Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Coil excitation: Current vs Voltage
Posted 16 lug 2010, 11:03 GMT-4 5 Replies
Please login with a confirmed email address before reporting spam
Hi,
I am new to Comsol 4.0 and I am trying to understand how to modelize an electro-magnet. I used the "Magnetic Fields (mf)" physical interface. I get two different results depending whether I use current or voltage in the coil excitation of the "Multi-Turn Coil Domain Settings". The problem is that both simulation should give the same result.
Here's my setup: 1 [in] iron core, 4 [in] copper coil and 2.5 [in] length. The coil is made of 3450 turns of 21 AWG wire (fill factor taken into account) rated at 1.2 [A] max and 42 [Ohm/km]. According to my calculation, the length of the wire is
688 [m] = N * L_average_of_1_turn
= N * 2Pi * {(R_coil - r_core)/2+r}
= N * Pi* (D_coil + D_core)/2
So the total inductance of the coil should be
29 [Ohm] = 688 [m] * 42 [Ohm/km].
With the relation V=RI, I should have a voltage of 35 [V] to have a current of 1.2 [A]. So the simulation shoud give me the same results whether I use voltage or current in the "Coil excitation" section of the "Multi-Turn Coil Domain".
The thing is that 35 [V] gives me 372.5 [Tesla] and 1.2 [A] gives me 0.061 [Tesla]. What could explain that big difference?
The file is attached to this message.
Thank you very much.
Alexandre
Student in mechanical engineering
I am new to Comsol 4.0 and I am trying to understand how to modelize an electro-magnet. I used the "Magnetic Fields (mf)" physical interface. I get two different results depending whether I use current or voltage in the coil excitation of the "Multi-Turn Coil Domain Settings". The problem is that both simulation should give the same result.
Here's my setup: 1 [in] iron core, 4 [in] copper coil and 2.5 [in] length. The coil is made of 3450 turns of 21 AWG wire (fill factor taken into account) rated at 1.2 [A] max and 42 [Ohm/km]. According to my calculation, the length of the wire is
688 [m] = N * L_average_of_1_turn
= N * 2Pi * {(R_coil - r_core)/2+r}
= N * Pi* (D_coil + D_core)/2
So the total inductance of the coil should be
29 [Ohm] = 688 [m] * 42 [Ohm/km].
With the relation V=RI, I should have a voltage of 35 [V] to have a current of 1.2 [A]. So the simulation shoud give me the same results whether I use voltage or current in the "Coil excitation" section of the "Multi-Turn Coil Domain".
The thing is that 35 [V] gives me 372.5 [Tesla] and 1.2 [A] gives me 0.061 [Tesla]. What could explain that big difference?
The file is attached to this message.
Thank you very much.
Alexandre
Student in mechanical engineering
Attachments:
5 Replies Last Post 2 feb 2011, 17:49 GMT-5
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
read carefully the doc and test out the three options in 2D axi, fo the same geometry
a) single rectangle normal surface current ,
b) single turn coil and,
c) multi turn coil.
There are subtile differences, as COMSOL adds in new variables to compensate for the induced voltage in the single and multiturn coil BC's. There are also differences w.r.t skin effects in harmonic frequency sweeps.
You will then probably better understand the differences, these have been common eror sources for coil FEM analysis
Have fun Comsoling
Ivar
read carefully the doc and test out the three options in 2D axi, fo the same geometry
a) single rectangle normal surface current ,
b) single turn coil and,
c) multi turn coil.
There are subtile differences, as COMSOL adds in new variables to compensate for the induced voltage in the single and multiturn coil BC's. There are also differences w.r.t skin effects in harmonic frequency sweeps.
You will then probably better understand the differences, these have been common eror sources for coil FEM analysis
Have fun Comsoling
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Alexandre,
I am also struggling to understand how to correctly setup up a single domain multi-turn coil that can models skin and proximity effects. Have you been able to figure out how the definition of the current density for a voltage-driven coil is derived? I would greatly appreciate any information you might be able to share.
Regards,
Johan
I am also struggling to understand how to correctly setup up a single domain multi-turn coil that can models skin and proximity effects. Have you been able to figure out how the definition of the current density for a voltage-driven coil is derived? I would greatly appreciate any information you might be able to share.
Regards,
Johan
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
I simply ended up defining the current density of the coil as it was the limits of my problem. I modelized a rectangle in a 2D axi-symetric setup and added en external current density with the mf physics.
I'm sorry I can't help you much.
If you feel what I did could help, don't hesitate!
Best luck!
Alexandre
I'm sorry I can't help you much.
If you feel what I did could help, don't hesitate!
Best luck!
Alexandre
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi,
as explained in another thread, I'd like to go even a step further and define turn by turn, using litz cable with individually isolated wire strands. Our coils probably have only a few turns (maybe 10), and I will have to see local effects.
So I am deeply interested as well, and will keep investigating and posting results.
Matthias
as explained in another thread, I'd like to go even a step further and define turn by turn, using litz cable with individually isolated wire strands. Our coils probably have only a few turns (maybe 10), and I will have to see local effects.
So I am deeply interested as well, and will keep investigating and posting results.
Matthias
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi,
I am new to Comsol 4.0 and I am trying to understand how to modelize an electro-magnet. I used the "Magnetic Fields (mf)" physical interface. I get two different results depending whether I use current or voltage in the coil excitation of the "Multi-Turn Coil Domain Settings". The problem is that both simulation should give the same result.
Here's my setup: 1 [in] iron core, 4 [in] copper coil and 2.5 [in] length. The coil is made of 3450 turns of 21 AWG wire (fill factor taken into account) rated at 1.2 [A] max and 42 [Ohm/km]. According to my calculation, the length of the wire is
688 [m] = N * L_average_of_1_turn
= N * 2Pi * {(R_coil - r_core)/2+r}
= N * Pi* (D_coil + D_core)/2
So the total inductance of the coil should be
29 [Ohm] = 688 [m] * 42 [Ohm/km].
With the relation V=RI, I should have a voltage of 35 [V] to have a current of 1.2 [A]. So the simulation shoud give me the same results whether I use voltage or current in the "Coil excitation" section of the "Multi-Turn Coil Domain".
The thing is that 35 [V] gives me 372.5 [Tesla] and 1.2 [A] gives me 0.061 [Tesla]. What could explain that big difference?
The file is attached to this message.
Thank you very much.
Alexandre
Student in mechanical engineering
Try V=IZ instead and it will work. Then again, I have no idea about how or what your core is doing as I did not see your attached file.