Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

3 Phase Rotor / Stator Swapping ?

  1. Jan 17, 2014 #1
    3 Phase Rotor / Stator Swapping ???

    I have been toying around with the following idea : (specific application project)

    Normally a stator is the external housing of a motor and is fixed to a structure - the rotor is the shaft through the middle of this stator etc. and is rotating at ...x... RPM !

    The application that I have been unable to find on the net is if I were to swop the rotor/stator around - i.e. the shaft (normally the rotating rotor) is static / standing still AND the stator (normally fixed and static) is now spinning at ...x...RPM ?

    Is there anyone that knows of such a design / motor ?

    Attached Files:

  2. jcsd
  3. Jan 17, 2014 #2


    User Avatar
    Science Advisor

    Yes that is possible, and indeed some machines are designed that way (rotor on the outside). There used to be style of bicycle generator (back in the day when generator lighting systems were used on bicycles) that operated this way. I've still got one somewhere. They had the permanent magnets on the outer part of the machine which rotated as it ran against the tire, and the windings were on the inner "axle" part, which was bolted (stationary) to the bike frame behind the cranks.
    Last edited: Jan 17, 2014
  4. Jan 17, 2014 #3
    Look up Hub Motor -- relatively common for vehicles like bikes and cars
  5. Jan 17, 2014 #4


    User Avatar
    Science Advisor
    Gold Member

    Aren't ceiling fans built this way?
  6. Jan 20, 2014 #5
    @ UART <---- If I'm not mistaken this is the old "dynamo" ?

    @windadct <---- Any URL's for this hub motor ? (I may have neglected mentioning it was for 3 phase?)

    @ Averagesupernova <---- I'LL google the ceiling fans ...........
  7. Jan 20, 2014 #6


    User Avatar
    Science Advisor

    The problem that arises is a question of bearing reliability. Since the heaviest part is now mounted on bearings, the bearings will need to be bigger than the original minimal diameter shaft bearings. That will lead to higher bearing surface velocities with significant reliability implications.

    If the windings are inside, the power will need to pass through the mounting bearing shaft which will require them to be larger with a lower RPM. Alternatively, the bearings could be moved to one end away from the power cables. If the bearings are closer they will have greater loading.

    If the windings are on the outside there will be very real balance problems and a need for slip rings.

    The 3 phase induction motor has evolved into an optimum design. Any change from that low cost commodity design needs to be very well justified.

    Navigational gyroscopes and some X-ray tube rotating anodes are built with an inverted design.
  8. Jan 20, 2014 #7
    @ Baluncore ......thanks for the bearing info ... for now that is not my main concern ..... the following is :

    I am trying to build a 3 phase motor - but the inner shaft is going to be the stator and the outside casing will be the rotor.
    The rotor (casing) will be turning at a constant speed of 3 meters/second. My interest lies at the fact that I would very much like to know how to wind/populate/magnetize this stator and rotor ?
    Factors to consider : qty windings; qty poles; thickness of conductors; lamination of conductors; spacing of conductors; layout of magnets and windings; types of magnets and conductors etc......
  9. Jan 20, 2014 #8


    User Avatar
    Science Advisor

    Ah, so you want to know everything.
    There are many books on the subject but it would be best to narrow down the choices first.

    You can make a three phase induction motor that will run at about 95% of synchronous speed, or you can use magnets and build a synchronous motor. The induction motor will be easiest.

    I presume the 3 m/sec is the peripheral speed of the motor if used say for a direct drive belt roller.

    Your supply frequency is probably 60Hz? (Because you spell the unit of length meter, not metre).
    The number of poles will decide the rpm. Then 3 m/sec will set the diameter of the roller.
    rpm = 60Hz * 60sec / poles. (These are synchronous speeds).
    Circumference; c = 3m/(rev/sec). Roller diameter, d = c / Pi
    2 poles, 1800 rpm, 30 rev/sec, c=0.1m, d= 31.8mm = 1.25” diameter
    4 poles, 900 rpm, 15 rev/sec, c=0.2m, d= 63.7mm = 2.5” diameter
    6 poles, 600 rpm, 10 rev/sec, c=0.3m, d= 95.5mm = 7.75” diameter
    8 poles, 450 rpm, 7.5 rev/sec, c=0.4m, d= 127.3mm = 5.” diameter
    12 poles, 300 rpm, 5.0 rev/sec, c= 0.6m, d= 191.mm = 7.5” diameter
    16 poles, 225 rpm, 3.75 rev/sec, c =0.8m, d= 254.6mm = 10” diameter
    24 poles, 150 rpm, 2.5 rev/sec, c = 1.2m, d= 382.mm = 15” diameter

    What horsepower do you require? How long can the motor be?
  10. Jan 20, 2014 #9
    Indeed I do (want to know everything) !

    Firstly : I am in South Africa (50Hz) (don't exactly know where you are from ?) ....I actually typed "metres" but the dictionary suggested meters - so I conformed to the dictionary ...lolol

    Secondly :
    The "motor" I want to design is indeed a "roller" .......

    I have an idler/roller with a diameter of 127mm-which equals 399mm circumference (if my maths don't fail me)

    A conveyor drives this roller (and others) at ± 3000 - 3500 mm/sec (3.5 metres/sec).

    ...which in turn equates to ± 7 ½ to 8 ¾ revs/sec.

    Now is the part where I need to know the best way of winding the coils and installing the magnets ?

    I intend sending the roller/idler for induction coil windings and magnets - but also not sure if the magnets should be on the stator (static shaft) or the rotor (rotating casing) ?

    What kinds of Voltages can I expect ?

    Does the rotational speed of the roller influence the voltage ?

    Does the conductor winding configuration influence the voltage or current ? (or both)

    Does the conductor thickness determine the current ?

    Does the quantity of coils (in multiples of 3) influence the voltage ?

    What is the best way of winding this "motor"/idler/roller of mine ?
  11. Jan 20, 2014 #10
    "I presume the 3 m/sec is the peripheral speed of the motor if used say for a direct drive belt roller."
    Yes it is indeed the peripheral speed of the roller.

    Supply frequency is 50Hz - but I dont think it is relevant at this stage - because the voltage generated by the roller will be totally disconnected from the supply source ?

    Horsepower : At this stage I cannot answer that because the voltage/current gained from this roller is not easily calculated - but as much Watts (Volt x Ampere) is always good ! lol

    The length of the roller varies - some are short and other longer - but for now assume the shaft (stator) is only 40cm (400mm) long ?
  12. Jan 20, 2014 #11


    User Avatar
    Science Advisor

    It seems you have switched from a drive motor to an alternator driven by a belt.
    The questions then become;
    How much power do you need to generate?
    Is it three phase?
    What voltage do you require?
    Why are you doing this?
  13. Jan 20, 2014 #12
    here is a pic of the roller/idler (note the hollow inside)

    Attached Files:

  14. Jan 20, 2014 #13
    I need enough power to supply 3 x floodlights (500W ea) ....alternatively I would connect more than one roller (parallel) to achieve this.

    I would like it wound as a 3 phase generator yes ! (even if the volts generated is less than required - then the next step would be to connect more than one roller to achieve this!)
  15. Jan 20, 2014 #14
  16. Jan 20, 2014 #15


    User Avatar
    Science Advisor

    There is no simple set of equations. To specify, design and implement electromechanical equipment of this nature requires an understanding of electromagnetic engineering only acquired after several years of training and experience. I would suggest you should study the field before embarking on what is a graduate level project. I am sorry but I have very real doubts about your ability at this stage to complete the design and prototyping of this project.
  17. Jan 20, 2014 #16
    Thanks anyway !
  18. Jan 21, 2014 #17
    I think we are the realm of "just because you can does not mean you should".... It seems that you are looking to take some excess energy from a belt(conveyor) and use it to generate 1500W of electricity. So I am left asking why are you doing this and why does the motor have to be the type you describe?
    Is this a product you are looking to design and market, a one off to fill a particular need - or is it a science project - just to prove that this type of motor will work here?
    To harvest power from a conveyor - link to a belt and drive a conventional (PM or induction) generator - and generate electricity.... making a motor from scratch to do this is on the order of 10 to 20X the time and cost, and will not be as reliable or effective. If a company were to build this - it would be a team of 3+ prople ( Mech E, EE and ?? - Product Manager etc).
  19. Jan 21, 2014 #18


    User Avatar
    Science Advisor

    There will be a problem of connection between the roller and the belt. Only at the ends of the conveyor is there sufficient wrap of the belt around the roller to reliably transfer power. At any other point the belt will slip on the roller, or require an idler roller and a longer belt.

    There is also the bearing problem, to grip the belt would require an additional load on the oversize bearings, that is critical on a conveyor belt where many bearings must function reliability for long periods of time.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook