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

Microsim Modeling - SRSG



Original poster: "Jeremy Scott by way of Terry Fritz <teslalist-at-qwest-dot-net>" <supertux1-at-yahoo-dot-com>

Hi,

I've been experimenting around in microsim trying
to figure out the optimal settings for a rotary
spark gap.

The problem which is usually not evident is that
unlike the real world, microsim ass-u-me-s unlimited
power. It doesn't know or care that a transformer is
designed for 60mA, it will draw whatever it takes from
the secondary to charge the tank capacitor. Throwing a
current probe on the other side of the transformer
will show in excess of 40A on the primary. That loud
snap was the tiny 15A breaker signalling it's
displeasure with the situation. That smell is the 14
gauge wire in the wall beginning to smolder.

So, do we fire on 60Hz peaks, giving 120BPS, or
should we try for more BPS in hopes to achieve
more spark length performance.

Well, to answer that question we have to drop
the assumption that unlimited power is available.
One needs to figure out how much current the
transformer can draw. Using ohms law we slap a couple
of resistors in the microsim circuit on the secondary
to make sure that no more current is drawn than
realisticly possible. For a 15/60mA NST, ohm's law
tells us that's 250K, so I put 125K resistor on
each side of the secondary's windings.

Now the story changes.

No longer is it acceptable to fire on 60Hz peaks:
because there is limited current available, the
capacitor may not yet be fully charged -- the voltage
across the capacitor is not necessarily in sync with
that of the line feeding it.

Generally, the faster the capacitor charges, the
better. The 60Hz AC half cycle is 8 1/3 ms long, which
means the capacitor has exactly that amount of time
to charge and discharge. At exactly 8 1/3ms there is
no voltage coming from the secondary, so the capacitor
dumps whatever it's got up until there into the core
of the secondary. This is stressful for the
transformer if it is done for many cycles because
it forces many amps into the secondary as the
capacitor discharges.

This is why LTR (Larger than resonant) capacitors
fire AFTER the peak, but before the 8 1/3 ms
deadline, and why it may not be advantageous to
have more than 120BPS with an LTR cap.

Resonant caps usually charge to their supply's
peak voltage at 13.4ms. At 8 1/3 ms the charge
on the capacitor is not quite enough to trigger
the spark gap, instead it dumps it's charge
into the secondary transformer windings. The
transformer converts it to an electromagnetic field
within the core which combines with that already
created by the primary windings. Then the field
collapses as the polarity reverses on the primary.

There is an extra boost to it now because
instead of just electromagnetic energy from
the primary, there is also energy added from
the partially charged capacitor in the core.
(From the last halfcycle.)

When all that collapses back into a reversed
current, it takes just 5.1ms to achieve full
potential on the capacitor now. This cycle
repeats until the capacitor voltage reaches
several magnitudes greater than that of it's
supply. (Resonant rise.) Sync gaps with resonant
capacitors usually don't improve performance,
instead a properly quenched static gap is best
if one wants to take advantage of resonant rise.

I've found in the LTR case that the best "time"
for gap conduction is this: Calculate gap conduction
time from the RPM and diameter of the rotating
electrodes. (disc diameter and also electrode
cross section diameter) Subtract that from 8 1/3
ms to find the best time to fire. Using trig, sin()
calculate the phase angle and adust the gap
accordingly. This ensures that the capacitor will
have the full halfcycle to charge minus the time
needed to oscillate the tank.