Simulate simple eddy current brake
Hi all,
I am very new to Flux and just acquired the student licence to try and parametrically develop a simple eddy current brake using permanent magnets.
For now, I have created a rotating disk and a fixed, simple magnet and am trying to get the resulting torque created around the center axis of my disk.
However, as soon as I add sensors to read out the data that I think is necessary, I get hit with a "The region [...] is in contact with the sliding zone." and I can't figure out what I'm doing wrong.
All help is greatly appreciated, and please have mercy with a newby asking stupid questions.
Cheers,
Tom
Best Answer
-
Hello Tom;
When you compute the magnetic force on the region, this one should be completly surrounded by air.
If you you use mechanical sets, then the mobile part should be surrounded by air too. So, the air surrounding the mobile part should be devided on two parts, one is mobile, and the other one is fixed. The common line or face is called sliding zone.
The error message you have means that your mobile part is not surrounded by air, and ferromagnetic part is in contact with the sliding zone.
I hope this will help you
Best regards
2
Altair EmployeeAnswers
-
Hello Tom;
When you compute the magnetic force on the region, this one should be completly surrounded by air.
If you you use mechanical sets, then the mobile part should be surrounded by air too. So, the air surrounding the mobile part should be devided on two parts, one is mobile, and the other one is fixed. The common line or face is called sliding zone.
The error message you have means that your mobile part is not surrounded by air, and ferromagnetic part is in contact with the sliding zone.
I hope this will help you
Best regards
2 -
Farid zidat_20516 said:
Hello Tom;
When you compute the magnetic force on the region, this one should be completly surrounded by air.
If you you use mechanical sets, then the mobile part should be surrounded by air too. So, the air surrounding the mobile part should be devided on two parts, one is mobile, and the other one is fixed. The common line or face is called sliding zone.
The error message you have means that your mobile part is not surrounded by air, and ferromagnetic part is in contact with the sliding zone.
I hope this will help you
Best regards
Hello Farid,
thank you for your quick answer!
I thought I had accounted for the "air-issue" by putting two seperate air volume regions (one fixed and one mobile) in between the fixed and mobile parts (exactly the size of the respective parts) and then sourrounding everything with the infinity cube (compressible air). But I'll definetly try out dividing my entire volume into two seperate air volumes.
Just to make sure I am on the right track in this regard as well:
If I want to simulate the braking moment in the center of the "flywheel" I'll have to add a sensor to the magnet volume region and determine the magnetic forces, right? This way I can simply calculate the resulting torque. Or is the "magnetic torque" that is outputted by default already the correct value?Thanks in advance,
Tom
0 -
You are welcome!
Actually if you use rotating mobile mechanical set, you can compute the torque directly on the mechanical set.
You can also create a sensor to compute magnetic force on region surrounded by air if needed.
Regards
1 -
Farid zidat_20516 said:
You are welcome!
Actually if you use rotating mobile mechanical set, you can compute the torque directly on the mechanical set.
You can also create a sensor to compute magnetic force on region surrounded by air if needed.
Regards
Sorry to bother you another time, but even after putting volumes of air arround everything, i still get the sliding zone error, which results in a missing output for any sensor/torque output.
Here's my setup in detail:Volume 2 is my rotating disk (Aluminium)
Volume 4 is my permanent magnet
Volume 5 is the air surrounding the disk (1mm in each direction)
Volume 1 is the air surround the magnet (1mm in each direction)
So to not have meshing errors due to multiple points being in the same space, I had to put V2 and V5 and V4 and V1 in one assemblies respectively. Between both volumes of air I left a .1mm gap.
I then create the infinte box surrounding everything.
All Volumes of air are compressible (even with the air surrounding the disk being part of the mechanical set of the disk itself doesnt change the outcome)The disk is part of a mechanical set that rotates around z and the magnet is fixed in space.
Only the disk is a "solid conductor region".
When I run the solver, I do get results in the form of the el. torque being displayed and I can use the isoval. and other graphical methods once the solver is finished, but at the end of each solve, I get a bunch of:
The region SCHWUNGSCHEIBE has a relative permeability close to 1.
The spatial group V_SCHWUNGSCHEIBE is in contact with the sliding zone.
*** Computing TMag(V_SCHWUNGSCHEIBE) is impossible. ***outputted for both the magnet (V4) and the disk (V2).
I have tried making the outer volumes of the infinite box fixed with no success, getting rid of all seperate air volumes, but the ones from the infinite box, with no success either.
I can only guess that maybe putting an object in the same assembly as the air that is surrounding it might be the issue, but if I dont do that, i cannot mesh at all. And when ditching the seperate air volumes (even only one) I still get the same error message.
I am starting to dream about disks colliding and sliding around, so any help trying to get around this upcoming trauma is greatly appreciated.
0 -
Hello Tom;
For a bigginer, start with 3D computation + Eddy current is not easy without training. What i can suggest is first to send us the project, so I can have a look on it.
In other side, you can may be try to follow one of the basic examples in 3D from the supervisor, for instance the following one:
Best regards
1 -
Farid zidat_20516 said:
Hello Tom;
For a bigginer, start with 3D computation + Eddy current is not easy without training. What i can suggest is first to send us the project, so I can have a look on it.
In other side, you can may be try to follow one of the basic examples in 3D from the supervisor, for instance the following one:
Best regards
Hi Farid,
after a bit of experimenting around, I found out that my issue was, that I was trying to calculate the magnetic forces on the magnet itself (for whatever reason i thought this would be easier to calculate with later on). Apparently trying to get forces from a fixed body isnt such a great idea...
When I changed the sensors to measure forces on the flywheel (or pulled the data from the mechanical set itself) everything worked flawlessly - even without my triple air layer monstrocity.Thank you again for all your help, I think I'm gonna try to take it from here
Kind regards,
Tom
0 -
Hi Tom;
Thank you for your feedback.
Enjoy Eddy currents
Regards
0 -
Hi Farid,
I thought I had figured out my issues with my simulation, but revisiting my project showed me i really haven't. The "surrounding air issue" appears to only be part of my problem, since my torque values are all over the place, and differ greatly whether I'm using periodicity or modelling the entire eddy current brake. In addition, my meshing results are all over the place too and i can't really figure out what I'm actually doing wrong here.
For now, I've been trying to sort of copy this method (p.92) to at least get some comparable results in order to simulate my own version of this brake. Looking at my graphed torque, I appear to have messed up big time.
Attached you will find my project file. I would greatly appreciate if you could look it over and maybe give me some advice on how to at least get some half decent looking results.
Thank you in advance!
Tom
0 -
Hi Tom;
I will have a look on your project. The first thing i can see from the first glance is that, the periodicity should be odd, in addition to that, the mesh should be improved.
I will let you know.
Regards
0 -
Hi Tom,
I worked a little bit on your project to improve the mesh, so i rebuilt it to have full extrusive mesh, and refined the mesh on the area where the magnet is mooving. The results are better now.
Please find in attchement the Flux project ready to mesh and to solve.
Regards
1 -
Farid zidat_20516 said:
Hi Tom,
I worked a little bit on your project to improve the mesh, so i rebuilt it to have full extrusive mesh, and refined the mesh on the area where the magnet is mooving. The results are better now.
Please find in attchement the Flux project ready to mesh and to solve.
Regards
Hi Farid,
sorry for the late reply.
Thank you so much for your help, I really appreciate it. It took me a while to get my hands on a 2021.1 student licence, but I got it to work nicely.
I'm going to try to "reverse engineer" your meshing method to hopefully get my future sims done by myself.
Best wishes,
Tom
0 -
Tom Hauer_22107 said:
Hi Farid,
sorry for the late reply.
Thank you so much for your help, I really appreciate it. It took me a while to get my hands on a 2021.1 student licence, but I got it to work nicely.
I'm going to try to "reverse engineer" your meshing method to hopefully get my future sims done by myself.
Best wishes,
Tom
Hi Tom;
Don't worry, and thanks for your feedback!
Enjoy using Flux and meshing features in Flux. However, the Student version is limitted to a given number of mesh elements.
Regards
0