> First is the change in magnetic flux if you > turn the rotor by hand (without current). > You feel the cogging. Second you get torque ripple > if the rotating magnetic field induced by > the current does not move smooth. Read some > papers about magnetic design of motors. > Even a 3ph ac motor - he has no cogging because > no permanent magnets - may have torque ripple, > if stator design is bad.
Thank you, I didn't understand correctly the difference. > Personally I have done some investigations with > small motors in the range of 10- 500W. > To drive a motor I bought the RDK-BLDC and RDK-ACIM > kit from Luminary Mirco (now TI). You get > the full source code. Plan is to measure the cogging > if you switch off motor current of a spinning rotor > and see what an attached 2ph encoder delivers. > In the past I have done this to see the stepper motor > vibrations. If you know the effect of cogging > (at some speed) operate the motor at this speed > and look again to the encoder signals. If minimum > and maximum values of encoder signal period during > one revolution inscrease, the induced rotating magnetic field > is not smooth. > Unfortunately servo motors operate in the frequency > range of some natural vibrations of mechanics > (I look at the picture of your webpage). Steel constuctions > suffers from lack of damping (steel!). BTW can > emc emit motor signals without special frequencies, > which stimulate machancal resonances? You sould change the code at pid.c, but changes are easy, I've already changed it to introduce controllers with arbitrary poles an zeros, so it could be done, I think that approach is called a notch filter. > If you use a toothed belt this can reduce the > vibrations from motor, but now you have additional > oscillation possibilities. You are right, but even if the belt is very stiff I think it can introduce damping -lowering vibration amplitude- in comparison with other approaches. I'm planning to measure in the linear guide, not in the shaft, so I presume that this effect would be not that important, or at least that I would be able to deal with it using appropriate control. Can you comment on this?. > To implement a feedforward method I see 3 > positions: emc, a microprocessor and a fpga. > Assume emc delivers some signal. Than a additional > module inside emc must put in the predistortion knowing the > rotor positon with high resolution.t, but even If a micro > (see Kirk's AVR page) can calculate a sine modulation > he can do some other function, because usualy > a table with sine values are used. And this table can be modified. > For very fast drives maybe a fpga solution is the way to go. I agree, the most important problem the seems to be to determine the correction that should be introduced to compensate for torque rippling and cogging. I think that this amount to obtain the fourier expansion coefficients terms of of two contributions: one independent of current related to cogging, an a second one, proportional to intensity related to torque rippling. > (I am working on my Raggedstone1 project.) > Is it that card already supported by EMC?. Thanks very much for your comments, Javier ------------------------------------------------------------------------------ Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)! Finally, a world-class log management solution at an even better price-free! Download using promo code Free_Logger_4_Dev2Dev. Offer expires February 28th, so secure your free ArcSight Logger TODAY! http://p.sf.net/sfu/arcsight-sfd2d _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
