[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

3D version of ETesla

Original poster: "Mark Broker" <mbroker@xxxxxxxxxxxxxxxx>

On Thu, 07 Jul 2005 00:38:34 -0500, Tesla list <tesla@xxxxxxxxxx> wrote:

Original poster: Terry Fritz <teslalist@xxxxxxxxxxxxxxxxxxxxxxx>

I'd love to be able to test this in Etesla, just for grins, but it uses
cylindrical symmetry and cannot simulate a conductive rod next to the coil
as in these experiments.  A 3D version of ETesla was on my TODO list about
4 years ago, but it quickly fell off the list because it required too much
effort and computer horsepower for too little gain.


I can run a 500 x 500 grid in about 45 minutes on a 3 GHz machine.  It is
hyperthreading so it still has about twice that available if it could be
done in two threads...  I was at the robot meeting tonight and they said
just to run two things after Main()....  I did not get it 0:-|...  But
"real" programmers would...  So that would be 45 minutes times say 250 = 8
days for  500 x 500 x 500 cube...  That is a "very" detailed array in a
realistic amount of time...  So maybe we have to start thinking about a
true "3D" version of E-Tesla...  But on the other hand...  Almost all coils
are served by the present version......  But a streamer off the toroid
would need 3-D.......  Would have to redo all the physical object
constraints to 3-D...


The constraints are trivial. Iterating through the grid will take "a while...." Fortunately there are some pretty easy things to do to speed up the math. For starters, dynamically-allocated arrays like ETesla uses can be "exploded" to run a lot faster. I believe Sue or some other female coiler had experimented with that when ET was converted to C, and the improvement was drastic. (IIR, it made the code harder to read, so speed was sacrificed in the name of readability.) It's also possible to use fixed point math using integers, which "may" crunch faster than floating point numbers.

Jim (?) mentioned a couple weeks ago that there are highly-optimized array-handling codes available, though I don't know if they are free.....

In my opinion, though, the most difficult aspect is displaying the results - "slices" would have to be created to create the nice 2D surface plots we're used to. It would almost certainly be better to have the application generate the plots than to load them into mathcad (Excel can't handle more than 255 columns). I was working on that particular issue a number of years ago, but "ran into trouble" and gave up. :o

To date I can only recall three instances where I could have used a 3D version of ETesla - I was wondering how the proximity of a grounded rod would affect things, I was wondering how a radial streamer/spark affected the tuning, and yesterday when I was wondering what the potential of a "floating" object would be.

Some of the "stuff" will be pretty easy, even to a fairly naive coder (me!), but there is some that would be pretty challenging. I guess I'm not sure if the result is worth the effort, but I could be talked into working on it. o:)

As an afterthought I had just as I was about to click the send button, this really is just a standard FEA problem. Although I don't know of any free FEA applications out there, there are discounted student copies (and illegal copies) of many of the commercial ones. Has anyone thought of and/or looked into this?

Cheers!

Mark Broker
The Geek Group