Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

Electric arc Furnace Comsol 4.1

Please login with a confirmed email address before reporting spam

Hello.

I am writing to you from Tallinn University of Technology, Estonia. My doctoral work thesis is titled „Electric arc furnace with controllable liquid metal circulation”.

My work is focused on electromagnetic fields and their effects. I have to make 3D computer model of liquid metal movement due to effects of electromagnetic field. I would need to make calculations of forces acting within molten metal in electric arc furnace due to electromagnetic fields.

However, I have some problems with my model. To know how Comsol 4.1 works I built up first one simple test model (Fig.1). There are two top electrodes and two bottom electrodes. Electrode pairs are connected to two separate independent power sources as shown in Fig.2. I want that electrode currents goes directly from top to bottom electrodes. For this reason I need two independent power sources. To build up such a scheme, I have used previously in Comsol 3.5 SPACE Circuit Editor, where I have written a program:

V1 1 2 sin(0 400 50Hz 0 0 0)
R1 1 0 1000
V2 3 4 sin(0 400 50Hz 0 0 120)
R2 3 0 1000
X1 1 2 3 4 ComsolModel
SUBCKT ComsolModel a b c d COMSOL: *
ENDS

Resistors R1 and R2 are introduced, because then I have some connection of the source to the ground. Otherwise SPACE Circuit Editor does not work at all. It wants that some points are grounded.

Now I have Comsol 4.1 ( we just got a newer version) and now I do not know anymore how to connect my circuit to the model ? There is no SPACE Circuit Editor anymore and tried many ways but don’t get it work….

Heigo Mõlder
Researcher, PHD-student
Tallinn University of Technology
Fundamentals of Electrical Engineering and Electrical Machines
Phone: +372 5264992
E-Mail: Heigo.Molder@ttu.ee


10 Replies Last Post 16.12.2011, 15:58 GMT-5
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 04.04.2011, 07:52 GMT-4
Hi

in ACDC V4 you have the "Electric Circuit" (CIR) physics as in 3.5, you must add it a s a separate physics.
Then in MEF or ES, EC, ECS you have the terminal node to be added manually, in MEF its under "Magnetic Isolation - Terminal" (it's not in MF since you do not solve for "V")

Within the Terminal GUI there is an option "Circuit" where the terminal name is the circuit port number

I agree it's slightly complex to begin with to find your way in here but rather logical once you have understood it

--
Good luck
Ivar
Hi in ACDC V4 you have the "Electric Circuit" (CIR) physics as in 3.5, you must add it a s a separate physics. Then in MEF or ES, EC, ECS you have the terminal node to be added manually, in MEF its under "Magnetic Isolation - Terminal" (it's not in MF since you do not solve for "V") Within the Terminal GUI there is an option "Circuit" where the terminal name is the circuit port number I agree it's slightly complex to begin with to find your way in here but rather logical once you have understood it -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 06.04.2011, 04:48 GMT-4
Thank you for your response.

I have plane to used MF module, because there I can calculate the pressure of liquid metal......but, what this mean: (it's not in MF since you do not solve for "V").....what this "V" means ?, how i can solv "V" before if I don´t have working circuit....?
I've tried allredy three days to connect my circuit to model but always some error comes :(

I can´t even normal resistor connected to a power source, some error comes all the time, see attachment.

Heigo.


Thank you for your response. I have plane to used MF module, because there I can calculate the pressure of liquid metal......but, what this mean: (it's not in MF since you do not solve for "V").....what this "V" means ?, how i can solv "V" before if I don´t have working circuit....? I've tried allredy three days to connect my circuit to model but always some error comes :( I can´t even normal resistor connected to a power source, some error comes all the time, see attachment. Heigo.


Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 06.04.2011, 05:29 GMT-4
Hi

V is the voltage, MF solves for A but not V, (MEF does for both) check the dependent variables, so it seems that you cannot link explicitly to the circuit port.

But there must be a way around with MF, perhaps by defining the circuit voltage as an external variable and using that in the MF, I would have to check (I'm stuck in some structural just now, I'll try some time later)

But careful when mixing MF and MEF in the same model (normally you need only one of them) to avoid getting confused with the similar variables and names

My way around when I get errors, is really to start fundamentally simple and build again up step by step

I believe you have missed something in the CIR model, your node 1 is "hanging"

--
Good luck
Ivar
Hi V is the voltage, MF solves for A but not V, (MEF does for both) check the dependent variables, so it seems that you cannot link explicitly to the circuit port. But there must be a way around with MF, perhaps by defining the circuit voltage as an external variable and using that in the MF, I would have to check (I'm stuck in some structural just now, I'll try some time later) But careful when mixing MF and MEF in the same model (normally you need only one of them) to avoid getting confused with the similar variables and names My way around when I get errors, is really to start fundamentally simple and build again up step by step I believe you have missed something in the CIR model, your node 1 is "hanging" -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 06.04.2011, 11:26 GMT-4
Hello...

I built a simple model of resistance of MEF module and I tried to connect the AC power source to resistor.
Please, could you tell me what is wrong with it ?
I just can not be connected to the scheme of model. All time same error comes. How difficult can it be ? :)

If you look at this model and correct the error could you please send it back to me. Then I can perhaps understand how to combine the model and scheme.

Thank you !

Heigo.
Hello... I built a simple model of resistance of MEF module and I tried to connect the AC power source to resistor. Please, could you tell me what is wrong with it ? I just can not be connected to the scheme of model. All time same error comes. How difficult can it be ? :) If you look at this model and correct the error could you please send it back to me. Then I can perhaps understand how to combine the model and scheme. Thank you ! Heigo.


Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 07.04.2011, 02:42 GMT-4
Hi

first of all, if you have "just" a large Cu cylinder, why use MEF?, EC should do, the magnetic field will anyhow be expelled from the Cu, and if you do not put some air around to allow the field lines to loop it will not make much sens

Then with a large block the resistance is small 1/(sigma[S/m]*pi*(0.5[m])^2/1[m]), and with 100V applied you will get a large current about 5[GA]

if you reduce the wire diameter to let's say 1 [mm], then remove the "Preserve aspect ratio" in the "Definitions - View 1" node, this will compress your model axis to make it easier to select the items, even if a circle becomes a square in he geom view, it will be better once solved.

Then try rather EC and CIR to start with, later replae EC by MEF and add some air around

--
Good luck
Ivar
Hi first of all, if you have "just" a large Cu cylinder, why use MEF?, EC should do, the magnetic field will anyhow be expelled from the Cu, and if you do not put some air around to allow the field lines to loop it will not make much sens Then with a large block the resistance is small 1/(sigma[S/m]*pi*(0.5[m])^2/1[m]), and with 100V applied you will get a large current about 5[GA] if you reduce the wire diameter to let's say 1 [mm], then remove the "Preserve aspect ratio" in the "Definitions - View 1" node, this will compress your model axis to make it easier to select the items, even if a circle becomes a square in he geom view, it will be better once solved. Then try rather EC and CIR to start with, later replae EC by MEF and add some air around -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 07.04.2011, 05:54 GMT-4
I know these things what you mention, but.....For me it is not important right now what kind of module i used (is it MEF or something else)....or what kind on dimension for resistor I used (is it "m" or "mm")....or how big current I get (some "A" on some "GA")....or what kind on material I used.

For me is important how to connect ELECTRICAL CIRCUIT (where I made power source) to the model !.
If i know this, then I can build up my Electric arc Furnace in right module (MF or MEF), but right dimensions and connect electrodes two deferent power source what i have or what I build before in ELECTRICAL CIRCUIT.
My question is….”how to connect ELECTRICAL CIRCUIT to physical model ?”….I tried to used Terminals but something is wrong and I don´t know what ? The resistant is just test model.

My model (Electric arc Furnace) works quite well Comsol 3.5, but I want to build up this now in Comsol 4.1, because I can see that there are many more options, like MF module I can calculate the Liquid metal pressure.

Heigo.
I know these things what you mention, but.....For me it is not important right now what kind of module i used (is it MEF or something else)....or what kind on dimension for resistor I used (is it "m" or "mm")....or how big current I get (some "A" on some "GA")....or what kind on material I used. For me is important how to connect ELECTRICAL CIRCUIT (where I made power source) to the model !. If i know this, then I can build up my Electric arc Furnace in right module (MF or MEF), but right dimensions and connect electrodes two deferent power source what i have or what I build before in ELECTRICAL CIRCUIT. My question is….”how to connect ELECTRICAL CIRCUIT to physical model ?”….I tried to used Terminals but something is wrong and I don´t know what ? The resistant is just test model. My model (Electric arc Furnace) works quite well Comsol 3.5, but I want to build up this now in Comsol 4.1, because I can see that there are many more options, like MF module I can calculate the Liquid metal pressure. Heigo.

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 07.04.2011, 11:17 GMT-4
Hi

I understand your concern, but for me it's also important that your model is coherent such that you do not get errors due to poorly defined physics BC that you mix up with errors due to the physics coupling.

there is also this one www.comsol.eu/community/exchange/146/

--
Good luck
Ivar
Hi I understand your concern, but for me it's also important that your model is coherent such that you do not get errors due to poorly defined physics BC that you mix up with errors due to the physics coupling. there is also this one http://www.comsol.eu/community/exchange/146/ -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 08.04.2011, 08:36 GMT-4
Thank you for your response, but there is not even any geometry defined in this model so it wouldn't have been able to teach me how to place a 3-D model into the electric circuit anyway.
My question is same: how to connect electric circuit to geometrical model ? I have to use terminals but it not works.

I'm sorry that I bother you so much.

Heigo.
Thank you for your response, but there is not even any geometry defined in this model so it wouldn't have been able to teach me how to place a 3-D model into the electric circuit anyway. My question is same: how to connect electric circuit to geometrical model ? I have to use terminals but it not works. I'm sorry that I bother you so much. Heigo.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 16.12.2011, 12:04 GMT-5
Hello.

Did you eventually find a solution to this problem ? There is still nothing in models.

Comsol 4 is really bad compared to 3.5 for these problems and for coupling with matlab.... What a pity!!!

Hello. Did you eventually find a solution to this problem ? There is still nothing in models. Comsol 4 is really bad compared to 3.5 for these problems and for coupling with matlab.... What a pity!!!

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 16.12.2011, 15:58 GMT-5
Hey

have you done the exercices/models in the model library ? there are a few examples on coupling circuits to physics

I find v4 far better than the v3, but there are a few new things to learn, and we must soemhow better respect physics, hence one get less non-physical solutions back

--
Good luck
Ivar
Hey have you done the exercices/models in the model library ? there are a few examples on coupling circuits to physics I find v4 far better than the v3, but there are a few new things to learn, and we must soemhow better respect physics, hence one get less non-physical solutions back -- Good luck Ivar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.