0
\$\begingroup\$

EDIT: As Bruce Abbott pointed out in his answer, it's not possible to connect 3-wire brushless motors in series, since you don't have access to both ends of each winding. Oops. So, the more relevant question is whether there are any potential problems with wiring multiple motors in parallel, other than the increased current draw.

ORIGINAL QUESTION: I'm working on a design that requires a large amount of torque in a very confined space. Due to this and other requirements, my best options seems to be to use two motors to drive a single load. The shafts would be connected together, so the two motors would always maintain the same relative orientation.

I would like to avoid using two motor controllers, so I'm thinking of wiring each phase of the two motors in series (i.e. phase 1 of motor A is in series with phase 1 of motor B, etc.). I would have a single encoder on one of the motors, which I would be using for position feedback and for sinusoidal commutation. I'm expecting that this would effectively allow the two motors to act as a single larger motor, with double the voltage required for a given speed. Is there any reason this wouldn't work, or any potential issues I need to watch out for?

One thing I'm worried about is that I would need to make sure that the connection between the two motor shafts is "clocked" correctly such that the rotors of both motors have the same position relative to the stator windings. There will inevitably be some error in this; I assume that this would just slightly reduce the torque for a given amount of current, but perhaps it would cause other issues?

Some additional information, in case it affects the answer:

  1. This is for a positioning application which uses a haptic operator interface. This uses the motor current to calculate the motor torque, which is then used to provide force feedback to the operator. Therefore, one of the main requirements is very low torque ripple.

  2. The motor speed will be very low (from 0-100 RPM).

  3. The motors I would be using are the EC-4pole motors from Maxon.

  4. I will likely be using the EPOS4 motor controllers from Maxon.

\$\endgroup\$
7
  • 1
    \$\begingroup\$ Why can't you pick a single motor to replace the two motors? You say small space, okay, but longer motors of the same diameter should exist. \$\endgroup\$
    – DKNguyen
    Dec 24, 2020 at 5:59
  • \$\begingroup\$ Will you be driving these with a controlled current, or with a constant voltage? If the command out of your controller is for current, you can just run two encoders and two motor drives, and gang the motors together any which way. Or just get a bigger motor. \$\endgroup\$
    – TimWescott
    Dec 24, 2020 at 6:00
  • \$\begingroup\$ @DKNguyen: If they exist, I haven't been able to find them. The Maxon EC-4pole motors are already a very long, narrow motor (32 mm diameter, 142 mm long, not including the shaft or encoder). I'm certainly open to suggestions, though! My choices are somewhat constrained by the fact that I'm looking for a motor with an ironless core, to eliminate cogging torque (see point #1 in my question). I was considering the Maxon EC-i52, which has a 52mm diameter and can provide enough torque, but it has an iron core and therefore suffers from cogging torque. \$\endgroup\$
    – Chris Graham
    Dec 24, 2020 at 6:19
  • \$\begingroup\$ That already is pretty long. Well, should be easy enough to align the rotors. Connect the phase wires in parallel and then feed current through two phases with a controlled source so the rotors lock but don't burn out and while its lock clamp both rotors together. \$\endgroup\$
    – DKNguyen
    Dec 24, 2020 at 6:24
  • \$\begingroup\$ @TimWescott: Current will be the controlled variable. Yes, I could certainly run two encoders and two drives, but I'd like to avoid this if possible, both to reduce expense and, more importantly, the number of wires that I have to run (the wires will probably be going through a slip ring, and I already have quite a lot of wires to deal with). \$\endgroup\$
    – Chris Graham
    Dec 24, 2020 at 6:25

1 Answer 1

Reset to default
0
\$\begingroup\$

so I'm thinking of wiring each phase of the two motors in series (i.e. phase 1 of motor A is in series with phase 1 of motor B, etc.)

To wire the phases in series you need access to both ends of each winding. Maxon EC-4 motors only have 3 external phase wires and no way to separate the windings internally, so you cannot wire them in series.

However you can wire identical brushless motors in parallel, so long as you match the rotor angles precisely and the shafts are tightly coupled. Use the encoder from one motor only and the pair will act like a single motor.

In parallel the total maximum current will be doubled, and your controller will have have to handle twice the current of a single motor. So choose high voltage motors whose combined current doesn't exceed the controller's rating.

\$\endgroup\$
9
  • \$\begingroup\$ Good point; in retrospect, that should have been obvious, but clearly I didn't think it through well enough. I had considered wiring in parallel, but wanted to avoid the extra current draw since it will exceed the continuous rating of the controller I was planning to use. Oh well, perhaps I can find a different one. \$\endgroup\$
    – Chris Graham
    Dec 24, 2020 at 17:17
  • \$\begingroup\$ I was also a bit worried about unbalanced current draw if the shaft angles aren't perfectly matched, but it sounds like you think it's feasible to match the angles well enough that this won't be a problem? Do you think the method mentioned by DKNguyen (supply current to the same phase of both motors, then couple the shafts together) is the best way to do that? I can't think of a better method, but clearly I have a lot to learn. \$\endgroup\$
    – Chris Graham
    Dec 24, 2020 at 17:21
  • 1
    \$\begingroup\$ @ChrisGraham I would jiggle the shafts while locked a bit to make sure they aren't sticking a bit off center. In self-calibrating systems, I've found that getting rotors to lock in position under their own power doesn't always result in them being centered over the pole. \$\endgroup\$
    – DKNguyen
    Dec 24, 2020 at 21:42
  • \$\begingroup\$ @DKNguyen: good advice, thanks. \$\endgroup\$
    – Chris Graham
    Dec 25, 2020 at 2:07
  • \$\begingroup\$ "wanted to avoid the extra current draw since it will exceed the continuous rating of the controller I was planning to use." - exactly which motors and controller were you planning to use? \$\endgroup\$
    – Bruce Abbott
    Dec 25, 2020 at 16:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.