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

Electric generator RPM

  1. Jun 11, 2015 #1
    What is the lowest rpm, currently, sufficient to generate an emf of a practical value?

    One site for generators states that even 1RPM is sufficient nowadays!

    Could anyone enlighten me, please?

    Thank you in advance.
  2. jcsd
  3. Jun 11, 2015 #2


    User Avatar
    Gold Member

    Hint: Low rpm generators are used for windmills as we then don't need a gearbox ( with a huge gear ratio ).

    Such generators are build with a lot of polepairs, thereby generating ac voltage with a sufficient frequency.

    Typically this ac voltage is rectified, thereafter inverted to the frequency matching the grid. You may say that the gearing is made electronically.

    AC generators are preferred as they have the greatets performance.
  4. Jun 11, 2015 #3
    Pole pairs are magnetic poles with a north pole and a south pole. A single pole pair in a generator would produce an AC signal matching the rotation rate. Two pole pairs would produce twice the signal rate and so on. So simply adding pole pairs increases the frequency. At some point the frequency gets high enough to run a rectifier/capacitor set to give something like DC.
  5. Jun 11, 2015 #4


    User Avatar
    Gold Member

    I considered that with a lot of pole pairs, a lot of phases could be made ( say 12 ).

    It wouldn't be nice to look at some hvac line with 12 phases/wires, but as all these wires in a windmill are just stuffed into a rectifier, nobody cares about the number of wires. Only the rectifier will be happy, requested to output a smooth dc voltage.

    Also the shape of the ac phase voltages could be made trapezoidal instead of sinusoidal. ( No need for a capacitor? ).
  6. Jun 15, 2015 #5
    Thank you for the answers.
    What about the minimum RPM necessary to obtain a useful EMF? Could the 1RPM do it?
    This is where I found the statement:
  7. Jun 15, 2015 #6
    Interesting idea. How would that be done?
  8. Jun 16, 2015 #7
    One way (hopefully not the best) would be to overdrive the cores into saturation.

    It might be possible to build a small anti-winding in the middle of each winding which would produce a flat spot at the top of the output voltage; again, hopefully not the best solution.

    But with the advent of ideal diodes (actually power transistors controlled to act like ideal diodes) there doesn't really seem a need. Diodes are cheaper than windings or power losses.
  9. Jun 16, 2015 #8


    User Avatar
    Gold Member

    In theory: Yes.
    In practice: I doubt it will be done. It will be very expensive to build such a ( slow turning ) rotating generator. But think of a linear motor used by railroads: Bend a piece of this railroad ( say 3.1416 km ) into a circular shape ( diameter = 1 km ), put some rotor inside: It will work. I think several kV could be generated, but 3 km of railroad is indeed expensive.

    Well, a BLDC-motor uses this trapezodial shape as for the phase voltages. It's a matter of distributing the windings and shaping the teeths in the stator.
  10. Jun 16, 2015 #9

    Hesch, actually the end-run of the idea is a linear generator!
    I started from the question of RPM to get a feel of how fast a magnet should slide along inside a tunnel in order to cause the necessary variations of the magnetic flux in the solenoid.
    I have a way to maintain that magnet moving back and forth.

    What would it take, practically, to generate usable (as voltage and amperage) electric current from such arrangement?
  11. Jun 17, 2015 #10
    Practically, what it takes is energy. How many watts of power could your tunnel supply?

    Then there is cost. The traditional solution to the linear motion problem is a solenoid. This means winding the entire tube with copper. While windings could be spaced with some loss of a flat output, there still need to be windings. Winding around something like a mineshaft would be wasteful. Making these smaller is generally better (up to a point).
  12. Jun 17, 2015 #11


    User Avatar
    Gold Member

    A linear motor is just like a rotational motor where the stator and rotor have been made straight.


    No need for a tunnel or tube. It could be made synchronous/asynchronous like a rotational motor.
  13. Jun 17, 2015 #12
    True. There's no need to wrap everything when a small rail will suffice. But most linear motion is on a much smaller scale, such as an actuator.

    With a large linear motor, power distribution can become a problem. It's solvable, but not simple or cheap. Typically we would want some sort of pulsed DC in the windings, but a high voltage AC for distribution to lower I2R losses.

    So a system that could be used in a gopher tunnel wouldn't be the best choice in the Chunnel.

    It comes back to, how much power is there to be exploited?
  14. Jun 17, 2015 #13
    The tunnel was for guiding the moving magnet (because of the peculiar way in which I provide the continuous momentum for the magnet).
    But I may be able to do away with it.
    The velocity of the magnet will vary with the size of the structure.

    Question is: can I make a generator out of this, be it in the shape of the linear motor or any other.

    All I am bringing is the ability to move the magnet.
    How would either of you build a generator from this?
    If this is doable, and you have the expertise, I will invite you to participate.
    Thank you.
  15. Jun 18, 2015 #14
    Yes it's doable, easily in theory. I could likely work out a small demonstration model, but you would do better with someone with better fabrication and power experience.
  16. Jun 18, 2015 #15
    Glad to hear, Jeff!
    For now, let me tread carefully and remove uncertainties as they loom in my mind.

    For now, the velocity of the magnet: would an average of 10-12 m/s be too low? (this is an estimate, not precise calculation)
  17. Jun 18, 2015 #16
    No, 10-12 m/s is fine. How many tons of force on the magnet? How long a tunnel?
  18. Jun 19, 2015 #17
    The magnet is a permanent one, not electromagnet. The source of its mobility must stay secret for the time being.
    Its magnetic moment is up to the educated choice of the experimenter. Its purpose is, of course, to excite the windings.
    These windings, and the guides for the magnet's tracking, comprise what I referred to as the "tunnel". The preliminary calculation of the velocity was done for a length of approximately 20 meters. The longer the tunnel, the higher the velocity. However, the structure then becomes quite large and suitable for mostly on-land use.
    Nonetheless, there is the goal of achieving maximum wattage in any given configuration!

    Interestingly, the converse is equally manageable: highly conductive materials (better than copper, of course, even nano-made!) running front-back through some stationary magnetic fields. Their movement is ensured by the same procedure as for the magnet.

    In all, the source of the front-back movement is the novelty. It is independent and makes the project possible on land, underground, in the air, underwater.
    Speaking of this, with the right configuration and choice of materials, the hope is to develop high enough wattage to airlift the structure. Here comes another concept not seen anywhere else, in which the wattage is employed in a way radically different from any airlifting anywhere.
    Either way, there will always be the usage on land, water, and underneath.
  19. Jun 19, 2015 #18
    Speed isn't important. Force and distance are important.

    I didn't ask the source of the force, I asked the amount. Since you don't have an answer, I'll assume perpetual motion. I'm not interested.
  20. Jun 19, 2015 #19
    Wrong assumption!
    Who could evaluate a secret, since it's secret!?
    It wouldn't have affected what you offered to do , and nothing was to be pro bono.
    But I understand your misgivings.

    Thank you for providing the main information: that it can be done with current technology.
  21. Jun 19, 2015 #20
    I just realized what you had asked for!
    It was not that I didn't have an answer.
    So, I owe it to you.
    Here it is:
    15-20 metric ton-force (in the above scenario).
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook