High voltage using car ignition coil

I have been contemplating my next move.  The SGFP in the NST I believe is a big problem. Here is a picture of it with the access plate on the bottom removed.  The potting compound over the transformer in the left compartment is epoxy-based and impossible to deal with.  As long as it is just the transformer in there, there should be no need for access as long as sufficient wiring comes out to feed the primary properly.  The compartment on the right hopefully contains all wires and circuitry needed for access as the potting compound there is very soft and easy to remove.  So it appears possible to perform an operation to remove the SGFP circuitry.  Is there an oscillator circuit that should be preserved?  If it is destroyed during potting compound removal, is the transformer ruined for a Smith Generator?  

I am betting the oscillator circuit is not needed, by looking at Marathonman's post of Tesla's spark gap circuit to power a secondary and tertiary coil.  But if I tear it apart will I be able to preserve complete access to the transformer part?  That has led me to look for some other options.

Comparing the Smith Generator to the Tesla circuit,  the L2 is really the tertiary coil counting from the power input perspective.  But how does a circuit spark at such high frequencies without any oscillator circuitry and diodes to gracefully feed a capacitor to spark voltages?   I am beginning to think the Smith Generator is probably not operating at 35.1kHz of an NST, but at the 220 MHz that Don mentions in his videos and book.  In the Resonant Energy Systems book on p42, he states,

"Let us then compare the 60 c.p.s. System with my 220 MHz Device."

Two questions need answered.

1. How can the "early device" we are trying to build actually be Don's 220 MHz Device, and operate at the 220 MHz frequency and not 35.1kHz?
2. How can the Tesla device operate without a diode/capacitor circuit to build voltages for the spark gap?

I would say what is needed to generate extra power is high voltage, high frequency, and resonance.  Keeping these principles on the table going forward, lets look at question 2 first.

Fellas,  Thanks for the switched  inductance info on the output and the Utkin PJ Kelly document.  I felt I was missing something in my adaptation of the DS circuit.  But I have noticed several pictures without them now in the Utkin Document, having just a bridge rectifier, as in my schematic.  The hand drawn circuit given early in the thread out of my source document for building a DS generator has both.  But it also shows using a voltage divider at 8000v which to me would be very "hairy".

I will be studying the Utkin document more.   

So we seem to be contemplating running the DS Generator L1 circuit without diodes (at least I am).  If you notice in my last picture of my disassembled NST, a GM ignition coil is lurking upper right.  These are pretty powerful and I have already show how they can provide powerful sparks, as I have shown with a picture and simple Tesla-like circutry, previously.  This model has dual HV output, to drive 2 spark plugs at the same time, and it would also spark at the top of every piston stroke, exhaust or power. Tthe DS PJ Kelly book on P95, has a diagram of a circuit dispensing with the NST and using ignition coils.  It shows a very high power rating, current and voltage.  It is shown below.  

Notice that it includes oscillator circuitry and uses the diodes.  Building a faster oscillator was discussed in the thread previously, to try to get up to a MHz range, but it is apparently unnecessary as long as you are using a spark gap that fires sufficiently fast.  With my L1 coil length of ~18', I am off the top of my head looking at maybe 18Mhz quarter wave.  I think this is a big step so far and I am confident it will work.

Success, sort of.  I am getting +10,000v with my new car coil, 2uF cap, and dimmer.  However I seem to have no control with my dimmer.  I sure would like to put out about 1000v here.  Well this bypasses the diodes and SGFP issues!  Whew, at least its sparking!

How to Make Uni-directional Wound Coils "Buck".

In Chris' partnered coils hyiq lead off photo, let us examine coil winding and current flow.  As we employ this circuit, we us AC current as Nikola Tesla and Don Smith did.  Lets take a snap-shot of current flow in one direction, as indicated by the black current arrows in the coil wires.  Due to the change in coil winding direction from Clockwise to Counter-Clockwise, these coils generate opposing North magnetic fields, cancelling themselves to give us current amplification.

In the video I posted back in Jan. on Kdkinen's work on the Smith Generator, he uses Barker-Williams stock coils hooked up in "series", just as Don shows in his display model, all wound in the same direction.  He notes he gets significantly better performance, however, when he has hooked the coils up in "parallel".    Then the question becomes, "Can you make two coils wound in the same direction behave as indicated above when they are wound either both CW, or CCW?  If so, how?"  I wish to illustrate what he means using Chris' diagram slightly redrawn to illustrate it.  In the following diagram, both coils are wound CCW.  The twist direction is a coils unique property.  It is the same no matter how the coil is rotated to examine it.  The right side coil is the same and in this snap-shot of DC current, with it exiting out at connection B, the same for Chris' coil.

In order for the left coil to generate an opposing N magnetic field, the current must be driven in the opposite direction if both coils are wound in the same direction.  Using the right-hand rule, the input point A of the left coil must be moved to the middle of the whole coil.  The center-point connection C must now be made from the left side of the coil to the left side of the right coil.  For a CW coil to generate cancelling N magnetic fields, the current flow in the above diagram must be reversed also.  That is how an off-the-shelf Barker-Williamson coil, or any set of coils wound in the same direction can be connected to achieve partnered output performance.