Menu

Discussion Forum

New Discussion
Filtrer

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 The information provided may be out of date.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.

Does anyone work with high-voltage electrostatic fields?

Posted 23 janv. 2011, 23:28 UTC−5 Low-Frequency Electromagnetics Version 4.0 11 Replies

Roman Tukanov
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi

I use COMSOL 4.0, Electrostatics (es) for simulating electric fields in high-voltage switchgear.
I faced with the following problem.

I have two porcelain isolators. Each of them has metal flange on the top and on the bottom.
First isolator stands on the ground, the second stands on the top of the first isolator.
So the top flange of the first isolator and the bottom flange of the second are connected together. These two flanges connected together create middle flange.
The bottom flange of the first isolator is grounded (potecial is 0). Boundary condition: Ground.
The top flange of the second isolator is under high potencial (about 1.55 MegaVolts). Boundary condition: Electric potencial.
So, I have the middle flange. What boudary condition should I define here to get a correct field?

As I think, the potencial from upper flange will "bring" some potencial to the middle flange, but giving the middle flange BC as Electric Potencial is not correct, because the potencial on the middle flange has another nature than on the upper flange. Grounding of the middle flange is not correct too, of course.

Will be very glad to get your replies!

Thanks in advance.
11 Replies Last Post 27 janv. 2011, 01:10 UTC−5
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24 janv. 2011, 01:17 UTC−5
Hi

I do not believe the "high" voltage is an affair, it's rather the ratio max to min voltage that is limited by the numerical representation of binary number (hence numerical precision). In any case, you talk about isolators (hence probably conductivities = 0), but do you have some "air" around, for me air does have a small conductivity, in the 1-10 [S/m] region, and that would help distribute more evenly your potential along the way from the HV to the ground

Another thing, find out with your sys-admin or your local COMSOL rep how to get updated to 4.1, its much better to use than the "early" 4.0 version, you will get less frustrated ;)

--
Good luck
Ivar
Hi I do not believe the "high" voltage is an affair, it's rather the ratio max to min voltage that is limited by the numerical representation of binary number (hence numerical precision). In any case, you talk about isolators (hence probably conductivities = 0), but do you have some "air" around, for me air does have a small conductivity, in the 1-10 [S/m] region, and that would help distribute more evenly your potential along the way from the HV to the ground Another thing, find out with your sys-admin or your local COMSOL rep how to get updated to 4.1, its much better to use than the "early" 4.0 version, you will get less frustrated ;) -- Good luck Ivar
Roman Tukanov
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24 janv. 2011, 01:43 UTC−5
Thanks for your reply, Ivar!

Of course, i have took into consideration the air around my construction.
The distribution of electric field is quite nice looking :) but it's incorrect as i think.

I have no problems with distribution, my problem is the correctness of distribution.
The potential, that appears on the middle flange is of capacitive nature, and it is incorrect to use Electric Potencial BC.
How can I take into consideration this capacity? Using another BC's? Or trying PDE mode?
Thanks for your reply, Ivar! Of course, i have took into consideration the air around my construction. The distribution of electric field is quite nice looking :) but it's incorrect as i think. I have no problems with distribution, my problem is the correctness of distribution. The potential, that appears on the middle flange is of capacitive nature, and it is incorrect to use Electric Potencial BC. How can I take into consideration this capacity? Using another BC's? Or trying PDE mode?
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24 janv. 2011, 02:07 UTC−5
Hi

but in static you cannot charge a capacitive system, or I'm I wrong ?

You need to have a transient or frequency drive solution then.

And with all my HV practical work, we never left a floating potential electrode in between a HV gap, without some resistors to balance the DC surface voltage you inevitably have on any "isolator"

--
Good luck
Ivar
Hi but in static you cannot charge a capacitive system, or I'm I wrong ? You need to have a transient or frequency drive solution then. And with all my HV practical work, we never left a floating potential electrode in between a HV gap, without some resistors to balance the DC surface voltage you inevitably have on any "isolator" -- Good luck Ivar
Edgar Kaiser
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 24 janv. 2011, 06:23 UTC−5
As soon as you assign a small conductivity value to your air domain nothing is floating any longer. I think continuity should be the boundary condition for the middle electrode. You may also need to assign a very small but nonzero conductivity to the isolators in order to avoid numerical instability.

Best regards
Edgar
As soon as you assign a small conductivity value to your air domain nothing is floating any longer. I think continuity should be the boundary condition for the middle electrode. You may also need to assign a very small but nonzero conductivity to the isolators in order to avoid numerical instability. Best regards Edgar
Roman Tukanov
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 26 janv. 2011, 01:17 UTC−5
Hi!

There is no resistors. My model is the isolation column, that isolates the arching chamber of high-voltage circuit-breaker from the ground. Inside this column isolating rod is situatad, which moves the contacts of arching chamber. The coloumn is separated on two smaller coloumns because of high wind pressure. One big coloumn can easily break. That's why the middle flange appears.

Why I ask about capacitance?
Solving my problem Comsol and Ansys gave not correct sollutions.
Results outgo from experimental data.
Comsol and Ansys gave uniform distribution of electrical potential between Electric Potential and Ground. It mens that if EP is 100kV, than middle flange has 50kV, Ground is 0kV.
In fact the middle flange will have about 33kV. It means that the upper isolator will take on him 2/3 of applied voltage, not 1/2 as solved by Comsol and Ansys.

How can I take into consideration this unevenness using Comsol?
Hi! There is no resistors. My model is the isolation column, that isolates the arching chamber of high-voltage circuit-breaker from the ground. Inside this column isolating rod is situatad, which moves the contacts of arching chamber. The coloumn is separated on two smaller coloumns because of high wind pressure. One big coloumn can easily break. That's why the middle flange appears. Why I ask about capacitance? Solving my problem Comsol and Ansys gave not correct sollutions. Results outgo from experimental data. Comsol and Ansys gave uniform distribution of electrical potential between Electric Potential and Ground. It mens that if EP is 100kV, than middle flange has 50kV, Ground is 0kV. In fact the middle flange will have about 33kV. It means that the upper isolator will take on him 2/3 of applied voltage, not 1/2 as solved by Comsol and Ansys. How can I take into consideration this unevenness using Comsol?
Edgar Kaiser
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 26 janv. 2011, 04:40 UTC−5
Is your physical setup using DC or AC voltage?
Is your physical setup using DC or AC voltage?
Roman Tukanov
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 26 janv. 2011, 05:37 UTC−5
DC.
I use Electrostatics (es), Stationary analysis.
I suppose this study type suggests using only DC voltages.
DC. I use Electrostatics (es), Stationary analysis. I suppose this study type suggests using only DC voltages.
Edgar Kaiser
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 26 janv. 2011, 06:44 UTC−5

The question was whether your physical setup (the real thing) is using DC or AC.
The question was whether your physical setup (the real thing) is using DC or AC.
John Peterson
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 26 janv. 2011, 14:47 UTC−5
hi,
if the two columns have exactly the same dimensions (radius, thickness,hight), and are made of exactly the same material, I don't see how the middle flange will give measurements different than half the voltage difference between the top and the bottom. (unless your bottom column is really "on the ground", and the high voltage is "charging" the soil around... then you have to model both columns and a piece of the soil around)

V_top
[ ]
[ ]
[ ]
flange <------(voltmeter)-----| V_bottom
[ ]
[ ]
[ ]
V_bottom
----------------------- Soil -------------------------------
hi, if the two columns have exactly the same dimensions (radius, thickness,hight), and are made of exactly the same material, I don't see how the middle flange will give measurements different than half the voltage difference between the top and the bottom. (unless your bottom column is really "on the ground", and the high voltage is "charging" the soil around... then you have to model both columns and a piece of the soil around) V_top [ ] [ ] [ ] flange
Roman Tukanov
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 janv. 2011, 01:07 UTC−5
Hi!

THANKS A LOT, JOHN!
Solving this problem, I totally forgot that my isolators were standing on the ground =)
Now I have a correct electric potential distribution.
Hi! THANKS A LOT, JOHN! Solving this problem, I totally forgot that my isolators were standing on the ground =) Now I have a correct electric potential distribution.
Roman Tukanov
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27 janv. 2011, 01:10 UTC−5
The voltge applied is Full Lightening Impulse.
I simplified it to DC.
The voltge applied is Full Lightening Impulse. I simplified it to DC.

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.

Suggested Content