[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] Terry filters (speaker/motor load modeling)

In a message dated 1/2/08 9:27:33 P.M. Eastern Standard Time,  
Gary.Lau@xxxxxx writes:

>It sounds like you have a far better understanding of motor theory  than I 
do (which truthfully is very 
>little); perhaps you can point out where my reasoning leaves the  tracks.

Thanks, Gary. I'm no expert by any means. I  always enjoy your posts because 
they cut right to the chase!

>My thinking is that in a loaded motor, the lion's share of the power  
consumed is delivered to the 
>mechanical load.  In a simulation model, this power must leave  the model, 
and the only mechanism 
>to do this is to burn off that power in a resistor to represent the  
mechanical load.  If the correct 
>model for a motor is predominantly inductive as you suggest, I don't  see 
where the power delivered 
>to the mechanical load is accounted for, as the power dissipated in  an 
inductor is zero.

OK, now I can see why that would be a problem for  modeling purposes!
    But resistance in the windings can't be the  answer, either - a 20 
horsepower AC induction motor can have a nominal  efficiency of 95%, so most of the 
energy isn't going into heat in the motor  windings. One would guess a 
superconducting motor would have only 5% better  efficiency. 
    But that doesn't help with the model! FWIW, I  think those nominal 
efficiencies are supposed to be at 35% load (energy  conservation legislation, 
average loading of motors in industry per extensive  research). Maybe a little more 
or less at full load. But if it spins free with  no load, it is doing *no* 
mechanical work and its efficiency is zero! If it  gets locked up, the rotor 
doesn't turn, yet the currents go through the roof,  it again does no work (the 
rotor isn't moving), so the efficiency is zero.  

>I agree that an unloaded motor is best modeled as inductive, but that  
adding a load must add 
>resistive loading.

    I think you're right, to be accurate we need both  inductive and 
resistive components. 
    Here's what I just found:
    Nice to see we were on the right track! 
    I guess the very simple version would be to say at  no load it's 
inductive, at full load it's resistive. If you need to model  in-between, you'll have 
to account for whatever load you put on it.
-Phil LaBudde
Center for the Advanced Study of Ballistic  Improbabilities

**************************************See AOL's top rated recipes 
Tesla mailing list