Re: Modeling and simulation

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: Terry Fritz <teslalist-at-twfpowerelectronics-dot-com>

Hi,

I guess I gotta say something ;-))  Having lived in all these worlds...

At 08:15 PM 9/18/2003 -0600, you wrote:
>To all those who trust in Spice and the like,
>
>The URL below is an interesting piece by Bob Pease (of National
>Semiconductor fame) regarding how "helpful" modeling is ... or isn't.  I
>would think his wisdom certainly applies to modeling of Tesla coils.
>
>--Steve Y.
>
>http://www.elecdesign-dot-com/Articles/Index.cfm?ArticleID=5632


99.9% of us are not blessed with "Bob's perfect God's eye view of the 
electronic world"(tm), "we" need all the help we can get ;-)))  Unlike Bob, 
"we" can't design highly complex circuits in a knowledge vacuum...

It is unfortunate that the situation Bob perceives is such an adversarial 
one.  Computer modeling is a helpful tool just like a soldering iron, 
pliers, and hammer...  I am surrounded by computer modeling gurus.  They 
hint at problems and disasters all the times in circuit designs.  However, 
"we" take their advice!!!  If they say a 50-J132 load is going to blow 
FETs, we don't argue.  We rush to the lab and test it out.  They seem to 
have an obnoxious habit of being right!!  And, if they are wrong, we help 
them all we can to find out why the model predicted the wrong thing!  Once 
their models are all optimized, they can test millions of situations in a 
day were we hardware guys can test three...  We trust them to do the work 
we can't...  A good find on their part can save tens of millions of 
dollars...  "We" love them ;-))    In just a few more years, I will be able 
to say that I was doing their job before they were born :o)))

Please don't think of computer modeling as some sort of evil "other side" 
to design.  It is just a very useful tool.  When hardware guys and the 
computer guys work happily together, helping each other find errors and 
reaching towards a goal of a good design, great things can be achieved!!

Today Bart introduced his JAVATC 10:

http://www.classictesla-dot-com/java/javatc.html

It is "just a computer model too"...  However, it can do in a day what a 
coiler would have to test in a few lifetimes...  It has technology behind 
it's results that are more accurate than any of us can measure given modern 
equipment...  It is not meant to "obsolete" building real coils :o)))  But 
only to "serve" the real coil builder...  LTR coils were based, optimized, 
and practically discovered by computer modeling...  The OLTC was "all" 
worked out on the computer before the first part was ordered...  Computer 
models can kick some serious power!!!  Now days, mistakes in building a 
Tesla coil can cost serious money!  You just have to listen to "all"...

In the past we have discussed those "seat of the pants coilers" and those 
"arm chair coilers".  I guess now we have coilers whose coils reside on 
their hard drives ;-))  Is one or any wrong??  Naw, they are all "right"!!

Some may look at only the sparks, some may look at only the scope captures, 
some may look at only the computer printouts...  But we all 
coiler's!!!    And we all learn from the others to help our particular 
effort.  In the end, we all win ;-)))

Cheers,

         Terry