New Transformer/Gap Quenching

["Malcolm Watts" <M.J.Watts-at-massey.ac.nz> (by way of Terry Fritz <twftesla-at-uswest-dot-net>)]

Hi Coilers,
            I've completed construction of a new power 
transformer which was designed to overcome limitations of the 
one I presently use with the 10" coil at work here. The 
principal limitation of the existing model was gross 
inefficiency due to high resistance in the windings and a high 
magnetizing current (about 6A) due to fewer than optimal 
primary turns for the core which was the result of a 
successful attempt to boost its output with a fixed secondary.
     The new transformer now has a secondary which occupies 
much more of the winding window, 0.25mm wire instead of the 
old 0.15mm, bolstered insulation and a couple of thousand 
extra secondary turns to boot. In fact each layer is wound 
right to the edges of the bobbin and air-dried varnished 
before applying the next layer. The reasoning here was that no 
air pockets is better than increased creepage distance between 
layers provided the bobbin is well insulated from the core 
legs.
     The new primary uses 1.25mm wire instead of 0.8mm, also 
occupies much more of the winding window and has a tap which 
allows me to operate the transformer in an added boost mode 
without bringing the core quite so close to saturation as the 
original as well as an optimal inductance at full turns to 
allow somewhat higher output voltage than the original with a 
relatively low Imag.
      The transformer is wound on a C-core with windings on 
opposite legs giving a leakage inductance of about 170H which 
serendipitously allows resonant operation with just one of my 
four 67nF MMCs. I can series-parallel all four to allow 
operation with resonant boost to 16kV.
      The 10" coil is a fixed arrangement which uses a 0.1uF 
primary cap so is LTR. If anyone out there is keen on winding 
their own transformer I heartily recommend placing the 
windings physically apart on different legs. No shunts are 
required to give a useful limiting/resonating inductance and 
pri-sec isolation is assured if the core is grounded. 
Operation with a static gap is smooth and well-behaved even 
with a LTR cap.
      I tested this transformer on the coil a couple of days 
ago and noted the following (apart from the obvious such as 
increased output from the coil and far less heating in the 
transformer windings: whereas operation had been previously 
very touchy with regard to airflow through the gap (too much 
and the gap couldn't fire; too little and the gap power-
arced), the increased stiffness of the supply coupled with the 
higher voltage at the same (best for the previous transformer) 
gap setting allowed me to use a much wider range of airflows 
including quite a stiff breeze for a variety of quenching 
effects. As expected, too much air did decrease output but 
smoother operation resulted with more than the gentle wafts I 
had been forced to use previously and the coil is a much 
better behaved overall. In fact the airflow can in effect be 
adjusted as yet another "tuning" parameter for best output.

     FWIW - the transformer secondary incorporates a variety 
of inter-layer insulation (mostly paper/varnish but several 
single layers of PTFE tape at more-or-less even intervals) and 
a reasonably thick coating of a rubber compound which comes in 
cans, contains a variety of nasty solvents and used to go by 
the name of "Colorguard" or something similar. Excellent stuff 
for coating high voltage components and I have used it for 
20kV EHT windings in the past. More conventional use includes 
dipping the handles of pliers etc. to insulate them. It needs 
to be applied in a number of coatings as applying it too 
thickly leaves gas pockets. There is also a double layer of 
PTFE tape between the phenolic bobbin and the core and which 
stretches an inch beyond each end of the bobbin so allowing 
self-suspension of the bobbin between the top and bottom core 
end legs. The core itself is ungapped.

Regards,
Malcolm