Need suggestions for 6v DC Gearmotor "tester"

WBA Consulting

I am in need of designing a test fixture to screen 6 volt DC gearmotors in volume (500 at a time). Other than a simple apply 6 volts and watch it spin type of method, I would like to somehow do some very basic performance tests on the motor. Using this motor from Pololu as an example, how should I go about setting up a Propeller based tester to verify gearmotor function?

http://www.pololu.com/catalog/product/1118/specs

I am thinking maybe just monitor free-run current and use an encoder to verify RPM is within the specified range?

Maybe use a brake to slow it down from the specified 45 RPM, to say, 35 RPM, and verify that the current change is typical of the rest of the motors in the batch? AKA, validate torque of the motor?

LoopyByteloose

Well, if you take TWO motors and tie the shafts together, a lot more info might be available.

The second motor can provide info about the ability to generate power under load. It's outputs can be loaded with various resistance. You can even use the second motor to brake and stall the first.

Bob Blick created this arrangement to demonstrate the durability of his H-bridge design. But the scheme has been around since dinosaurs made electric motors.

whicker

If you just measure the no-load characteristics, it's going to throw a lot of failures, because ultimately you're looking at the uncontrollable factors, especially temperature. temperature is going to affect current draw and peak RPM. (temperature affects resistance, lubrication viscosity, bearing friction, etc)


you should probably find a way to put a consistent load on the motor and geartrain...

gravity is a good load if you have the space to lift something and reach constant velocity. F = ma, you know mass and you know acceleration due to gravity so you can find force, and thus torque... torque and current are the useful values.

you can also accelerate a significant mass (significant would be at least the armature and motor shaft mass times the gear ratio) and directly measure the motor current and RPM of the output shaft. high pulse count encoders are expensive, though. but it just depends on how much time you have to run each motor. If you have minutes, then you can find an acceptably precise speed and travel distance with even just one pulse per rev using a proximity sensor or photoeye target.

If you do try to backdrive a brushed dc motor, you're going to encounter wear issues. but spinning some sort of mechanical to electrical device is probably the cleanest way of going about this task.


Oh, and what you're trying to do is probably patented:
http://www.google.com/patents/US7044003
yep. Completely obvious what literally anyone would do as a first step. granted anyways.

erco

For a DC motor, peak power is typically at half the no-load RPM and half the stall torque. Verifying that may be a reasonable indicator of overall motor condition. If you can come up with a consistent load (weight/string/pulley, etc) that approximates the torque at peak power, you could simply measure RPM and/or current to see if they are within spec.

kwinn

You could use a suitably sized stepper motor as an alternator and use the ac output to measure the RPM. A variable load on the stepper outputs would allow you to measure speed, voltage, and current draw at various loads. By doing some simple calibrations you could also measure the torque and power output of the motor.

Don M

I made a gear motor test fixture once using a magnetic particle brake for a variable load. They work pretty nice but are kind of pricey.

Genetix

I don't know what these motors will be used for but considering they are so cheap how can you justify doing all this testing. Why not just make a simple fixture where you drop in a motor and basically make sure it works. The specifications are general so you don't what the limits are or how much variation there is.

I personally think it's better just to spend more on a quality part made by a reputable vendor and skip the testing.

WBA Consulting

Everyone,
Thanks for all of the suggestions. The end result is a free run current draw test. The free run current is very stable (+/- 2mA) unless an issue exists (then it increases 20+mA).

Genetix: the justification for the motor testing comes from the fact that the only other way to ensure the motor works is at final assembly which means a 12 minute teardown/rebuild process to replace the motor. The easily negates a 4 second motor test. The motors in use are the best available from both price and quality as well as being from a reputable vendor.