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

Re: ScanTesla program - basically working!!



Original poster: "Malcolm Watts" <m.j.watts@xxxxxxxxxxxx>

Hi Terry,

On 10 Jun 2005, at 19:13, Tesla list wrote:

> Original poster: Terry Fritz <teslalist@xxxxxxxxxxxxxxxxxxxxxxx>
>
> Hi All,
>
> I have the program basically working now ;-)))
>
> The executable and C program (just a text file that any text program
> can open) are here:
>
> http://drsstc.com/~terrell/modeling/DRSSTC-Combined-200.ZIP
>
> Right now, the program just scans about 7000 models over a coupling
> from 0.01 to 0.99 and the primary inductance 60 - 120uH.  It runs 130
> models / second on the 3GHz computer.  Anyone who has spent all day
> trying to do the some thing buy hand with MicroSim will just cry ;-))
> It has easy output files that can be fixed go to text or spread sheets
> with whatever data you are interested in.
>
> Right now, the output just lists the time (always 1mS), primary
> inductance, coupling, and resulting power / burst.  It updates
> whenever a more powerful burst is found.  You can add output stuff
> like in Antonio's original version to plot the output waveforms and
> all real easily too.  One could, of course, add fancy GUIs too if you
> know how ("I" don't)...
>
> There is no way to change the input yet unless you modify the scan
> loop ranges by hand in the program and recompile the program, but all
> the hooks are right there for playing.  The program is a little
> "hacked up" right now, but it will get cleaned right up once things
> are more figured out.  It should read the values from say a text file
> in the future.  I use the freebee C compiler LCC:
>
> http://www.cs.virginia.edu/~lcc-win32/
>
> ftp://ftp.cs.virginia.edu/pub/lcc-win32/lccwin32.exe
>
> It's only a 5Mbyte download.  E-Tesla is written with it too.  The
> program uses only real simple C stuff so just about any compiler
> should work too.  One should be able to figure it out pretty fast even
> in this form.
>
> This first version (one hour old ;-)) certainly is not ready for prime
> time or anything.  Only the C programmers can really "play" with it
> right now....  But it does show interesting things.
>
> At first it just lowered the tuning to draw vast primary currents to
> get high streamer power.  Steve Ward and the old models noted this
> effect.  If it can draw infinite current, it can supply infinite power
> ;-))  So I limited it to 500 primary amps.
>
> Then it shot the coupling up real high to just basically turn the coil
> into a high coupling transformer drawing 500 amps all the time for a
> nice burst power of 2300 watts PER BURST!!!  A 340V buss at 500 amps
> can theoretically supply 170,000 watts and the program just drove it
> right to that!!  So the next constraint is to input a BPS to give a
> maximum energy/burst.  The program needs constraints added to get it
> to real world situations where everything is not infinitely possible
> ;-))
>
> I was wondering if there would be an "ultimate answer".  Like a coil
> to produce 4 foot streamers would just have one set of perfect
> numbers...  but there are so many different coil constraints depending
> on what parts we have around that I think the program will be very
> useful and not just need to be run "once".  I was worried we would
> find that it would say use values like 34nF, 123uH, k=21.423, L2 =
> 120mH, CT = 23.485 pF and that is just the end of it....  Any other
> values would always be worse...  But there are enough odd things
> varying the coils in all of our situations that I don't think "one
> solution" will ever be the final answer and take all the fun out of it
> ;-))  All our coils have too many odd personal quirks for one solution
> for everyone to ever work.

I sincerely hope that no program proclamed to be useful spits out k =
21.423

;)
Malcolm

> The program is all real simple real time stuff so you can change any
> parameters by the nS if you want. Right now I am using a time step of
> 100nS. but that number could be anything you wish... the program has
> turned out to be very easy to modify to your whim...
>
> I think Antonio's variables for X0 are VC1, VC2, VC3, IL1, IL2... If
> not, the program is all screwed up, but trivial to fix ;-)) The
> "Antonio Engine" is wonderful!!! It would have taken me forever to
> figure that stuff out!!! It just happily takes the coil's parameters
> and generates the outputs just like MicroSim but now we can control
> everything and run it super fast!!
>
> So it is all very sub alpha version, but the concept is proven now
> ;-))
>
> Of course, vast thanks goes to Antonio!!!!!!!!!!! I changed the
> variables to match yours. I changed "i" and "ii" loops to real time.
> It slows things some, but makes it easier to fiddle with. I am
> assuming maximum power to the streamer load R3 is "good" and the
> program searches for that.
>
> So, It is looking super cool!!! ;-))
>
> Cheers,
>
> Terry
>
>
>