1. Attachments are working again! Check out this thread for more details and to report any other bugs.

What is MG2's type and construction?

Discussion in 'Gen 2 Prius Technical Discussion' started by Berkeloid, Jan 5, 2019.

  1. Berkeloid

    Berkeloid Junior Member

    Joined:
    Jan 5, 2019
    11
    4
    0
    Location:
    Australia
    Vehicle:
    2007 Prius
    Model:
    N/A
    Hi all,

    I'm having a bit of trouble finding some technical specs on the Prius electric motors, as in what type they are (induction, permanent magnet), how many phases, etc. The sort of thing you look for if you want to know whether they would be suitable for using in another project, unrelated to cars.

    Does anyone have any detail about what type of motors they actually are?

    I have only been able to find photos that show three or four connectors - so I'm guessing they might be three phase - and I'm presuming that they aren't induction motors as these seem like they'd be more difficult to reliably switch between using and generating power. Some people mention magnets when removing the rotor during replacement, so that makes me think they might be permanent magnet motors.

    Since the Toyota manual mentions possible damage if the car is towed incorrectly, this also leads me to suspect a permanent magnet motor, which will generate electricity any time the shaft is rotated, potentially causing excessive heat build up in a towing scenario.

    Am I on the right track? Has anyone used any of these motors in any other projects outside of EVs? If there's anywhere that lists the type of info I am after let me know as I haven't had much luck with Google.

    Many thanks!
     
  2. Leadfoot J. McCoalroller

    Leadfoot J. McCoalroller Senior Member

    Joined:
    May 12, 2018
    7,438
    6,920
    1
    Location:
    Pennsylvania
    Vehicle:
    2018 Prius c
    Model:
    Two
  3. ChapmanF

    ChapmanF Senior Member

    Joined:
    Mar 30, 2008
    24,948
    16,246
    0
    Location:
    Indiana, USA
    Vehicle:
    2010 Prius
    Model:
    IV
    Hmm, that's a long thread, and kicks things off with a bold misstatement in the third post. It gets better (if nonmonotonically) from there, but just to get the short answer into this thread so it doesn't have to be chased down:

    MG1 and MG2 both have permanent magnet rotors and three-phase, delta-wound stators.
     
  4. TomB985

    TomB985 Member

    Joined:
    Apr 20, 2014
    137
    68
    0
    Location:
    Isanti, MN
    Vehicle:
    2006 Prius
    Model:
    Two
    OK, now I’m curious. Care to enlighten us on what your project is?
     
  5. Berkeloid

    Berkeloid Junior Member

    Joined:
    Jan 5, 2019
    11
    4
    0
    Location:
    Australia
    Vehicle:
    2007 Prius
    Model:
    N/A
    Many thanks for the helpful info! Through the other thread I was able to find a "Report on Toyota/Prius Motor Design and Manufacturing Assessment" which I would include a link to, but it tells me I don't have enough posts here yet. Here they broke the motor down into its components and I was able to ascertain that the Gen2 MG2 is:
    • Three-phase permanent magnet synchronous motor
    • 8-pole
    • Operating voltage of 200-500 VDC
    • 50 kW total power, when driven at 500 VDC (less overall power at lower voltages)
    They didn't mention the windings so thanks @ChapmanF for letting me know they are delta-wound.

    I haven't been able to ascertain whether the motors are driven on pulsed DC (e.g. PWM) such as from the output of three-phase VFD controllers, however given that they are AC three-phase but the power is always reported in volts DC, I am guessing that the Prius inverter may output PWM DC rather than sinusoidal AC. Please chime in if you know for sure.

    My "project" hasn't started yet, but I am working towards moving to a rural property where there will be no mains power. So I am investigating some ideas for generating and storing electricity beyond just solar and batteries. I am thinking that the MG units could be a very inexpensive way of producing up to 50 kW of three-phase AC power, as buying second hand motors of this size are rather expensive, at least in my area. I don't talk much about my unconventional ideas in this respect because the first thing most people say is "oh why don't you do X instead, it would be much easier" completely missing the point that half the fun for me is in the challenge of getting it working!
     
    TomB985 likes this.
  6. Leadfoot J. McCoalroller

    Leadfoot J. McCoalroller Senior Member

    Joined:
    May 12, 2018
    7,438
    6,920
    1
    Location:
    Pennsylvania
    Vehicle:
    2018 Prius c
    Model:
    Two
    For homebuilt generation, I never found a better recycle than exercise treadmill motors. Lots of people give them away or sell them off cheap, so they're easy to get hold of. The motors are easy to extract from the treadmills, they're lovely high quality permanent-magnet motors with great sensitivity/efficiency as generators, and they're a useful size for hobby/home wind & hydro.

    Good luck!
     
  7. TomB985

    TomB985 Member

    Joined:
    Apr 20, 2014
    137
    68
    0
    Location:
    Isanti, MN
    Vehicle:
    2006 Prius
    Model:
    Two

    I'm far from an expert, but I just thought of a document I found awhile back regarding the first generation Prius. You might find what you're looking for in here, most of it is over my head. Hope this helps!

    Dropbox - 1st Gen Prius Motor.pdf
     
  8. ChapmanF

    ChapmanF Senior Member

    Joined:
    Mar 30, 2008
    24,948
    16,246
    0
    Location:
    Indiana, USA
    Vehicle:
    2010 Prius
    Model:
    IV
    The drive voltages are applied by giant, switching transistors in an H bridge (one per phase), so each pulse is either full DC, or zero, or full −DC. You wouldn't want to be doing anything in the linear region at those power levels. But you can certainly pulse at a high frequency and modulate the pulse widths to have the motor windings see something as close to sine waves as you can get.

    Somebody (probably hobbit) looked at some of the transistor drive signals on an oscilloscope and saw that at high RPMs, the driver kind of gives up and does square-ish waves, one per pole, too fast to try to be fancier, but at lower RPMs, it does switch to higher-frequency pulses and modulate them to get a sinusoidal result.

    I don't know if they're using a technique exactly like Don Lancaster's magic sinewaves, but they certainly could be.

    An interesting thing about each motor is there's a "resolver" on it, a shaft-position encoder with a little magnetic pickup read by a special IC that gives real-time, absolute shaft position to one 4,096th of a revolution. That information is used to time the pulses. Adjusting the resolver is a critical operation that's done at the factory, and there are no instructions for doing it in the field; some people replacing their MG2 have gotten into trouble by loosening the wrong screws there and having it go out of adjustment.
     
    #8 ChapmanF, Jan 6, 2019
    Last edited: Jan 6, 2019
    Robert Holt and Berkeloid like this.
  9. JimboPalmer

    JimboPalmer Tsar of all the Rushers

    Joined:
    Apr 14, 2009
    12,470
    6,871
    2
    Location:
    Greenwood MS USA
    Vehicle:
    2012 Prius v wagon
    Model:
    Three
    In any chat, some will believe wrong information. PriusChat does quite well in correcting that.

    The Weber auto channel has many videos on the internals of the Prius, these two I am most familiar with*:




    Prof John Kelly has newer videos, called Deep Dives, with more details, but these two combine to give a good overview.

    *Later Prius get more complicated.
     
    Berkeloid, TomB985 and Raytheeagle like this.
  10. ChapmanF

    ChapmanF Senior Member

    Joined:
    Mar 30, 2008
    24,948
    16,246
    0
    Location:
    Indiana, USA
    Vehicle:
    2010 Prius
    Model:
    IV
    I had not seen the PDF that TomB985 linked in #7 before, but it's written by the motor design team, and it describes this RPM-dependent change in modulation approach; it's in section 6-3 "one-pulse switching control" and illustrated in figure 15.

    They actually get better torque, at the higher RPMs, when they leave behind the fastidious PWM and just feed the motor a square wave.
     
  11. Berkeloid

    Berkeloid Junior Member

    Joined:
    Jan 5, 2019
    11
    4
    0
    Location:
    Australia
    Vehicle:
    2007 Prius
    Model:
    N/A
    Unfortunately they are very scarce in my area, as three-phase power outlets are not common here so they are all only small single-phase units for home use. I am looking for something that can produce a decent amount of power (at least 24 kW to match an on-grid service is the goal), and the Prius motors have been spot on in terms of power and price compared to everything else I have seen so far.

    @TomB985: That PDF was quite interesting. More about how the transmission works and how to run the motor, but nonetheless some interesting detail there.

    Very interesting. With that and the PDF, it does indeed look like the motor is controlled via PWM and then eventually just a low(er) frequency square wave at higher RPM. I presume it still generates a sine wave when run as a generator though?

    Interesting note about the resolver, that could also make it much easier to calculate the rotor speed as opposed to measuring the output frequency separately.

    @JimboPalmer: Thanks for the videos. I had seen one before but not the other, and they are quite interesting, although like most things I've found they are more focused on how the system works as a whole rather than the specifics of how the electric motors work. But it's still interesting to see all the same.
     
  12. ChapmanF

    ChapmanF Senior Member

    Joined:
    Mar 30, 2008
    24,948
    16,246
    0
    Location:
    Indiana, USA
    Vehicle:
    2010 Prius
    Model:
    IV
    Sure, if you had it hooked up to three light bulbs and spun it, you would get three sine waves.

    In the car, though, the ECU is still in charge of how much either MG will generate, and controls that by taking PWM sips, just like driving it as a motor but with the phase flipped.
     
    Berkeloid likes this.
  13. ChapmanF

    ChapmanF Senior Member

    Joined:
    Mar 30, 2008
    24,948
    16,246
    0
    Location:
    Indiana, USA
    Vehicle:
    2010 Prius
    Model:
    IV
    ... and if the idea of taking discrete sips from a giant inductor makes you think "hey, wait, but doesn't that ... ?", the answer is yes, it does, which Toyota manages to use as a kind of innate boost-conversion scheme and get more regen at low speeds.
     
  14. Berkeloid

    Berkeloid Junior Member

    Joined:
    Jan 5, 2019
    11
    4
    0
    Location:
    Australia
    Vehicle:
    2007 Prius
    Model:
    N/A
    Ha, I hadn't even thought of taking sips from the generator as a way of boosting the voltage. I wonder what the maximum voltage the system has to deal with really is?

    Any idea how much voltage the MGs might produce when used as a generator? I'm thinking that if they are run at 500 VDC PWM when used as motors, then running them as a generator might produce 340 VAC (which will hit +/- 500 V at the peaks) but that presumably means you can't get the full 50 kW out of them as a sine wave won't let them spend enough time at 500 V.

    340 VAC is only a guess though, as this assumes a single phase and the MGs are three phase. I'm not exactly sure how to guess what AC voltage they might produce because I'm not sure entirely where the 500 VDC pulses are measured from - these must be phase-to-phase voltages as I don't think there is a neutral connection on them.