Re: [TCML] Spark gap Resistance

["Chris Swinson" <list@xxxxxxxxxxxxxxxxxxxxxxxxx>]

Hi all,

pondering some thoughts about spark gap losses too at the moment......

If the spark gap works better at the first few cycles, then I assume there 
is more energy across the gap. More energy making a better ionised flow 
across the spark gap ? If so, then why would adding more inductance reduce 
the losses in the gap ? surely the primary will limit the flow across the 
spark gap making less energy and less efficient sparks ?

On one hand it reads that faster and higher bang size the better, then 
saying more primary turns seems a contradiction as it will limit the energy, 
or at least slow it down, however you look at it, more primary turns I see 
as an overall loss ( It has to be?)    If you use more primary turns then a 
higher inductance is needed to get the frequency down for an increase in cap 
size...

I look at it that if you double the tank cap, which gives a lower frequency, 
you would need less primary turns ?  to keep the frequency the same?. It 
sounds to me more like a pulse current limitation of the tank cap more than 
anything ? It seems a bit daft to increase the tank size and increase the 
primary turns aswell ? It would lower the frequency and surely the spark 
would spend more time discharging over the spark gap and over all increase 
the losses ?

I tend to think that a large tank cap is great for a bigger bang size, 
though increasing primary turns is just increasing the primary losses ? I 
would have thought 100khz for example would discharge across the spark gap 
twice as fast as 50khz. so half the time over the spark gap should be a good 
thing, right ?

If we get onto notches, it complicates matters further. The spark gap will 
need to quench twice as fast from 50khz to 100khz ? If we assume first notch 
quench for 50khz then progress 100khz, then the wave is half the speed so 
100khz given the same RSG will quench at the second notch and waste a lot of 
energy in the process. So overall, it "looks" like 100khz does not work 
aswell, when in fact the problem is elsewhere in the system..

I just have a hard time seeing why everyone keeps going for high inductances 
all the time which just takes twice as long to discharge across the spark 
gap so should "burn up" more energy. I suppose it depends on if you class 
the spark gap as a loss or not and how much of a loss you class it as. Even 
so, more primary turns is a higher resistance, so more losses. I take the 
spark gap as a general loss of 10% per notch, I don't know the exact figure, 
but really, does anyone ?

I think overall its a trade off of many factors.... If you can transfer the 
tank energy in 100uS  for example, this is 100uS it is conducting across the 
spark gap and primary coil. Now double the frequency and you transfer the 
energy in 50uS instead, so half the time over the spark gap, will need less 
primary turns which should be better ( providing the tank cap can pulse a 
high enough current in the first place)

As a side note, I did crunch some figures the other day taking into account 
HF  losses and even though losses maybe double at a higher frequency, all 
other factors taken into account also, it still seemed to me to be overall 
much more efficient running at a higher frequency.

It makes me wonder if the main problems are RSG off times (to slow and too 
many notches etc) with also that most tank caps cannot pulse the current 
fast enough in the first place. So a trade off for poor RSG quench and poor 
tank cap results in a lower frequency being used.  It also makes me ponder 
on how higher power coils are not as efficient as smaller ones?, The larger 
they are the lower the frequency overall has to be and more energy burnt up 
across the gap per cycle maybe ?

Sorry for the amount of "ponders" in this mail.  It is just my 2cents worth 
that a higher frequency with less primary turns and a faster RSG would 
overall reduce losses far more than anything else.

Chris


_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla