Me and a bunch of other members of team 1073 are going to be designing and building a new shiftable transmission with the help of our mentors. I’m just curious what other teams who had them think about them. Are they worth it, are they helpful for only certain times or helpful in every match. I don’t need blue prints, (if you want to you can post them though :o ). I just want to know if this is going to be worth the trouble, thanks to all those who post, and see you all come next season.
I find it to be very useful. The reason why I like it is because, sometimes you want the extra speed and sometimes you want nimble maneuverability, this gives you both. Also, you can help keep your battery up longer for the match, instead of running high RPM’s the whole match, you can save some for the end of the match. If you need help with the design, some good people to talk to would be, Andy Baker, Paul Copioli, JVN and Andy Brockaway.
We have never built our own shifting transmission, but this year we took the plunge and bought the Andy Mark 2-speed shifting gearbox’s. They were great gave us good speed and power. Easy to mount, very little to worry about as long as you install them properly (lets just say that we killed one :yikes: ). They are defiantly a lot easier than making your own, but they do come at a price :o. Take a look at them: http://www.andymark.biz/am-shifter2speed-gen2.htm
But making your own is always more fun :D.
The head programmer of 1155 and i have been looking into designing our own type of gearbox for the 07 season that we hope to prototype in the preseason in order to educate the new members and get warmed up on building. one thing that we have found very helpful is to look at all the plans and pictures of other two speed transmissions that have been used in FIRST and borrow ideas from them(for example the usage of dog gears in the AndyMark gearboxes).
We did a 2 speed dog shifting transmission both this year and last. While I think it is useful, you need a driver who can use it. There’s no point in building it if your driver won’t shift.
for the past two year’s 340 has had a multiple speed transmission. last year we worked on the Dr. Joe DeWalt XRP transmission (read this for suggestions on how we suggest how do it) while this year we went with the Andy Mark two speed. we find them very useful to the games an love having them. if you need help with the design or anything like that let me know. i have helped out the students with the designs for mounting, custom plates, and other components during both of the years.
My team used a pair of AndyMark two-speeders. We had a pneumatic shifter, but had to take it off to make weight, so we just jammed the shifter into low gear. However, about halfway through our second regional, we added a shifter using servos, which worked ok.
Shifting is very useful. It allows you to quickly traverse the field and still push, pull or climb. I think the design of the game dictates how much you use it though.
I think teaching the students by designing and building the transmissions is a great idea. I think the knowledge gained is invaluble. However just to be that guy and make sure you know what your getting into I want to let you know some of my experience.
You will spend more money to make them than AndyMark’s costs. Just know that right off the bat it has been discussed in many threads here at CD.
Plan out your gear meshing carefully. There are a lot of good white papers and threads to search here about suggest DPs for dog shift vs direct meshing transmissions.
If you use dogs to shift, unless you have REALLY good CNC programers and machines, just buy them from AndyMark. Trust me on this, it took two years of trying before we made the switch.
I suggest trying to work out a servo shifting mechanism as part of the project. The weight payoff should be worth it.
Pete
Not unless you are paying for manufacturing. If you do all the manufacturing yourself, you will save money.
Andymark’s prices on the individual components are really good, but who says they are the size you need? Also, a dog shifting transmission can be made with a milling machine and lathe (if you go square dog and 716 style) or with the addition of rotary table or perhaps indexing head, you could do a 3 “finger” shifting dog. Of course CNC is better, but it isn’t absolutely necessary.
We have a 2-speed dog-shifting transmission that we’ve used in 2003. Back then, our robot was a beast and people often commented on the power of our tiny 14 inch high robot.
It’s got nice power in low-gear and nice speed in high-gear in at 15 lb (for both) package
No, really we did the pricing. I can’t get the raw materials for less. If I can find the thread where we had the discussion a few months back I’ll link to it.
I just meant to suggest designing around a purchasable dog (note I only meant the dog) to make life easier. The mating patterns in the gears are actually not too difficult.
EDIT: I had a total brain fart on this. We couldn’t make a dog it because we couldn’t broach the hex slot.
We made our own 2-speed dog shift transmission this year. It worked great. No break downs. We made all our parts at our school machine shop, no CNC. I’ll post some pictures in the future. One advise, just watch for the weight when designing a transmission.
There are three approaches to using a shifting transmission:
- Buy your own (AndyMark)
-Yes: No effort needed
-No: Expensive - Make your own using another team’s design
-Yes: Many different designs to choose from
-No: Access to a machine shop and material - Design from scratch, figuring gear ratios and sizing
-Yes: Build the perfect transmission- No: Requires a machine shop, material, complex project needig time
I should mention that in the last two years teams have been using modified DeWalt XRP cordless drill transmissions on their robots. These are 3-speed shifters are worth researching. Search the white papers for Nothing But Dewalts (.pdf) 
Everything I was going to say has been said. except:
If you build a transmission that uses a dog mechanism to switch gears, I would suggest purchasing a couple extras (or making them as the case may be) so that if you are at a competition and one breaks, you will have a spare handy. Just a thought.
It might be worth the weight, but is it worth the delayed shift time and possibility of breaking?
Our team is planning to work on a 2 gear transmission as well in the off season… the details haven’t been worked out yet but it’s floating around
I dont have much experience with shifting tranys, but can you ellaborate on the delayed shift time that occurs from servos? 
Our team is also planning on designing and prototyping with a 2 speed this offseason. We bought the AndyMark 2 speeds and we found them to be expensive, yet very reliable. We have the gear ratios and calculations for torque and what not down. And now just - CAD TIME.
We’re hoping to have a sponsor machine most of the stuff…that is unless we get a CNC Mill sometime soon.:rolleyes:
If you can perfect the shift on the fly and have your low gear have high torque and your high gear have good top speed, you will most likely be a force to rekon with. Reason why? Most 2 speed transmissions have pretty bad low end torque and good top speed. the problem with this is that if you get into a push match then your extra gear is pointless because you cant get away to use it. so a high torque low gear would help you get out of push matches which of this year we all know from now on will happen, and a top speed high gear will help you get away from your oppentes with ease. so the choice is up to you. personaly that is a lot of stuff that has to work in order to just get the robot moving not even to mention the rest of the robot that has to complet the game. my advice, buy a 2 speed transmission from andy mark or some other robotics company, build it and see how it works. then if you think that you can do what a professional company could do it and make it run well in 6 weeks then I say go for it. But as for me and my team will let the big first dogs do that and we will stick with our simple yet effective transimission which has so far got us to the finals two years running. Good luck with your transmission and hope it al works well in the end. 
one cause for delay would be the small difference between servos, so that one side shifts before the other, and kinda causes the robot to go in a not-so-good direction.
I developed my own two speed recently and i’m currently busy integrating it into a mock up drive train in autodesk. The transmission looks good on paper (or in this case LCD), but i’ll have to see how it works in reality, with weight and costs and such. I’ve thought of my own little tricks for shifting and all, but those are my little secrets. As for the advantages to multi-speed transmissions, i think it is a great direction to go in. They destroy the generalization of function on the robot and open up the window of specialized functionality available in multiple scenarios on the same robot. In other words, if you’re in a situation where you need to have detailed control over the robot, or you need power, you can get exactly that and nothing extra to get in the way. And then if you need the highest possible speed with no power, you can have that as well. Simply put, the two-speed’s fuction has no downfall, it’s only the approach towards the two-speed that can result in lost time, money, and weight.