Discussion Forum

Hi

I'm trying to solve very interesting task but don't know how.
A permanent magnet (Neodymium as material) is surrounded by circular coil both packed in aluminium case, so it's a magnetic sensor.
There is a sur gear under the sensor (exactly under the magnet, air gap 3..4 mm). When the gear is rotating the changing magnetic field is obtained and this field induces a voltage in coil around magnet.

I've tryied to use "Magnetic fields" physics to provide values of remanent flux density in magnet, coils parameters and other. And also I've used a "Multibody dynamics module" to rotate the gear.
When COMSOL calculates results with stationary solver there's normal results (magnetic flux density, e.g. which is max in the magnet, "Coil geometry analisys" is also used).
But when I enable Time-Dependent study step and "Multibody" module (to rotate the gear) there is a message like "Can't find consistent initial values" or "Resudial is more times higher (2.1e+02 e.g.) than the relative tolerance" or smth about this.

Well, then I've tryied to use only "Rotating Machinery, Magnetic" module to set physics conditions and rotation of gear but there still message "Resudial is more ...".

The question is: what physics interface shouls I use to solve this task?

--
There'is some interesting thing. First time I used "Magnetic Field, No Currents" physics to create permanent magnet. Then I addded "Magnetic Fiels" physics (no coils, just blank interface). But the plots of magnetic flux density ontained from both these physics weren't similar! The plot from "MF" shew zero in all donains and the same time plot from "MFnC" shew some values of magnetic flux. Are these physycs devided in solving process?
--

Thnx so much.