strong composites for belly pan

We have decided to look into changing our method of building from extrusion, to a welded/ riveted box frame, with a bellypan keeping everything together. The only problem is that we do not have the capabilities to machine a full sized aluminum belly pan. We are however getting a techno router installed by january, on which we are able to machine composites, plastics and woods. Does anyone have experience with the belly pan design and the strength needed for the bottom plate? Ive been looking at peek and noryl with 30% glass, as these are the strongest plastics that i have found. If we go with 1/4 " thickness and pocket it it similar to what other teams have done (1323, 254), would we see about the same strength for the same weight (peek is about 1/2 the density of aluminum)? We are completely new to this method of manufacturing, and are unsure how much force the belly pan actually sees. The only problem is that peek is fairly expensive, are there any cheaper alternatives that would also be strong enough?

thanks Sven

For a few years we have run 1/16" thick garolite G-10/FR4. It’s not too expensive on mcmaster, and pocketing isn’t necessary. If you put it on properly (so it is spaced right and not oil canning), it is very strong.

It’s also real easy to drill to mount electronics, and is nonconductive.

Oh, the glossy black also looks fantastic.

hi Adam
thanks for the quick reply.
Unfortunately we are located in canada, and i believe mcmaster will no longer ship here. I will try to locate some other retailers here in canada though.
When working with this material, what precautions should be taken? Is the dust similar to that of fiberglass?


Lexan can be used for an electronics belly pan at 3/32 or 1/8th. However, unlike an aluminium belly pan it won’t provide much structural reinforcement / rigidity to the chassis, as lexan is very flexible.

It is not hard to laminate a layer of s2 glass or carbon fiber to a sheet of 1/4" plywood and gives a very strong and ridged board for your bottom. We planned on doing this for 2010 but the design did not allow for it. S2 glass is a value solution and is almost as good as carbon for this app unless you get some areospace grade carbon. A good grade of birch ply works well.

That’s rough. I’d find a way to get it! It’s great stuff to work with, and is really a pretty ideal solution to the problem.

It can be cut with hand tools easily, and also cuts nice on wood tools. The dust is fiberglass, so vaccum while you cut and don’t breathe it.

I am pretty sure McMaster still ships to Canada. Mabye not all parts, but you can get it to a major centre anyways (we can make orders in Waterloo, but Kincardine is perhaps another story).

Have you considered cutting the pan out of metal mesh? I have seen it done by RAWC before, and the results look fantastic. Its possible you could weld it right in too. That wouldn’t require much heavy machining by my reckoning.

You might ask yourself if the belly pan needs to provide rigidity, or if it just needs to house electronics. In the past years, we have always used a box al welded frame and 1/8" lexan. The frame is solid without the lexan.

If you decide you still want to go with a rigid “belly pan”:
I can give a +1 on garolite. I’ve always heard that it is about 90% the strength of carbon fiber. Machines just like carbon fiber or fiberglass in my experience.

As per your techno, if it is similar to the new one we got here, it can cut aluminium using a carbide endmil, and if you want some suggestions on speeds and feeds, I can give you first hand exerpience with a techno router. It’s not fast, but it can be done if you decide that is what you want.

One other thing I remember hearing while lurking the forums was a suggestion to use a high quaility 5 ply 1/2" plywood. It’s rigid and easy to work with.

The RAWC actually uses 1/8" 6061 Aluminum and gets it waterjetted/laser’d.

Sven, we’ve used belly pans like 254 did last year.

We actually had these guys cut on a 4x8’ Techno Router. Our team has owned one for about 3 years now. If you have any questions or need help, feel free to shoot me an email.


In my time on 330, we had some really nice plywood bottoms. 2005, we took aircraft-grade plywood, cut some pieces to match the openings in the frame that we wanted to cover, and cut a piece to match the entire bottom area to cover. Glued them all together, and had a strong belly. 2007, we looked solid from the top. The bottom of the belly pan (1/4" ply, I think) was a spiderweb of strength and thin areas. Attachment points were full thickness, supports for the attachment points were full thickness, everywhere else was thin. A skilled hand with a router was all that was needed.

McMaster is a bit strange with regard to shipping to Canada (provided nothing has changed in the last several months). They’ll happily and promptly ship to existing customers, but I’ve heard several times from potential Canadian customers that they’re being turned down.

Since you’re in Kincardine, maybe one of the Bruce Power folks can arrange for you to ship it to their dock? (Surely they’ve got a prior relationship with McMaster.)

As for belly pan materials, laminated carbon fibre and plywood was just what I was going to suggest. Companies like DragonPlate sell it, and it’s expensive but excellent. I’ve never been able to afford enough of it to use on an FRC robot though.

Garolite can work too, though I’ve never seen it laminated with a lightweight core. You may have to build your own using a high quality epoxy and some appropriately smooth plywood or other material.

With any of these composites, machining can be a bit of a pain (not to mention the dust is nasty). If you still have a way to get some parts cut on a waterjet, that may be the thing to do, provided that any wood core is marine grade (or similar) so the glue won’t dissolve, and doesn’t expand too much when wet. (I seem to recall that DragonPlate says they’ve tried this with their plywood-cored material, with decent, but cosmetically poor results.) Otherwise, they’re usually cut with a CNC router (but the material wears out cutting tools quite quickly).

I’d actually be interested in seeing how progressively-stiffer robots perform on the playing surface. There’s probably a point where more stiffness becomes inconsequential in any given design, and it might be worthwhile to experiment with several materials to try and find it.

One final thought: in most designs using one, the belly pan is probably most useful as something that keeps the frame rectangular, rather than something that keeps it from bending in the vertical direction. In that sense, buckling resistance in 2-D is probably more important—and that has a lot to do with the way you attach it to the frame rails. (Lots of rivets are good!)

G10 is incredibly abrasive and will destroy even carbide end mills/router bits in no time flat. You will probably want a diamond coated bit if you intend to do extensive machining to it.

The dust is very very nasty. If you breathe it, imagine essentially ending up with microscopic glass particles deep in your lung tissue. You want an appropriate respirator for sure, if you produce any dust.

G10/FR4 is not cheap. To make a full size carbon/s2 glass plywood panel for a robot bottom can be done for relatively cheap. 20$ for a 1QT. epoxy kit. 14$ for home center 1/4" plywood 4 x 4. About 35$ for Baltic birch. 8 oz. S2 glass about 8$ per yard. 660 KSI carbon about 35$ a yard. This is all you need to do a basic lamination using weight to smash it down. Better results can be had with vacuum bagging but, would require more investment in materials. A fine tooth metal blade will cut the finished product just fine. CONTROL THE DUST!! Do not allow epoxy on skin and ventilate well. Be aware that there are some people that have amine allergies. Laminating is not that difficult. The are many videos on your tube.

A 3x4 piece on mcmaster (with their high prices) of 1/16" g10/fr4 is only $60. We’ve been using it for years, that gets you a bellypan and plenty of spare material.

make the belly pan heavy, and you’re robot won’t be so likely to fall over…

But if you’re going to put electronics directly on it, then making it insulating is also a good idea.

We’ve used 1/16" aluminum and 1/8" plywood to good effect in different years. Aluminum is easy to rivet to the frame (assuming the frame is aluminum or fiberglass or similar), and wood is easy to staple/glue to a wood frame.

I’d still like to be able to use 1/4" steel plate for the bellypan…it’s a life goal…

Just to throw another idea in there… what about making the bellypan in pieces? With out design last year, our pan had to be pretty deep to hang off the frame like we wanted. As a result, we couldn’t make it out of a single sheet of aluminum - we ended up bending up two sheets, and pop-riveting them together down the middle of the robot. It worked great, and was plenty strong. So if you can work smaller pieces of aluminum and build the pan out of 2-3 sheets instead of one, that could be a viable alternative.

A couple ideas:

Carbon Fiber with balsa core.
Carbon Fiber with honeycomb core.
Fiberglass with balsa core.

You can find almost any composite/core combination you’d like. We used a piece of fiberglass/honeycomb core for our 2005 robot electronics board. It is very stiff, light, and strong. Can be a little pricey though. Taking the 1/2" balsa-cored carbon fiber as an example, it has a density of 0.0125lb/in^3, about 88% less dense than aluminum.

If you get a little ambitious you could make this sort of stuff yourself, it’s not terribly difficult.

Any thoughts on perforated plastic sheet (e.g., McMaster-Carr 9293T56)?

I don’t know about perforated plastic sheet, but corrugated plastic is fairly strong and fairly light. If it’s properly supported (around the bottom of the frame, say) it can hold a full-weight robot. I don’t remember what thickness we used on our ramps in 2007, but I don’t think it was more than about 1/4"-3/8", on about a 3’ square.

In case honeycombed composites are considered…

Our electroncs boards have been made from honeycombed fiberglass made by PlasCorp since 2007. The combination of a stiff board, insanely light weight vs. square footage, and insulating factors made it ideal for us to use. We even mounted the compressor and battery on the material for 2010.

The material is somewhat easy to cut with hand tools, but I would recommend a table saw and/or band saw. It is extremely easy to drill through, and not too difficult to punch through with a hand punch + hammer. Finally, spraypaint works great for color.

However, I would not recommend using it for a belly pan that provides support to the frame. The gaps from the honeycomb structure are not strong enough to withstand the implied lateral forces. I also recommend against mounting the board vertically unless the bottom edge has a lateral support since gravity will have the same lateral-force effect.