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

RE: Magnifier conversion



Original poster: "Terry Fritz" <teslalist-at-qwest-dot-net>

Hi All,

I was trying some side by side computer models of my coil and it's 
magnifier conversion:

http://hot-streamer-dot-com/temp/NewMag/NEWvsMAG00.gif

With no quenching, the top voltage and RMS voltage for the first 100uS 
looks like this:

http://hot-streamer-dot-com/temp/NewMag/NEWvsMAG01.gif

The magnifier has higher peak voltages and the initial RMS voltage is also 
far higher.  The much higher peak voltages continue for some time, but the 
RMS voltages tend to be similar over time.

When we add good quenching in...:

http://hot-streamer-dot-com/temp/NewMag/NEWvsMAG03.gif

The situation stays much the same but the magnifier gains some RMS voltage 
advantage.

http://hot-streamer-dot-com/temp/NewMag/NEWvsMAG04.gif

So it would appear that if everything else is the same, the magnifier 
should do better regardless if streamer length is a peak voltage or RMS 
voltage function (it would/will be VERY interesting to know which!))  So, 
the computer says my observation Sunday night must have been incorrect 
;-))  The computer tends to be right ;-))  I can think of four 
possibilities for the discrepancy.

1.  The tuning was wrong. - Possible since I had no real test equipment to 
be sure.  We played with some things but there was a lot going on and my 
"pals" can be a detriment to fine science ;-))  But the tuning is not 
critical and I don't "feel" this was a major problem.

2.  Spark gap dwell. - The computer says the "optimal" spark gap dwell 
between the two systems should differ.  With my blowing the MOVs, fuses, 
dwell problems and all, the dwell got "adjusted" to goodness knows 
where.  However, judging from the consistent safety gap firing between the 
two systems, I think it was "ok".  I ended up using the same dwell for the 
two systems (a good spot for the conventional setup), but it is a source of 
uncertainty.

3.  Quenching. - If the coil were quenching well with the standard system, 
but quenching poorly with the magnifier system, that could cause a dramatic 
difference in performance.  This is the gap voltage chart assuming perfect 
quenching for both:

http://hot-streamer-dot-com/temp/NewMag/NEWvsMAG05.gif

The magnifier (top green) needs to quench far sooner and then has far high 
voltages peaks to deal with.  This may make quenching much more difficult 
for a magnifier as many have reported in the real world.  But note how the 
voltage at quench is zero as opposed to 20% Vfire in the conventional 
system!  But since energy is proportional to V^2, we are only loosing 4% 
((1/20%)^2) of coil power and streamer length to that.  Hard to say how all 
that played out.

4.  Corona loss. - Was all that pretty blue glowing stuff loosing all the 
power?  "Maybe"  but the conventional setup had some pretty good corona 
going too at the top turn of the secondary.  It had a big pretty blue sheet 
"shroud" effect going on.

I would guess it was a combination of a few of these...  "Knowing", that 
the computer IS "right"...  It would appear that a magnifier should do 
maybe 20% better streamer length than a conventional system, but it takes a 
little care in getting everything just right.  So I will work on 
eliminating the possible problems and then seeing how it does again.

I will also get the pinger going so Paul's program can tear always 
searching for losses and other interesting details.  I have a pulled back 
muscle thing, so it is a little hard to climb around doing real testing at 
the moment...

Had a very nice talk with Jeff Parisse today on the phone.  He had many 
interesting ideas and observations on this ;-))

I replaced Q4, D5 and D6 on my ailing Kill-A-Watt meter but that did not 
seem to help the calibration problem.  I have two new ones coming so I can 
maybe I can probe around and try comparison things when I have a working 
one too.  I am still hoping they can be "hardened" to Tesla coiling 
use.  They are just too cool!!!  I don't think the calibration problem is 
in the new LM324 since it stayed the same with the new chip.  I have yet to 
go through the resistors with the HP34401...

BTW - If you type "Kill-A-Watt" into froogle:

http://froogle.google-dot-com/

Spytown-dot-com has them for $29!  I didn't get mine from there (I didn't 
know...) but this Froogle thing sure is cool!!!!

Cheers,

	Terry