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

Re: Weird safety gap behaviour



Original poster: "Gerry Reynolds" <gerryreynolds-at-earthlink-dot-net> 

Hi Ed,

I'm thinking my case may be a different phenomenum.  My capacitance is 26.5
nf and I'm at 2.5* Cres.  The thevenin inductance of the tranny is 663H and
the thevenin resistance is 15.9Kohms.  The calculated the steady state
voltage on the capacitance is about 14500 Vpeak (assuming 21200 Vpeak for
the Vs_oc).  The safety gaps are now adjusted to not fire so no transients
present.

Today, Terry and I measured the PF during this event.  Driving a mostly pure
capacitive load and before runaway, the PF was 0.17 (small as one would
expect).  After runaway, the PF was 0.97 and the line current droped
slightly.   The power seems to be mostly dissapated in the NST.  The NST was
buzzing but not too violently.   I did note that the sine wave output was
relatively clean before the event and became very distorted after runaway.
The distortion apears to be odd harmonic in nature (symetry seemed to be
preserved).  A scope capture of the event shows a peak to peak growth until
30KVpeak  is reached.

My current thinking is that what ever the nonlinear effect is,  it is
causing the effective inductance to decrease to cause the resonant rise.  I
am studying ferroresonance to see what I can learn.

Gerry R.


 > Original poster: "Ed Phillips" <evp-at-pacbell-dot-net>
 >
 > That's not what I intended to say.  At least in my experience the
 > inductance INCREASES as the voltage is raised from zero, so to get this
 > phenomenon the capacitance would be less than that required for
 > resonance at normal operating voltage (the mythical "matched"
 > capacitance). If you look at a typical B-H curve you'll see what happens
 > to the permeability as the flux density increases.
 >
 > The first time I encountered the effect was when I had a big 0.0062 ufd
 > transmitting mica capacitor connected to a 15 kV, 60 ma transformer.
 > That's a lot smaller than the "matched" capacitance of around 0.0106
 > ufd.  I hooked the capacitor and a Simpson meter on 5000V scale across
 > the transformer and brought the voltage up from zero with a variac.  At
 > somewhere around 30 volts, as I recall, the meter suddenly pegged,
 > taking out the 5000V multiplier.  Fortunately the rectifier and meter
 > weren't damaged.
 >
 > Ed
 >
 > P.S. This was one of those stunts which, right after things went wrong,
 > caused me to wonder "why did I have to do that!".
 >
 >