Vex Motor Speeds - New vs Old and Shaft Encoder vs IME
Let me start this off with a disclaimer. This is not meant to start any issues just to show facts. The way I tested this was by setting up a motor with a shaft encoder on the motor output. For the first test I only used the shaft encoder. For the second test I compared the shaft encoder to the Integrated Motor Encoder (IME).
The first test was performed in order to determine if motor velocity decreased with age. When testing a new motor, the average velocity for forward was 123.75 RPM and backward was -126.3 RPM. When testing an old motor (avg 5-10 hrs of use), the average velocity for forward was 126.35 RPM and backward was -123.975 RPM. When testing another old motor (avg 20-25 hrs of use), the average velocity for forward was 112.475 RPM and backward was -118.3 RPM. These results seem to suggest that the motor speed decreases as the magnets weaken with prolonged use.
A second test was performed in order to verify the velocity of a new 393 motor as recorded by both a shaft encoder and an IME. According to the product page, a new standard Vex Motor should spin at 100 RPM. In my test, it actually spun at 122.625 and -121.775 according to the Shaft Encoder and 122.75 and -122 according to the IME. The new high speed Vex Motor should spin at 160 RPM. I found that it actually spun at 207 and -202.125 according to the Shaft Encoder and 205.475 and -203.075 according to the IME. Similarly, a new turbo Vex Motor should spin at 240 RPM. I recorded the velocity at 312.45 and -312.675 according to Shaft Encoder and 311.35 and -312.675 according to the IME.
All data and graphs available on Dropbox
What was your measurement technique? What was the battery voltage? The nominal speed is at 7.2V but free speed is generally a little higher even at that voltage.
I agree that this is a critical question to be answered if we want to get a real assessment. I would even like to see a test with the same motor using batteries charged to different levels.
Another thought would be to assess multiple motors as the same time and repeat the tests and switch the ports into which specific motors are plugged.
This is a great start, though. Thanks for getting the research started.
I don’t necessarily think the root cause is the magnets weakening; that’s not really what magnets tend to do. It is probably other mechanical wear in the system, such as the gears themselves, the motor bushings, windings, etc. And without testing the same motor as it wears, you can’t totally eliminate the variance inherent to brushed motor manufacturing (roughly +/-10%?). I’m also curious to hear how you controlled battery voltage across these tests.
Another significant factor. These are brush type DC motors and not particularly expensive.
Both the brushes and commutator will wear. Interestingly the brush contact area might very well increase a little during initial run time as they wear to the commutator shape. Forward/reverse characteristics are often affected by brush wear and brush holder precision. It’s also possible to position brushes to bias performance in one rotational direction by compensating for armature reaction but very unlikely the EDR motors are designed this way.
Summary: Expect variation in your motors from many sources.
Thank you for all the responses. This is how I tested the motors. The IME was only used for the second test. The motor was swapped for the first test. The tests were performed on a motor with no load. I didn’t think to record the battery voltage so I don’t have that. The batteries were charged and our average fresh battery is about 8.0 Volts. I will see if I can do more tests but I can’t make any promises.
I had heard way back in 2012 that the motors had a relatively low expected lifetime to make sure they were affordable. I think they have gotten more affordable since then too.
I seem to remember a 20 operating hour expected life. This test may validate that.
If you have access to a bench power supply you could keep the voltage much more constant. I know others like @jpearman have used them in the past for tests. And more motors to test against would help validate the results too. But you will end up burning through your motors best part of their useful life in the process. Is it really worth a few hundred dollars of motors in tests?
I agree that motors do, in fact, weaken over time, and I don’t think the sensors are off, I think that the some sensors have more friction than others.
