pic: 2003 TechnoKat Ball Drive : Patent Pending

[Clark Gilbert]

Prototype: Here is the prototype Andy talked about. This picture was taken the night of kickoff.
http://www.pictars.com/032003/DSCF1676.JPG

While I think this is great and very cool, I just wonder if this years task was so simple that we can afford such lavash things. What about other teams who can barely move? Were the TechnoCats right to spluge on their drive while many teams still don't move?

We do help people, by donating our robots to other teams, and by producing our "white papers" that we post here every year.

http://www.firstrobotics.net/03galle...1024-1_jpg.jpg


Slipping/Traction: Traction really wasn't meant to be an issue with this robot. Sure it needs some traction to push a stack around or move around, but it's meant to avoid situations where high amounts of traction are needed. Trust me, you probably wont see this robot trying to push anyone off the top of the ramp.

Like Andy said, hopefully our main robot does well enough (like we have much doubt in it), but if a perfect alliance could be made by using the ball drive, I'de bet we'de be all for it.

If you want to see the Mighty Mouse, come to our pit at St. Louis, Pittsburgh, Midwest Regional, or the Championship and you will be able to either drive it (if there are not too many people, or see a demonstation of it by someone).

[Austin]

I think he means dropped from a significant height. The 1/4" thick urethane coating, which he mentioned, would absorb any small forces acting on the balls.

Yes, there is a very dented ball at the shop that we play around with sometimes, and try to dent it further. To put any significant dent at all in it you must THROW it down pretty darn hard.

[Beast314]

Well done. This will give me another thing to check out in the pits at St. Louis. I have also worked on a similar idea in the past along with a few classmates, using a much smaller wheel though. The idea was lost in the wind when we saw no use for that type of system last year, but I'm glad to see a workable variation of it.

[Mimi Brown]

wow. thats all i have to say. oh yeah...and i cant wait to see it in pittsburgh!

[abeD]

i remember someone in the thread saying that they saw a video of that, is there any way we can get a link to some type of video?(especially for those of us that won't get to see it in person)

[Cory]

Originally posted by JosephM
Why not take some of the money that would have gone to building another bot or a more complicated drive system and give it to a struggling team or a new team? [/b]

How far would 700 bucks really get you towards helping another team?

TechnoKats: super cool. Im amazed that you could pull that off. I hope I end up at Nats and can see that baby work.

My Hat is off to you =)

Cory

[Ricky Q.]

Andy, Clark, DJ and all the TKats

Sweet! I had everyone around me in my History class (when we we're typing an essay test ) just amazed, I think it made some of them consider joining the team ....Excellent job, as we've all come to expect...I'm looking forward to looking it over and seeing you all in Evanston....

We all tip our cheeseheads to you

[Clanat]

Congratulations on a very innovative design! I want to see this thing in action now

[Clark Gilbert]

I'm not personally going to post any video of the ball drive till after the Pittsburgh regional. Having 2 regionals in the next 2 weeks makes me very busy, along w/ various school work. There is video, and there will be video. For now you will just have to enjoy the pictures. Sorry

[Joe Matt]

Originally posted by Cory
How far would 700 bucks really get you towards helping another team?

TechnoKats: super cool. Im amazed that you could pull that off. I hope I end up at Nats and can see that baby work.

My Hat is off to you =)

Cory

AHAHAH!!! I like it too, it's just that mabey they could....

oh, forget it...

[Ellery]

Andy and Team #45,

My hat goes off to you guys once more being the drives fanatics that you guys tend to be year after year! It's definitely an interesting design that's worth investing extra resources if available. Especially this year's game definitely pushed our team to chose the extreme end of simplicity compared to our extreme complexity we had last year. As long as it keeps students involved and free from boredom it's a great excuse to make.

See you guys soon!

Ellery W.

[Dima]

well actually this is probably the BEST soluction to this years game: Have two robots each one optimized best for what its supposed to do!

Technokats i think it Will be woth it for you to buy another controll system and a set of motors so that before any given match you can have a choice of what to bring to your alliance!

Imagine being paired with another good kind of the hill robot. Obviosly your alliance need to have a box stacking bot to make that optimal stack. so you just put your other bot in that match!

now that you have both bots if i were you i would do just that!

Oh thats aside from the point that you win the Xerox Creativity Award! (or at least have very high chanses to do so)

[Andy Baker]

OK... I'll try to answer some questions and explain some things further.

First of all... the patent.

Our management here at Delphi saw the prototype of this design and strongly suggested that we patent this design... not to protect us from other FIRST teams to use, but rather future companies. This will be a Delphi patent, once it goes through it's process. Currently, it is a "record of invention", and we are protected to show it to the public.

Natchez is wise. The biggest reason we are looking to protect ourselves is for the use of something like the Segway. Imagine a one-ball balancing Segway. This is exactly why our management wants us to protect it. We currently have an agreement with Segway (of the details, I have no idea) that encourages Delphi to help develop technology for the Segway... so this is a good thing (every Segway has about $400 of Delphi electronics in it).

Rickersen2 - you are exactly right... it is the opposite of a mouse. This is the easiest way to explain it.

JosephM - you are right also. Your concern about the 2nd level of friction loss is very valid. This is really the biggest hurdle of this design. The fact that the small drive wheel has friction losses to the ball and the ball has friction losses to the floor is not trivial. This makes the "ball drive" definitely not a very efficient mechanism. Our prototype would stop and start because of these losses (and an out of round ball). JosephM - your questions are valid, and you have good points. I appreciate your comments.

Katy - the balls dent if we drop them from about waist high, but they have not dented while driving up the ramp. If we ran into a 4x4 at full speed, they might dent... but we have not tried that yet.

Ricky - it is great to hear that other people might get on your team by seeing this. If find that suprising, but if it is true, then our goal for this drive base has been achieved.... it's all gravy from here.

M.Krass and others - I am not sure what you mean by the stall issue. This device works REALLY well when both small drive wheels are moving. The urethane ball drives nicely at a 45 degree angle, but it drives less efficiently at 0 degrees or 90 degrees. It works, but has less power. Also, at straight forward or backward, the lateral drive wheel does not spin. The main reason we decided to do this was to use the shifting gearbox for the front & back drive direction. Optimally, we would rather use 4 gearboxes (one on each aluminum drive wheel) and have each drive wheel positioned at a 45 degree angle. That way, if both wheels are turning, the robot will go straight forward and backward. But, we could not afford (weight, cost, time, etc.) 4 gearboxes. Thanks for your comments.

Dima - good ideas... but FIRST has alot of rules against this. Update #4 was a killer. It really set us back. There is actually a post on the FIRST Q+A forums that says a team should not re-deploy a motor to another subassembly, even if it is the same motor (and not a spare). This ruling is ridiculus, in my opinion. We will abide by the rules, but we will need explainations.

For the rest of you: thank you. Come by our pit, and I will introduce you to the guy who came up with the main part of this design: Mark Koors. He told me of this idea last summer and I simply laughed at him. My first concern was the same as JosephM's (the 2 levels of friction issue). But... the more Mark and I worked the issues, the more they cleared up and the simpler the design got.

Simplicity - Krass is right... this omnidrive is extremely simple, not only in the mechanics, but especially in the software. No chain linked axis turns, no lookup tables... simple. You who are wrestling with crab drive software/hardware would cry to see how simple this is (yet another reason to patent it).

More to come... sorry about rambling. I surely did not heed my own words to be concise.

Andy B.

[Ryan Albright]

how does the balls work ont he ramp are you shure the ramp wont puncture them caust he ramp ripped apart are wheels from last year

[Madison]

Originally posted by Andy Baker
M.Krass and others - I am not sure what you mean by the stall issue. This device works REALLY well when both small drive wheels are moving. The urethane ball drives nicely at a 45 degree angle, but it drives less efficiently at 0 degrees or 90 degrees. It works, but has less power. Also, at straight forward or backward, the lateral drive wheel does not spin. The main reason we decided to do this was to use the shifting gearbox for the front & back drive direction. Optimally, we would rather use 4 gearboxes (one on each aluminum drive wheel) and have each drive wheel positioned at a 45 degree angle. That way, if both wheels are turning, the robot will go straight forward and backward. But, we could not afford (weight, cost, time, etc.) 4 gearboxes. Thanks for your comments.

Andy B.

Andy, thanks for taking the time to answer so many questions

"Stall" was probably the worst choice of words in the history of the English language, but I still can't come up with anything better. I guess that's what happens when you're discussing something that's never been seen before.

Akin to what you mentioned to JosephM. . . .there exists the possibility that the losses in the two unique components of the drive train aren't equal. That is, whatever losses the omniwheels experience in the gearing, chain, sprockets, contact with the carpet, etc. may not be equivalent to the losses that exist in the ball mechanism; the six points of contact on each ball, the gearboxes, etc.

As the ball's forward rotation is tied to the omniwheels rotation since they're in powered by the same gearbox, I was wondering how you accounted for any imbalance in losses that might exist. After all, if there are more losses on the ball than there are in the omnwheel, wouldn't it spin slower? ...and then, wouldn't it just be dragging and do nothing to "power" the robot?

It just seems like whenever there exists a chance that one "wheel" in a side of 4WD drive system might turn slower than the others, it'd be better off to make that "wheel" unpowered, or otherwise find some way to balance the losses so that they all spin at the same speed.

Does that make more sense?