[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [TCML] Spark gap Resistance
Dr. Resonance,
John has hit on a very important point here. With classic RSG type coils
2nd thru 4th notch quenching seems to work best. As he suggested the
problem lies in deionization or interruption of the spark gap. This is
problamatic due to the very high temperatures involved in the spark
channel, especially with all the capacitive energy behind it.
High current make harder quench ? This would support john in saying high
current across the gap is bad... though it was why I mentioned about running
a much faster RSG to decrease the dwell time. I am not saying its a fix but
really can't hurt ?
So is the problem the quench, and/or actually passing more current across
the gap ?
I would have thought higher current would conduct better, but also makes
quench a lot harder. Easy example, double the current and double the quench
time, over all gain zero....
It is what brought me onto the idea that if the actual conduction of current
across the gap is not a factor, but the only factor is the higher current is
making the gap harder to quench, then it was the idea to reduce dwell time..
use 30" dia RSG run 100bps, 1" dia electrodes, pass a lot of current easy
over the large electrodes, will help keep them cool also... plus the large
dia RSG will reduce the dwell time a lot. maybe if you double the current
you have to reduce the dwell time 4 times to make it work.... just pluggin
at ideas.....
This is one reason why experimenters are now turning to the dual resonant
solid state coil drivers to replace the less efficient spark gap type
switches. The IGBT type switches, replacing the open air spark gaps, are
able to turn on and off at higher rates providing more rapid dI/dt rates.
Due to the fact there is no open air spark, excessive heat does not
provide the shut-off conductive problems that a spark gap endures.
But is the spark gap "really" a loss.... IGBT can turn off whenever you
want, though *IF* you could turn off the spark gap just as fast, then what
would be the results then ?
The switching off part is much more efficient, less heat and no light
energy is wasted, and the IGBT type switching allows the experimenter to
keep 30% more energy in the secondary coil to ring up without trying to
dump the energy back into the primary system. First notch quenching is
achieved with efficiency. Spark lengths are exceeding the classic 1 ft
per kVA range typical with RSG systems.
Sounds fair enough... and I agree... But how much is lost in the spark gap
heat& light wise ? which is the greater loss, heat & light or turn off
conduction problems ? (and I just know you are going to say both...)
For this same reason, classic type coils like a coeff. of coupling in the
range of 0.18 to 0.14 while solid state coils, with much more rapid
switch-off characteristics, operate in the coeff. of coupling range of
0.16 to 0.20. At present, research suggests that 0.18 is usually a very
efficient design goal for max power transfer. Smaller coil systems can
tolerate up to 0.20 while larger coils seem to operate best around 0.18.
Tight coupling is where you get this "double hump problem" I saw some data
on this ( TCBA news maybe) it looksed to me like a lower coupling would be
better, though the frequency range seemed to be a lot narrow.. I may have
got it backwards... though the chart had like 50khz to 200khz for a double
hump 0.4K and 80khz to 120khz for 0.2K. I am not sure about the frequency
aspect of the chart but it looked to me that lower coupling would be better.
As I was pondering also.. lower coupling would be less inductance on the
primary and will greatly increase the current... which has to be good......
though when you factor in park gap losses, it becomes a bit of a problem!
To me, use a low coupling to get best primary current, and solve the longer
cycle/conduction time by just using a higher frequency. for example.
0.2K = 100khz = 10 cycles = 100uS
or
0.1K = 200khz = 10 cycles = 100uS
I know there are other factors to take into account.. though just trying to
improve on the spark gap problems... though in itself could generate more
problems. Spark gap current and conduction time was what I was trying to
indicate...
Bart Andersons excellent computer design program, JAVATC, allows one to
experiment with different sec heights and primary inner and outer radii to
obtain the desired coeff. of coupling range. It's easy to use and very
accurate. It works great with both classic RSG coil designs and dual
resonant solid state coil designs. Just note the different design goals
for
Yes, I use JavaTC all the time, I think I broke the RSG bit, email sent to
bart about it (pokes bart)
the coeff. of coupling as you adjust your inner and outer primary radii
and also the height of the sec lower winding above the primary. I usually
use 0.25 inch elevation of the sec above the horizontal plane of the
primary for high coupling. Pri-sec spacing usually around 0.5 inch.
I had a thought about that also... say if you had a 24" dia secondary and
used a 12" dia primary and raised the coil maybe an inch above the primary..
how would that work out over a normal primary coil which is a few inch's
wider than the secodnary (normal type setup)... ?
Use as large of a dia secondary coil as practical, typically around 10 to
20 inch dia range, to obtain max. sec potential. This also helps keep the
res freq low while helps the IGBTs switch more efficiently. They are
rated at 75 kHz, but operate more efficiently below 50 kHz.
I would like to use a large dia secondary to get high inductance (which is
good, right?) but also use a large dia wire to get resistance down. I did
plot some ponders and charts on this, to my workings resistance goes up
almost exactly the same as inductance so Q factor never really alters... I
think the best coil would be a wide dia secodnary, and using a thick cable
to get resistance down to less than 1 ohm. This is my dream coil, though it
also clashes with number of turns so I am not sure if it would even be
physically possible to build one... I think it must be as Tesla seems to use
less than 1 ohm coils.
I read ( in CSN I think) that resistance was more of a factor than
inductance for getting long sparks. I do not think anyone even considers the
resisitance of the coils and just goes for high inductance all the time ?
I also have a second ponder ( getting off topic a little ) that Tesla said
the large toroid is only there to counteract the large inductance.. While I
gave up a few weeks ago trying to work that one out, It sounds to me like
Tesla says use a small dia toroid, and get the coil resistance down to 1 ohm
or below. Maybe use a larger dia tube to get the inductance as high as
possible... Though I think resistance increases pretty much at the same rate
as inductnace does... which makes it pointless and just forces the use of a
large toroid for no reason.....
Experimenters who have seen a solid state coil in action immediately
notice the very loud "mean sound" from the spark as the much higher
currents tears the air molecules apart in the spark discharge. The spark
is also white hot and not blue or purple like the lower secondary coil
currents in a classic RSG design. Keeping the energy in the secondary
coil is the name of the game for maximum performance.
I built a SSTC, though this was just a basic H-bridge. I have a video of it,
only got 6" spark output and was very poor. I tried many coils and I found
the more inductance there was the worse it was. It could have been the
inductance on the primary acting up though I was pulling about 10amps at the
time... also it was CW and still gave purple sparks. I can only account for
this for the secodnary resistance, I think it was running 60khz and DCR was
120 ohms 0.2mm wire. I think I ran another coil , 0.4mm wire, probably
60ohms 120khz (guessing) worked better... I do remember a small 1" spark in
the early days of testing which was white hot with a hiss.. but as things
progressed to longer sparks they went purple.
some info here if you want to take a look-see
http://www.future-technologies.co.uk/IMPULSE/hotstreamer/sstc/sstc.htm
Many happy Holiday sparks!
Holiday sure, sparks maybe not ;-)
Chris
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla