Symmetry problems

[Zelioo]

Hello everybody.

We've been using pyleecan in order to plot magnetic field lines of various synchronous reluctance motors using built-in pyleecan models and custom-made materials.

Yet, even when using the default features, we've stumbled on an intriguing matter : The field lines don't look symmetrical at all. Does anyone know if this behavior is normal or if there are ways to improve the 'look' of the field lines.

Best regards.

[Zelioo]

I'm sorry if it's a basic mistake from my side but I'm just a beginner with Pyleecan

[BonneelP]

Hello,

Sorry for the answer delay, I wanted to investigate by simulating the machine on my side but I didn't have the time. My guess would be that the machine has p=2 => 2 pole pairs which means that only a single pole is drawn in FEMM and then I expect FEMM to have set anti-periodic boundary conditions. To check if the simulation is correct, you can remove the symmetry to draw the full machine (simu.mag.is_peridodicity_a = False)

Best regards, Pierre

[crowdnidhi-design]

**Dear Pierre,

We tried to update the simu.mag.is_peridodicity_a = False to True in your code example( notebook ) below but it seems it get stuck

look at the error below ( if we skew the rotor does it create problems ?.)**

** Exception Traceback (most recent call last) Cell In[5], line 1 ----> 1 output_ref = simu_ref.run()

File c:\users\haris\appdata\local\programs\python\python38\lib\site-packages\pyleecan\Methods\Simulation\Simulation\run.py:52, in run(self) 49 self.init_logger(results) 51 # Compute the simulation ---> 52 self.run_single() 54 if self.index is None and self.layer == 0: 55 msg = "End of simulation"

File c:\users\haris\appdata\local\programs\python\python38\lib\site-packages\pyleecan\Methods\Simulation\Simu1\run_single.py:31, in run_single(self) 29 self.elec.run() 30 if self.mag is not None: ---> 31 self.mag.run() 33 if self.force is not None: 34 self.force.run()

File c:\users\haris\appdata\local\programs\python\python38\lib\site-packages\pyleecan\Methods\Simulation\Magnetics\run.py:46, in run(self) 44 if Nslices > 1: 45 self.get_logger().info("Solving slice 1 / " + str(Nslices)) ---> 46 out_dict = self.comp_flux_airgap(output, axes_dict, Is_val=Is_val, Ir_val=Ir_val) 48 is_loop = True 50 if Nslices == 1:

File c:\users\haris\appdata\local\programs\python\python38\lib\site-packages\pyleecan\Methods\Simulation\MagFEMM\comp_flux_airgap.py:169, in comp_flux_airgap(self, output, axes_dict, Is_val, Ir_val) 146 ( 147 B_elem, 148 H_elem, (...) 165 filename=self.import_file, 166 ) 167 else: 168 # Without parallelization --> 169 B_elem, H_elem, mu_elem, A_node, meshFEMM, groups, A_elem = self.solve_FEMM( 170 femm, 171 output, 172 out_dict, 173 FEMM_dict=FEMM_dict, 174 sym=sym, 175 Nt=Nt, 176 angle=angle, 177 Is=Is_val, 178 Ir=Ir_val, 179 angle_rotor=angle_rotor, 180 is_close_femm=self.is_close_femm, 181 filename=self.import_file, 182 ) 184 # Store FEMM_dict in to avoid drawing the machine several times 185 output.mag.internal.FEMM_dict = FEMM_dict

File c:\users\haris\appdata\local\programs\python\python38\lib\site-packages\pyleecan\Methods\Simulation\MagFEMM\solve_FEMM.py:217, in solve_FEMM(self, femm, output, out_dict, FEMM_dict, sym, Nt, angle, Is, Ir, angle_rotor, is_close_femm, filename, start_t, end_t, Nmess) 214 femm.mi_analyze() 216 # Load results --> 217 femm.mi_loadsolution() 219 # Get the flux result 220 if is_sliding_band:

File c:\users\haris\appdata\local\programs\python\python38\lib\site-packages\pyleecan\Classes_FEMMHandler.py:2297, in _FEMMHandler.mi_loadsolution(self) 2296 def mi_loadsolution(self): -> 2297 self.callfemm("mi_loadsolution()")

File c:\users\haris\appdata\local\programs\python\python38\lib\site-packages\pyleecan\Classes_FEMMHandler.py:280, in _FEMMHandler.callfemm(self, myString) 278 # print(myString) # uncomment to see all the calls made to femm while working 279 if is_windows_os: --> 280 x = self.callfemm_Windows(myString) 281 else: 282 x = self.callfemm_Linux(myString)

File c:\users\haris\appdata\local\programs\python\python38\lib\site-packages\pyleecan\Classes_FEMMHandler.py:218, in _FEMMHandler.callfemm_Windows(self, myString) 216 elif x[0] == "e": 217 ErrorMsg = x.replace(",", " ").replace("1j", "I") --> 218 raise Exception(ErrorMsg) 219 else: 220 x = eval(x)

Exception: error: Some regions in the problem have been defined by more than one block label. These potentially problematic regions will appear as selected in the initial view.

[Zelioo]

Thank you very much, Setting the parameter to true ( or in my case unticking the angular periodicity in the femm simulation window in the gui ) now allows me to plot, the full machine. Yet, for some reasons when I get the femm file after running the simulation ( instead of getting it from the Drawing Summary window) it seems that the flux density is equal to 0 everywhere.

[BonneelP]

Great :) I think the reason is that we save the .fem file just after drawing the machine. The current are set in a second step so maybe you have the original fem file without current ? Normally when running the simulation from the GUI, in the result folder you should find both .fem and .ans with the current defined in the last step of the GUI.

[Detest1]

but I think we don't get current info please. We were also confused sometimes.

[Detest1]

We tried to open the .fem file using FEMM 4.2 and check the circuit properties but did not find the current info

[Zelioo]

Well, after launching the simulations from the GUI I get a femm file which when opened in FEMM doesn't show any magnetic flux. In the other hand, when I open the femm file given before simulation ( and before defining current ), it does show flux density.