Hey, Guys.
A while back I posted that we say a fair bit of variance in some batches of motors. I could pull a Co27 out of the box and see 28,000 RPM or I could see 30,000+ RPM. In the most recent batch(es) that we've seen this has not been the case. Here's a good post from Big Daddy off of RCTech:
A couple more things on the tuning/voodoo motor front that came to mind after reading the replies:
People underestimate how important the sample size is when you're making decisions about setting up a motor or making an adjustment to your current tuning regime. That's the other thing that can be counter productive with the stand-out motors from a batch. The notion that all of the other motors can be "tuned up" to match the best one from a batch or a hypothetical target does not work in practice because of the variances in winding of the armatures and magnets in the can and so forth.
We try to mitigate these differences with things like zapping to ensure the magnets are all at full strength before assembly as well as unifying variables like brush hoods, spring tensions, and so on. Other things like the position of the magnets in the can, and the winding of the armature are effectively parts of the motor's "DNA" that will be carried with it for it's lifespan.
If you use the scientific method of changing one variable at a time across a broad set of samples, you can see the effect that your ideas have on the motors much better than fixating on what makes a particular motor the fastest, because often, the "DNA" is the major contributing factor and you can't change that in the shop.
By using a specific process across a broad sample range and experimenting with very specific ideas, we can shake down what actually has an impact on the performance of a group of motors, which is a better indication of long term performance characteristics. The odd freakshow motor is interesting to see what you can glean from it, but more often than not my experience is that it's the "DNA" of the motor as opposed to anything that I would claim I have done to it.
The other thing is that the plot of all this looks incredibly similar to the "bell curve" which we've all seen before. On typical run, we get a large number of motors in the same range, the odd one above, and the odd one below expectations.
On the flip side, we have a lot of loyal customers that come back because the motors are "always fast", and that's exactly what we're shooting for by monitoring trends on the larger sample sizes.
The other big difference is that we run our own dyno software which I wrote for doing this exact thing.... It actually tracks every pull I've ever done for each type of motor we produce. This is an awesome database for me to draw on when I need to look at the overall performance expectations for a given type of motor and so forth.
T
It's so true! We regularly receive calls from guys asking for the "best" motor we have and, well, when they've been through the process, many actually have
identical numbers in some parameters while the vast are/were very, very close to one another. This is a new development from what we have seen and speaks to the sample size issue that Trent speaks of.
We've also had guys asking for *special* brushes and *special* springs - honestly. One race I was at recently, everyone was after these magic black springs - turns out they were the OEM springs from a while back and the spring rate was the same or close to the same as many I had in my own pitbox but, nope - they were *special*. The brush thing was funny. We have actually spoken with the manufacturers about this and it's just bizarre.
Interesting!
P.
