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

Safety Gap Importance (long-sorry)



Original poster: "Ted Rosenberg by way of Terry Fritz <twftesla-at-qwest-dot-net>" <Ted.Rosenberg-at-radioshack-dot-com>

Hello everyone. For the experienced coiler this may be boring. For the
newbie, please make notes!

I built a 900 Watt coil in 2000 for use in a local haunted house. It worked
flawlessly for 2 years straight. Due an apparent build-up of dirt and
conductive material on the NST protection circuit board, AKA the Fritz
Filter, an MOV shorted when I started it up about 4 weeks ago. I decided to
rebuild the entire circuit board from the start.

The coil is conservative. It uses a 15KV/60ma Allanson NST and a t-gap. The
MMC consists of 2 strings of 11 C-D caps for an LTR of .028mF. The primary
is 1/4" refrigerator tubing, 24-inches in diameter while the secondary is
about 1500 turns of green #26 wire on a 6-inch PVC form.

I obtained another set of 14 MOVs from Digi-Key (USPS Priority-$20) and
constructed a new board using 1/8" Plexiglas. It was much larger (11x12
inches) which allowed me to space out all components. Looked good too.

When I brought up the Variac, the coil worked. But in less than 1 minute, it
stopped and I knew something was wrong. I looked at the MOVs and they seemed
to be askew on the board. As I removed the board I realized that all the
MOVs were HOT, not just warm. I turned the board over and saw that all the
soldered leads were mostly open and the solder melted. The Plexiglas beneath
was slightly deformed from heat.

With assistance from the circuit's designer, it was soon determined that my
safety gap was much too wide. I had estimated the setting when I completed
this board. The distance from each HV side to the center GND ball was about
1/4-inch. It turned out to be greater than that but I didn't know it then. I
thought that is what I had on the first board. I was wrong. The safety gap
never fired. It was far too wide to fire at all. Instead the excess voltage
took out the MOVs, literally cooking them. The solder on their leads melted
at 320 degrees.

I ordered another set, (another $20) increasing the string to 8 on each side
of ground. I apparently was getting more available voltage from the addition
of a PFC box ahead of the NST. The amperage was less but the voltage was
more.

This time I made a "daughter-perf board" with just the string of MOVs on it
and bolted it between the three brass 1/4-20 screws I have for the HV and
GND input on the Plexi board. That allowed easy removal and re-attachment
before and after I set the safety gap. And I setup the gap from scratch!
When I was through, it turned out that the right distance was slightly less
than 1/4", not slightly more! There must have been at least 1/8-inch
difference.

The moral of this long story is simple. Do not take that safety gap for
granted. It serves a very important purpose and it's setting, while not
super critical, must be set carefully or you will clearly pay $20 over and
over to Digi-Key or the supplier of your choice. Step by step safety gap
info can be found on the pupman archives at: <Terry...could you put in the
pupman archive listing you sent me??>

And the coil is working well. The gap fires occasionally and briefly. And
the output is fantastic once more. 

Safety First

Ted