I tried to design this drivetrain to be as cheap and easy to manufacture as possible. Required tools are a drill and a saw, although drill press and bandsaw would be preferable, and the cost of all of the components comes out to around $400 (not including the toughboxes).
Looks easy to make, the only concern I would have is can it take a hit? Aside from that, what is the weight and where do I mount bumpers? (assuming they are legal) And how do you tension the chain?
Just a thought but could you save weight by using thin sheet metal as your belly pan?
It is not completely clear from the CAD, but remember that:
The middle wheels need to be offset lower, or the robot won’t be able to turn easily. We are using a .5 cm offset this year, but I have read about teams using 1/8 inch.
Many teams use omni wheels on one or both ends (not the middle) to aid in turning. The omnis compromise pushing power a bit, but make turning much easier.
Doesn’t look very stable. Maybe its the wheels or the actual frame. What would happen if it were hit on an angle? Tip over?!
I assume the sides and ends are made of some hardwood, such as oak? and the bottom and corner braces are made of a thin plywood? Looks strong! would the weight be about 15 lbs for the wood parts? (just a guess)
Perhaps you could add some vertical pieces of aluminum angle to the corners, to easily mount bumpers to?
What do you mean by hit on an angle? Do you mean if it were hit while going up a ramp?
I would also have to disagree and say that stability would most likely not be an issue unless they put a lot of weight up high but that would be an issue for most chassis. This looks very stable, I am assuming those are 6" wheels which puts almost all of the weight of this chassis below 6". This should be pretty stable unless they pick up like 5 of the old tetras and swing them around 8 feet in the air it should be pretty stable.
This is basically an all wood live-axle version of our 2008 chassis, which worked great (except perhaps the strange chain path I used…). Looks good, but I think it would be preferable to mount the transmissions in the center, so you can use corner plates like on the front on all three corners.
I’m not sure how well it would survive compared to other frames, but with bumpers I don’t think it would have any problems. As squirrel pointed out, bumper mounts on the corners are easy to make and work well. For chain tensioning, I like the nylon-circle-with-offset-hole-approach. It has worked well for us in the past and is easy to do.
A sheet metal belly pan would probably save a bit of weight and could be stiffer, but I laid out this frame trying to minimize the number of parts that need to be bought online. All of the wood parts can be easily found at a home improvements store. (although if we built a chassis like this, it would probably have an aluminum base plate. )
It’s hard to see in the rendering, but the center wheels (5" banebots wheels) are lowered 1/8".
Actually, having the transmissions in the back has some neat advantages. Since the center wheel is dropped, the rear wheels don’t carry as much load. This is probably a lot nicer on the transmissions since the wheel is is about 1.5" from the transmission bearing. Also, with the chain routed this way (one on the inside, one on the outside) the wheels can sit in closer to the frame.
Can you describe the power transmission components in more depth?
Assembling reliable power transmission schemes is the most difficult part of building a FIRST robot’s drive train. Without learning more about why you’ve selected an unorthodox series of chain paths and wheel locations, I would be hesitant in recommending this to any teams with few resources.
I don’t see your reasoning. The way I’m looking at it, the rear wheels carry more load than if the center wheel were not lowered (up to half the robot’s weight or even more, depending on the center of gravity). They just do it intermittently – which seems likely to cause greater shock loads on the axle when the robot rocks backward.
Overall I like it but why wouldn’t you just run all the chain on the inside? Also Maybe look into 8wd with the transmission centered and chained to the center pair of wheels. It would increase ramp-climbing ability while keeping the same center of gravity.
Well I wouldn’t recommend it yet… I would hope that others could offer some criticism and improvements.
Starting at the transmissions, we have two AM toughboxes with the extended output shaft option. The face of the transmission is attached directly to the wooden framerail. On the output shaft, there is a spacer, then a sprocket (12 tooth AM), and then a 5" Banebots wheel. These are held on with a cotter pin and a washer.
The other axle assemblies are similar, with the sprockets and wheels held onto the 1/2" keyed axles by washers and cotter pins. At the moment, I have the flanged bearings sitting right in a 1.125" hole in the hardwood, but this part that worries me the most at the moment. (maybe an aluminum insert?)
That’s where I have it at the moment, if you see any problems please point them out.
I was making the comparison between two 6WD frames with a dropped center wheel (I might not have made that clear enough in my first post). With this configuration, the end wheels support a much smaller portion of the total robot weight, especially if it’s center of mass sits right over the center wheels.
That is a good point about the shock loads to the transmission, though.
I considered this too, but for this particular version I wanted the transmission face to be flush with the framerail and the wheels to be close to the frame. In its current configuration, the rear wheel is about as close as you can get it to the transmission. (although, it could probably be made to work as you stated as well)
Running all of the chain on the inside would not increase the distance between the wheel and the bearings at all. It would simply require a longer axle on the inside of the frame.
How fast do you estimate it will go?
Direct drive off of ToughBoxes seems a little fast.
Could the wheels be any smaller?
I like the chain on the outside.
It makes it easy to get to the chain and it simplifies the ToughBox mounting since you don’t have to make spacers between the frame and the gearbox.
There isn’t a bearing in the frame at the rear wheel, so the wheel is supported by the transmission bearings. A bearing could be put there though, and then the chain run on the inside. I’ve always thought it was a good idea not to run three bearings on the same shaft though.
I would estimate about 8 ft/sec. A 5" wheel on the toughbox gives about 9 ft/sec at the free speed of the CIMs, but there is going to be some frictional loss.
I am talking about not having the transmission output shaft directly driving a wheel. Maybe put the transmission between a pair of wheels and chain them to the transmission and then chain the middle one to the last unchained wheel. As you have it right now there is a large cantilevered load on the output shaft which I would be wary of.
I am also wary of a bearing bore in wood like you and the few wood designs I have worked on have all used aluminum bearing blocks to hold the bearings.
Wood bases are a totally viable option. Many teams have done them successfully in the past (173, 61, 195 jump to mind).
From my experience, the holes tend to “stretch” over time. We used T-nuts to anchor everything to the base to help with this. Therefore, pressing bearings into the side plates seems pretty risky.
What type of wood are you planning on using? We used 1/2" (9 ply) Baltic birch.
It would be very easy to put a 1/16" wall aluminum tube in the bearing holes, which would eliminate the risk of the framerail cracking/widening around the bearing.
I think the transmission would like it much better if there are bearings supporting the output shaft in the frame.
Tom, the current thought is that the frame members would be oak, and the sheets would be thin plywood.
I don’t see how that would fix the problem. Maybe I’m envisioning this incorrectly?
Also, I’d advise running the chain on the outside of the frame. It’s easy to lose things in this kind of “bucket” frame, and I see bolts, loose wires, and all sorts of things getting tangled in there.
I don’t think using another tube in the hole will really help the problem. You need to have the bearings in some type of plate and then the plate fastened to the wood with bolts or t-nuts to spread the load. The best thing to do that I can think of is to have to plates of aluminum sandwiching the wood and then have them bolted together with bearings in each plate. Also, why use hardwood? Baltic birch is more than strong enough and is a lot more uniform than hardwood lumber. You can always layer 2 pieces of plywood to get a thicker piece.
On a side note. I have been designing a ton of drivetrains with wood and I just keep coming back to aluminum. Wood although cheaper, seems to be a bit more difficult to build and design with to get the same functionality. Maybe I just haven;t come up with the right thing yet but so far my aluminum designs are winning in my mind.