Motor model for Starting

[HamidEle]

Hi,

Some maufacturers can provide Cicuit data and Speed-torque Data. I am not sure which one is more accuarate for the simulation. From the field test feedback, It looks like speed-torque caracteristic is more appropriate for the simulation purpose. I don't really understand the reason behind. I would like to listen to the comments from any experts on this. Any comments will be appreciated.

 

[electricpete]

By conservation of energy, the circuit model calculation would give the right answer IF you had the exactly right circuit paramater for every point in the curve and IF those parameters didn't change as slip changes.

But the parameters do in fact change with slip. A few effects to mention:
1 - deep bar effect makes the effective resistance of the rotor higher at high slip
2 - the magnetic permeability is not constant... it depends on the flux (which changes with slip in various parts of the machine) in a non-linear manner with the most pronounced changes occuring near saturation. The toothtops may saturate changing the leakage reactance. Also the bridge over the rotor bars in closed slot rotor (random wound machine) will surely be deep into saturation at high slip (which corresponds to high current).

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

also the circuit model has to be corrected for additional losses beyond the rotor losses.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[aolalde]

I agree with Electricpete. The speed-Torque curve is more accurate for simulation since the equivalent circuit in reality has variable resistance and reactance parameters from zero to synchronous speed.
May be a computerized equivalent circuit program that allows adjustment of the parameters for every speed change will give more realistic results.

 

[LionelHutz]

If you have the true speed vs torque and speed vs current curves then you can easily simulate the motor start. You also need the speed vs required load torque and inertia of the rotating assembly.

If you know the torque the motor produces and the torque the load requires then you subtract and get the accelerating torque. You use a version of

accel time = (WK^2 x speed change)/(308 x accel torque)

Solve for speed change and then calculate the speed change for small dt steps.