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

Re: 43" First light! - Dead Transformers :-(

Tesla List wrote:
> 
> Original Poster: Tony Lekas <tony-at-lekas-dot-org>
> 
> Here are the stats for our first coil.  I have several questions.
> 
> 3-15kv 30 ma NSTs
> 0.028 30KV Pulse Discharge Cap - Surplus
> Primary -  8 turns 1/4 fridge tubing, angled up at 15 degrees
> Ground - ~10' heavy copper clad steel rod, #4 stranded wire
> Secondary - 8.4" SDR35 Sewer Pipe, 45" long, 1500 turns #22, 1/4 wave freq
> 77.5
> Khz
> Corona Shield - 36" across, 7" cross section, flex heating duct and
> aluminium tape
> 
> Spark Gap - Single gap set at ~1/2" Made of 1/8" brass fittings.  There is
> a flat
> square plug with a small hole drilled in it facing a round plate.  A 1/2
> horse air
> compressor is kept busy blowing air through the small hole.  Lots of
force and
> flow.
<SNIP>
> How much better performance should we expect to be able to get from a
> 15KV/90ma
> supply?
> 
<SNIP>
> 
> I am looking for suggestions for a good protection system for this coil.  I
> will
> have 4 more NSTs available so I should be able to operate with 4-7
(120-210ma)
> depending on how many I can repair.  I am considering unpotting them and
> puting
> them in oil.
> 
> I am also looking for general suggestions on solving the self strike
> problem and
> increasing the performance of the coil without destroying it.
> 
<SNIP>
> 
>     Tony

Tony,

Sounds like your coil's getting there! 15 KV at 90 MA should get you
into the 4-5 foot range... and it sounds like you're getting close.

Some suggestions:
A conical primary will tend to attract more secondary strikes than a
flat one. Either type should have a grounded strike rail to prevent
"hits" into the primary circuit. Make sure you have safety gaps on each
HV leg of the NST going to ground in order to safely shunt any high
voltages that may come from secondary:primary hits or from overvolting
the main gap. The combination of a 1/2" air gap (you may want to reduce
this a bit) and the primary strikes was probably too much for the NST's.
Reducing the gap spacing will improve performance as well, since getting
the BPS rate into the 300-400 range will do wonders for streamer length.

A conical primary couples more strongly than a flat primary, so you may
also want to try elevating the secondary versus the primary to reduce
coupling a bit. The secondary winding is a bit long (which would
normally tend to reduce coupling). However, the symptoms you saw (i.e.,
inter-turn flashovers) when you raised the primary strongly suggests
that you were, in fact, overcoupled. 

It's not clear how well a single air-blast gap will perform versus a
multiple (series) gap. After making some of the other changes, you may
want to try doing a comparison test with an air-cooled/vacuum series
gap.

A 1500 turn secondary is on the high side (most secondaries in
well-running systems tend to be in the range of 450 - 1000 turns)- the Q
will be less than optimal, which will tend to reduce performance
somewhat. The added height should provide better voltage withstanding
capability once you've got the system running. The 0.028 uF tank cap
will have a hard time getting fully charged with only three 15/30 NST's
in parallel - you'll need around 150 or 180 MA to fully utilize the cap.
Unpotting and putting the NST's in oil should increase their strength...
but I wouldn't do this unless the transformer needed to be disassembled
for other reasons. BTW, I've found (the hard way!) that 60 MA
transformers seem to be built tougher, and seem to take over-current
abuse better than 30 MA ones (especially in resonant charged systems). 

I won't get into the LC, RLC, RC, etc debate on protective networks -
check the archives for suggestions... Oh Heck, yes I will! Overall, an
RC network on each HV leg sounds to me like it makes the most sense. Use
500-1000 ohm 50-100 watt wirewound resistors (to handle the voltage
spikes and power level)in sereis with each HV leg, and 500 pF ceramic
bypass caps to ground (transformer side of the resistors).  These should
block the any high voltage VHF gap/wiring parasitics from getting back
into the transformers.  :^)

Hope this helps!

-- Bert --