pic: Another Cheap Wood Frame Idea

[MrForbes]

The wood chassis by itself seems to be plenty strong and stiff, and like I said before it took us an hour to build it, and cost $20 for the materials.

If you think it could use improvement, I suggest you build one yourself, and see where the weaknesses are, and let us know how to make it better! After all, it only costs $20 and an hour of labor. (yes, I'm making a big assumption, that you have the machinery necessary to build it....but that is mostly the same machinery that has been required to build bumpers the past two years)

[Chris Fultz]

You might want to put on a "marketing hat" as you talk with people about this, or similar ideas -

Instead of a "Cheap Wood Frame", maybe a "Low Cost Wood Frame" or an "Inexpensive Wood Frame" or a "High Value Wood Frame".

The word "cheap" can imply junky, low value, and turn people away - when I think you really mean it is Low Cost..

[MrForbes]

Good idea! "Quick, Affordable Wood Chassis" does sound much better.

[Rick TYler]

I've built three small boats and been a project manager for a Boy Scout project which built seven canoes -- all out of plywood/epoxy/fiberglass composite. I'm 100% with Bruce, if you decide you need fiberglass use epoxy resin, not polyester. If you want some solid advice and a good source of materials, check out the boat-building forum at bateau2.com. The folks there don't have much experience with robots (except me...) but what they know about strong, light, easy-to-make composites is pretty impressive.

As for materials science, there's a great article in "The Nature of Boats" by Dave Gerr on materials commonly used in small-boat building. The weakness of wood is that it is labor-intensive (not an issue in robots), eventually will rot (not an issue in robots) and has fairly low impact resistance (an issue for Battlebots, but probably not for us). According to Gerr, Douglas fir is stiffer than steel, aluminum or laid-up fiberglass for the same weight of materials.

Fiberglass does not add stiffness in a wood/fiberglass composite. Epoxy-saturated fiberglass is a great low-weight fastener for joints, and the epoxy seals the wood to reduce water penetration (not an issue for robots, unless we get the long-awaited aquatics game!). Fiberglass itself is not stiff -- the wood is the component in the lay-up that adds stiffness. The glass holds it all together, protects from weather, and adds some impact and abrasion resistance. There are additives for epoxy, like graphite powder, that can improve abrasion resistance.

Generally, I'd avoid Kevlar cloth (it fuzzes and is a pain to sand smooth), or carbon-fiber cloth (very hard to work with, fuzzes, and supernaturally expensive) (unless you make carbon-fiber poles as arm material, in which case I'd like you to send me pictures). Other cloths like Dynel are more for abrasion resistance than strength and will do nothing for a robot chassis than add weight.

A chassis made from 6mm okoume plywood with joints taped with 9-ounce fiberglass set in SystemThree or WEST epoxy would be amazingly strong and light. My sons and I built a 17'4" canoe with 4mm okoume, 6-ounce glass, and SystemThree epoxy that weighs less than 60 pounds without seats and hardwood trim. I think a robot base could be made that is less than 10 pounds.

One really important downside to composite materials (like a chassis) is that repairs during a tournament will be nearly impossible. Epoxy sets up in a chemical reaction that takes days for maximum strength. If you break a joint in a competition you can't just whack another piece of stock off with a hacksaw, drill holes in it, and bolt it into place.

With some planning you could get the same effect with careful material selection (okoume plywood is more expensive but a lot lighter than birch, for example) and glued corner blocks. Those gussets are big and heavy. Epoxy and glass might impress the judges, but wouldn't really add much value in a structure that doesn't have to stand up to weather and waves.

I wonder if FIRST would approve the use of wood in FTC?

[artdutra04]

If you are going with dead axles, why don't you just support the axle/bolt from both sides with two chassis rails on each side of the robot chassis? This would take a lot of the stress off the single hole and spread it out between two holes.

Drilling precise (though not necessarily accurate) holes can be easily acheived by clamping frame members together (possible putting some screws through to guarantee they won't slide), and drill all of them at once on a drill press.

The outer chassis rail can either be permanently attached, and have the axles able to side out of the robot frame, or you can press threaded wood inserts into the wood to have a removable outer chassis rail with bolts.

As for the existing design, the only frame members that you will really have to worry about are the front and back lateral members in high speed impacts, which may splinter the wood. To help protect against this, use a biscuit joiner and drill biscuit holes in the tops and bottoms of these laterial members, and attach a 3/8"-1/2" by 1.5-2" wide piece of wood across the top and bottom to make a C-channel. If you use a lot of biscuits and wood glue, this should form a joint stronger than the original wood itself.

Actually, the more I think about it, any team which may lack a machine shop or CNC sponsor can create a really impressive robot as long as they have at least one parent/mentor with a bunch of woodworking tools and know-how.


Onto this being a "cheap" way out, I wouldn't really think of it as such. Perhaps we've gotten so used to aluminum and polycarbonate in FRC that we forget that wood can have a lot of interesting material properties that make it ideal in some applications. And if done right with careful attention to detail, using a router to make all the edges nice, remembering how to position the wood grain to maximize strength, and staining all the wood with a nice, rich color, then the final result can really be quite impressive. In fact, if a team went this far, they might even just carry on the theme and make everything steampunk.

[gorrilla]

Originally Posted by artdutra04

If you are going with dead axles, why don't you just support the axle/bolt from both sides with two chassis rails on each side of the robot chassis? This would take a lot of the stress off the single hole and spread it out between two holes.

Drilling precise (though not necessarily accurate) holes can be easily acheived by clamping frame members together (possible putting some screws through to guarantee they won't slide), and drill all of them at once on a drill press.

The outer chassis rail can either be permanently attached, and have the axles able to side out of the robot frame, or you can press threaded wood inserts into the wood to have a removable outer chassis rail with bolts.

As for the existing design, the only frame members that you will really have to worry about are the front and back lateral members in high speed impacts, which may splinter the wood. To help protect against this, use a biscuit joiner and drill biscuit holes in the tops and bottoms of these laterial members, and attach a 3/8"-1/2" by 1.5-2" wide piece of wood across the top and bottom to make a C-channel. If you use a lot of biscuits and wood glue, this should form a joint stronger than the original wood itself.

Actually, the more I think about it, any team which may lack a machine shop or CNC sponsor can create a really impressive robot as long as they have at least one parent/mentor with a bunch of woodworking tools and know-how.


Onto this being a "cheap" way out, I wouldn't really think of it as such. Perhaps we've gotten so used to aluminum and polycarbonate in FRC that we forget that wood can have a lot of interesting material properties that make it ideal in some applications. And if done right with careful attention to detail, using a router to make all the edges nice, remembering how to position the wood grain to maximize strength, and staining all the wood with a nice, rich color, then the final result can really be quite impressive. In fact, if a team went this far, they might even just carry on the theme and make everything steampunk.

i was thinking about doing the same thing, i am stuck on how to attach it though,

[MrForbes]

Originally Posted by artdutra04

If you are going with dead axles, why don't you just support the axle/bolt from both sides with two chassis rails on each side of the robot chassis? This would take a lot of the stress off the single hole and spread it out between two holes.

That is another way to build the robot, and if you feel like making up a sketch or Inventor rendering of it, that would be great! It sounds like what team 173 has done in the past.

The cantilevered axle design has some advantages and some disadvantages. One advantage is that it's pretty easy to work on it, another is that the holes don't have to be aligned, another is that it saves the weight of the outer piece of wood, and since there's a bumper there anyways that piece of wood is kind of redundant. On the downside, it takes a bigger axle with a strong attachment to equal the strength of a non-cantilevered design.

We used a chassis layout almost identical to this one last season, with pultruded fiberglass C channel for the side and end rails, and sheet aluminum for the bellypan and gussets. Axles were 1/2" threaded rod with big flange nuts holding them to the fiberglass. It worked well, and we felt it was easier to work on the drivetrain with this design, than with the double-supported dead axles we used the previous two years. Also, building the chassis took less time. We are beginning to think that getting a chassis built and running quickly is a good thing, so the team can concentrate on the game-playing parts of the robot. That's one reason we are taking a lot of time playing with new chassis ideas now, hopefully we'll have enough design ideas ready to go that we can pick one and built it by the end of the first week.

[gorrilla]

Originally Posted by squirrel

The cantilevered axle design has some advantages and some disadvantages. One advantage is that it's pretty easy to work on it, another is that the holes don't have to be aligned, another is that it saves the weight of the outer piece of wood, and since there's a bumper there anyways that piece of wood is kind of redundant. On the downside, it takes a bigger axle with a strong attachment to equal the strength of a non-cantilevered design.

of the two, which do you think is the easiest to build and most durable?

and were do you guys get pultruded fiberglass channel

[MrForbes]

I think we got the fiberglass from Creative Pultrusions, we bought a bunch our rookie year. I think that cantilevered axles are easier to build, but do need to be made stronger since they are only supported at one end.

[gorrilla]

Originally Posted by squirrel

I think we got the fiberglass from Creative Pultrusions, we bought a bunch our rookie year. I think that cantilevered axles are easier to build, but do need to be made stronger since they are only supported at one end.

on a 6wd base, why not just direct-drive two of the wheels on each side(using say, a bb transmission), then run a chain the the last one? that way you would only need to really make one dead canilevered axle

Originally Posted by gorrilla

in my expeireance with it, fiberglass only becomes brittle if you dont mix the resin right, or wait to long to apply it,


they switched to epoxy resin in surfboards because, polyester resin dosent stick to the new type of foam most companies use.


you dont even have to use the fiberglass mat, you could just take a paintbrush and dip it in the resin and "paint" it around the corners and edges


we're getting off topic now.......

My experience with composites is mainly in the realm of Longboarding, and building customs out of most available types of composites. Most new users to fiberglass end up with a goopy, brittle mess, assuming they just lay the fabric and slap some resin on. You're correct about using Epoxy resin, it's much stronger (and more expensive).

Originally Posted by Rick TYler

Generally, I'd avoid Kevlar cloth (it fuzzes and is a pain to sand smooth), or carbon-fiber cloth (very hard to work with, fuzzes, and supernaturally expensive) (unless you make carbon-fiber poles as arm material, in which case I'd like you to send me pictures). Other cloths like Dynel are more for abrasion resistance than strength and will do nothing for a robot chassis than add weight.

I've had nothing but success with both Kevlar weave, as well as Carbon Fiber. Granted, these are in Longboard applications, but those tend to have a little more force involved (a 200lb rider going through a 90 degree turn at 60mph). As long as you take the time to lay out all your steps right, and then don't dawdle when working with the actual glass, you'll be fine with both these materials. If a team has the budget, there are some IMPRESSIVE things that can be done with composites. Heck, I'd LOVE to build an entire robot out of the stuff. Foam molds and vacuum bags galore!

[Rick TYler]

Originally Posted by CraigHickman

I've had nothing but success with both Kevlar weave, as well as Carbon Fiber. Granted, these are in Longboard applications, but those tend to have a little more force involved (a 200lb rider going through a 90 degree turn at 60mph). As long as you take the time to lay out all your steps right, and then don't dawdle when working with the actual glass, you'll be fine with both these materials. If a team has the budget, there are some IMPRESSIVE things that can be done with composites. Heck, I'd LOVE to build an entire robot out of the stuff. Foam molds and vacuum bags galore!

My background is in amateur boat-building where the object being made is too large to bake and usually wildly impractical to vacuum bag. If I needed the penetration protection of Kevlar, I'd cover it with a layer of glass cloth to prevent surface fuzzing, or, even better, get an expert to design a layup schedule that included some biaxial cloth for strength mixed with Kevlar. Usually, in amateur boats you find carbon fiber cloth in spars and blades -- not usually in hulls. Longboards use a lot less material. Carbon fiber in quantities big enough to build 20- or 24-foot boats is really pricey. Carbon fiber is also a pain to cut, it floats in resin (which is why it is usually vacuum bagged), and fuzzes if it's on the surface. The carbon fibers are usually considered dangerous to breathe, so you want to wear a respirator or at least a NIOSH dust mask when cutting.

I do think carbon fiber tubes are The Bomb for robot arms, but haven't had a chance to try it due to the famously conservative students on the FRC team with which I used to work.

We need some pictures of your longboard(s). Here's a link to pictures one of my boats: http://www.jemwatercraft.com/images/...ahPictures.htm, made from 4mm okoume and 6-ounce fiberglass. And other stuff!

[MrForbes]

This discussion is making it look pretty sensible to just cut up some plywood on the table saw, and glue it together!



The carbon fiber tube idea is nifty, but kind of pricey for those on a limited budget.

http://www.carbonfibertubeshop.com/large%20tubing.html

[gorrilla]

Originally Posted by squirrel

The wood chassis by itself seems to be plenty strong and stiff, and like I said before it took us an hour to build it, and cost $20 for the materials.

If you think it could use improvement, I suggest you build one yourself, and see where the weaknesses are, and let us know how to make it better! After all, it only costs $20 and an hour of labor. (yes, I'm making a big assumption, that you have the machinery necessary to build it....but that is mostly the same machinery that has been required to build bumpers the past two years)

could you tell us the dimensions?

ill see if i can get to home depot today or tomorrow

did you apply the glue then staple it?

luckily gorrilla glue is cheap

[MrForbes]

Originally Posted by gorrilla

could you tell us the dimensions?

The dimensions are based on a few things that you can probably be the best judge of....length and width a bit less than the maximum allowed, the inward offset of the sides is enough to allow the wheels and chains of your choice, and the height is enough to fit the transmissions in. We chose 26" x 36" x 5.5" overall size, and 2.25" side offset. You can do whatever you want!

did you apply the glue then staple it?

Yes, we used Titebond premium wood glue, because there were three bottle of it in the fab shop, and we used 1/4" crown x 7/8" staples because they were in the pneumatic staple gun, and it seemed like they'd work. You could use nails, brads, small wood screws, other sizes of staples, etc. Just beware that using a fastener into the edge of plywood is not very strong, so getting a good glue joint is kind of important.

And thanks for all the great info Rick!