Herrick's variable primary

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: "K. C. Herrick" <kchdlh@xxxxxxx>

Herrick is still at it, albeit fitfully.  Ref.

http://hot-streamer.com/temp/TCH-NEWPRI.JPG

My postings of some time ago, regarding the s.s.t.c. primary I'd been 
building inside an 18-gallon bucket, may be remembered.  I threw out 
the scheme featuring an internal primary: too much electrical 
stress.  Also threw out my paper-design for a variable-inductance 
primary coil and have -- finally! -- almost completed the new 
design.  I post this, well-prior to firing it up again since I'm 
facing other reworking tasks, thinking that a) it may offer an 
interesting idea or two and/or b) someone may be pleased to tell me 
in what ways it's a bum design.

If you will ignore the clutter in the photo...  At the left are 2 
knurled 1/4-20 nuts for the H-bridge input leads.  The 
series-paralleled capacitors are split into two strings, with the 
coil connected to those strings at the right side, underneath.  The 
coil is made from two stacked layers of 1/4" Cu tubing, formed in 
McMaster-Carr "multi-slot routing mounts" (their no. 7659A11 and if 
you didn't know what to call those you'd play hell finding them in 
their catalog without the no!).  The routing mounts, actually 
multi-position plastic tubing-clips, are arrayed on a Lexan disk and 
the upper coil is fastened to the lower one with pop-rivets from 
underneath so its top surface is smooth.  The disk, in turn, is 
mounted against a flange of the bucket.

Nine sliding contacts are made from 0.04"-diameter phosphor-bronze 
spring-wire.  They are clamped between aluminum guide-plates along 
with the end of a piece of 3/4"-wide copper braid.  The braid is led 
down through the inside of the hollow driving-shaft at the center.

Rotated by an offset pin fastened to the shaft is another, smaller, 
Lexan disk which not only rotates but also slides on the top coil  -- 
back and forth along the axis of  the phenolic stiffener that is 
fastened to it.  Its lateral position is determined by a pair of 
nylon guide-screws to be seen near the right end of the 
stiffener.  The nylon screws project down between turns of the upper 
coil so that, when the disk is rotated (via a plastic timing-belt 
sprocket underneath), the ends of the contacts accurately track the 
turns of the coil.

The brass-tubing shaft/braid assembly is stiffened by the insertion 
of a solid metal rod inside the braid.  Underneath, a 4-turn coil of 
the same copper braid -- having minimal diameter when it is wound the 
tightest -- connects the shaft/braid to the appropriate end of a 
capacitor-array.  (The outer end of the copper tubing coil-pair 
connects similarly but directly since it does not move.)  I've wound 
the braid-coil in the direction that will cause it to contribute to 
inductance as it expands, for the rotation-direction that increases 
coil-inductance (and vice versa, of course).  I measure, for the 
entire coil assembly, ~8 m-ohms dc + 7 uH at one end and ~13 m-ohms + 
22 uH at the other -- for a travel of a little over 3 (inside) 
coil-turns.  I want to ultimately accommodate 2 secondaries, 
resonating at ~100 and ~130 KHz.

The 12"-diameter secondary has fastened inside its lower end an 
identically-shaped bucket, with its top section cut off down to the 
(identical) flange.  Such buckets are designed to stack, so this 
yields a convenient way of holding the secondary firmly upright atop 
the primary--kind of like a bayonet-mount.  It serendipitously 
happens that the bucket is a perfect fit inside the secondary's 
sonotube coil-form.

I may expect corona from some of those capacitor-screws, or the 
copper straps; but when & if that happy event should occur I could 
likely add some small hollow aluminum balls to smooth out the 
fields.  McMaster-Carr stocks them.

KCH