PDA

View Full Version : [Question(s)] Motor sizing for 4 wheeled robot



alphaCentaur
01-21-2014, 11:34 AM
Hello all !

This is my first post here and I would appreciate your help.

I am trying to find suitable motors for a four wheeled robot and doing the following calculations:-

Robot Characteristics
==============

weight(w)=50 kg
wheel diameter(d)=8 inch=0.2032 m ==> wheel radius (r) = 0.1016 m
required velocity (v) = 1 m/s
required acceleration (a) = 0.1 m/s^2
inline angle=0 deg
no of wheels = 4
expected efficiency = 0.85 ( 85 % )


Calculations
=======

required force (F) = m x a = 5 N
required Torque (T) = F x r = 0.508 Nm
required Torque per wheel (Tw) = T/4 = 0.127 Nm
required torque for 0.85 efficiency Tw=0.149 Nm ==> Tw_kgf= 0.0152 kgf-m
required rpm = (V*60)/(pi*D) = 93.989 rpm ==> rps= 1.567 rps

I have been told by my supervisor that these calculations are not correct.

I cant figure out what I am doing wrong ...

jwatte
01-21-2014, 02:59 PM
At a minimum, you are not compensating for friction (unless that's baked into the 0.85 efficiency factor.)
Also, you are assuming an absolutely flat floor -- no incline climbing.
That being said 0.1 m/s/s is a very slow acceleration rate. Are you sure that's what you need?

tician
01-21-2014, 08:56 PM
If you trust your assumptions, using one of these (http://support.robotis.com/en/techsupport_eng.htm#product/dynamixel/ax_series/ax-18f.htm) per wheel would be overkill by an order of magnitude.

...but there is little chance you will get 85% system efficiency. Motors alone, maybe 85% efficient if decent quality, but smaller motors tend to be less efficient unless you are willing to pay a very high price. Rolling resistance of wheels, gear train losses, etc. can drop system efficiency quite a bit. Also, there is little chance the ground will ever be completely level, so definitely design for a minimum of 10~15% grade to overcome smaller obstacles (would need ~18N/wheel just to hold position on 15% incline). And I expect you are really going to want to be able to stop the robot a lot faster than 0.1m/s^s for safety, so either brakes or much larger motors.


I consider the four 17W gear motors powering the lab's ~4kg omni-wheel rover to be significantly underpowered (~26N at each wheel at stall). It was able to get up to ~0.6m/s in a second or two carrying (and driven by) a DARwIn-OP, but have not tested that since the chassis (~2kg) and battery (3kg) upgrades. Fully loaded, it would be well below 50kg and I am not sure it could achieve a pathetic 0.1m/s^2 acceleration.

alphaCentaur
01-22-2014, 12:27 AM
thanks for the replies ...

@jwatte ... yes this is my requirement at the moment, friction is being ignored

@tician ... i only need to make sure my calculations are right at this point