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

Why crawling happens in induction motor?

  1. Mar 29, 2015 #1
    It is said that crawling occurs mainly due to 7th harmonic component. Due to 7th harmonic the speed decreases by the ratio of 7. However speed is directly proportional to frequency. So speed should get multiplied by a factor of 7 instead.
    Should the change in frequency affect the number of poles formed?If yes, how?
    Are not the number of poles independent of frequency or harmonic??
     
  2. jcsd
  3. Mar 29, 2015 #2

    Hesch

    User Avatar
    Gold Member

    Asynchronous motors will try to accelerate to their synchronuos speed, but cannot reach this speed because the emf in the rotor, and therefore the current in the rotor will be zero at that speed. The rotor will not be magnitized = no torque. Due to self-inductance in the rotor, the current in the rotor will rise as the rotor gets closer to synchronus speed as the frequency of the rotor current decreases. So the result of torque as a function of speed is shown here:

    https://www.google.dk/search?q=asynchronous+motor+torque+curve&biw=1366&bih=635&source=lnms&tbm=isch&sa=X&ei=miIYVeiMOIffU-v9gNgG&ved=0CAYQ_AUoAQ#imgdii=_&imgrc=nfF0mWSNX_TTEM%3A;W9RnDBZT3tD1WM;http%3A%2F%2Fupload.wikimedia.org%2Fwikipedia%2Fcommons%2Fthumb%2Ff%2Ff6%2FVariation-couple-uf.svg%2F2000px-Variation-couple-uf.svg.png;http%3A%2F%2Fwww.eevblog.com%2Fforum%2Fcrowd-funded-projects%2Fimpossible%2F15%2F;2000;1600

    Higher harmonics in the motor arises due to quantifying noise in the magnetic curcuit in the motor ( e.g. grooves in the stator for windings ). During acceleration the rotorwindings will pass many of these harmonics, and synchronizing to one of these, the rotor will stick to them. In the attached torque curve, you can see that at lower speeds the motor has a lower torque. So if the mechanical load torque of the motor is close to the torque curve of the motor, the motor will crawl at this speed. Changing the curve for the load torque may result in, that the motor will stick to a different harmonic.

    The number of poles (pole-pairs) is fixed and depends on how windings are placed in the stator. As the magnetic field from the rotor is created by induction, the rotor will have as many pole-pairs at the stator.
     
    Last edited: Mar 29, 2015
  4. Mar 29, 2015 #3
    Thanks for the reply.
    Can you clear the point that when we know that speed is directly proportional to frequency then why does the speed becomes one-seventh for the seventh harmonic which is the main reason for crawling in induction motor.
     
  5. Mar 29, 2015 #4

    Doug Huffman

    User Avatar
    Gold Member

  6. Mar 29, 2015 #5
    thanks for the link. It was informative.
    But unfortunately my question still remained unanswered in the link. I want to know that why does the speed become one- seventh for seventh harmonic when in formula synchronous speed is directly proportional to frequency?
     
  7. Mar 29, 2015 #6

    Hesch

    User Avatar
    Gold Member

    #3: . . #4 has given an explanation.

    In practical motors, bars are cast in grooves to form this squirrel cage in the rotor (typical in aluminum). But to dampen the amplitude of the harmonics, the bars are not made parallel to the center axis of the rotor, so that a bar will "slide" from one magnetic tooth to the other, when turning inside the stator, instead of in steps. By adjusting the angle of the bars, the manufacturer of the motor is able to balance the dampening of various harmonics in different ways. So maybe a manufacturer can make the motor crawl at other speeds (I don't know).

    If you take apart an asynchronous motor, you can see this skew squirrel cage.
     
    Last edited: Mar 29, 2015
  8. Mar 30, 2015 #7

    uart

    User Avatar
    Science Advisor

    It's actually a really good question nand. :) It turns out that this is just small misunderstanding of what is meant by "harmonics" in this context, and there is a very simple explanation.

    You are thinking in terms of the harmonics of a time domain signal (and by this I really mean the harmonics you'd get by transforming a time domain signal like the supply voltage or current into the frequency domain). The harmonics that they are referring to in this context however are those of a "spatial domain" signal, specifically the spatial distribution of flux around the machine.

    Imagine for example that you had a machine with one pole pair and a very high degree of saliency, such that the flux under the poles was a uniform maximum, and that between the poles was zero. If you looked at the flux as a function of angular position it would basically be a square wave, and so have a Fourier decomposition in the same way that a square wave voltage does.

    Notice that in this case however, the third harmonic (for example) doesn't correspond to a something that is changing three times as fast in the time domain, it instead corresponds sinusoidal to a flux component that undergoes three completely cycles as we move one complete rotation around the motor! Do you see the difference.

    Hopefully it's now becoming clear, this 3rd harmonic flux component is precisely that which we would get in a machine with 3 times as many pole pairs! Hence 1/3 the speed for the same applied voltage/frequency.
     
    Last edited: Mar 30, 2015
  9. Mar 30, 2015 #8
    Thanks for the reply.
    I got it.
    So, we can say that in a way number of poles increase and thus speed decreases.
     
  10. Mar 30, 2015 #9
    Can you please refer me some book or source where i can study crawling in detail.
    Where i can study the concept you mentioned in detail.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Why crawling happens in induction motor?
  1. Induction Motors (Replies: 1)

  2. Induction Motor (Replies: 1)

  3. Induction Motor (Replies: 1)

Loading...