22-01-2010 02:20

dtengineering

Re: pic: 1885 Drive Train 2010

I'll be interested to see how the prototyping pans out, but wood is great for whipping up some models to test on the bump.

We have an 8 wd mock up that we tested in both wide and long orientations before deciding to give up on the wide orientation.

It looked good on the computer... but having something physical to manipulate made all the difference.

Jason

22-01-2010 09:06

Raul

Re: pic: 1885 Drive Train 2010

Are you planning on actively articulating the wheel pods or just letting them conform to the ramp?

22-01-2010 09:42

hipsterjr

Re: pic: 1885 Drive Train 2010

My, that frame looks familar

22-01-2010 09:52

JesseK

Re: pic: 1885 Drive Train 2010

The current plan is to let them conform to the ramp. If there are a few pounds left over at the end of build season, we may add some sort of simple tension system to put a traction bias on the inner 4 wheels, yet all 8 wheels would still stay on the ground.

The chassis is based off of this concept that was thought of before the C-Base towers and Toughbox Nano's came out. That concept was based off of this concept of a natural terrain-hugging suspension.

22-01-2010 10:11

MrForbes

Re: pic: 1885 Drive Train 2010

It's a neat design. I wonder about the turning ability though...we are working feverishly to get our 8wd prototype running (hopefully today) so we can drive it around on carpet and on the bump. We are keeping the wheels fixed, but with spacing kind of like what you have--not evenly spaced, center wheels further apart longitudinally than the end wheels. With the end wheels raised just a bit.

We really like how 6wd robots drive, but were concerned about the bump, and we think the irregularly spaced, drop center wheel 8wd might be a good compromise between all the designs.

22-01-2010 11:39

scott

Re: pic: 1885 Drive Train 2010

HEY! I've seen this before...somewhere...

22-01-2010 12:55

CHapstack

Re: pic: 1885 Drive Train 2010

We made pretty much this same prototype last week. We figured that the main difference between this and a 4wd like the kit bot is that the fulcrum at each corner stays closer to the ground than the axle in a 4wd would when going over the bump (so the robot doesn't tilt as much). If you spread the 2 wheels at each corner apart more, the fulcrum will stay even closer to the ground.

22-01-2010 14:51

artdutra04

Re: pic: 1885 Drive Train 2010

If you are planning on using a free-spinningn tri-wheel design (all three wheels are powered, with the carrier automatically rotating over obstacles), make sure the input shaft and wheels rotate in the same direction and make sure the sun gear is significantly larger than the planetary gears (e.g. gear the wheels faster than the input shaft).

This means you will have to gear the carrier input slower (only to gear it faster at the wheels, I know it's an ugly solution), but the torque exerted on the individual wheels needs to be less than the torque needed to rotate the entire tri-wheel carrier assembly. If this is inverted (small sun, large planetary), the robot won't go anywhere with the tri-wheel carriers just rolling over themselves.

22-01-2010 14:58

Madison

Re: pic: 1885 Drive Train 2010

If you're confident that the innermost wheelbase is short enough to allow unfettered turning, you might consider putting omniwheels in all of the outermost locations instead of developing the tension system you're talking about.

It's largely a trade off between time and money, I guess.

22-01-2010 15:09

MrForbes

Re: pic: 1885 Drive Train 2010

I was thinking about using slick wheels on the outer ones, but omnis are a better solution.

22-01-2010 15:13

Madison

Re: pic: 1885 Drive Train 2010

In your case, however, be careful with liberal application of omniwheels. If y'all are using a fixed 8WD and if climbing the bump is your thing, you'll have some circumstances where only omniwheels are contacting the ground.

22-01-2010 15:22

acidrain2012

Re: pic: 1885 Drive Train 2010

looks like someone else had almost exactly the same idea we did

22-01-2010 15:34

MrForbes

Re: pic: 1885 Drive Train 2010

That's true. In our case we have the end wheels raised up (which you can't do very easily with the articulated design) so traction wheels all around is best.

22-01-2010 15:53

Tom Line

Re: pic: 1885 Drive Train 2010

A word of caution. If, during a pushing match etc, the torque of your wheels causes them to rotate the assembly and lift your bot up an inch or two, your bumpers will be out of the legal zone. This may be deemed illegal and you may be DSQ'd (if someone decides to protest). You may want to look at some sort of active actuation to prevent it.

22-01-2010 19:18

Travis Hoffman

Re: pic: 1885 Drive Train 2010

Indeed....

23-01-2010 03:06

Valkyries_698

Re: pic: 1885 Drive Train 2010

we are doing something similar but with a slick wheel at the front,

We have been researching a system called Rocker-bogie

23-01-2010 18:20

fordchrist675

Re: pic: 1885 Drive Train 2010

I think you should ship the robot as is! I am sold on the wood chassis haha!

Looks like a ingenious idea.

23-01-2010 19:12

hyperdude

Re: pic: 1885 Drive Train 2010

Mars rovers, right? Funny thing, I came up with something almost exactly the same (prototype design).

23-01-2010 19:15

Re: pic: 1885 Drive Train 2010

Tanks have been using a Rocker Bogie system since long before the rovers. Go take a look at the old Shermans, those are neat pieces of rolling coffins!

23-01-2010 19:42

Valkyries_698

Re: pic: 1885 Drive Train 2010

well this rocker-bogie was worked on by someone that is on the game design team

24-01-2010 17:48

Raul

Re: pic: 1885 Drive Train 2010

I believe the main advantage to articulating the wheel pods is to maintain a more level chassis while going over the bump.

Look at the wooden model in the picture Jason provided. If the rear wheel pods are articulated (rotated) so that only the rear-most wheels are touching, the angle of the main frame will be at a lesser incline and provide a greater safety factor to prevent the center of mass from getting close to the tipping point.

24-01-2010 21:01

JesseK

Re: pic: 1885 Drive Train 2010

Interesting point, especially considering that the bumper zone rules aren't as strict while on the bump. Our primary goal with the conforming articulation (not sure what to specifically call it) was simply to maintain a controlled climb, peak, and descent. Doing so, we believe, will reduce the risk of tipping over. If anything, the heightened clearance requirements of this natural articulation may elevate the c.g., yet we are doing some very careful calculations (one of the mentors understands them ... I don't, heh...) to ensure we're under the tipping threshold.

For those interested, the actual drive train weighs in at 45lbs. Yet that's also with a subframe capable of withstanding some pretty high stresses due to high tension. We feel the extra weight is a good trade-off for stability & reliability.

29-01-2010 01:23

waghalter

Re: pic: 1885 Drive Train 2010

jesseK is right,

as the student leader of the design team building this, along with our mentor's help, we found that the pivot has some extremely strong forces acting on it. the potential for the pivot axel to get bent or damaged is high if the robot's design does not build in enough support to stiffin the axel.

so please consider this if you are doing a similar design. best wishes team 698. Good luck !

29-01-2010 12:58

Daniel_LaFleur

Re: pic: 1885 Drive Train 2010

Incorrect. The bumperzone is defined with your wheels touching the ground.

If you are pushed in such a way that your wheels are not touching the ground (like being pushed up a ramp) then the bumperzone no longer applies.

If it were as you stated, then any time a robot went over the bump they would gather a penalty.