seen those guys out in composites doing layups all day. [For context, in our composites department I do see the workers spending a tedious amount of time laying down layers upon layers of carbon cloth.] For custom parts you need to make a mold of course. But there's also many tedious steps including mold validation to ensure the mold surface is clean enough for laying carbon fiber, and there's a very specific technique for laying down all the cloths. Then you need to be able to vacuum bag it and put it in an oven to cure. Overall it could be useful if you have the capability to do it or the necessity, but it seems like you're probably going to just be able to mount most of your things to off-the-shelf tubes so you probably don't need to go through the extra effort of making custom pieces.

Q: What's the best weave pattern to use for a member?
A: Obviously depends on simulation results and what you know about the directions of the forces. But in general, if you know a piece is always going to be in tension, you could definitely just have a unidirectional piece there, and you can save weight on it because you only need to account for that one direction. But for something like a member on the drivetrain where it could be in tension or compression, it's probably better if you just use a tube of multidirectional layers. Often you'll see those tubes have the layers offset 120 degrees off from each other so that it averages out and gives you more or less uniform strength. It's just easier to use a multidirectional tube so that you can account for any directions. It especially makes calculations easier.

Q: If we do decide to use tubes, what's the best style of tube to use, circular or square?
A: Structurally speaking, circular tube is always the best, it has the heighest strength to weight ratio. Square tube naturally has stress concentrations on the corners because they're bent there compared to the rest of the tube. However, they're not THAT much worse, square tubes are still fine, only marginally worse. Also for your purposes, mounting to square tubes is probably a lot easier than mounting to circular because of the flat face. You might even have a lower strength to weight ratio by using circular tubes because you have to add extra weight with whatever mounting solution you develop to account for the round face. [Asha's comment: Then maybe we should use circular tubes for the drivetrain members but use square tubes for the arm and chassis because we will need to mount things on to the arm chassis but not so much with the drivetrain members.]

Q: How bad is it to machine holes into carbon fiber parts?
A: Although it's better to avoid if possible, it CAN be done if necessary; people do it. Obviously it does weaken the carbon fiber but not necessarily to a point where it's unusable. A "knockdown factor" can be calculated based on the geometry of the hole and of the piece you're machining. I can't give a specific number because it all depends on those parameters, but there are procedures for tests and calculations to be performed that would tell you exactly how much strength you're losing as a result of cutting out pieces of a carbon fiber part. [Asha's comment: we should run some of these tests to quantify how much strength we would lose! But also ideally we just design good so we don't have to drill holes in the first place.]

Q: Is galvanic corrosion a risk for us when bonding carbon fiber to aluminum?
A: Not really, especially considering you're competing in a dry desert. It's more of a concern for systems that need to last many years and mostly ones exposed to moisture. Aircrafts for example because they will need to fly through rain.

Q: Should we consider using another non-metal like fiberglass?
A: Not unless cost is an issue. Carbon fiber is better all around.

Comment: [after I mentioned that we would need to have separate mounting solutions for each wheel cause we can't mount them directly to carbon fiber] Keep in mind that even though you're saving weight by using carbon fiber, you now have to add extra weight to make these separate mounting solutions to mount things to your carbon fiber, whereas mounting to aluminum is a lot more intuitive. So if the amount of weight you add to make your carbon fiber members work is the same as the weight you would have had if you just used all aluminum, it might just cancel out and not be worth it. Just don't forget to factor the mounting solutions into your weight savings calculations to see if it's really worth it.

Comment: Carbon fiber's main weakness is sharp objects. It can withstand distributed loads all day long, but one tiny focused penetration can ruin an entire piece. To get around this, people often laminate the outer layers of carbon fiber parts. The lamination protects the carbon fiber from sharp impacts so that the carbon fiber can do its job bearing distributed loads.