CST EM studio and other simulators / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

CST EM studio and other simulators

1 2 next

Move Thread

LAN_403

Patrick

Thu Jun 10 2010, 06:27AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

does any one know of these types of software packages? (CST EM et. al)it also needs to be 3-D.
field intensity is what im reall after.

i need the ability to simulate carona/voltage gradients, on physical plates and spheres, coils and such.

the CST EM one is like 4.3 GB on download. are there smaller, cheaper, easy to use ones?

-Patrick

Nicko

Thu Jun 10 2010, 06:32AM

Registered Member #1334 Joined: Tue Feb 19 2008, 04:37PM
Location: Nr. London, UK
Posts: 615

4.3Gb? Sounds like a DVD ISO image rather than just the kit. Where's the download from, or shouldn't I ask?

Cheers

Patrick

Thu Jun 10 2010, 06:36AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

no comment on the source, as the fuzz maybe listening. but i think youire right about the iso image.

wylie

Thu Jun 10 2010, 07:14AM

Registered Member #882 Joined: Sat Jul 07 2007, 04:32AM
Location:
Posts: 103

Can ANY simulator really provide better data then corona-fying structures yourself and measuring the results? Hell, we still can't predict the weather more than a few days in advance, with supercomputers.

Paraphrasing SteveM: Sims are no replacement for actually slapping the circuit together and seeing what it really does.

(for example, LTspice told me that using 2n3904s and 2n3906s to source and sink (or vice versa) just wouldn't work unless my supply was 12v. Reality disagreed.)

EDIT: apparently Simulation is great, except when it isn't. Is is that much easier to model electrostatics than umm, i guess, electrodynamics?

Patrick

Thu Jun 10 2010, 07:23AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

its ball-bearings under oil, i need to reduce my proyotypes from 431 device trials to like 9 trials, numeric simulation is good for reducing random crap to a few well designed attempts, which are then refined in the real world with less time, fewer attempts, and reduced cost, while yielding better results then experiments alone. this is how the wise use technology, while pansies and cowards merely screw around with flashy new stuff.

-Patrick

Mattski

Thu Jun 10 2010, 09:57AM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

Quite right Patrick, simulation will get you to a good design much faster. And electromagnetism simulators generally provide very good agreement between simulation and practice, it's mostly a matter of how detailed your model is and whether the materials involved are well characterized.

This sounds like DC/electrostatic, correct? I suspect that CST is way overkill for something like this. FlexPDE is something that I have used for simple electrostatic simulations in a class once. The completely legitimate student copy has a limit on the number of nodes, but may work for your application. It's a generic PDE solver, so it works for E&M, heat, fluids, and more. You just need to set it up with the right equations. As I recall they have help files and examples that should get you on your way quickly, if you have trouble I could look up one of my past assignments that I used it for.

jpsmith123

Thu Jun 10 2010, 04:02PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

IMO, simulators are tools, just like oscilloscopes, multimeters or anything else.

Used properly, they can help give you some insight into what's going on in a particular situation and in some cases save you time and money...but you have to have an idea of what you're doing...which I admittedly don't always have. Sometimes I seem to get an unphysical result and don't know what I did wrong.

Anyway, I've experimented with both CST and Ansoft Maxwell, and they both have their flaws and downsides.

What's nice about CST is that the geometry editor is easy to use.

The automatic meshing generally works fairly well for simple problem geometries, but for some things, e.g., where there are some fine features, you may need to manually adjust the mesh.

One thing that irritates me is that the field plots sometimes look crappy, especially when using the hexahedral mesher and default plot settings.

Then there's the issue that the various optional solvers don't always agree with one another. Although the differences are generally small, sometimes they're different enough that you want to know which result is the more accurate one.

Another thing I find fault with is their documentation (or lack of it, I should say); some features just seem poorly documented, and it would've been helpful if they'd given an example problem or two and/or provided a more comprehensive explanation.

Then there's their lousy database of materials. You'd think that they'd include the commonly encountered plastics and ferrites, e.g., 3C90, N27, etc., but no, they don't. You can add these materials to your simulations if you know the relevant properties, but it's an unnecessary PITA.

Lastly, some types of problem geometry are inherently difficult if not impossible to model; e.g., I've tried several times to model flat spiral coils, to no avail.

BTW, Field precision has a freeware 2D field solver available here:

Patrick

Fri Jun 11 2010, 03:44AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

the sate one is hard to figure out. how do i draw 2 sphere?

jpsmith123

Fri Jun 11 2010, 09:09PM

Registered Member #1321 Joined: Sat Feb 16 2008, 03:22AM
Location:
Posts: 843

I downloaded it a while ago, but I've never done anything with it, so unfortunately I can't answer your questions about it.

Supposedly it can import DXF files, so maybe you could lay out the problem geometry in a CAD program you're familiar with and then import it?

Patrick

Sat Jun 12 2010, 12:11AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Ooooo, im good with autocad (LT, R12, R13 ,2000-2010) ill try that.
(I hope i havent dated myself)

EDIT: i dont see anything related to DXF files, and ill have to study this for a while, its complicated i cant even get the examples included to work yet.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob