[TCML] Quench time

[Chris Swinson]

> I am not out to rake Chris over the coals, or anything like that. In
> fact I was not really thinking about his coil at all with my last
> post. Mostly I am looking at the large number of posts on this topic
> where the accounting of energy needs a bit more clarity of thought.
>
> As for conduction loss at the spark gaps, in wires, and between tank
> capacitor plates. One can make a general remark regarding
> thermodynamic efficiency. As we slow down the flow of energy across
> dielectrics and imperfect conductors we will find that our entropic
> losses decrease This may sound trivial, but it is not. In theory if a
> spark gap could take forever to act, it would have no heat loss.  Of
> course it is also true that a spark gap that took forever to act would
> process very little power. (the catch- 22)
>
> As a general trend, slow reactions equate to higher efficiency.
>
> Jared Dwarshuis
> _______________________________________________


Yes that is correct. Though there was actually 3 coil designs been talked 
about all at once which is most of the confusion.

The problem is if you reduce the primary turns, it pulls more amps from the 
tank, so you need a better tank cap to start with. more current over the gap 
which is bad... so with the gain in amps and the gain in losses over the gap 
you really do not gain anything. If you replace the SG with solid state ( ok 
just assume a 20KV 50KA SS switch )

The second problem is frequency, skin effect, resistance over the primary, 
so you make the primary much larger in area to help counteract the losses 
related to frequency.

Once you manage those 2 things, then you can design a very high Q system 
which will give greater voltage gain.

Even though you *may* loose a lot of energy related to higher frequency and 
higher losses, also do not forget the secondary Q is also many times higher.


So in relation to my first posts, I was talking about speeding up the RSG 
dwell time to help quench the gap faster. This is a must if you increase the 
amps otherwise you will gain nothing. Though as everyone knows, its all 
"catch-22" and "swings and roundabouts".

overall, it is not hard to accept that a high frequency tank circuit will 
overall results in more losses than lost over a spark gap, though with a 
higher Q system you do not have to worry too much.

Each point I have made, could take 20 posts to talk about, though I think 
everything has already been talked about so far over the past week.

The first step is to assume the RSG *CAN* quench at 1st notch regardless of 
amps.... I really do not want to get into how this cannot work (again) but 
this is the first step in a new design as you are not restricted to RSG 
conduction problems... Higher amps are now accessible , how would you make 
use of that single factor ? 1uF tank cap maybe charged to 20KV 200Joules! 
Tank cap must also be capable of pulsing in the KA range, all the wiring 
needs correction also.. keep thinking this way and you will start to design 
"new" higher Q systems...

I think all the factors have already been talked about over the past week so 
no need to go over old ground again.... All is "WIP" so I am not casting 
anything in stone.,,,, *yet*....

The overall first idea, was that current causes the spark gap to turn off 
slow. As more current is more spark length, then this is a good thing. 
Though as the spark gap suffers with higher current , after a point of over 
200nF or so, you run into problems. It was my first initial idea to still 
run at 100bps, but make the RSG maybe 30" in dia to reduce dwell time.. 
electrical dwell time is probably 2 or 3 times that value, so again the need 
to increase the dia of the disc... The basic idea that say if you built 100" 
dia RSG the dwell time would be so low that you *should* be able to push 
more current over the spark gap and quench a lot faster...

Though we have come back to the problem already talked about anyway... no 
need to keep going over old ground...  I have started a high Q design using 
a spark gap though this will use 20KV 5nF so the RSG is not really a factor. 
This design will be high Q to help explore just the high frequency 
aspects...I am also building a high Q solid state system using low voltages, 
though all this is really a new area to tesla coiling so "new developments" 
will take time..

I think everyone has got stuck with low frequency that higher frequency's 
never even get a proper look at. Higher frequency = more losses = poor 
coil... which is proven fact.. though there are ways to counteract these 
problems to increase efficiency elsewhere.... it is the same as the RSG, 
more current is good, but you have to limit it to gain efficiency 
elsewhere... IMHO, not trying to slate anyone here... though it seems to me 
everyone just thinks lower the frequency the better but there are so many 
factors involved has anyone *really* thought about gaining efficiency in 
another way other than lowering the frequency ? I think everything has just 
become "in general" over the years... I think I will probably get some hate 
mail for suggesting higher frequency higher Q systems...

I am not to hung up on how well it does or does not work in the end... 
Though it is just a area I wish to research. just as much can be learnt in 
success or failure! *if* I do get very good results then maybe more people 
will start to design high Q systems... the aspect of winding 100turns of 
thick cable rather than 2000turns of 0.2mm wire is worth a look in my book!


/end rant

Chris