Hey all,
I’m looking at getting the Predator 6 wheel sumo kit, but I’m wondering - would it be possible to put PGHM-20 motors on it instead of the GHM-0X motors it comes with? I know power isn’t everything with a robot, but I need to beat my computers teachers robot, and the thing has enough torque to move the labs laser printer 
Thanks!
Grecko
Well, yes and no.
First off, the Predator is designed to use the shorter motors. One motor per wheel. On the inside, the motors are opposite to one another. If you replace them with the PGHM motors, they won’t fit anymore. They are more than twice the size (in length) of the GHM motors. It will be like trying to fit 2 motors in the same space.
My champion, Pete’s Minion, sumo is the only U.S. made robot to beat a Japanese Sumo in an official competition, and it uses 4 of the 50:1 7.2 volt GHM motors. I run them at 8.4 volts using relays for the motor controller, so I don’t see any voltage losses from mosfets.
I just looked on Jim’s website, and I don’t see the same motors any more. The ones I see that has the same specs have a shaft encoder, which again will get in the way. But there are ways around that.
Buy the 30:1 gear motors and the 50:1 gear motors, and swap the gear heads. It is easy to do that. Jim will probably hate me saying this, buy some higher gear reduction gear heads from Jameco and swap them with the 30:1 bodies that Jim sells. You want torque here, not speed. I believe they have a 78:1 gear head. Just make sure you get the same size (i.e. diameter). The gear heads are easily interchangable. 3 screws is all that needs to be changed.
Then to get speed, run the motors at a higher voltage. Jim’s 7.2 volt motors have a better current handling capability than the other voltage ranges.
Now if you are set on the PGHM motors, I would recommend the PGHM-13 over the PGHM-20. Far better torque characteristics. I would run them at 8.4 volts. Speed would be down a bit, but the torque will be still better than the PGHM-20 motors. The problem you will run into is you will have to make special axels for the robot because the motors will be in different locations. Every other wheel would connected to a motor. I am not sure if you will have all of the space.
The other thing to look into is traction on the wheels. Either use the Urethane wheels that Jim sells, or smear a thin layer of RTV on the wheels the night before the contest. One of the reasons your teacher can move that printer is that his robot has good traction.
I would be interested in knowing more about that robot. I like to what the competition looks like.
Pete
What about using the Cobalt 400 motor? I assume you would want to use some kind of gearbox with this - where would a person find one to fit this motor?
Grecko
I have used Lynxmotion PGHM 8 - 10 gearboxes on Speed 380 motors. I think that the Speed 380s are very similar to cobalt 400. But, I don’t see PGHM 8 - 10 on Lynxmotion site anymore. I have not run that combination much, so I can not comment on reliability.
In any case, the resulting motor/gearbox would be too wide for a stock Preditor chassis.
Are you thinking of a legal 3 Kg sumo or just a robot for push?
What kind of mechanical or electrical aptitude do you have for converting a standard Lynxmotion product to something unique? Is it to be Autonomous or R/C?
For 3 Kg sumo robots, I don’t think that you could do a whole lot better than one of the Preditor or Terminator kits with the GMH02 motors and molded urethane tires. Buy the kit, build it, play with it, then decide about modifications.
If you want more power, I’d suggest the GMH-13. Even that motor is probably too long for stock Preditor/Terminator chassis, but it is very close.
Rick BrooksThe problem here is that I have one chance to convince the school that a robotics club has a chance of working. I have until the end of the semester (about another 8 weeks) to convince them that students can build robots for a reasonable price (basically under a grand) without a huge amount of background knowledge, that can then actually compete.
So I need to follow all the standard 3kg sumo rules, and I need to do it in a fairly short amount of time. Hence using a prebuilt chasis.
I have a fair amount of knowledge with embedded machines, microcontrollers, and that end of the project. The robot I’m a little fuzzy on, but I’m a quick learner with way too much time on my hands.
So I need a robot with extreme power, traction, etc that I can then install one or more of my controller boards onto and make autonomous. The predator seemed like a good choice, but if the motors are so limited, maybe not.
Is there another chasis out their that could meet these requirements?
Thanks!
Grecko
I think you are looking at this all wrong.
I have been competing in Sumo for about 10 years now (even wrote a book on it). The most important thing about sumo is having the robot work, function properly, not drive off the ring, reliably seeing the other robot, not falling apart, not resetting, etc.
In most of the competitions I have competed in, and have seen, the most successful robots are not the most powerful robots.
You need to make sure that the robot will:
You do this, and it will generate interest in the event/club. People who have never seen a sumo robot are amazed when they can see it say on the ring, and follow an opponent (or even a hand or a broom). If it constantly attacks, they are impressed. Believe me, I know this is true because I take a lot of sumo robots to different types on non-robot related events, to show off, and they are impressed with just seeing them stay on the ring.
I have seen too many people so focused on power/speed, they fail to make sure that items 1-4 above are working at 100% reliability. In fact, they just assume that it works without any real testing. And these people all have one thing in common, they lose.
Forget your focus on power and spend all your time on just making the robot work reliably.
If you have the funds, I would suggest that you get at least one more sumo kit, a Viper, kit. This will show different style of robot bodies to help people’s thinking. The Viper kit is my favorite, and they are very competive. I have a couple, and they do quite well, and they are under weight.
All of my 3kg sumo robots are under weight by at least a 1/3. This is because they work, and work well. Most of their competition has been focused on power, and my robots exploit them for it.
So, forget about the pushing power of your teacher’s robot. Make your robots smarter. Build several. Have a small tournament with them. This will get people’s attention. Even if they loose, it doesn’t matter, if the meet the 4 items above, it would be a great success.
Pete
Grecko,
If you want to produce a working 3 Kg sumo robot in 8 weeks, then the Stomper Sumo Combo Kit for Basic Atom is for you. It is the most complete 3 Kg sumo kit that I know of. Buy some sensors of your choice (there are pros and cons to any sensor) and build a sumo that can find and push a 3 Kg block of wood. With some microcontroller experience and some practical knowledge in sensor integration, you should be able to accomplish all of that in your time frame. No electronics or mechanical aptitde is required.
Once you get it running, then Molded Urethane Tires will give you more traction that you can handle. Remember that the amount of push that a sumo can exert is limited by the amount and stability of the tires’ traction, not by the torque available from the motors.
Then you will have a good, basic sumo robot. Refine the basic robot by attending a lot of events and working on the robots operation/response and reliability in various situations encountered. And, don’t forget to have some fun with all of this stuff.
Rick BrooksJust a thought:
If you’re going to be building with the goal of going up against a specific bot, are you going to have a chance to observe it in action ahead of time? If you know the placement, coverage, and types of sensors it’s using, and the behaviors it exhibits while running, perhaps you can use those to your advantage while designing yours.
Admittedly, designing your bot to counter a specific bot’s design isn’t a good general strategy in a competition type of scenario, but since in this case you know ahead of time that you’ll be going up against only one specific bot, with a very limited timeframe to work in, knowing these things might allow you to focus on some specific areas that you know will be important, while saving others for later.
Incidentally, why in the world would a teacher pose this sort of challenge so close to the end of the school year? It’s hard enough to get students to wrap their heads around anything as technical as a sumo bot when given an entire year to plan for it.
I just transferred into his class a few weeks ago, and we started talking about robotis (I’m huge on computers and always wanted to get into robotics). He said he had a bot, and wanted to start a robotics club, but that the administration didn’t think it was feasible. It went downhill from there 
If I can beat his bot, the administration will allow us to try it next year. If not, oh well, we move on… for the time being
On the idea of making the robot speicifc to one competitor - if your competition was going to be smaller than you, would conventional sumo rules allow you to use an arm on the bottom of the robot to pickup your opponent?
Its just a though
Grecko
Grecko,
The CIRC Central Illinois Bot Brawl on May 6 was won by a Stomper Sumo Kit with 12v 30:1 motors and urethane tires.
My three sumos couldn’t beat him.
Have you ordered your kits from Lynxmotion yet?
RickI haven’t ordered my parts yet, no.
I’m really thinking of going scratch built - moreso for the other competitions (minesweeper especially), haivng a scratch built robot gives you a huge advantage, not only in terms of bonus points, but also in the regards that you can design the robot to be adaptable to a certain task.
What I’m thinking is a robot that starts out as a U, with tracks along the two outside edges. Then when the competition begins (for sumo), it folds down, giving it a width of about 30cm. This way, the front and back can fold down into ramps, and the middle can be completely hollow. If the opponent is not too big, it could be ‘swallowed’ up and carried to the edge, to be pushed out by a large track or arm. Even if the other robot is too big to be swallowed up, having a hole in the center of the bot should make it much harder to detect by conventional means.
For the minesweeper challenge, the center would be a perfect place to store the tags and the collected objects, as well as equipment for the second and third stages.
For just having around the house, the storage space means he could go grab me a coke from the fridge and bring it to me… or maybe even with two good arms and some good programming, he could make me a latte… 
Grecko
Certainly an interesting idea, and one worthy of trying out, but remember the old adage “Quick, cheap, or powerful. Pick two.” With only a few weeks to design, build, test, and refine your robot, it is probably better to go with a tried and true design that is already very solid and well supported. Trust me when I say that you WILL run into unexpected problems that will hamper your progress.
Mike
I’m not to worried about time. If I don’t get it done this year, it’ll get done next year. Either way, the robot club is never going to move fast enough for me to join in (I’m finished next year), so as long as I get it done before I graduate… everyones happyish
I’m now just trying to find the best way to do it…
Grecko