Custom Two-Speed Transmission

[tennispro9911]

Our team is going about designing a two-speed transmission. We are planning on designing it similar to an AndyMark supershifter, with some AndyMark gears and some gears that are purchased from other suppliers.

At this point we are trying to decide on gear ratios for our high and low speed, and what teams have decided is their "optimum" high and low speed. We used AndyMark 8" mecanum wheels on our robot for 2008, and they worked beautifully for us. We plan on using them again next year. Because we are using mecanum wheels, we will have four independantly driven wheels and four transmissions. We intend to direct drive our wheels off of the output shaft of the transmission.

Right now we have a 4:1 ratio between our high and low speeds. We are using the specs of the CIM motors at "Normal" load to give us a fairly realistic maximum speed that our robot should be able to reach in a competition setting.

We are assuming a .90 efficency for our gearbox because all the stages have bearings and it uses external spur gears. We are adding .002" center to center distance for our gears to reduce friction. Is this a good distance?

This will be our third season as a team, and our first foray into both custom transmissions, and shifting transmissions. We considered AM transmissions, but we did not want to purchase four of them, we hope that we will be able to make them lighter, we want the knowledge from building it, and we wanted to be able to customize the gear ratios for our robot and our wheels.

[Dolan]

That sounds like a cool idea and good luck

[tennispro9911]

Thanks

[Alan Anderson]

In order to give a good answer to your question about "optimum" speeds, I need to ask a question in return.

Why do you want two speeds on a mecanum drivebase?

[thefro526]

It seems like you guys have a good start. One of the things I would suggest that you further research is your need for a second speed in your drivebase. From my experience, and that of other teams, Mecanum drives do not really play enough hard contact defense to need two speeds. Now if you wanted 2 speeds to have 2 different levels of control I could understand. In a few of my mecanum prototypes I made high speed drives that were very hard to control with out a closed loop system but, lower speed ones were easier to control.

But for the gearbox, it sounds like a really good plan but it might not be suited for a mecanum application.

[AndyB]

Originally Posted by Alan Anderson

In order to give a good answer to your question about "optimum" speeds, I need to ask a question in return.

Why do you want two speeds on a mecanum drivebase?

Seconded.

[tennispro9911]

Well, to answer that, I'll tell you what we did this year. This year we attempted to hurdle the ball, but honestly, we failed miserably in doing that. So in the middle of our regional we changed to being a lapbot. We could go about 13 ft/sec so we were above average for speed, and we were selected for eliminations on the 4th seeded alliance. In eliminations, we played hard contact defense of the form that, in your words, mecanum bots don't usually play. Although we had never planned for it, we ended up being really good at it. We won the FLR with 20, 191, and when 191 broke, 174.

This is what Steve Warren from 188 said about our defense:

"I believe that I saw some of the best defence played this week at FLR. The team that did a fantastic job happened to be on the winning alliance. Team 2053 played the part perfectly shutting down some great robots on the way to victory. They were so good that their opponents got penalties trying to play around them. If and when videos become available watch the elimination rounds and especially the matches that Moe played against 2053."

http://www.chiefdelphi.com/forums/sh...759#post715759

1126 was the first team selected on Curie and we played defense against them during the finals of our regional. They also won the Buckeye regional.
67 was also at FLR and did not reach the finals of the regional, but did read the finals of the championship event. HOT also won the Great Lakes Regional ... the only regional 217 lost.
1507 was the first seed at FLR with two loses. They also competed at the GTR where they only lost once in qualifications ... to 1114 AND 2056 on the same alliance. Somewhat understandable. They paired up with 365, MOE and we played defense against MOE.

We were very lucky to have some great alliance partners in 20, 191, and 174. However, in the first round our opponents had a Offensive Power Ranking (OPR) of 49.4, but averaged 34 against us. In the semis, the first seeded alliance averaged 92 points in their first round but 67 against us. In the finals, the alliance averaged 81.2 points previously and had an OPR of 85.9, but averaged 65.3 points against us.

Mecanum bots may not have a reputation for playing effective defense, but even though we were geared for 13ish ft/sec, there were very few times we were pushed around. Also, in this year's game there wasn't nearly the emphasis on physical defense that there was in some previous years. Even if we choose to have an offensive robot next year, we will almost always need to be able to hold our location against teams that play physical defense.

Thats why we would like the capability to push others around. Never know when it will come in handy.

[tennispro9911]

So what gear ratios/speeds do teams use, disregarding the mecanum wheels?