07-07-2007 12:25

=Martin=Taylor=

Re: pic: DeWalt drive base

Correct me if I'm wrong, but isn't it risky to use two shifters on the same side?

Suppose one of the servos stops shifting... Then you'll have two gearboxes chained together running at different speeds. The DeWalts will strip themselves, and you'll lose half your drive train.

Yes, it is an unlikely scenario. But its the kind of disaster that could ruin your day.

07-07-2007 12:59

Jeff 801

Re: pic: DeWalt drive base

and why 8 40 amp fuses and 10 victors

07-07-2007 14:42

Richard McClellan

Re: pic: DeWalt drive base

It wouldn't be quite as risky if you didn't chain all 6 wheels together...just connecting the middle wheels to the back wheels...I would probably try that first and see how it drives before added a second set of chains. On the other hand, I have seen other teams like 418 use this configuration successfully with no shifting problems

07-07-2007 16:16

Re: pic: DeWalt drive base

I'm intrigued by your frame design. What made you do the two triangles?

This would make manufacturing it take a bit longer, as well as welding it slightly more difficult. I personally would have just run the bars straight across, which would have also given more area to mount to for other mechanisms.


Looks like and interesting idea, good luck with it!

07-07-2007 20:46

dbell

Re: pic: DeWalt drive base

This is a possibility. If the servos do not shaft at the same time one or more of the following might happen:
The Dewalts might break (they are very strong)
the CIM might burn out (unlikely)
the sprockets might break
the chain might break

If this happens it will be bad.
The plan is to have another 1 or 2 spare side modules ready to go on. It will be able to come on and off in just a few minutes, to we will be ready to drive again in the next match.

If the drivers realize that this is happening during the match they might be able to shift one dewalt into "neutral" (between 1 and 2 or between 2 and 3) and use the other one.
I really don't know how likely this is.


This might help. If the wheels didn't shift at the same time the wheels would slip a bit, but still much less harmful.
I chained them all together because that way each of the 6 wheels has the power of 2 CIMs. If you did what you are saying you would not have as much power to each wheel itself.

I haven't finished the electronics yet.

Yeah, the triangles do add some complexity. The reason i didn't put one bar across is because that way there is a rectangle. I usually try to use triangles when possible because they are stronger in general. Your idea may be better though, I'll think about it.....


DB

07-07-2007 21:44

newton418

Re: pic: DeWalt drive base

Slick design, I'd love to see pictures of the real version, and perhaps some battle reports, so make sure to keep us updated after next season.

418 has worked with a similar design for the last 3 years, and the only time we've had a shifting problem is when the backup battery wasn't charged... so none of the DeWalts shifted.

IF one of them didn't shift, then the DeWalt that didn't shift would just be powered by the one that did, so it shouldn't create any significant damage (though it will be quite obvious something is wrong). If this happened, I'm sure the driver would just shift back and stick with that gear for the rest of the match. But again, this is a very unlikely scenario if you follow the whitepaper word for word and your code does what you want it to.


We did this our first year with the DeWalts, and learned a valuable lesson.
If you have one DeWalt powering 2 wheels, and another powering 1 wheel, you will have a significant power bias. Think of it this way, you get into a pushing match, your back wheels have their own CIM-DeWalt to themselves, and the middle and front wheels share another. While the back DeWalt easily delivers enough torque to slip your back wheels (creating significant wear on the thread, and reducing your traction) your other DeWalt struggles to provide enough torque to the the middle and front wheels. I'm sure you COULD distribute the weight such that the traction of the back wheels is the same as the combine traction of the middle and front wheels, but that seems unnecessarily constraining.

Another advantage is that if one of your back wheels isn't contacting the ground, the wheels that are contacting the ground still have the power of both CIM-DeWalt assemblies. This might seem like an unlikely scenario, but if you lower your center wheels (as most teams with 6 wheel drive do), and your center of mass changes (you pick up a heavy game piece, for instance) then before you realize it you're driving around on your center and front wheels... and half your power is uselessly spinning wheels that aren't contacting anything.

08-07-2007 09:51

ctorloey

Re: pic: DeWalt drive base

This is a very cool, interesting design.
Do the Dewalts work good on your robot?
Was it hard to make the Dewalts and put them on?

08-07-2007 12:48

Dan Petrovic

Re: pic: DeWalt drive base

How strong does this frame have to be for a FIRST Competition?

1/8" wall box tubing is very strong. A rectangle would work just fine. You've seen our robot drivebase this year, right? All rectangles and we never had any problems with it bended (even when people stood on it for various reasons).

11-09-2007 21:55

robostangs548

Re: pic: DeWalt drive base

That CAD looks amazing, nice work!

11-09-2007 22:18

StevenB

Re: pic: DeWalt drive base

We haven't built this design - it's just a concept at this point. I'm not even completely sure it's possible to do (we may not be able to fit the third stage sun gear onto the CIM).
It took a fair bit of work to modify the DeWalts, but it was still less than any other 3-speed transmissions I know of and that we were capable of making with our limited resources.

11-09-2007 22:57

AdamHeard

Re: pic: DeWalt drive base

If you can't keep that stage, it wouldn't be too much extra work to move each dewalt inward, and have a sprocket reduction to the outer wheel. Then keep the center wheel connected in the same fashion.

Keep refining this until it is 100% ready for production, it will save your team a lot of time.

12-09-2007 11:28

Qbranch

Re: pic: DeWalt drive base

I'm not a big time drivetrain guy... but can a DeWalt transmission take the torque of two CIM motors? Perhaps you could combine their torque before going in to the transmission?

Or, to avoid the 'losing half the transmission and stripping them out' problem if one shifts and the other doesnt (or other reasons why one might sease) can't you just turn the torque clutch down on the drill to a reasonable number so that in case you start to backdrive the transmission too hard it just slips? I don't know how much torque the clutches go up to in those things before they go direct drive, just a thought.

-q

12-09-2007 12:42

AdamHeard

Re: pic: DeWalt drive base

118 Ran four small CIMs and two FPs into one dewalt this season.

12-09-2007 13:00

newton418

Re: pic: DeWalt drive base

I got to talk to some of the guys from 118 at LSR last season. They told me they'd had some problems with the gears in the DeWalt shattering, but I don't know exactly why this occured, or how frequent it was. Also, I remember them having several backup V6's. Nonetheless, they definitely proved that it can be done. Hopefully one of the guys that worked on the '07 118 drive train could fill us in on what they found out about the DeWalts.

The DeWalts have been shown to handle the torque of two CIMs, but since only a few teams have done it, we don't have much information on how well they hold up. So many teams have followed the whitepaper and have never touched their DeWalts after that, and that reliability is part of what makes the DeWalts so valuable. Also, in dbell's design he's keeping the first stage of planetary gears, so there's already 3 times more torque on the transmission than what most people have used.
As for the torque clutch, it's typically removed with the CIM configuration, since we remove the first stage of planetary gears. Since this design is keeping those, I'm not sure if he's going to keep the clutch or not. Either way, in a high torque application (like the drive train), you probably wouldn't want to mechanically limit your power unless you really knew what you were doing.

dbell - are you planning on keeping the clutches? If so, you should read this: http://www.chiefdelphi.com/media/papers/1680. They kept the clutch in and found that it bent after a couple of matches, causing them to service the transmission frequently. If you aren't going to keep them, how are you going to keep the first ring gear in place?

16-09-2007 19:27

dbell

Re: pic: DeWalt drive base

It might save time.
This would be a possibility, but a significant change to my current design. But part of me wants to try using the extra stage. If it works it could be great. Using a chain reduction would a significant amount of weight and space.

The DeWalts would almost certainly be strong enough. But it would give up these advantages of my current design:
-weight balance of CIMS
-if any chain were to break, only one wheel would lose power
-does not require an extra gearbox to gear the 2 cims together

I am not planning on using the clutches. I'm not sure what you mean by "how are you going to keep the first ring gear in place?"

DB

18-09-2007 17:10

newton418

Re: pic: DeWalt drive base

I didn't spend time trying to figure out exactly how the clutch works, but from what I remember it seems to me like the first ring gear in the DeWalt transmissions are kept in place by the clutch. When too much torque is applied to the transmission, the clutch gives in, allowing the first ring gear to slip. If you plan on using the first gear, but you are going to remove the clutch, there is nothing stopping the first ring gear from spinning freely. Thus, you would need to secure the first ring gear in place.