FEMM Help needed

hillzofvalp

100 kW
Joined
Dec 25, 2010
Messages
1,887
Location
Somewhere over the rainbow, Canada
I can do basics of motor model in FEMM with no scripts or add ons. I'd like to simplify the workflow and get a motor spinning in FEMM. If anyone can show me or teach me how it'd be nice... OR perhaps do final analysis on my motor for me.. even better.

Farfle is busy. I heard miles may be able to help as well.

Small 2.5Nm outrunner <50mm diameter.
 
I've only used FEMM for magnetic circuits. Now I do all of my motor design in Emetor. Unless your design requires an IPM rotor, I'd recommend you try it. https://www.emetor.com/
 
Well, it depends what you are trying to do but generally you will group the entire rotor into one group. Then you can select the group and then go up to "Rotate" under the Edit menu. The angle you need to rotate it for each step is going to depend on a number of things (number of poles, number of slots, speed, etc.) but basically you want to make sure you have enough resolution. That's how you actually get the motion. From there, you can do a number of things. If you are looking for cogging torque, you can select the entire rotor group in the magnetics output screen and get the torque at each angle you move through. Plot it out and you get cogging torque. The back-emf takes a little more work. If you go to circuit properties in the magnetics output screen, one of the outputs is flux linkage. Back-emf is just the derivative of flux linkage with respect to time, so if you get the flux linkage at each angle, then you can approximate the back-emf if you relate each angle to time (based on the speed of the rotor).

You can do this with or without Lua scripts. There are plenty of examples on the FEMM website that you can use to modify Lua scripts.
 
Miles said:
I've only used FEMM for magnetic circuits. Now I do all of my motor design in Emetor. Unless your design requires an IPM rotor, I'd recommend you try it. https://www.emetor.com/

That's what I've been using.. over 200 simulations.. I can't get my torque ripple below 12.5%. I'm also unsure if I should bother trying to skew the stator half a tooth.


I guess I'll just stick with emetor. Wish the mesh count was higher than 7000. In femm I have it up at like 50000
 
learningrc said:
Well, it depends what you are trying to do but generally you will group the entire rotor into one group. Then you can select the group and then go up to "Rotate" under the Edit menu. The angle you need to rotate it for each step is going to depend on a number of things (number of poles, number of slots, speed, etc.) but basically you want to make sure you have enough resolution. That's how you actually get the motion. From there, you can do a number of things. If you are looking for cogging torque, you can select the entire rotor group in the magnetics output screen and get the torque at each angle you move through. Plot it out and you get cogging torque. The back-emf takes a little more work. If you go to circuit properties in the magnetics output screen, one of the outputs is flux linkage. Back-emf is just the derivative of flux linkage with respect to time, so if you get the flux linkage at each angle, then you can approximate the back-emf if you relate each angle to time (based on the speed of the rotor).

You can do this with or without Lua scripts. There are plenty of examples on the FEMM website that you can use to modify Lua scripts.

I've been rotating it manually by that method you describe. but it is very time consuming to check every 1 degree and plot it in external program. Wish I could just let FEMM run for a few hours and do it all for me.
 
hillzofvalp said:
I can't get my torque ripple below 12.5%. I'm also unsure if I should bother trying to skew the stator half a tooth.
If you post your simulation here, we can make suggestions. I got torque ripple at nominal torque down to 0.3% when simulating my design. There are a lot of factors.... Anyway, low torque ripple in a simulation is not a guarantee that you'll have a quiet motor.
 
Miles said:
hillzofvalp said:
I can't get my torque ripple below 12.5%. I'm also unsure if I should bother trying to skew the stator half a tooth.
If you post your simulation here, we can make suggestions. I got torque ripple at nominal torque down to 0.3% when simulating my design. There are a lot of factors.... Anyway, low torque ripple in a simulation is not a guarantee that you'll have a quiet motor.

As you alluded to Miles, getting the BEMF deviation from a perfect sine wave as low as possible is a substantially bigger factor in noise abatement than torque ripple.

The funny thing is, when you start to get the BEMF dialed in, the torque ripple also becomes dialed in automatically.
 
Miles said:
hillzofvalp said:
I can't get my torque ripple below 12.5%. I'm also unsure if I should bother trying to skew the stator half a tooth.
If you post your simulation here, we can make suggestions. I got torque ripple at nominal torque down to 0.3% when simulating my design. There are a lot of factors.... Anyway, low torque ripple in a simulation is not a guarantee that you'll have a quiet motor.


Would you be willing to view privately? I'm all about sharing knowledge, but only once I'm in production. Already taken on too much risk and I can't afford others to copy my stator. Feel free to copy my stator after I've reduced or eliminated all my other risks and balanced out my losses
 
Back
Top