John in CR
100 TW
Looking at Astroflight's motor data changing the turn count of a brushless motor seems to have results that are easy to predict http://www.astroflight.com/pdfs/3210WEB.pdf.
Their charts show proportional relationships of turn counts with Kv, no load current, current at maximum torque, but they compare them all at the same rpm, torque, power, and efficiency. They just vary current and voltage inversely. Maybe I've read something wrong into some of Biff's posts, but it seems like some of the performance relationships don't correlate as cleanly as Astro indicates.
I bring this up because I want to do something different. I want to free a smaller direct drive hubmotor from it's wheel, and rewind it to a Kv that is 3 to 4 times higher, with the intention of running it at much higher power and using a simple and quiet chain reduction to the wheel. I'll ventilate the motor so the extra heat won't be an issue. Easily obtainable and cheap controllers aren't available in high enough voltage or I'd just skip the rewind and triple the voltage.
One question I have is about saturation of the stator. If a stator with 8 turns on it's coils reaches saturation at 50A, does it require 100A for the same motor at 4 turns (assuming equal copper fill)?
Is it reasonable to think that the rewound to a high Kv hubmotor is likely to be more efficient? It's my understanding that due to their low speed operation the majority of a hub motor's losses are copper losses, and I get a double benefit of both thicker wire on each turn and a shorter length of wire. That tells me that cutting the Kv in half reduces copper losses to 1/4th, so even if iron losses double due to double the rpm, I should come out ahead since the iron losses were smaller to begin with.
Am I choosing a good motor to try this with, because it's different from my other hubmotors? The other hubbies have a higher slot count than the number of magnets, which seems to be the trend for newer hubbies like 9C and the new Xlytes with 51 slots and 46 magnets. This older model motor has 40 magnets and 36 slots on the stator. It has a longer stator too, with the stack of laminations 40mm thick, so the end copper as a % is less than a motor like a 9C that has only a 27mm thick stator, but a larger diameter to make its torque. The main reason for the choice is this motor will be only 6.75" in diameter once I cut off the spoke flanges, a full 2" smaller than the 9C without flanges, but capable of the same power in-wheel.
Other than the physical difficulty of a rewind, are there any pitfalls to the approach?
John
Their charts show proportional relationships of turn counts with Kv, no load current, current at maximum torque, but they compare them all at the same rpm, torque, power, and efficiency. They just vary current and voltage inversely. Maybe I've read something wrong into some of Biff's posts, but it seems like some of the performance relationships don't correlate as cleanly as Astro indicates.
I bring this up because I want to do something different. I want to free a smaller direct drive hubmotor from it's wheel, and rewind it to a Kv that is 3 to 4 times higher, with the intention of running it at much higher power and using a simple and quiet chain reduction to the wheel. I'll ventilate the motor so the extra heat won't be an issue. Easily obtainable and cheap controllers aren't available in high enough voltage or I'd just skip the rewind and triple the voltage.
One question I have is about saturation of the stator. If a stator with 8 turns on it's coils reaches saturation at 50A, does it require 100A for the same motor at 4 turns (assuming equal copper fill)?
Is it reasonable to think that the rewound to a high Kv hubmotor is likely to be more efficient? It's my understanding that due to their low speed operation the majority of a hub motor's losses are copper losses, and I get a double benefit of both thicker wire on each turn and a shorter length of wire. That tells me that cutting the Kv in half reduces copper losses to 1/4th, so even if iron losses double due to double the rpm, I should come out ahead since the iron losses were smaller to begin with.
Am I choosing a good motor to try this with, because it's different from my other hubmotors? The other hubbies have a higher slot count than the number of magnets, which seems to be the trend for newer hubbies like 9C and the new Xlytes with 51 slots and 46 magnets. This older model motor has 40 magnets and 36 slots on the stator. It has a longer stator too, with the stack of laminations 40mm thick, so the end copper as a % is less than a motor like a 9C that has only a 27mm thick stator, but a larger diameter to make its torque. The main reason for the choice is this motor will be only 6.75" in diameter once I cut off the spoke flanges, a full 2" smaller than the 9C without flanges, but capable of the same power in-wheel.
Other than the physical difficulty of a rewind, are there any pitfalls to the approach?
John
Now all I need is a 250V controller and I won't have to do a rewind. 
