RE: Variac turn-on surge solution

Hi Gary,

	Comments and testing results below...

At 10:32 AM 10/28/1999 -0400, you wrote:
>One thing I failed to mention - When I was using the fusebox, the route to
>the fuse was long and through an extension cord as well.  Significant
>resistance.  Now with the circuit breakers, I have an outlet in the garage
>directly off of the breaker box, not more than 2 feet away from the breaker
>itself.  Perhaps it was the length of the old wiring that was cushioning the
>turn-on surge, rather than a difference between the fuse and circuit

This is VERY possible.  Perhaps a little inductance or resistance would help...

>I did go through the Home Depot circuit breaker isle looking for a slow-blow
>version, but didn't see any different breaker options other than current
>ratings. Perhaps a "real" electrical supply distributor might have had more
>options.   I did try both a 15A and 20A breaker, same result.  I don't think
>the 20A unit was defective.

Very unlikely that two breakers would be bad.  I work with some industrial
breakers were you can pick all kinds of trip ranges and delays.  Perhaps
house stuff is not as flexible.  Perhaps the following will shed insight...

I hooked up my fiber optic probe the my variac.  It is a 15 amp job with
lots of MOVs and dual RFI filters in it but it is basically just a
contactor hooked to the variac and then to the output sort of deal.  It is
hooked to lots of extension cords and house wiring to thermal breakers.  I
have a bit more primary resistance than you do.


This is the scope wave form of many turn-ons of just the variac with
nothing connected to the output and set to zero volts.  Looks like ~100 amp
spikes of 4mS duration.  


Same thing with just the neon added.  Variac still set to 0 volts.  Really
no difference.


Turned the dial to 120 Volts and switched in on and off many times.  Still
just the neon is hooked to the variac.  Note that the neon has 200uF of PFC
caps in the primary and another filter/MOV set.  Now we get thin (0.5-1.0
mS) spikes of 80 amps and long duration charging current of ~15 amps.


As above but with 27.5 nF cap added to output of neon.  This sounded bad
and the lights dimmed (I turned if off FAST but the trusty scope caught
it).  No wonder.... 50 amps continuous peaks of perhaps 30 amps RMS.


Same thing but I turned the variac up slowly.  The current grew nicely as
expected until I hit about 100 volts.  Then something seemed to go into
saturation and the current suddenly jumped!  Got the 50 amp peaks again.
Something "funny" is going on there...  I'll have to look into that.
Probably the variac, PFC caps, neon, and primary cap are hitting some
saturation or resonance.  With a spark gap in the system, this does not
seem to appear.  A MicroSim problem ;-)

Hopefully this info will help you out.  I bet with low resistance wiring
those spikes my be getting higher than my 100 amps spikes.  



>Regards, Gary Lau
>Waltham, MA USA