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

Salient and Non-Salient pole Synchronous Generator

  1. Jul 8, 2016 #1

    I was wondering if some kind soul could explain to me why salient pole machines are preffered in hydropower? From what I've read salient pole machines are preffered for applications with "low" rpm, while non-salient poles are used for applications that runs at higher rpm. O

    What is the reason for this; that salient poles are preffered for low rpm and non-salient poles for high rpm applications?

    I'm not sure if I'm asking in the right section of this forum, perhaps this thread also belongs in the mechanical section, I'm not sure.

    Best regards.
  2. jcsd
  3. Jul 8, 2016 #2


    Staff: Mentor

    Good question. Actually, salient pole generators would be preferred in all cases. The problem is that the centrifugal forces at high RPM (i.e. 3000 or 3600 RPM) are too much for those salient windings, so salient pole rotors become impractical. They use non-salient (i.e. round) rotors instead.
    Low RPM hydro rotors work fine.

    By the way, for the benefit of others. In a 60 hz system, the electrical rotation speed of a synchronous generator must be 60 hz (377 radians per second). That can be done with a rotation speed of 3600/N RPM, where N is the number of pole pairs. Many steam turbines have N=1. Many nuclear turbines have N=2. Many hydro turbines have N=10, 12,...24 pole pairs.

    http://ips.us/cleveland/images/rewind-2.jpg [Broken]
    Hydro rotor with each distinct salient pole clearly visible.
    Last edited by a moderator: May 8, 2017
  4. Jul 8, 2016 #3
    Thank you for the answer.
  5. Jul 8, 2016 #4
    Actually, this got me wondering; why are salient pole generators preferred over non-salient pole generators? Ok, fine that centrifugal forces at high rpm are to much for salient pole rotors, so non-salient rotors are used instead. But you say, that salient pole generators would be preferred in all cases, why is that?

    Does it have anything to do with that salient pole generators can continue to run(depending on load) after loss of excitation?
  6. Jul 8, 2016 #5


    Staff: Mentor

    I am not an expert on generator design. Perhaps other PF members are. My belief is that with salient poles, the designers have more freedom to determine the distribution of flux both radially and axially. Round rotors are very constraining design-wise. Remember that all power plant synchronous generators are extremely efficient electrically. Various designs give manufacturing cost improvements, overspeed tolerance, and ease of cooling. There is not much leverage in playing with the electrical characteristics. Hydro plants in particular must tolerate up to 100% overspeed, sometimes even more.

    Your assertion about operating without excitation is news to me. In all cases I know, loss of excitation leads to undervoltage, or underexcitation trips, and would result in loss of synchronism if there is no trip. However, I can't exclude the possibility of circumstances where continued operation is impossible. Since you raised the point, you probably know more than I do on that subject.

    What I have seen are some very dramatic demonstrations of bringing a turbine generator up to partial speed with no excitation, then closing the breaker and gradually raising excitation to synchronize. What happens makes one's hair stand on end (reasonable men would flee for their lives), but if carefully managed it is survivable. If readers are interested, I might be able to locate a video of the demo.
  7. Jul 9, 2016 #6
    I doubt that I know more, perhaps I've mistaken something, because I can only find information on this working on salient pole motors, if one google the term "reluctance power".

    However, I would like someone with experience with this kind of topic, so we can avoid any confusion.
    I was looking around and found this equation for salient pole machines:

    [itex]P = \frac{E_f V_t}{X_d + X}sin\delta + \frac{V_t^2}{2}(\frac{1}{X_q+X}-\frac{1}{X_d+X})sin2\delta[/itex], If [itex]E_f[/itex] becomes zero, we still have this term: [itex]\frac{V_t^2}{2}(\frac{1}{X_q+X}-\frac{1}{X_d+X})sin2\delta[/itex], but [itex]X_d[/itex] is always less than [itex]X_q[/itex] i suppose? If that is the case, it will just work as a motor, right?

    I tried to find a source talking about this, but was not successful. Also I do not have much experience whitin this field, so I guess I should have not said what I said in last post without anything to back it up with, I appologize.

    That would be interesting to watch, if you could locate it.
  8. Jul 9, 2016 #7


    Staff: Mentor

    When you use equations like the ones you showed, remember that they are linear. There are always nonlinearities, limits, and ranges of applicability to consider. Relays that trip the breakers are an engineered way to enforce limits.

    My flight simulator may have a linear equation that says I can fly my plane through the center of the earth. That doesn't mean that it is really possible.

    But don't let me discourage you from thinking about the implications of equations. That kind of thinking is very healthy.
  9. Jul 9, 2016 #8

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Intuitively that sounds backward to me.

    Air gap is largest between the salient poles so quadrature X is much smaller than direct
    check this guy's treatment it appears pretty practical

    i'd have to see what is every term in that equation before i accepted it
    is the sin2δ term reluctance torque ?
  10. Jul 9, 2016 #9
    I found the equation mentioned above, referred to as "Eq. (5.69)" in the text below. It's derived from the same phasor diagram shown in the PDF document you linked in figure 4-41, just note that [itex]r_a[/itex] is assumed to be negligible.
  11. Jul 9, 2016 #10

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Aha so i guessed right, that term is reluctance torque which exists because of the gap between field poles.
    Thanks !
    i think the answer to your question is of mechanical nature not electrical.

    It does not make sense to me that one would attempt to make very much power using reluctance torque.

    It does make tremendous sense that it becomes exceedingly difficult to manufacture a salient pole rotor for high speed.
    because of the difficulty of keeping the poles attached to the rotor at high speed.
    Bolts, or dovetail joints as used on turbine blades, simply don't have the cross sectional area to handle the tension produced as centrifugal force tries to pull the massive pole piece off the rotor. Salient rotor pole pieces are laminated .

    High speed rotors are machined from a solid forging. That spreads the tension over the whole area of the pole not just its hold-down bolts.

    That last sentence - Energy lost to windage is significant over the forty year life of a generator. Big ones are cooled with hydrogen which being light is easy to buffet about. A smooth rotor churns it less.

    Any help ?

    old jim
  12. Jul 9, 2016 #11
    Yep, your answer was helpful, thank you!
  13. Jul 9, 2016 #12

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Wow look at all the bolts on this salient rotor !

    Copy and paste it into Paint and expand..

    How many square inches of shiny bolt hold those pole pieces to shaft ?

    I'm no mechanical engineer
    but it looks to me like they gusseted it to turn some of the tension from radial to circumferential , is "hoop stress" the proper ME term ?

    Here's another link
    I never saw anything like that. Ours were all machined from a solid cylinder which i believe is more typical.

    I admire Mechanical engineers. They do the hard part , they make our coils move through the magnetic field.

    old jim .
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted