Robot Dynamics Tutorial- Units query

[timstirling]

I've been trying to find out what sort of motors would be required by my robot in order to carry around a laptop and some cameras. The tutoiral was helpful in revising my physics knowledge but I have a question about the units used.

Specifically with the Robot Motor Factor.
The equation examples have the Torque in lb/in but the acceleration and velocity are in ft/s, i.e there is a mix of inches and feet going on. Is this correct, doesn't seem right to me unless I'm missing something.




These are my numbers:
Mass: 6-8kg/13.2-17.6lbs
Motor Choice 1: 200rpm, 4.6kg/cm = 25.8lb/in
Motor Choice 2: 120rpm, 8.87Kg/cm =  49.7lb/in

Motor 1 RMF = 15.3 kg.... / 86   lb...
Motor 2 RMF = 17.74kg..../ 99.4 lb...


Now, using similar velocities as in the example equations= 3ft/s and 2ft/s^2
~= 90cm/s and 60cm/s.
Metric: 6*90*60/(2Pi)= 4751....
with your imperial equatgion and ft/s: 11.6.......


Why is the velocity in feet and the torque in inches?
 

[Admin]

Oops! I wrote inches where I should have written feet in several places. I even had another mechanical engineer check before I posted this and she didnt even notice it . . . I guess I was inspired by NASA 

Anyway it should be corrected now.

I probably should have written this using the metric system . . . its just Im american and although it is easier I believe to calculate in metric, it is easier for me to think in the english system.

As long as you stay consistent with unit use, your RMF will always be:
Torque * rps > = Mass * Acceleration * Velocity / (2 * pi)


So I did the calculations, and I got the same motor RMF's as you:
86 lb in rps
99.4 lb in rps

And finding the minimum RMF required, I got:
english
(13.22lbs * 3ft/s * 2ft/s^2) / (2 * pi) = 1817.88 lb in rps
metric
(6 * 91.44 * 60.96) / (2 * pi) = 5322.95 kg cm rps

Those motors are way below your spec . . . you might just want a velocity of 1 ft/s and accel of 1 ft/s for a minimum RMF of 303 lb in rps. Then find a motor about 3x better RMF than motor 2. The problem you will find is that stronger motors are expensive and weight on your robot means money out of your pocket . . .

Anyway let me know if I pulled a NASA again . . . g'luck!

[timstirling]

Ok, thanks. So I haven't gone bad.

This is one time when i wish i was wrong though. Looks like it is going to be very difficult to get hold of motors powerful enough to lug the laptop around. And if I do get motors powerful enough then  the motor controllers I was hoping to use wont be able to take the current.

back to the drawing board! 

[Admin]

use a microcontroller <cough> 

[timstirling]

use a microcontroller <cough> 

i may well use a microcontroller anyway, but the main purpose of the robot is to do some advanced vision processing, sound localisation and recognition and neural-SLAM type stuff.

[Admin]

Why dont you have the robot communicate wirelessly with a computer instead? They got many wireless webcams out there. Your PC can then just transmit the highly processed data back to a MCU on ur bot . . .

Search this forum for wireless, webcam, and bluetooth. Others have asked bout it too . . .

My next tutorial is bluetooth wireless communication with a robot . . . should be done within a week.

[Kohanbash]

while not always the best advice but if you are on a tight budget if the current is above the motor controllers rating (within reason) you can usually heat sink it and not have any problems.(I have a controller rated at 1.8A and it has routinely handles 3A for the past 7 months)

[timstirling]

I've had major problems finding appropriate DC motors in UK. I'm thinking that if I replace the SLA battery with NiMH I'll save close to 4lbs weight. The other thing is that the robot will basically always be 2WD and so the torque requires can be divded between the motors.

With this I should just about be able get the laptop powered robot to travel at ~1ft/s.

http://www.robotmarketplace.com/marketplace_machinelab.html
The Machine lab MMP-5 has 4wd, each motor I'm pretty sure is identical to the motors I can buy in the UK (under the motor section they reference the use in their MMP robots, the spec is almost identical to those I can get in the UK).

The MMP5 robot weighs 4.5lb and has 4lb rated pay load. That weight will be similar to the weight of my robot if I use NiMH cells. They rate a speed of 3.5 ft a second. If I want a mere 1ft/s and have only 2 motor this should work out. Although it does ignore the fact that the 3.5ft/s is most probably unladen.

Of course, there is always the 4wd option.....

[Admin]

Consider lithium batteries, they are lighter than NiMH, but will cost more . . .

Yea with 2WD, each wheel just needs half that torque . . . Just note that it is possible to have an extremely low torque yet still have a high peak velocity. When I first started making robots I had trouble with two things more than anything else - finding motors, and motor drivers - so I completely understand.

But from what you mention, it should all work anyway. Let me know how it goes!