Re: Smoke(d) Filter

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Terry Fritz" <twftesla-at-qwest-dot-net>

Hi Matt,

The MOVs turned on, overheated, the internal solder connections melted, the
solder wicked around the MOV elements, the breakdown voltage dropped, the
current through them shot up, the power went way up, and the heat toasted
them.  They did what they are supposed to do in trying to save the NST form
over voltage.  But with 3600 watts behind it, they didn't take it very long :-(

At 60 or 120 mA the MOVs can take a little abuse, but with 3600 watts online,
they will crank on full short to try and save the NST.  You may want to
consider adding another MOV to each side to give them more margin before they
turn on.  Be totally sure the safety gaps are set correctly.  You may want to
check out this post:

http://www.pupman-dot-com/listarchives/2002/March/msg00614.html

How high was the variac set (140VAC?) or is it a resonant system?  It sounds
like the voltage across the MOVs got too high.

They do have bigger MOVs but they get to be really expensive.

I wonder if I should go to two more MOVs in the strings to give greater margin
for 15kV NSTs...

Cheers

        Terry


At 10:07 PM 3/14/2002 -0500, you wrote: 
>
> ARGGGGGH!
>          Brought my coil with incremental power supply on line full today
> with disaster! My "BigTerry-style"  filter has worked fine for 6 months at
> 15/60, worked great for a week at 15/120 and for a day at 15/180. Today I
> kicked it up to 15/240 and the MOVs joined in a mass extinction, quietly
> releasing volumes of choking, magic smoke. A post-mortem revealed the 100 W
> resistors were cold, the caps were slightly warm, and the MOVs were scalding
> hot; several had ruptured. Safety gaps set at ~ 8 mm each side of ground;
> SRGS w/ phase control and 2 gaps set to about 6 mm -at-120BPS. Any suggestions
> before I build/burn another one?
>
> Matt D.
> G3-1085 
>
> And God Said:
> Del . Fe=rho(ve)/epsilon;  Del . Fm=0; Del x Fe + d Fm/d t =0;
> Del x Fm=mu*J + mu*epsilon*(dFe/dt)  . . . and there was light !