gac3rd wrote:This is a branch from the Programming the P&G controller thread.
JoeC wrote:I'm looking at the possibility of using Roboteq on the GB motors of INVACARE.
Did I say that? From what I know now, I wouldn't expect it to work. The Invacare Gearless-Brushless motors are kind of unique. I don't know exactly how they're put together or how signals are managed, but there are things happening that are a bit more sophisticated than your average brushless motors. Based on discussions I've had with the company that built the originals (UQM) plus things I've read in patents, it appears that the motors undergo a factory calibration in order to achieve smooth response. My understanding is that imperfections in the windings and supporting structure causes fairly significant fluctuations in torque throughout a rotation. This variable torque is stored in a table in an EEPROM, which sends this data to the motor controller for use in modulating the output power to achieve smooth torque and smooth speed. There might be other tricks happening that would have to be identified and either ignored or replicated.
What I am still thinking about is using the Roboteq with the Quickie P222 (not SE) motors. These have three power wires and some number of sensor wires. I haven't had the time or workspace to really map it all out to ensure that the arrangement is compatible with the Roboteq, but I would be surprised if it isn't.
Roboteq has a dual channel version of the brushless controller, the HBL2350 -
http://www.roboteq.com/brushless-dc-mot ... all-inputsIt's only 75 amps, but that's OK for my application. In sports use my wife is unable to take advantage of much over 80 amps due to traction limitations. Her present controller only does 80 amps, and it's taken careful distribution of weight (and hundreds of hours of practice by her) to minimize wheel slipping during during certain precision maneuvers. We've experimented with different tires as well, but haven't found anything practical that grips the court better and survives the miles of pavement travel to get to and from practice. A person who weighs much over 100 pounds might feel a loss from only 75 amps, but by increasing the voltage and gearing I think the chair would still have more than enough thrust for sports purposes. I wouldn't consider using this arrangement without increasing the battery voltage to at least 36 volts. Since mosfet failure isn't an issue in the same way it is for brushed motors, I would be more comfortable pushing the voltage closer the the controller's rated voltage.
If nobody noticed, my whole interest in high performance controllers is for sport use. My wife is tiny and doesn't often have the opportunity to drive through sand or mud or terribly uneven terrain, and so she doesn't ask for power improvements to her everyday chair.
