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

Re: 1st Motorized Spark Gap!



I can offer a reply to some of these queries....

> 1)  I have a computer program RotJit which says a large disk with lots of
> bolts in it would be better than a small disk. What sizes (diameter) are
> most of your spark gaps? Whats the largest diameter I should use? Would
> an 8 or 10 inch diameter disk wobble too much? I have no machine tools,
> so the tolerances won't be exact!  

You _must_ use a properly balanced wheel or you'll end up with 
equivalent of a grenade. A large disk means more distance between the 
contacts for a given number of contacts which = reduced dwell time as 
the wheel gets larger for a given speed.
     I recall from the Rotjit analysis that the main concerns were 
with avoiding break rates that allowed very non-uniform cap charges
which depended on the portion of the mains cycle at the time each 
break happened (_much_ worse than a static gap at slow breaks - see
comments by Jim Fosse elsewhere).

> 2) Is there a certain speed-range that most of you find works best? I
> realize that speed probably affects tuning, but what would be considered
> to slow or to fast?

That depends entirely on the charging characteristics of the power 
supply (which includes the transformer, chokes, and primary cap). For 
example, a large limiting inductor charges a given capacitor much 
more slowly than a small inductor. This is really a can of worms for 
a simple explanation. The cap, if not hindered by a high break rate, 
will charge to a voltage allowed by inductor energy storage which is
dependent on inductance and peak current according to 0.5LI^2 (cap can
reach :  Vcap = Ipk*SQRT(L/C) assuming no additional transformer push 
behind it - usually not the case). You can see from this that varying
break rate can seriously influence cap voltage assuming significant
energy storage in the inductor. Greg Leyh used this to good effect in
his big coil which I recommend anyone interested in issues surrounding
the use of rotaries and resonant charging get a hold of.

Malcolm