pic: Carbon fiber monocoque drivetrain Top view



This is a drivetrain that I CADed in the last few days. I think it is the most aesthetically pleasing drive that I have CADed so far. Post any questions or comments here.

Guess I’ll take the obvious first question: How much does it weigh? :rolleyes:

That’s hard to say, as inventor doesn’t have a material for carbon fiber in its default library. By setting the carbon parts to the closest material I found, it says 28.5 lbs.

Very neat. Any idea on how one would build this? Have you done any simulations to check the stress resistance compared to the “standard” aluminum chassis?

I have ideas. Part of the reason I CADed this was so that I could learn how to make it easier to manufacture. Team 2471 recently had a visit from a new sponsor who might have the resources to make this chassis. I used the information they gave us abut what is easiest for them to make, but that doesn’t mean that it is optimized for ease of manufacture.
currently, it is made of five peaces. One C channel on each side, and a dropped center piece that would be vacuum molded into a “top hat” shape.

Have you done any simulations to check the stress resistance compared to the “standard” aluminum chassis?

No.
Based on the sample peaces I have seen from our sponsor, I think that this would be very strong.

No.
Based on the sample peaces I have seen from our sponsor, I think that this would be very strong.

It will also be a major pain to drill any holes in.

Definitely. Make sure you get all the holes in the original design!

And, OBTW, what’s plan B if it does actually break? I was nervous about our welded aluminum chassis in 2013 (though not as nervous as I should have been now that I know more about aluminum grades and welding and heat treating), but this is a whole 'nother level of “no way to repair”.

I’m guessing a patch kit, some really strong epoxy, and a good deal of faith.
But really, carbon fiber is quite the science. I’d recommend checking out acpsales online as they have decent guides on carbon fiber. Consider a filler material like foam or a hex shaped layering for some added strength along the rails. You want carbon fiber to have some flex, but not too much. Also be careful about hole placement.

From what I remember from talking to a team that does a majority of their robot out of carbon fiber, that seems to be the main drawback.

Any clue on how expensive this would be compared to if it was made out of standard drivetrain steel or aluminum?

Extremely expensive.

There are a number of ways to fabricate carbon fiber structures.

The first is to use dry cloth and saturate with epoxy. Generally speaking this is what home-builders use. It’s going to be heavier than a well made vacuum bag or pre-empregnated setup, but doesn’t require expensive materials or specialized building techniques.

Your next choice is how to form the parts. Will you make a male mold? A female mold? Use a spacer material like polystyrene that can be chemically removed?

Fastening things in carbon fiber can be tricky. The strength in a fiber structure comes from the fiber itself. The epoxy is merely there to hold the fiber in the shape you need. Fastening to the carbon fiber incorrectly will crack the epoxy and destroy the structure’s strength. You also tend to need backing plates to spread the pressure out.

Then there’s cure times, cure temperatures, mixing ratios, substrate materials…

Starting around 2:10:

Wet Layup - what many FRC teams would do in their shops:

Vacuum Infusion:

The last two months of my life finally coming to a close (and why I know so much about fiber construction):

http://s15.postimg.org/63uvhzbp7/20151024_051737.jpg

Assuming they’d be doing their own layup, and depending on the number of layers it could be anywhere from $100 to $500+ (only for the carbon fiber, not counting the other supplies like epoxy and such). You can get 50" wide carbon fiber cloth for as little as $20 a yard if you look around a bit.

Personally if I were to build something like this, I’d use a high compression strength foam core material rather than building it monocoque (plus it’s a bit easier to do the layup). Then simply use a lightweight wood for the center material in areas you need to drill holes (sitka spruce is a good option) and then use just 1-2 layers of carbon fiber cloth. You could also make the walls of this design much thinner using such a method.

IMO, vacuum bagging gets a bad rap for being too complicated for the average user. Our team managed to build a very simple vacuum bagging table using some non-stick plastic bag material, PVC, sticky tack, and a basic shop air compressor with a vacuum valve.

None of the parts we made used a mold, we simply applied the cloth directly to the core material, added epoxy, vacuum bagged, and repeated. This won’t work for all shapes, obviously, but for most FRC uses it works fine.

As long as you’re smart with what kind of core materials you use and where you use it, backing plates are largely unnecessary, when they are, you can usually get away with some 1/8" polycarbonate. Again, for most FRC uses.

Examples:
www.chiefdelphi.com/media/photos/27823
www.chiefdelphi.com/media/photos/31319

I’m not the expert, but frame parts would likely be vacuum bagged in a female mold.
The gearbox plates would be cut from stock plate on a CNC router.

Fastening things in carbon fiber can be tricky. The strength in a fiber structure comes from the fiber itself. The epoxy is merely there to hold the fiber in the shape you need. Fastening to the carbon fiber incorrectly will crack the epoxy and destroy the structure’s strength. You also tend to need backing plates to spread the pressure out.

The axles in this design are supported by aluminum plates that would distribute the load to the carbon structure.

Then there’s cure times, cure temperatures, mixing ratios, substrate materials…

I don’t know anything about what would be used for these. That might be intentional on the part of our sponsor.

This could be done in house with the coaching of our sponsor, although that’s not what I was imagining when I designed it. I was surprised at how affordably our sponsor could create carbon components.

Personally if I were to build something like this, I’d use a high compression strength foam core material rather than building it monocoque (plus it’s a bit easier to do the layup). Then simply use a lightweight wood for the center material in areas you need to drill holes (sitka spruce is a good option) and then use just 1-2 layers of carbon fiber cloth. You could also make the walls of this design much thinner using such a method.

I’m not able to picture how this construction method would be used here, but maybe I’m missing something.

So, for example, if you were using a foam center material, the process would look something like this:

  1. Cut/sand foam to desired shape (for this example, a simple square box)
  2. Find a clean flat surface (the cleaner and flatter the better), this will be the vacuum table
  3. Lay down a layer of disposable absorbent material on the table (there are lots of options for this, you can even use common paper towels if you’re working on the cheap)
  4. Lay foam part flat on the table (on top of the aforementioned absorbent material)
  5. Lay carbon fiber cloth on top of foam part, preferably cut so that it hangs just over half way down the sides of the part
  6. Lay down a layer of “breather cloth” (basically a non-stick cloth material to prevent the absorbent material from sticking to the part and wicks away excess epoxy)
  7. Add more absorbent material (paper towels) to the top of the part
  8. Cover with plastic
  9. Insert vacuum hose under plastic
  10. Seal plastic completely to table using sticky tack (or similar), ensure the area around the hose is also sealed
  11. Turn on air pump, run overnight
  12. In the morning, remove part from bag, peel off breather cloth and absorbent material
  13. Flip the part over and repeat steps 3-12 again
  14. ???
  15. Profit!

You can also use a space heater and half of a cardboard box to speed up the epoxy setting process if you want (don’t leave it unattended though, for obvious reasons).

In the case of your design, you could use this method to basically make the whole thing with 5 pieces (or as one big piece if you were feeling really ambitious) and you wouldn’t have to make the frame walls so wide (leaving more space for electronics and such).

And now for some related old pictures…

1, 2, 3, 4

Edit:

After re-reading your previous comment a few times, I’m thinking you might have been referring to how the construction method would be used in this situation, and not so much how it works (correct me if I’m wrong)… >_>
Answering that question might require some CAD work to explain unfortunately.

This is how I’ve personally done a few of our parts, it uses really only 4 ±2 layers of carbon fiber and is much easier to shape your parts. We just get the desired shape we want and make it out of foam core, plan around hole placement and how much carbon fiber needs to be in the pockets, and then lay the carbon fiber. I’m not sure how it would work on a frame of your size, but it can help with rigidity and cut costs.

Like some of the other posters, I’m no expert; but I immediately wondered why the design includes so many 90 degree corners and edges.

Aren’t sharp corners/edges the exact places where structures made from carbon fiber, or similar materials, are most likely to fail, when the entire structure flexes?

Isn’t the key to success allowing many fibers to share a load instead of letting the load get concentrated onto a few fibers (at corners/edges)?

Color me curious.

Blake

From what I understand of this material. It is important to give the structure three dimensional aspects so that it can’t have strength and rigidity in the important directions. This can be achieved in a number of ways. The two ways that have been discussed in this thread are using a filler material, and forming a structure with flanges, corners, and box shapes. In this design, I have gone the flanges and corners route.

Imagine a flat piece of carbon fiber that is 4" wide. In some directions it is strong, but in others it is very flexible and not very strong. Now imagine that same amount of carbon fiber is instead a 2" by 1" C channel. In that configuration, it might break along its corners if enough force is applied, but it would be a lot stronger and more rigid for the same weight.

Tubes/cylinders are strong/rigid three dimensional shapes.

Carbon fiber bicycle frames are made of roundish tubes, not box shapes.

Antenna towers, crane booms and frames, car roll cages, are generally made from (round) tubes, not boxes.

Rounded shapes might be worth investigating.

Definitely ket us all know how the eventual frame holds up, if one gets built.

Blake

This is for a number of reasons. Chief among which is probably aerodynamics.

Antenna towers, crane booms and frames, car roll cages, are generally made from (round) tubes, not boxes.
Rounded shapes might be worth investigating.

I agree. There are definitely merits although making sure that there are still adequate mounting options may be difficult.

Definitely ket us all know how the eventual frame holds up, if one gets built.

Because this was mostly an exploratory design exercise, I doubt that it will be built. If it is, I be sure to post about the experience.

This depends a lot on how you do the layup. Most of the time, in my opinion, rounded parts are used because it’s a bit easier to fabricate, at least for the layup process (in addition to what Bryce2471 said already). Inevitably though, even if you do have 90 angles, the inside angles usually end up having to be slightly beveled just due to the nature of how the fabric connects the parts in most situations. Outside corners are a bit more forgiving but are still nice to round off a bit.