pic: 6 speed?? Great in theory - lets see how it works!

[Rod]

Something to think about.
With all that spinning mass could this become impossible to turn due to gyroscopic forces?

[JVN]

Originally Posted by Andy Baker

That three speed shifter was from team 968, with Travis Covington as the main designer.

Andy B.

Ya sure he's not talking about 222?


It sure sounds like it.
But 968's 4-speed was pretty sweet too.

[Matt Attallah]

For the shifting..

what if you use a globe motor? They have lots of torque and using a rack and pinion style mechanisim - you shouldn't be backing out anytime soon...?

Maybe even have some limit switches above the positions of the gears so you don't have to worry about timing?

Looks good though. Mind if I subsuite it for some of the nastier tranny's I see here in the automotive world?

[Madison]

I'm going to chime in on the side of people who know a lot more about these things than I do and agree that a ball screw seems like the best option for shifting to multiple positions. However, since I despise programming and electronics, I'm hesitant to rely on switches and sensing devices to accurately place the shifting collar within the gearbox. That's not to say it's impossible to do it, but only that I do not have the experience to make it work myself.

Instead, to reduce the accuracy in rotation needed by a conventional motor to spin a ball screw and accurately position the shifting collar, you might consider using a geneva mechanism (see also) as an interface between the shifting motor and the ball screw. An indexing device such as that should help to alleviate concern about the shifting collar's position.

Also, it's really cool.

[Sean Schuff]

Originally Posted by Tytus Gerrish

why would you ever need a 6 speed transmition in a First Robot?

Just a proof of concept. The reality is you could change it from a 6 speed to a 10 speed to a 2 speed all by removing a few gears and shortening the shaft lengths.

We're going to prototype it with 2 gears for starters to see how it works and slowly add more gears.

This will prepare us for when the field quadruples in size and we want to get to the other side in 6 very theoretical seconds!

More screen shots coming soon.

Sean

[suneel112]

In theory, it is possible to have an infinite stage gear. Put threads on a cone and put a FIRST gear (the sideways ones in the kit) on top of it. Adjust its position with a weak motor, and you should be able to deliver the exact amount of torque that you want. I will put up drawings later when I scan them in (yep, i'm a paper and pencil designer), and I may even put it in a white paper. whether it will work.....

basic idea (very ghetto):
/
O/ The O is the sprocket, and the < is the conic gear.
/
/
\
\
\
\

[Rickertsen2]

If you had free reign to buy whatever you wanted, shifting this thing pneumatically would be very easy. With the parts available to FIRST, i would forget the idea of pneumatic shifting. I can tell you now that if you go that route and try to position a cylinder in that many places soo close together and still have reliable shifting it will not work. It would be much easier with something like a rack and pinion or leadscrew mechanism with a globe motor or similar. The one thing you need to watch out for if you use a motor is that you need to make sure that you are never stalling the motor. You will need to add a spring or something that will compress until the gears mesh.

[Katie Reynolds]

Woo more pictures

[Adam Y.]

In theory, it is possible to have an infinite stage gear. Put threads on a cone and put a FIRST gear (the sideways ones in the kit) on top of it. Adjust its position with a weak motor, and you should be able to deliver the exact amount of torque that you want. I will put up drawings later when I scan them in (yep, i'm a paper and pencil designer), and I may even put it in a white paper. whether it will work.....

It's possible to have an infinite stage gear system but Im not sure if what you say will work. It's called a cvt or cct. I forget.

[Matt Attallah]

Originally Posted by Adam Y.

It's possible to have an infinite stage gear system but Im not sure if what you say will work. It's called a cvt or cct. I forget.

The pitch and number of teeth would change as you move down the gear...

You really can only acheve infinante gears with a belt - which isn't too good when you have to deal with FIRST rules (Even in vehicles - after so much torque the belt starts to slip. That why you don't see more of them in V-6's and V-8's)

[Sean Schuff]

Originally Posted by Adam Y.

It's possible to have an infinite stage gear system but Im not sure if what you say will work. It's called a cvt or cct. I forget.

CVT - continuously variable transmission.

See http://www.andersoncvt.com/ for a pretty slick variation. I'm not sure if I found this site by poking around or if someone linked to it in a post on CD. It looks good but seems fairly large. For FIRST applications you would have to scale the design down enough to make it practical yet keep it as large as necessary to be robust.

Lots of awesome input! I love how these boards work for all of us!!

Thanks all!

Sean

[ahecht]

Maybe I'm missing something, but all the gears in your rendering look the same size.

[Hinkel Y.]

Originally Posted by Sean Schuff

CVT - continuously variable transmission.

See http://www.andersoncvt.com/ for a pretty slick variation. I'm not sure if I found this site by poking around or if someone linked to it in a post on CD. It looks good but seems fairly large. For FIRST applications you would have to scale the design down enough to make it practical yet keep it as large as necessary to be robust.

Lots of awesome input! I love how these boards work for all of us!!

Thanks all!

Sean

A similar cvt is this by Pat Roche posted up on CD a few months ago...http://www.chiefdelphi.com/forums/pictures.php?s=&action=single&picid=7945&direction =DESC&sort=date&perrow=4&trows=3&quiet=Verbose

Wouldn't there be a problem with the FSB (?) that was described on the A+CVT that the belt just slip right off or shear if it push another robot? If it uses a chain like the A+CVT, the FSB would have to be really strong if it hits something at high speed cuz it could bend or snap.

[sanddrag]

Originally Posted by ahecht

Maybe I'm missing something, but all the gears in your rendering look the same size.

I was about to post the same thing when I realized it about an hour ago, but with such detailed discussion (by myself included) into the actual method of shifting and such, I was embarrassed that I simply was missing something. But if it doesn't seem right to you either, maybe this is seriously flawed, or we both are missing something.

Even if the gears were different sizes, the bright green gear in between the pink "slider" would have to be of varying size each time it shifted.

A transmission of this style can difinitely be fabricated, but it woul dhave to be done so in a matter similar to that of 222, that ball and plunger design.

[ahecht]

Originally Posted by sanddrag

I was about to post the same thing when I realized it about an hour ago, but with such detailed discussion (by myself included) into the actual method of shifting and such, I was embarrassed that I simply was missing something. But if it doesn't seem right to you either, maybe this is seriously flawed, or we both are missing something.

Even if the gears were different sizes, the bright green gear in between the pink "slider" would have to be of varying size each time it shifted.

A transmission of this style can difinitely be fabricated, but it woul dhave to be done so in a matter similar to that of 222, that ball and plunger design.

I thought the same thing you did, and actually sat on it for 24 hours before I realized what the designer was trying to do and posted my comment.

However, the transmission is possible with different sized gears and minor modifications. Lets call the input shaft "shaft 1" and the other shaft "shaft 2". Now, if you stripped the hubs off of all the gears on shaft 2, and attached a larger gear to each of the six existing gears (with all six larger gears being the same size), the gear in the slider could interface with the larger gears by sliding back and forth.

Although that isn't what is in the drawings, as far as I can tell, that may have been the design intent.