[Prev][Next][Index][Thread]
Re: Arc length and Gap Dwell Times
Tesla List wrote:
<BIGGG Ssip>
Jack Couture said (in response to Dan Gowin):
> > I ran your new info thru the JHCTES computer program. For 120 volts, 14 amps
> > = 1680 watts input with a 19.0 mh secondary the program shows 640 watts per
> > foot of spark and an overall efficiency of 31.2%. Your coil appears to be
> > similar to a typical coil with this new wattage.
> >
> > You said you did not believe your coil was 50% efficient and that the laws
> > of physics were broken. How do you calculate the efficiency of your Tesla
> > coils? Also, how did you determine the laws of physics were broken?
> >
> > What are you using for a primary capacitor that gets hot during operation?
> > If you are careful building a homemade capacitor it might work better than
> > the one you are using now. Richard Hull can give you plenty of good
> > instructions for building your own capacitors.
> >
> > Jack Couture
>
To which Richard Hull responded:
> I'll have to go with jack on this one and even go one better. I would
> kill to obtain 50% efficiency in a Tesla coil just once in my life before
> I die!!!!
>
> If one looks at the losses on a casual basis:
>
> Transformation ~2-4%
> Current limiting ~10%
> Tank circuit and switching losses ~10-25%
> Magnetic primary/secondary coupling factor K=.15 (quiet heavy) 85%
> secondary losses ~5%
>
> These losses cannot be just summed, either. They are taken off the
> preceeding losses remaining energy.
>
> Example: 1000 watts input energy
> watts remaining:
> best case worst case
> Transformation 980 watts 960 watts
> ballasting 882 864
> Tank/switching 794 648
> coupling 119 97
> secondary 113 92
>
> Whittles it down doesn't it? Now these figures are based on good
> esitmates of averaged energy from power plug to flaming spark output. We
> see that 10% efficiency of conversion is incredibly good. 25% would be a
> Godsend. Magnifiers can triple this output figure, but only in the
> higher power ranges. Finally part of that output energy is RF radiation
> and not spark!!!! A well designed coil will have very little radiative
> losses however, as its output is designed for hopefully high local air
> loses from spark.
>
> The above losses are probably not all the loses attendant, but probably
> covers many of the important ones.
>
> Richard Hull, TCBOR
AND, in another dialog, Steve Roys said in response to Richard Hull:
> >In an over-coupled, and, or, under-quenched coil. All the energy from
> >the capacitor which can be magnetically coupled is in the secondary after
> >the end of the first 1/2 oscillation in the primary. Immediately
> >following this, the secondary is pumping some of its valuable, just
> >received, energy back to the primary, if the gap is still conducting.
> >
> >Precision quenching is what I'm looking for in the H2 thyratron
> >experiments.
>To Richard Hull (primarily) -
>I went "straight to the horse's mouth" ("Vacuum Tube Tesla Coils" by the
>Corums) and now I'm confused. In appendix IV, they discuss coupling,
>spark duration, and output voltage, and they come to the conclusion that
>"The primary energy will be transferred to the secondary in one half of a
>BEAT period....In spark transmitters, it is desireable to quench the
>primary spark when the energy has been transferred to the secondary."
>Their graph of secondary voltage vs. spark duration (fig 8, appendix
>IV) also points to a theoretical ts = 1/(2 * deltaF) primary spark
>duration for "optimum spark".
>This relationship is what I've been trying to get across, but this
>doesn't seem to be what you're saying. Are you saying that the gap should
>be quenched after only the first half-cycle of primary oscillations at
>the natural free-resonant frequency? If so, then I'm confused and I have
>more questions about the applicability of the Corum's findings in appendix
>IV. If you are talking about quenching the gap after one-half cycle of
>the BEAT frequency, then we've been saying the same thing.
>Steve Roys.
Interesting dialog!!
The losses are probably not quite that bad. The coupling coefficient is
not really a loss, but instead a measure of the portion of magnetic flux
lines that are common to both the primary and secondary. This is also
tied into the question and dialog that Steve had regarding quench times.
Primary flux lines that are not common to the secondary simply don't
couple to the secondary - When these collapse, the energy they had is
still returned to re-charge the primary cap.
Richard is correct in part: All the energy stored in the primary
capacitor IS transferred during the first quarter-cycle of the tank's
coupled frequency. The initital electrostatic energy ("bang"),
previously stored in the cap, has been transformed to magnetic field
energy stored in the primary, along with a portion that has been
transferred to the secondary (less any losses). During the next quarter
cycle, the magnetic field collapses, reverse charging the primary cap,
but also transferring some more energy to the secondary. While the gap
continues to fire, this process continues, building energy in the
secondary, and removing it from the primary (less losses). In fact, if
you could ignore gap, resistive, eddy current, and secondary discharge
losses, all of the energy initially stored in the primary cap would
eventually be coupled into the secondary circuit in 1/2 of the beat
period that Steve talks about.
The coupling coefficient merely governs how long this process will take,
not the portion of energy lost in the transfer process. The longer this
takes, the more other losses can reduce the energy left to transfer to
the secondary/toroid. The overall transfer efficiencies should be
significantly higher with higher coupling coefficients. The above losses
provided by Richard are reasonable IF you remove the coupling
coefficient as a loss.
As usual, flames, brickbats, and rocks are welcomed! :^)
-- Bert --