How fast r u gonna go?

[Bruceb]

Don Don Don... I have the motor curves and I have the JVN design calculator.
I know the free speed is 5500 rpm. I am not an engineer. I don't know how to figure what I can realistically expect the CIMs to be turning at at nominal load at speed. Sorry, my calculus is nonexistent, I never had physics and I am just trying to help some kids build a robot. Now, if you are willing to help, great.
Bruce

[theprgramerdude]

Realistically expect them to turn at a speed drawing below 40 amps while cruising, or you're going to have a very dead robot. The data sheets explain 27 amps; use that as a base point.

[JVN]

Originally Posted by Bruceb

Don Don Don... I have the motor curves and I have the JVN design calculator.
I know the free speed is 5500 rpm. I am not an engineer. I don't know how to figure what I can realistically expect the CIMs to be turning at at nominal load at speed. Sorry, my calculus is nonexistent, I never had physics and I am just trying to help some kids build a robot. Now, if you are willing to help, great.
Bruce

If you're geared "reasonably" and your drivetrain doesn't have some ridiculous friction load (or is a Mecanum), the CIMs will run at about 81% of their free speed when at top speed.

Reasonably geared is defined for a drive 4 motors to be "less than 13 fps @ free speed."

-John

[BIGWILLI2081]

team 2081 is estimating they're going to go this fast:

http://www.youtube.com/watch?v=SiMHTK15Pik

[Captaindan]

no matter what this year is about speed strength and the exact manipulation of both a two speed transmission is going to put a robot at an advantage if the drivers can handle the speed ive seen the andymark supershifters go too fast for drivers

[lbarger]

Originally Posted by Captaindan

no matter what this year is about speed strength and the exact manipulation of both a two speed transmission is going to put a robot at an advantage if the drivers can handle the speed ive seen the andymark supershifters go too fast for drivers

It is important to remember that humans do need to control the robot and that they can be made to go too fast to control effectively.

However, it is not quite fair to say that AndyMark SuperShifters go to fast as you do have options in the final gear ratio by selecting different sprockets or even different size wheels. There are also software options to limit the top speed should you find your hardware outrunning your drivers.

[Tristan Lall]

Originally Posted by JVN

If you're geared "reasonably" and your drivetrain doesn't have some ridiculous friction load (or is a Mecanum), the CIMs will run at about 81% of their free speed when at top speed.

John, do you have any information about the test conditions used to establish the 81%? And how much does that vary? (I've always made a similar approximation, but I don't have any hard data to back it up.)

[JVN]

Originally Posted by Tristan Lall

John, do you have any information about the test conditions used to establish the 81%? And how much does that vary? (I've always made a similar approximation, but I don't have any hard data to back it up.)

About 8-9 years ago when I was teaching myself this stuff I designed a drive using 81% as a magic number (I think someone on the forums mentioned this) then tested it -- the experimental matched the theoretical. It has just sort of turned into a "rule of thumb" for the typical style of drivetrains. I know some will be a touch slower and some will be a touch faster, but I've never been pissed off if I design at 81%.

I'm humiliated by how unscientific this is, but it seems to work for me.
I don't have hard data... just 9 year old, soft data...

-John

[ks_mumupsi]

From the last 11 years of seeing a lot of different drive trains, I have to say practice makes one of the biggest difference in ability of drivers to handle speed.

In Overdrive, we used 4 CIMS with a 7:1 reduction using the AM Stackerbox with an ackerman steering, which seemed to work really well, we approximated our drive speed to around 22fps.

Last year, we went a completely opposite direction and designed an 8 wd using 8" skyway wheels, we were an even match for even some of the lowest geared robots. It all depends on whats appropriate for competition. This year on a flat field I see advantages to a 2 speed transmission in gaining speed but I dont think it will help with anything if the drivers cannot learn the change the speeds well on the fly. The shifters now have gotten far better than they were a few years ago.

[team 3311]

here's and idea try making a programing shiter aka dumb down your contlers imput values then have a easy byttion to go to full speed

[team 3311]

here's and idea try making a programing shiter aka dumb down your joysticks imput values then have a easy byttion to go to full speed

[Bruceb]

Well, the reason most folks use a shifter is to have a LOT of pushing power and what you are suggesting will just slow it down.
Bruce

[lbarger]

Originally Posted by Bruceb

Don Don Don... I have the motor curves and I have the JVN design calculator.
I know the free speed is 5500 rpm. I am not an engineer. I don't know how to figure what I can realistically expect the CIMs to be turning at at nominal load at speed. Sorry, my calculus is nonexistent, I never had physics and I am just trying to help some kids build a robot. Now, if you are willing to help, great.
Bruce

Bruce, I find it better to think about how fast the robot SHOULD go than trying to target a specific speed and working out the gear ratio. What is often not mentioned is the fact that all drive motors run through circuit breakers that limit current to 40 amps. If you look at the CIM motor you will see that it's stall current is more like 130 amps. If you stall the motors, you will trip the breakers. If you trip the breakers you are not moving until they reset.

I encourage my team therefore to design the robot so the wheels will slip before the motor current exceeds 40 amps. To do this requires knowledge, or reasonable estimates, of both robot's weight and the coefficient of friction from the wheels.

It will take too long to describe the process in detail here. I have given a presentation on this subject and send it if you PM me with contact information.

The short version (which will likely not help without a background in engineering or physics) is that you need to determine the torque required at the wheel to cause slippage as well as the available torque from the motor at or below 40 amps. The ratio of the two gives you the desired maximum gear ratio. Each stage of gear reduction (gear, sprocket or pulley) could be roughly estimated to reduce the maximum motor speed by roughly 95%. Most gear boxes have 2 stages of gears and then chain to drive the wheels (0.95^3 is about .85). So the speed of the robot SHOULD be something like (motor free speed)*(~0.85 efficiency)*((motor torque at 40 amps)/(wheel torque for slippage))*(wheel circumference). Of course it gets more complicated with multiple motors, multiple wheels and such. Generally safe to assume symmetry and that one or two motors drive half the robot weight for tank or skid steer drive trains.

Now I've probably gone and scared anyone from asking for the 'long' version.

As I said, contact me and I can break it down into manageable size bits of information.