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

Re: SG questions cont'd + power factor

At 01:11 AM 01/04/2000 -0500, you wrote:
>Reinhard, thanks for the detailed response!
>
>>No. not quite. Quenching refers to the gapīs possibility to
>>switch the current on and off at the *proper* moment. I say
>>proper here, because a while ago there was a discussion
>>about the necessity of 1st notch quenching.
>
>I'm assuming a SRSG here. So how does one figure how much time is needed
>for the cap to fully discharge?  Is it a function of the V and I across
>the cap and the inductance of the primary?   

The time to the first notch (after the gap fires) is 1/(2 x Fo x K).  The
second notch is 3/(2 x Fo x K)...  Note that this is not an exact relation
but based empirical data from John Couture and others.  It is "fairly"
close.  To get "really" close you need to model it with MicroSim or some
such program.  To be "exact" you have to check it with a scope.

F is the fundamental frequency and K is the primary to secondary coupling.

>In most cases is the power
>supply stiff (think thats the right word) enough to charge the cap fully
>between electrode presentations almost no matter what speed the motor
>spinning them is (1800RPM, 3600RPM, etc)? 

Not quite.  We select cap sizes to be as large as possible to get the
biggest arcs.  So the cap is so large that the transformer is just barely
able to get it back to full voltage before it fires again.  This is life
and death stuff in neon systems but pig system have a lot more room to play...

>It seems that this would be
>have to be true in order to explain the evidence that a higher BPS
>yields more instantaneous power (I'm assuming it reduces the dt in
>dw/dt).  

Hi BPS systems us smaller sized caps that can charge very quickly given the
power source.  Neon systems us resonant charging for this and pig system
just turn up the current...

Hi Stan,

>Is the theory behind an LTR setup that the larger than res cap
>takes advantage of the PSU's 'extra' capacity?  How does reactive power
>factor into all this?

The large inductance (I am not sure if we agree at all what the inductance
is ;-)) of the transformers secondary stores a large amount of energy in
the form of current.  Just like a capacitor stores energy in the form of
voltage.  LTR systems are timed with sync gaps (some have gotten it to work
with static gaps too) and have carefully chosen primary cap values to
optimize their ability to use as much of the stored inductive energy in the
transformer as possible to help charge these large caps.  Basically this
inductance wants to keep the current at a constant level so it "pushes"
extra current into the cap.

>
>
>While I'm asking dumb questions, I have one about power factors.  Would
>it be possible to use pf correction on the secondary side of the NST?  I
>would assume i would be lagging or leading v on the secondary side too. 
>So wouldn't putting correction on this side give more bang for the same
>power output of the NST?

Such a capacitor would alter the output impedance of the transformer which
may help if the transformer was poorly impedance matched to the rest of the
system.  In practice, this function is automatically taken care of when we
select the "right" primary cap value.

>
>What's the best book/site for having the math thoroughly explained that
>governs this stuff to someone fully competent in calculus but not in
>EE--as I'm sure is made obvious by my questions :)

Since I sort of know ("think" I know ;-)) these things I have never really
looked for such sites.  I have noticed that the suggested values for PFC
caps never seem to be optimal for me.  I end up using MicroSim to predict
the best values (and confirm them by a bit for trial and error on the real
system.  MicroSim has always been right)).  The formulas assume things are
nice and linear and they really are not.  But the suggested values don't
hurt, they just are not the "best".  Others probably know more about PFC
selection than I do.  I just (kind of blindly) do what MicroSim tells me...

Cheers,

	Terry

>
>-Stan
>