Re: New secondary, John Freau efficiency theory

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "by way of Terry Fritz <twftesla-at-qwest-dot-net>" <FutureT-at-aol-dot-com>

In a message dated 6/27/02 2:49:01 PM Eastern Daylight Time, tesla-at-pupman-dot-com
writes:


>
> Original poster: "Mr Gregory Peters by way of Terry Fritz
> <twftesla-at-qwest-dot-net>" <s371034-at-student.uq.edu.au>
>
> All,
>
> As afore mentioned, my old 12" secondary just destroyed itself. I now
> think that this may have been partly caused by very poor tuning as the
> coil has survived several other runs with no breakdown, all other things
> being equal except tuning (I retune every time I use the coil). Anyway,
> I need to make a new ~12" (315mm) secondary. I have been reading John's
> efficiency theory, particularly regarding secondary construction. While
> it goes against everything I have been led to believe over the last 
> seven years, the ideas do seem feasible, 


Greg, all,

I've had no way to test my ideas at higher powers, but I'm happy
to report that a well known commercial coiler has reported that he's 
been using about 1300 - 1400 turns lately on his larger coils, and 
has reported improved results compared with 1000 turns.  (I'm not
sure if he wants to be identified on the list.)  A number of other
coilers have also reported good results.  From a theory/practice
point of view, there are many factors that can work for/against 
long sparks.  For those who are not familiar with my suggestions,
I'll relist them here, and give comments on each point:

1)  Use about 1300 to 1500 turns.  Some reasons for doing this
may be; to lower the frequency, to raise the primary surge 
impedance (thereby lowering gap losses, as Malcolm has
explained), and to permit the secondary to be narrower in 
diameter for a given power input.  If the secondary wire is too
thin, or too many turns are used, the losses will be too high
and any benefit will be lost.  This method can only be taken
so far.  It's a trade-off between various factors.

2)  Use a narrow secondary.  It seems that the energy that
it takes to charge up the capacitance of the secondary, is 
not available to charge up the toroid.  By keeping the secondary
narrow, it's capacitance is reduced, and more energy makes it
to the toroid.  Energy in the toroid is more effective at producing
long sparks, than energy in the secondary.  A narrow secondary
also helps because it lets more of the toroid's capacitance 
be "expressed".  If the secondary is too narrow, the inductance
and coupling will be too low and any benefit will be lost.  However
if a low of extra power is available, then it's OK to use a wide
secondary, because it will store some energy, and a larger
toroid can be tolerated, but the overall coil may be less efficient
at producing long sparks.  If a wide secondary is chosen, then it 
may be OK to have fewer turns.  Still 1300 turns or more may be
beneficial.

3)  Use a low resonant frequency.  If the secondary is narrow,
the only way to obtain a low frequency is to use thinner secondary
wire with more turns.  A low frequency may improve the spark
growth, and help the gap to quench faster.

4)  Use a high primary surge impedance, and a low break rate.
When a low break rate is used, it means a large capacitor must
usually be used (or a very high input voltage), to get a large
enough bang size.  If the capacitor is large, the primary inductance
tends to be low.  To make it higher, to increase the surge impedance,
use more primary turns, and therefore more secondary turns.  
A wider diameter will also increase the inductance, but a wide 
primary will trap too much energy in the secondary capacitance
(see (2) above).  There's no real value in making the surge
impedance very high.  It need only be high enough to keep the 
gap losses low.

5)  Use a low breakrate of around 120 bps.  I've found the best
overall spark lengths at this breakrate, for a given input power.
A low breakrate requires a large capacitor (or very high input
voltage), to obtain a suitably large bang size. 

6)  Terry has produced a theory that a high impedance secondary
(many turns), gives a better impedance match to the sparks.
Others have pointed out that the impedance of the sparks is
determined by the impedance of the coil.  Still it would seem
to me that a high impedance spark may be desireable.  I think
a narrow secondary also helps to keep the secondary impedance
high.  We know that low impedance sparks tend to be short as for
example in a tube coil.  We also know that a coil with very few turns,
such as 200 or 300 turns, usually does not perform that well.  What's
magical about 1000 turns?  Why should 1000 turns be best?
It seems likely to me that folks in the past chose 1000 turns
for a number of reasons of convenience.  The thicker wire is
easier to handle and wind, and the fewer turns that result for the
primary are easier to deal with.  Of course the biggest reason was
that folks were concerned about wire losses.  It's true that thinner
wire has higher losses, but one must consider the entire coil
design as a system... as a whole.  Too much of a focus on one
parameter will not result in an optimal design. 

In your proposed design Greg, you are obtaining your inductance
increase mostly by making the coil longer, rather than by using
thinner wire.  Thus, you won't get as much of an increase in
inductance.  Still, I think the new design may give a benefit, and
I do like the idea of tall coils.  It's possible that the primary will
have a perfectly adequate surge impedance anyway (I didn't do
any calcs).  The taller coil may demand more
power, due to the larger secondary capacitance, so it may not
really be a lot more efficient.  There's a trade-off there of more 
secondary capacitance vs. more secondary inductance, which
makes it hard to analyze.  In any case, I'm sure it will be at
least as good as before.  Hopefully you have enough extra
primary turns to tune the coil.  If I were designing such a coil,
I might make the secondary narrower, such as 10" wide, but
keep the tall height.  It's possible that no benefit would be
seen though.  I forget what your power input and spark length
are.  I like to either start with the power supply and design from
there, or start with the desired spark length and work backwards
to the power supply.  Each parameter tells me what the next
parameter should be in a sense.  I can't give a complete answer
or recommendation without knowing the input power and desired
spark length.  Even then, my answer would lack the practical
experience of those who have built larger coils.  My answer would
be based on how I would build such a coil.  Actually, if I had
the space, I would try a variety of different large coil designs,
and compare them.

To try the theory exactly as I advocate it, you'd have to keep
the secondary the same total height as before, and use thinner
wire to fit the 1300 or 1400 turns.  I don't know which way will
turn out best for sure.  Overall, not much research has been 
done on these issues.  I think a lot more work needs to be
done before we can be really sure about it.  In the past, some 
folks who were the most "sure" about things, also turned out 
to be the most wrong.  I'd better end here, or this post is in
danger of going on forever   :)

Cheers,
John


>  and seing as though my last

>
> secondary just died, it has given me motivation to try something
> different. My old secondary was 315mm wide, with a 950mm long winding of
> 0.95mm wire for 1000 close wound turns. For the new secondary I propose:
>
> Diameter: 315mm
> Winding length: 1400mmm (4.4:1 aspect ratio)
> Wire diameter: 1mm (1400 turns)
>
> I thought I would run this by the list to see what you all think. I am
> open to suggestions. I would particularly like John's input on this matter.
>
> Cheers,
>
> Greg Peters