Learning Forum

[fatim119]

Hello,

I am simulating rotor core loss in ANSYS Maxwell 2D using a transformer-based excitation setup, similar to the ring core method used in experimental validation of lamination materials. In this method, the rotor of an induction motor is excited via a primary winding, and the induced voltage is measured on a secondary winding, allowing estimation of flux density and core loss through transformer principles.

I attempted to replicate this setup in Maxwell by energizing the primary winding and modeling the secondary winding using the "End Connection" configuration. Material properties are based on the manufacturer’s lamination datasheet, where—for example—at 0.1 T, the corresponding current is 0.11 A.

However, when I use this value (0.11 A) in the simulation:

• The total core loss is extremely low (e.g., ~21.21 µW),

• The flux density at a probe point in the yoke is only ~0.964 µT,

• The magnetic field intensity (H) is also negligible in the yoke.

Interestingly, when I move the probe closer to the rotor slot, the values of B and H increase, but as I move it into the rotor yoke, they drop significantly. This is inconsistent with my expectations based on experimental results, where flux density in the yoke is non-negligible and contributes significantly to the overall core loss.

My questions are:

1. Is it valid to directly use the current value from the manufacturer’s lamination data (e.g., 0.11 A at 0.1 T) in such a winding-based model?

2. Why is the flux density so low in the yoke, despite reasonable excitation in the primary winding? I understand that the excitation is not strong enough however, comparing to experimental data it should be enough to get the desired value.

3. Is Maxwell integrating rotor core loss across the entire rotor lamination volume, or is it relying solely on the probe point?

4. How can I better replicate the ring method (transformer principle) in Maxwell to accurately simulate flux penetration and loss behavior in the rotor?

Any guidance on correctly configuring the excitation or interpreting the B-field behavior in this transformer-type setup would be very helpful. Thank you!

[MirandaH]

Is it valid to directly use the current value from the manufacturer’s lamination data (e.g., 0.11 A at 0.1 T) in such a winding-based model? Yes, you will also need to enter the core loss information as well.

[MirandaH]

Why is the flux density so low in the yoke, despite reasonable excitation in the primary winding? I understand that the excitation is not strong enough however, comparing to experimental data it should be enough to get the desired value.

We will need to take a look at the model, unfortunately, this is beyond the scope for Ansys employees to answer on a public forum as it requires to verify the model. Ansys employees cannot access any attachments on the forum. We encourage other participants on the forum to help with your problem. 

If you have Ansys Service access please raise a service request.

[MirandaH]

Is Maxwell integrating rotor core loss across the entire rotor lamination volume, or is it relying solely on the probe point?

Maxwell calculates core losses using a combination of material properties, field solutions, and solver-specific methods depending on the simulation type (e.g., frequency-domain or time-domain).

[MirandaH]

 

How can I better replicate the ring method (transformer principle) in Maxwell to accurately simulate flux penetration and loss behavior in the rotor?

Please refer Examples->Maxwell->Transformer Electronics/Power 

Or create a Maxwell model from RMxprt examples

 

[fatim119]

Hello MirandaH,

Thankyou for your response. They were helpful in identifying some of the errors. So i) I will be then keeping my current same as the experimentatal value. 
2) I noticed one thing that the windings are across the teeth in my simulation but in experimentation the windings are across the yoke of the rotor lamination. so now it justifies that a small value of current is having an impact on the lamination level core loss. So I don't know how to do the winding across the core (yoke) more like a toroidal winding in 2D maxwell?
3) To validate core loss vs flux density and inductance vs flux denisty I am running a solver Transient. Is it suppose to be Transient to get such results or Eddy current solver?
4) I want to share model with you. There is no way to share? 
5) Experimentally I got Permeability vs flux density graph as well but in Ansys I am unsure how to attain permeability vs flux density and inductance vs flux density. Please know that I build my design in SolidWorks then imported in the Ansys So I am not doing in Rmxprt. Only Maxwell 2D Ansys. Please help.