We are in the process of evalutng PM-SynRM.
To understand if our design makes sense and are able to verify our design assumptions and results we are using PrepoMax.
For current analysis the motor rpm would be around 9000.
I have attached the model we have developed can you please review and provide us your feedback if we are doing the right thing.
From our model we would like to understand
Max displacement and stresses
a) Due to placement of magnets
b) and without magnets
c) structural/mechanical failures ( including magnet bonding)
For tied contact it’s necessary to define friction as well (value is irrelevant).
Leaving one rotation unconstrained won’t work here because you have to use rigid body constraint to operate on rotational degrees of freedom (like in that case with gears).
Don’t select all surfaces for contact pair, use only those surfaces that are expected to come into contact with each other.
That tie constraint seems to be unnecessary.
If you want to simulate rotating disc coming into contact with magnet, use prescribed rotation (again like with gears) instead of centrifugal load. If you just want to check the stresses due to pure rotation of the assembly, use centrifugal load like here: PrePoMax (CalculiX FEA) - Tutorial 6 - Rotating disc - YouTube
The second model is fine but try using different boundary conditions and see how the response changes. Use the video linked above for reference.
Please find an updated rotor with Adhesive and Magnet. Can you review and let us know if our model is correct?
Our objective is to build an FEA model to
Understand the stress/strain/displacements of/on the rotor to ensure it does not touch the stator( Air gap between the rotor and stator is 1 mm).
Understand at what RPM the adhesive will give away the magnet bonded to the rotor
Stress/strain/displacement on/of magnets based on RPM
Do we have to include momentum ( torque) to understand the effects or we can use Torque in the final test as stated below?
/4. Finally, we want to insert the rotor shaft and study the results./
/Please note from a Magnetostatic and Electrostatic analysis standpoint we are using other tools and models./
Based on the analysis we will make modifications to the rotor/design to achieve desired RPM, torque and etc…
NGO steel, Adhesive, and Ferrite specs are obtained from the below link
The model is underconstrained - magnet and adhesive separate from the rotor. Define contact pairs only for the surfaces that are in contact or may come into contact during the analysis (as suggested before). Use tie constraint or merge the parts if you don’t want to allow for any separation or slipping of the surfaces. Otherwise use standard contact but add boundary conditions to the magnet so that it can’t fly away.
Maybe you could start with a 2D analysis - will take less time to solve and then, once the setup is correct, you can proceed to 3D simulation if needed.
Thank you for your feedback.
Let me update the model and send it for your review and feedback.
One quick question. When you say merge the parts are u saying to create a compound ?.
If yes can you elaborate on how can we measure the outputs ? i.e. stress/strain/displacement/shear of adhesive and magnets ?..
Yes, touching parts can be turned into a compound if you don’t want them to move relative to each other during the analysis. Tie constraint has the same effect but merging is usually safer.
However, if you want to account for the relative motion of the parts and verify the stresses (including pressure and frictional stress at the interface) and displacements caused by this motion then regular contact will be necessary. Of course, tie constraint can be used at the same time for parts that should be bonded permanently.
You should be able to detect, at least, in a first order of magnitude, if your model is minimally set up or if there is something weird on the results. You are asking for review without even set up the contact parameters. Your questions are surprising to me. If your model is going to be part of a project with commercial purposes (I hope not) I strongly suggest you look for an external advisor. You can also evaluate by hand and very accurately the centrifugal force acting on your magnet and which is the expected contact pressure.
@Jakub
I really appreciate your help to the community, me myself I enjoy helping others very much too, but I don’t think it’s convenient nor safe to encourage someone involved in a commercial design project to use a tool for which they do not have a minimum of training/understanding. Sorry to say that but most likely he/she will end up shooting himself in the foot.
Thank you for your feedback.
I watched your videos ( as a matter of fact-all) and It seems to me, as we are not well versed in the tool usage… which is resulting in creating some dumb mistakes.
Performing finite element analysis is not about learning to use the simulation tool, as PrePoMax. It is first abour understanding the finite element method and understanding the physics of your problem. That is why it takes some time and experience to learn to apply the method correctly.
