RE: [TCML] JAVATC PFC Query

["Lau, Gary" <Gary.Lau@xxxxxx>]

Hi David,

I disagree with your rationale for when PFC caps are needed.  I think it has nothing to do with a coil being well designed or what kind of spark gap it has.  It's probably more likely to be needed with LTR caps than with smaller ones, but I can't say that with certainty.

However, the only _real_ reason that one would need to use PFC caps is if the line current being pulled from the wall is approaching the point of tripping the breaker.  If you're pulling 20+ Amps as I was, nuisance tripping gets old fast, and being able to reduce that figure to 12-14 Amps with no loss of performance is an easy sell.  Reducing the amount of power billed by the power company is not a concern for such a low duty-cycle hobby.

Regards, Gary Lau
MA, USA



> -----Original Message-----
> From: tesla-bounces@xxxxxxxxxx [mailto:tesla-bounces@xxxxxxxxxx] On
> Behalf Of David Dean
> Sent: Wednesday, March 18, 2009 6:49 PM
> To: Tesla Coil Mailing List
> Subject: Re: [TCML] JAVATC PFC Query
> 
> Hi Phil
> 
> Short answer 86uF across the three.
> 
> Slightly longer answer:
> 
> The PFC value given is the correct value to be resonant (parallel) with the
> transformer at line frequency with the transformers load shorted. This is fine
> for a Jacobs Ladder, but in a Tesla coil things are not that simple. In a well
> designed Tesla coil with an efficient spark gap the main load seen by the
> transformer is charging the tank capacitor. If a value of tank capacitor is
> chosen that is resonant with the transformer at line frequency, power factor
> during the charge part of the cycle will be unity. However resonant tank cap
> size is not recommended as problems can occur like very high voltages which can
> destroy the NST, tank cap, or both. For that reason LTR cap size is recommended
> for use with NSTs. JavaTC gives both the LTR values for static and for sync
> rotary spark gaps. If you have an efficient spark gap and the proper LTR tank
> cap value, you will find that the power factor of the system is about 0.9 or
> 90%. Somewhere in that range, overall, average. If you are using a LTR cap
> value the power factor during the charge part of the cycle will be leading, and
> adding power correction capacitors may actually make the overall power factor
> worse. Ultimately you may find you need to add some inductance across the input
> to the transformer in order to improve the power factor.
> 
> In most cases PFC is not needed in a well designed coil. If you do need PFC, it
> is most likely due to running a STR tank cap or a poor spark gap that is power
> arcing, etc..
> 
> In any case, it is most difficult to do power factor correction for a non linear
> system. The best you can do is to try different values and see what happens by
> trial and error. Start small and work your way up little by little.
> 
> Best advice I can give is just don't use power factor correction unless you have
> to. And if it looks like you are going to have to, try looking at your spark
> gap first to see if it can be made more efficient. Then look at adding some PFC
> if it is still necessary.
> 
> Most often PFC is used for motors where the load current is sinusoidal, or
> nearly sinusoidal. Also used frequently in magnetic ballasts for HID lighting,
> etc. where the load is mostly resistive and the ballast is mostly inductive.
> 
> If you want you can look at http://deanostoybox.com/pfc.html for some more info.
> This page is incomplete and I hope to add more info soon. So many projects, so
> little time.
> 
> later
> deano
> 
> 
> 
> 
> 
> On Wednesday 18 March 2009 04:21:35 pm Phil Tuck wrote:
> > Hello.
> >
> > JavaTc gives the PFC value that is required based on the transformer
> > information you have entered. In my case I have entered 10,000v 150 m/a and
> > it gives 86uF (240 volts (50hz). But I am not using one single transformer
> > supplying this, as the program presumes. Instead I have three NST's each of
> > 50m/a in parallel. So do I put a 86uf across the three ? or single 29uF
> > caps across each NST? Or either?
> >
> >
> >
> > Regards
> >
> > Phil
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > _______________________________________________
> > Tesla mailing list
> > Tesla@xxxxxxxxxxxxxx
> > http://www.pupman.com/mailman/listinfo/tesla
> 
> 
> _______________________________________________
> Tesla mailing list
> Tesla@xxxxxxxxxxxxxx
> http://www.pupman.com/mailman/listinfo/tesla
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla