A s.s.t.c. problem (fwd)

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: Gerry Reynolds <greynolds@xxxxxxxxxx>



---------- Forwarded message ----------
Date: Wed, 20 Dec 2006 14:09:35 -0800
From: K. C. Herrick <kchdlh@xxxxxxx>
To: Tesla list <tesla@xxxxxxxxxx>
Subject: A  s.s.t.c. problem

Steve Ward & any others interested-

I solicit comments on a problem I now have:  I have the following 
elements connected into a feedback loop:  A) a series-resonant primary 
circuit, set presently at ~120 KHz Fr when unconnected to the system; b) 
an IGBT H-bridge driver, itself driven by sundry circuits operating at 
both 5V and 16V; c) my "signal conditioner" circuit as shown in 
http://www.pupman.com/current/kcherrick/tch-synctest4a.jpg; and d) a 
100-turn current transformer on one output leg of the H-bridge, driving 
that circuit.  I'll prevail upon Chip to post a photo of the 
assembly--as, I suppose, 
http://www.pupman.com/current/kcherrick/tch-assy_12-06.jpg.

In the photo, the primary is at top center: capacitors arrayed 
circularly and the flat variable-inductance coil centrally.  The 
12"-diameter secondary can be seen set off to the side at the left.  The 
current transformer is just above right-center, around one leg of the 
H-bridge drive going to the primary; its leads go to the red connector 
at the lower left.

I've tried two 100T current transformers: one made on a ~1 1/2" diameter 
ferrite toroid of unknown provenance and the other (the one shown) wound 
toroidally on an air-core former.  Within the low-voltage circuits I 
have a polarity-reversing switch to set the proper polarity for positive 
feedback.

With the air-core current transformer especially, I get expected 
waveforms throughout (except for a bit of ringing when the H-bridge 
switches, as I've commented on already).  But the problem now is this: 
firing up the system, initially with the mains variac set moderately 
low, I find that I get those gratifying waveforms at about 106 KHz 
rather than the 120 KHz at which the primary resonates by itself.  As 
the man says, go figure; there doesn't seem to be enough additional 
inductance in the system (note the short, wide copper-braid connections 
from the H-bridge IGBTs at center in the photo and the same short braid 
interconnecting the mains-supply capacitors, left and right) to bring 
the Fr that far down.

So does this kind of thing ring a bell with anyone?  A) the loop should 
resonate essentially at the Fr of the primary, absent inordinately long 
leads; and b) the loop shouldn't operate at all with the wrong polarity 
of feedback -- as it does with the ferrite-core current transformer 
where I get an operating frequency of ~90 KHz with one polarity of 
feedback and ~160 KHz with the other.  Because of that, I'm certainly 
inclined not to use the ferrite transformer; but I still have the odd-Fr 
problem with the air-core transformer.

Help, anyone?

Ken Herrick