Crab drive question

[Tazlikesrobots]

I've been playing with a EDU kit and I am trying to make a crab drive, but only using two motors. Is this possible? If so, does anyone have a drawing they can share?

The reason for the two motor restriction...My remote control only has 4 channels, and I need two of them to operate the arm and the claw.

[sanddrag]

You can use omni wheels or casters on two of of them (assuming you plan to have 4 wheels).

[mechanicalbrain]

two motors for what part? two for the wheels?

[NoodleKnight]

Originally Posted by mechanicalbrain

two motors for what part? two for the wheels?

Read what he said:

Originally Posted by Tazlikesrobots

I am trying to make a crab drive, but only using two motors

Originally Posted by Tazlikesrobots

The reason for the two motor restriction...My remote control only has 4 channels, and I need two of them to operate the arm and the claw.

He's trying to build an entire co-axial/crab swerve drive with only using two motors (powering the wheels and swerve mechanism with two motors). I can't give specific drawings, sorry, but I will vouch that it's possible. I made a single crab module (then I sort of gave up making the drivetrain altogether) with the 2003 edubot kit. The bevel gears were sort of tricky, I just mashed two gears together, lol, the module was sort of ugly looking too.

Or you could just have one single powered module and 2 or 3 omniwheels.

[mechanicalbrain]

Yeah if you want one motor to run your drive train than you are going to have to do some fancy gearing, especially where steering is concerned. Is it a matter of a lack of motors, too much weight, or just to see if its possible. I know you can do it with three motors pretty easily but it wont be very powerful or it wont be fast (or both will be mediocre). On a side note i want to see a drive train that doesn't have a direct motor to wheel system (what i mean is having a system where you have a torque to speed ratio).

[sanddrag]

Oh, I didn't realize he was trying to make it centrally powered. In that case, yes it is possible but I think you need bevel gears. Also, you will not be able to have differential (tank) steering.

You may want to look at Chief Delphi's 2000 robot http://www.firstrobotics.net/00galle.../047-1_jpg.htm I think it is similar to what you are trying to do.

[NoodleKnight]

Originally Posted by mechanicalbrain

Yeah if you want one motor to run your drive train than you are going to have to do some fancy gearing, especially where steering is concerned. Is it a matter of a lack of motors, too much weight, or just to see if its possible. I know you can do it with three motors pretty easily but it wont be very powerful or it wont be fast (or both will be mediocre). On a side note i want to see a drive train that doesn't have a direct motor to wheel system (what i mean is having a system where you have a torque to speed ratio).

I believe "Tazlikesrobots" was making an EDU bot for prototyping/tinkering, it doesn't have to meet any weight specs and so forth. He's bound to two motors because of his controller (which reminds me, do you have more motors? Because you can do away with the coaxial junk and just make each wheel have their own motor and use a pwm cable split 4-ways).
And I don't exactaly understand by "a drivetrain that doesnt have a direct motors to wheel system" -- do you mean a shifting transmission? or some drivetrain that transfers power to the wheels via viscous fluids?

[Jeff K.]

You can probably pull it off using the EDU kit. It would just be really complicated and also A LOT of the little plastic chain. You can make a little rotating platform for all of the four wheels and then from there, have their rotating be controlled by one motor that is connected to all of them by chain, and then on the rotating platform....wait...nevermind.
I just realized that it would be overly complicated and it is late...
But maybe you can instead use three servos for drive, one on each wheel to drive and then one to control the rotation of it, but instead of having to use three channels, you can use a y-cable for the PWM cables so they both are on the same channel.

Well...that's my two cents.
That projects sounds really cool. Good luck and I am interested in seeing some pictures on the finished product.

[mechanicalbrain]

Originally Posted by NoodleKnight

I believe "Tazlikesrobots" was making an EDU bot for prototyping/tinkering, it doesn't have to meet any weight specs and so forth. He's bound to two motors because of his controller (which reminds me, do you have more motors? Because you can do away with the coaxial junk and just make each wheel have their own motor and use a pwm cable split 4-ways).
And I don't exactaly understand by "a drivetrain that doesnt have a direct motors to wheel system" -- do you mean a shifting transmission? or some drivetrain that transfers power to the wheels via viscous fluids?

I mean like on a car. i believe its because of the toque slipage something or another. Basically its the reason you can have RPM stay the same even though your car increases in speed. im not real familiar with the mechanics of cars but ive been going into depth with a cars drive train and its really nice!

[NoodleKnight]

Originally Posted by mechanicalbrain

I mean like on a car. i believe its because of the toque slipage something or another. Basically its the reason you can have RPM stay the same even though your car increases in speed. im not real familiar with the mechanics of cars but ive been going into depth with a cars drive train and its really nice!

That's for car's with a CVT drive -- the transmission control module tells the transmission to slowly increase the gearing inside the peak of the car engine's torque curve, so the car appears to accelerate even without varying RPM's.

[mechanicalbrain]

I'm not talking about CVTs all cars can do it (i think). this isn't during gear change just normal running of the car!

[NoodleKnight]

Originally Posted by mechanicalbrain

I'm not talking about CVTs all cars can do it (i think). this isn't during gear change just normal running of the car!

Maybe your talking about an RPM range of 0-1000 rpm? Cars equipped with automatic transmissions have a torque converter, which transfers power via viscous fluids and other fun mechanics. Other than that, I don't see how you could accelerate without increasing RPM's, unless you happen to be going downhill, or you brake launched your car.

[mechanicalbrain]

Originally Posted by NoodleKnight

Maybe your talking about an RPM range of 0-1000 rpm? Cars equipped with automatic transmissions have a torque converter, which transfers power via viscous fluids and other fun mechanics. Other than that, I don't see how you could accelerate without increasing RPM's, unless you happen to be going downhill, or you brake launched your car.

YES YES thats what i was thinking of! i asked my dad to explain the mechanics of automatic transmissions and this was one of the things he pointed out. if it wasnt for viscous fluids (which im sure are illegal) it would be cool to see this on a bot. Anyhew back on topic, as i said the problem with using one motor to drive with crab drive is that i imagine the weight of the crab part is going to slow your bot to a crawl. of course i could be wrong and if i am im sure i will soon find out!

[Tazlikesrobots]

I think I might have someting working now. I used two omi wheels placed in the center of the base in the form of a T and put casters in each corner. It works for the mos part, but I need to tweak the software an slow down the ramp up speed of the motors...