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

Re: 20 gauge wire



Hi Adam, Leslie, all,

> Original Poster: Adam <adamsmith-at-mediaone-dot-net>
> All the coil data I've collected is leading me in the opposite
>direction from what Nick has suggested above. I wound
>my 8" secondary with 20 AWG, but I'm now making a new
>22 AWG secondary in hopes of improving performance. My
> database of other coilers' results shows that those getting
>the best performance in my power class have high inductance
>secondaries wound with #22-#25.


Hmm, Iīm not sure I can agree with such a simple statement
(no offense meant). Just for the fun of it, I did a little calculating
below:

Hypothetical coil dimensions:
-------------------------------------
8" diameter
36" winding length
Cself: 15.5 pF

AWG 20: // 1040 turns // 43.71 mH // Bare FRes: 192.9 kHz
AWG 21: // 1198 turns // 57.98 mH // Bare FRes: 167.5 kHz
AWG 22: // 1299 turns // 68.17 mH // Bare FRes: 154.46 kHz
AWG 23: // 1515 turns // 92.73 mH // Bare FRes: 132.44 kHz
AWG 24: // 1607 turns // 104.34 mH // Bare FRes: 124.85 kHz
AWG 25: // 1886 turns // 143.71 mH // Bare FRes: 106.38 kHz

The above numbers make something quite clear: Going from
AWG 20 to AWG 25 will more than triple your inductance, but
it will NEVER triple, not even double, your output spark length
for a given input power.

Just from looking at Lsec, you will get NO idea, what kind of
performance you will receive (for equal input power). You just
canīt foresee a coilīs performance this way. If this was possible,
everyone would be building with "fixed" dimensions. One reason
against using a super thin wire is the dramatic increase in the
DC resistance of the coil. My 8" coil is wound with "AWG" 19.5
wire. Lsec is around 36mH and I get a 63" discharge using only
1275VA of input power. On the other hand, one of my testbed
coils was a 2.5" x 20" cardboard secondary, very sloppily wound
(as it was for test use only). With around 800VA of input power,
I was able to achieve some 60 cm (~24") discharges. I find this
pretty amazing, because this coil was wound with AWG 26
(1111 turns). Cself was only 9.48pF and Lsec was a low 9.1mH.
I WAS using a large topload (for the former size), it being 4.1" in
minor diameter and roughly 25" in major diameter (made from Al
ducting). 800VA was pretty much the limit for this coil. I DID
increase the power, but I also killed the secondary in the process.
Oh well, that IS what testbeds are for ;o)).

It should be noted that the results above where achieved by using
a low voltage (I have no other choice) xformer (7.5kV). However,
I did design the coils in such a fashion, that I get high primary
currents. My 8" coil (at this stage) is running on some 600+A of
peak primary current. I also use a KISS type flat static gap w/o
ANY additional cooling except for convection cooling to the
surrounding air (meaning NO blowers, vacs or fans).

Real output performance increases are to be found in the top
load. Every time you increase the topload, you WILL increase
the output spark length of your coil. There is of course a limit,
where your setup wonīt be able to charge up that huge toroid
any higher and a further increase in topload size will lead to
less output, but other than that, the topload size is the real
give-away to grand sparks. Those few pF at the top donīt
sound like much (and they are not, from the pure viewpoint
of C size), but charging those 50+ pF to 500kV will get you
6.25 J, which is enough to knock you clean through a brick
wall ;o). Spark length IS NOT related directly (in any way)
to the output voltage of your coil. It also should be noted, that
as you increase the topload, fine tuning gets to be quite a bit
more tricky. Moving my primary tap more than a 1/4 of a turn,
lets the coilīs output drop like a skydiver w/o a parachute.
Moving it only an 1/8 of a turn will drop the output by some
~25-30% and I am NOT using the biggest topload possible (I
think I should be able to squeeze around 65-70" from this setup
using a *properly sized* toroid). The reasoning behind this is
actually quite simple. As long as your topload is sufficiently
small, it doesnīt make a real difference if your coil is 100% in
tune or not, because even though you are already losing energy
to the sparks, your setup can still feed them. In other words, the
coil isnīt loaded to the max, yet. (Hmm, not quite sure if I made
this point clear enough). Your coilīs output voltage drops like a
rock, once the discharge has been initiated. To keep up the spark,
(or make it reach out longer) you need a coil with the ability to feed
the plasma channel (stored energy in Ctop). So in the too small
toroidīs case, it doesnīt make a difference if your coil is putting out
200kV or 300kV. However, as soon as you get close to the max
topload your coil can handle (also meaning bigger Ctop), you need
to tune it the best you can, because then it WILL make a really big
difference if your coil charges that topload to 200kV or to 300kV.
As a matter of fact, the closer you get to the maximum toroid size,
the more aware you must be of "arc loading" the coil. In other
words, you must tune your setup to the length of the output spark
during a run. This is also the reason why you CANNOT tune your coil
to the sweet spot using a signal generator and an o-scope. It will get
you near the tuning point and it IS great fun and a good learning
experience, but it will never tune your coil 100% to the sweet spot.

Bottom line: Donīt expect a great performance increase with your
new coil former being wound with AWG 22 instead of AWG 20. If
you see more than a 10% increase in output, it would *really*
surprise me.

Coiler greets from Germany,
Reinhard