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 The information provided may be out of date.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.

Current not being conserved in AC/DC.

Please login with a confirmed email address before reporting spam

Hello all, I am having an issue with current conservation using the AC/DC module and a stationary 3D model.

The model is two parallel pieces of copper and a small piece of aluminum between them. I add either a voltage or current source at the end of one piece of copper and a ground or current sink at the end of the other. Basically I am making current from through one piece of copper, through the aluminum block and out the other copper piece.

I have been setting the normal current density source at the ends such that exactly 100000 amps (arbitrary) runs into and out the ends of the copper. I then solve the system and do a surface integral of the normalized current density at the interface between the copper and aluminum pieces. I get a total current of over 100000 amps, Something 1.5 times as much. If I do a cut plane in the middle of the aluminum piece however and integrate over the current density there I get something like 99999 amps (good enough since its a numerical solver). I tried cranking up my mesh at the boundries of the aluminum and copper interface but it is still the same.

According to Kirchhoff I should be getting 100000 amps at any slice through my model. There is no way I can have 150000 amps running through anywhere in the system if all I input is 100000. Or if I do a voltage source at the end of the pieces I should get the same integrated current through any slice of the model.

Any ideas? Thanks

3 Replies Last Post 30.04.2013, 02:24 GMT-4
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 30.04.2013, 01:18 GMT-4
Hi

no particular idea as for my cases so far the current flow is mostly within 5-10% ,
so perhaps some singularity (sharpe edge that overestimates grossly the current calculation), or are you in 2D-axi and you have added manually and via the selection markl the 2*pi*r (but then the units should not match) ?

Without seeing the model it's difficult to say much more
--
Good luck
Ivar
Hi no particular idea as for my cases so far the current flow is mostly within 5-10% , so perhaps some singularity (sharpe edge that overestimates grossly the current calculation), or are you in 2D-axi and you have added manually and via the selection markl the 2*pi*r (but then the units should not match) ? Without seeing the model it's difficult to say much more -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 30.04.2013, 01:34 GMT-4
Hello Ivar, thanks for responding.

There is definitely current build up on the corners/edges at the interface between the parts. But it is something I want to see and model accurately. Is there a way to get comsol to make more accurate calculations there and not over calculate? I am away from my workstation but can post screenshots/the model tomorrow.
Hello Ivar, thanks for responding. There is definitely current build up on the corners/edges at the interface between the parts. But it is something I want to see and model accurately. Is there a way to get comsol to make more accurate calculations there and not over calculate? I am away from my workstation but can post screenshots/the model tomorrow.

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 30.04.2013, 02:24 GMT-4
Hi

try filleting off and rounding corners hen the derivatives along these round corners will be better defined and the integration should give less errors. But you cannot always easily "round-off" all corners, all depends on the model

--
Good luck
Ivar
Hi try filleting off and rounding corners hen the derivatives along these round corners will be better defined and the integration should give less errors. But you cannot always easily "round-off" all corners, all depends on the model -- 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.