A sheet resistance simulator could use squares, lots of them, but the solution wouldn't be to count them. Counting squares is useless if the current density is not uniform, or if the squares aren't aligned in the directions of current flow.
This one is not bad, and free:
The user interface is a bit clunky, input is via a .BMP of your model. It does give a numerical output as well as graphical, the numbers match my hand calcs if you get the BMP scaling right. It doesn't cope well with large or highly detailed models, it runs out of RAM and crashes.
I had a play with Sonnet, much more powerful but also a much steeper learning curve. I may give it another go if I have a problem I can't solve with idc2d.
I tried Sonnet once and learning it exceeded my attention span. I think I'll get one of my young things to learn it and show me how, or better yet run my geometries for me. The Youtube tutorial doesn't make it look that bad.
We sometimes use ATLC, also a BMP i/o thing, and that is a tad klunky.
Looks like you can download Sonnett Lite for free, that may be enough so that you don't have to spend real $$. Of course for all of these choices the real cost is going to be the time needed to learn the software. Femm is free and handles planar 2D (and axisymmetric 3D) cases so would handle one layer but not vias to another layer. It's pretty easy to learn, you just have to forget all you already know about keyboard shortcuts and what a mouse click will do :-). I've used it for a fair bit of permanent magnet simulation and each time I go back to it I have to relearn the interface, sigh. It's a very general heat and electromagnetic package but does specifically include current distribution calculation as one function. OpenFoam is another free package, fully 3D, originally written for CFD work but people have added solvers for all kinds of work including electrostatics (but I don't know if that includes current distribution). I haven't tried using it yet, but it seems the most powerful/general, there is commercial training support available, and lots of people are currently using and developing it. The advantage of it would be that once learned, you could use it for heatsink/chassis airflow and heat transfer work, as well. Oh, it does require Linux, the other open source packages are available for Windows and Linux. Someone else mentioned idc2d which appears to be 2d only, and there is Elmer, a freeware 3D FEA package that could probably be used for this, but both of those seem to be pretty old and no longer being developed or actively supported (my take, could be wrong). Anyway, those are the choices I know of.
----- Regards, Carl Ijames
"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...
On Mon, 19 Oct 2015 09:34:02 +0100, RBlack wrote:
I tried Sonnet once and learning it exceeded my attention span. I think I'll get one of my young things to learn it and show me how, or better yet run my geometries for me. The Youtube tutorial doesn't make it look that bad.
We sometimes use ATLC, also a BMP i/o thing, and that is a tad klunky.
Thanks, Carl! Nice list of solvers! Googling idc2d, it's unclear if it runs under Windows? Could be quite useful for coping with current densities on-chip... some of my recent LED billboard driver chips end up in the AMPS range :-) ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
it is if you make them small enough. The bigger, the less precise
I'd morph the shapes into something easier to work with. Sketch approx isopotential lines across the route of current flow. Redraw these shapes as rectangles in a line - each the same width & length as the curent flow sees it. Dunno if that's clear enough.
as above, but if it's not a large %age of total R I'd be far more tempted to just eyeball guessimate the width & length of current flow and run with that.
NT
t the squares. I expect you could easily write a little routine to do it.
opotential lines across the route of current flow. Redraw these shapes as r ectangles in a line - each the same width & length as the curent flow sees it. Dunno if that's clear enough.
NT
to just eyeball guessimate the width & length of current flow and run with that.
Is there a users manual for idc2d? I haven't been able to find it by surfing. ...Jim Thompson
-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at
I think that would be me.
There is no PDF manual or help file AFAICS, all the documentation that exists is linked from the project's Sourceforge page:
HTH
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.