Page 255 - Robot Builder's Bonanza

224 CHOOSING THE RIGHT MOTOR

• If the lever is 1 inch long, and the weight successfully lifted is 2 ounces, then the motor is
said to have a torque of 2 ounce- inches, or oz- in. (Some people reverse the “ounce” and
“inches” and come up with “inch- ounces.” Whatever.)
• Or the torque may be stated in grams, not ounces. In this case, a lever calibrated in centi-
meters may be used. This gives you grams- centimeter, or gm- cm.
• Torque for very large motors may be rated in pound- feet, or lb- ft.
• Becoming more popular is the newton- meter unit of torque, slowly being adopted by
motor manufacturers. You may see it as N- m or Nm. One N- m is equal to the torque that
results from a force of 1 newton (no, not the fig kind) applied to a lever that is 1 meter long.
(If you’re interested, the newton is equal to the amount of force required to accelerate a
mass that weighs one kilogram at the rate of 1 meter per second per second.)

STALL OR RUNNING TORQUE
The typical motor is rated by its running torque— that is, the force it exerts as long as the
shaft continues to rotate. For robotic applications, it’s the most important rating because it
determines how large the load can be and still guarantee that the motor turns.
Manufacturers use a variety of techniques to measure running torque. The tests are imprac-
tical to duplicate in the home shop, unless you have an elaborate dynometer and sundry other
tools. Instead, you can empirically determine if the motors are sufficient for the job by con-
structing a simple test platform, as described in the next section.
Another torque specification, stall torque, is sometimes provided by the manufacturer
instead of or in addition to running torque. Stall torque is the force exerted by the motor when
the shaft is clamped tight. The motor does not turn.

JUDGING THE TORQUE OF MOTORS

If the motor(s) you are looking at don’t have running torque ratings, you must estimate their
relative strength. There are formulas you can use to rate a motor for a specific task, and these
have been detailed in numerous books on robotics; try Building Robot Drive Trains (Clark and
Owings, McGraw- Hill, 2003). In truth, this is what most people do: they mount the candidate
motors on a makeshift platform, attach wheels to them, and have the motors scoot the test bot
across the floor. If the motors support the platform, start piling on weights. If the motors con-
tinue to operate with the additional load, then you know they’re suitable for the job. (This is
called empirical evaluation. It’s how Thomas Edison did most of his inventing.)
You don’t need to build the whole robot just for this test. Employ a temporary construction
using lightweight materials, such as heavy- duty cardboard or artists’ foam board. See Chapter
14, “Rapid Prototyping Methods,” for more information. With the techniques outlined in that
chapter you can quickly test various motors and robot platform designs.
Such crude tests make more sense if you have a standard by which to judge others. If
you’ve designed a robot before and are making another one, you’ll already know what kind of
motors work for a robot of a general size and weight.

Testing Current Draw of a Motor

You can often just look at a motor and know it’ll have enough torque for your robot. Less
ensured is knowing how much current the motor demands when it’s running. It’s not possible,
or even advisable, to infer the current draw of a motor just by its size, shape, or type.

21-chapter-21.indd 224 4/21/11 11:50 AM