{"id":252654,"date":"2023-03-06T06:52:33","date_gmt":"2023-03-06T06:52:33","guid":{"rendered":"\/forum\/?post_type=topic&p=252654"},"modified":"2023-03-06T06:52:33","modified_gmt":"2023-03-06T06:52:33","slug":"maxwell-2d-permanent-magnet-bldc-motor-using-injection-molded-neodymium-magnet","status":"closed","type":"topic","link":"https:\/\/innovationspace.ansys.com\/forum\/forums\/topic\/maxwell-2d-permanent-magnet-bldc-motor-using-injection-molded-neodymium-magnet\/","title":{"rendered":"Maxwell 2D permanent magnet BLDC motor using injection molded neodymium magnet"},"content":{"rendered":"

Hello,<\/p>\n

I am working with Electronics desktop 2023R1 Maxwell 2D, trying to analize a permanent magnet BLDC motor. The permanent magnet that I am using for my model is a “injection molded neodymium magnet” and I am struggling with the setups and its results. I will try to explain what I wanted to perform with Maxwell 2D, what I did and results comparisons. This will be a bit long paragraph. I would like to get some help if there are anyone expirenced with similar model.<\/p>\n

Background<\/span><\/p>\n

I already have a physical motor that I can test and I am trying to compare between simulation and measured data.<\/span><\/p>\n

To make simulation results realistic, I am trying to matching up the shape of magnetic flux as sinusoidal or trapezoidal form using plastic injection cylindrical type magnet.<\/span><\/p>\n

Because permanent magnetic flux shape of this motor is trapezoidal.<\/span><\/p>\n

Normally, in material set-ups on <\/span>Ansys<\/span> Electronics, I set the magnetization direction of a permanent magnet either radial (Cylindrical coordinates) or parallel (Cartesian coordinates) as pic below:<\/span><\/p>\n

\"\"<\/span><\/p>\n

\"\"<\/span><\/p>\n

In the model, magnet separated by number of poles<\/span><\/p>\n

 <\/p>\n

With above set-up (Radial magnetize), No-load BEMF shapes looked as [Fig. A] while real sample measured BEMF shapes looked like [Fig. B].<\/span><\/p>\n

Results shows both the shape and value of BEMF data was different.<\/span><\/p>\n

\"\"<\/span><\/p>\n

[Fig. A] No load BEMF simulation results at 1000 RPM<\/span><\/p>\n

 <\/p>\n

\"\"<\/p>\n

[Fig. B] No load BEMF measured results at 1000 RPM<\/span><\/p>\n

 <\/p>\n

 <\/p>\n

Questions<\/span><\/p>\n

\"\"<\/p>\n

[Fig. C] Permanent magnet B-H curve<\/span><\/p>\n

 <\/p>\n

\"\"<\/span><\/p>\n

[Fig. D] Permanent magnet material<\/span><\/p>\n

 <\/p>\n

1. Double check on data input<\/span><\/p>\n

My input data is following: <\/span>Hc<\/span> (#A) as coercive force <\/span>Hcb<\/span>, Br (#B) as residual induction and Mu (#C) calculated value using <\/span>Hcb<\/span> & Br & u<\/span>0<\/span>. Those values automatically goes into relative permeability, magnitude of magnetic <\/span>coercivity<\/span> from Mu (#C) and <\/span>Hc<\/span> (#A) respectively. Regardless of magnet shape, is there any error input on permanent magnet material set-ups?<\/span><\/p>\n

 <\/p>\n

2. Magnetize <\/span><\/p>\n

Assuming my set-up was correct, to form magnetic flux as trapezoidal, I used parametric equations since Maxwell support mostly everything.<\/span><\/p>\n

I made trapezoid graph on dataset (Project > Dataset) as shown [Fig. E]<\/span><\/p>\n

Using this dataset and internal function called “<\/span>pwl_periodic<\/span>”, I modified magnitude value from a single number to parametric equation. Also, I united the magnet model as single part as well (separated by number of poles <\/span>à<\/span> one single part)<\/span><\/p>\n

\"\"<\/p>\n

–<\/span>Coordinate<\/span>: cylindrical <\/span><\/div>\n
–<\/span>Magnitude<\/span>: <\/span>–<\/span><\/u>356000<\/span><\/u>*<\/span>pwl_periodic<\/span>($ds1,2*Phi*180\/pi<\/span>)<\/span><\/div>\n

 <\/p>\n

The results looked like [Fig. F] when I use magnitude as above equation. It shows BEMF shape pretty similar to real measurement; refer to [Fig. B]<\/span><\/p>\n

\"\"<\/span><\/p>\n

[Fig. F] BEMF results using parametric magnitude (magnet)<\/span><\/p>\n

 <\/p>\n

\"\"<\/span><\/p>\n

[Fig. E] Data set<\/span><\/p>\n

 <\/p>\n

Question is: is this the right way to magnetize when the magnetic flux has specific periodic shapes?<\/span><\/p>\n

\"\"<\/span><\/p>\n

 <\/p>\n

3. Amplitude of magnitude value<\/span><\/p>\n

Obviously, the higher amplitude creates the higher BEMF values. However, using given <\/span>Hc<\/span> value did not make the results similar to measured value.<\/span><\/p>\n

–<\/span>Magnitude<\/span>: <\/span>–<\/span><\/u>356000<\/span><\/u>*<\/span>pwl_periodic<\/span>($ds1,2*Phi*180\/pi<\/span>)<\/span><\/div>\n

To make BEMF value similar to each other I had to somewhat increase scale value. Now, about 9 % error exist with respect to simulation results but I would like it to be less then 5 %. <\/span><\/p>\n

Is there right way to select proper amplitude value?<\/span><\/p>\n

\"\"<\/span><\/p>\n

 <\/p>\n

4<\/span>. <\/span>Permeance coefficient of magnet<\/span><\/p>\n

I believe one of the reason why existing gap between simulation and real measurement is a permeance coefficient of the permanent magnet. In my opinion, the Permeance coefficient has something to do with surface magnetic flux.<\/span><\/p>\n

Is the Maxwell able to calculate the coefficient in any way? If so, is it possible to share simple manual?<\/span><\/p>\n

 <\/p>\n

Those quesitons are pretty self-related each other. Again, if any help on these problem, I would be really appreciated.<\/p>\n

Thank you! <\/p>\n","protected":false},"template":"","class_list":["post-252654","topic","type-topic","status-closed","hentry"],"aioseo_notices":[],"acf":[],"custom_fields":[{"0":{"_bbp_author_ip":["23.192.164.14"],"_bbp_subscription":["271461","186775","302002"],"_btv_view_count":["1675"],"_bbp_topic_status":["unanswered"],"_bbp_status":["publish"],"_bbp_notification_enabled":["115069","115069"],"_bbp_topic_id":["252654"],"_bbp_forum_id":["27793"],"_bbp_engagement":["186775","271461","302002"],"_bbp_voice_count":["3"],"_bbp_reply_count":["4"],"_bbp_last_reply_id":["317354"],"_bbp_last_active_id":["317354"],"_bbp_last_active_time":["2023-11-13 00:51:44"]},"test":"hyukjin-leecoavis-com"}],"_links":{"self":[{"href":"https:\/\/innovationspace.ansys.com\/forum\/wp-json\/wp\/v2\/topics\/252654","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/innovationspace.ansys.com\/forum\/wp-json\/wp\/v2\/topics"}],"about":[{"href":"https:\/\/innovationspace.ansys.com\/forum\/wp-json\/wp\/v2\/types\/topic"}],"version-history":[{"count":0,"href":"https:\/\/innovationspace.ansys.com\/forum\/wp-json\/wp\/v2\/topics\/252654\/revisions"}],"wp:attachment":[{"href":"https:\/\/innovationspace.ansys.com\/forum\/wp-json\/wp\/v2\/media?parent=252654"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}