Quantcast
Channel: Solving, solving\ - Latest Discussions - COMSOL Forums
Viewing all 3988 articles
Browse latest View live

'ec'-'ht' Multiphysics error and impractical outcome of Temperature distribution for a Power Cable

$
0
0
Hello all members,

There are 2 physical problems that I'm required to solve simultaneously;
1] the distribution of electrical current (due to an application of high-voltage)
2] the distribution of temperature

My model under analysis has a copper conductor (simulating a 500kV Cable) which is enclosed inside a cylinder having a top and bottom Torus. This complete arrangement is then enclosed inside SF6 gas chamber.
This complete equipment is a high-voltage component operating at 500kV and therefore I'm required to apply a voltage of 500kV to the top Torus and Ground to the bottom Torus of the Cylinder (and also high-voltage to the top boundary of copper conductor and Ground to the Cable bottom) for computing the Electric fields for the optimisation of this equipment.

However, when I apply this high-voltage, 500 kV to the copper conductor (of the cable, which is also a high voltage cable), COMSOL is looking at a very high current through the conductor resulting in a very high temperature.
As per my calculations, I have a rated current of 2621.4 A through the copper conductor (considering the cable thermal effects from the data sheet). The resistance (from the data sheet) is 1.848e-5 ohm. This gives a very small voltage of 48.44 mV (Ohm's law) which is correct because it is simply the voltage drop over the cable.
But how do I make COMSOL understand this effect and at the same time apply 500kV for simulating the electric fields to my equipment and hence get a realistic temperature distribution with the temperature difference being 30[degC] between the conductor and the ambience

Can anyone please suggest something on simulating this issue??
Many Thanks (in advance)

Kind regards
Tanumay Karmokar

Microwave heating of a Split Ring Resonator.

$
0
0
Hello good people!

I have modeled microstrip line coupled with a split ring resonator(SRR) and have included in the attachment.
Resonance of the SRR is achieved at 2.06GHz.

Now I want to simulate one more thing:
Microwave heating of the SRR.
I.e I want to see the heat distribution of the SRR while passing microwaves with power through the ports.

Just like this poster:
www.comsol.com/paper/download/183071/nkale_poster.pdf
In the computational methods section he says:
Then using Microwave Heating Module, we subjected the SRR to different power power levels.

How do I do this?

GaAs Solar Cell; Semiconductor Module + RF Module

$
0
0
Hi there ! So this is my problem ... I get this Error Message (Error message is "Maximum number of Newton iterations reached."); "Doesn't Find Solution... I'm new with this software and i just throw myself in this new version, it's seems very friendly... but i can't figure out how to get the solution...

The file of my work is attached, PLEASE HELP...
this results are for credits in a summer research program ...

Re-using physics interface as a study step

$
0
0
Hi,
In my study I use EMW interface to calculate a set of parameters (let's say, polarization-'p')
and then implement these results in the next step, which uses a different EMW interface.
Schematically:
Step 1(EMW1)--> Step 2 (EMW2(EMW1.p))

Now I'm trying something a bit more complicated, but can't find an elegant way to do so-
I want to calculate the first step twice, each time with one parameter changed (let's call it 'd'), and then implement both results in the same advanced step.
again, schematically:
Step 1(EMW1(d=1)), Step 1(again) (EMW1(d=2)) -->Step 2 (EMW2(EMW1(d=1).p,EMW1(d=2).p))

How can I do that without duplicating the first EMW inteface? (which is done manually, can't find a way to duplicate it automatically)
Can I use Parametric sweep just for one step, and then implement it in the next one? If so, how can I refer to the results in the previous step, and choose which of the results sets to use?

Thanks,
Shay

How to couple physics

$
0
0
Hi,

I've seen several tutorials in which the demonstrator is able to take the solution from one solver and use it as a constraint for the next solver.

In every case, the demonstrator already knows the expression for the output that he wants to use. For example, if I want to couple a pressure simulation to something else and need the pressure, I'd use the expression "p" because that is the one for pressure.

My problem is that I do not know the expression for all of Comsol's outputs. Is there a list that Comsol stores for such expressions?

Thanks.

transport of diluted species and laminar flow

$
0
0
Hi all!
I want to simulate a cylinder moving in the water, and it can generate some species on the bottom of the cylinder. So the move of the cylinder will influence the transportation of the diluted species. I set the moving wall boundary condition to the outer surface of the cylinder, but when i set the magnitude of the velocity too big, there is an error says"failed to find a solution. Maximum number of newton iterations reached, returned solution is not conberged"
can anybody help me, thanks a lot!

Recoverable error in residual, but solver could not recover.

$
0
0
Recoverable error in residual, but solver could not recover.

What is the meaning of this error ?

Fluid flow in variable zeta potential

$
0
0
Hello,
I'm trying to simulate the velocity profile of fluid in a long and narrow column. For wall boundary condition I'm using "Electroosmotic velocity". I want to have variable zeta potential on the wall, what is the best way to achieve that?

Eigenvalues from different study steps

$
0
0
Hi everyone,

I am trying to calculate an overlap integral between two modes of different origin -- a common problem of electrooptics or optomechanics. So I have two physics interfaces (say electromagnetic and structural mechanics) and I need to solve them and compute overlap integrals between all modes.

If I use two distinct studies (one for electric eigenvalues and other for mechanic eigenvalues) I get two different solutions (data sets) that cannot be brought together.
If I use two study steps in one study I again get two data sets. Though the first one has one of the modes of first problem, it looks like there is only one of them. I tried to use the with() operator but I still can't address to other modes.

Is there any way to do the thing?
Thanks.

Using lower limit node

$
0
0
Hi,

I'm trying to use the lower limit node, but I can't select any variables because the variables window is not there. I can select variables in the segregated step nodes. See attached file for screenshot.

Stiffness error

$
0
0
Hi Ivar,

I am having a problem (written below) to solve my model. could you please suggest me what I should do? So far, I have defined all parameters but still having the same prob. My model is attached herewith..

Undefined value found.
- Detail: Undefined value found in the stiffness matrix.
There are 1813 equations giving NaN/Inf in the matrix rows for the variable comp1.C_c.
at coordinates: (-1), (-0.9975), (-0.995), (-0.9925), (-0.99), ...
and similarly for the degrees of freedom, NaN/Inf in the matrix columns.

Thanks in advance..

Noor

coupling of two physics

$
0
0
Hi guys

My simulation has two physics in it. The second physics is piezoelectric and I am trying to apply the line integral of the dependent variable from the first physics as the input to the Voltage of the piezoelectric physics.

I have gone through the forum comments on applying the solution of one physics to the dependent varibales of the second physics but here i need to perform a line integral operation on it first before applying it to the dependent variable of the second physics. So i am a bit confused.

Also what is the correct syntax of the line integral in COMSOL.

thanks

Cannot model losses in RF module

$
0
0
Hi

I've been trying to model a lossy dielectric in the RF module, using the transient electromagnetic waves interface. However, if I set the model to be based on the refractive index, COMSOL appears to ignore the imaginary part in my material definitions. If I then try and model the losses by introducing an imaginary component to the relative permittivity, COMSOL always treats it like a gain medium regardless of the sign of the imaginary component.

Is there anything I can do to fix this?

Thanks.

Parameteric vs. Batch vs. Cluster sweeps

$
0
0
Hi,

I have a time-dependent model and I would like to sweep through a range of values of one of the model parameters. The problem is an 'embarrisingly parrallel problem' and very much similar to the example given in this COMSOL blog post:

www.comsol.com/blogs/added-val...task-parallelism-batch-sweeps/

However after much reading on the support pages and in the reference manual, I am still not quite sure how COMSOL handles the various sweep study options:
1. Parametric Sweep
2. Batch Sweep
3. Cluster Sweep

I'm a bit confused because I have managed to use the Batch Sweep to solve the model on our computational cluster. So then how is a Cluster sweep different to a Batch sweep? When is it advantageous to use over the other?

Regards
Amr

Permanent magnet model produces no field + relative error to high

$
0
0
Hello all,

I am a relatively new COMSOL user attempting to model an electric motor design (of the out runner type) with the Rotating Magnetic Machinery physics. Currently I am working on the permanent magentas in the rotor, and have been unable to get them to work properly. I have modeled the permanent magnet using the Magnetic Conservation node with the constitutive relation . The magnet is attached to a steel ring which is the body of the rotor. Currently I am only modeling one magnet but will eventually have 14 magnets arrayed periodically within the ring. I have two problems:

1) When I perform the study I get the message that the relative error is greater than the tolerance. I have been able to play with things so the relative error is below ~50 each time I run the study, however this is still much greater than the tolerance (.001).
2) Despite the previous issue, the study does go to "completion," and I get some results that I can look at. However the results show a minuscule field within my magnet, much lower than the 1.3 T I have assigned to the magnet. Essentially the magnet does not display any field.

I am assuming the two issues are related, though I am not positive. Does this problem have to do with my meshing? Or am I implementing the Magnetic Conservation node improperly?

Any assistance will be greatly appreciated!

-Eli Zoghlin

Duplicating variables with component couplings

$
0
0
Hello All,

I am using a "control variable field" in the optimization module to define some variable 'p' on a boundary. I have an identical boundary offset by a few cm, but I would like that variable 'p' defined on it as well. I know I have to use a model coupling but I'm not exactly sure which one to use, as I found the documentation on it slightly confusing.

Thanks,
Aaron R. Shifman

Two simultaneous COMSOL instances on one machine can save time

$
0
0
Dear community,

today I tested something I had in mind for quite some time. On my workhorse computer, a 3.6 GHz Core i7 with 4 cores I always observe that a COMSOL calculation brings up CPU use to a maximum of 50%. Only starting a second COMSOL instance brings up CPU usage to 100%.
I was wondering if dividing extensive parameter studies into two instances would yield a benefit or the increased CPU usage would suffer from increased overhead.
Here is the result for a small AC/DC parameter study:
1 instance: 2min 58 s
2 parallel identical instances: 3min 48 s
The direct MUMPS solver was used.
So this is a significant benefit. I still have to check if it is the same for large models that take much more time for a single parameter step behave the same way. Of course the machine has to have enough memory for the two parallel instances.

Cheers
Edgar

--
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com

constraining dependent variable

$
0
0
Good day:

I am trying to find the most effective method of implementing a constraint on a dependent variable.

I am solving a classical diffusion equation (using Equation based modeling)of concentration 'c' and I do not want my variable 'c' to exceed some value 'cmax' on the entire domain I am modeling on.

What is the most effective way that COMSOL allows this to be done?

Thank you for any help given in advance.

Initial conditions (steady state) to time dependent model

$
0
0
Hello All,

I have a model where I want to find the steady-state solution to a time dependent model, I let it run with a large time block. Now I want to use this as the initial condition to the next study step. The issue I have is that there is a massive solution stored in memory. Is there a way of taking the last time step of a study, saving it as a steady state solution and forgetting the rest (or better yet, as the time dependent solver progresses, only keep the previous solution in memory while solving)?

Thanks,
Aaron R. Shifman

Java Runtime Environment error

$
0
0
Getting following error,

A fatal error has been detected by the Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x0000003da3efbc6c, pid=6909, tid=46912497239488
#
# JRE version: 6.0_20-b02
# Java VM: Java HotSpot(TM) 64-Bit Server VM (16.3-b01 mixed mode linux-amd64 )
# Problematic frame:
# C [libgtk-x11-2.0.so.0+0xfbc6c] gtk_icon_set_render_icon+0x58c

Does anyone know how to resolve it...

Regards
Viewing all 3988 articles
Browse latest View live


<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>