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

RE: Tesla coil Parts and plans

Original poster: "Lau, Gary" <gary.lau@xxxxxx>

Hi Paul:

Welcome to the List.  These are all good questions, and while they have
been discussed in the past, there are no doubt others who would benefit
from this, so I'm happy to reply on-line.  My comments interspersed:

> Original poster: "Paul B. Brodie" <pbbrodie@xxxxxxxxxxxxx>
> I have been scrounging parts to build a SGTC and so far I have: 15 kv
30 mA
> NST, 3 lbs. of 26 AWG and 9 lbs. of 25 AWG magnet wire, a 8 ft x 4
> piece of thin wall white PVC (not foam core), a 0 - 140 v 20 A variac,
> length of 3 inch diameter flexible dryer duct formed into a toroid
with 2
> pie plates forming the center, 50 ft. of .25 inch soft copper tubing,
> .003 mF 35 kv film caps, two .03 mF 35 kv pulse discharge caps, 4 680
pF 20
> kv door knob caps, 2 .01 mF 40 kv poly caps, eight  5 kv 2 A diodes,
an 8
> ft copper clad ground rod, and a 4 series gap made with 1.5 inch hard
> copper pipe inside a 5 inch piece of PVC with a cap that allows me to
> connect the shop vac to the top and suck fresh air through the spark

Your .03/35kV caps sound like the Maxwell units sold on eBay, and they
should be OK.  I couldn't comment on your .01/40kV or .003/35kV units
without knowing a good deal more about them.  The doorknob caps are good
for an R-C filter between the main gap and NST.  After things are tuned
well, you may find that you can use something thicker than a 3x??

> I also have Wintesla, Walt Noon's designer software, and JavaTC 9.1.
> all give very similar results, which is reassuring. I still have 1001
> questions. My knowledge of electronics is virtually all self taught
and I
> worry that I am guilty of some misconceptions. Also, I am extremely
> conservative and I want to build in as many protective features, both
> safety and protection of the coil itself.
> Before describing my plans for the controls, tank circuit, and
primary, I'd
> like to get the secondary out of the way. I have cut off 21 inch
length of
> 4 inch diameter PVC and sanded off all imperfections and ink. I then
> it with multiple coats of polyurethane, inside and out. The following
> have been a mistake. I covered all the area to be wound with wire with
> double sided carpet tape to help hold the winds as I apply them. I
want to
> use the 26 AWG magnet wire and wind 18 inches. I have calculated that
I can
> get approximately 1000 winds. I also plan to seal both the top and
> of the secondary and provide a small hole in the bottom to equalize
> pressure. I plan to leave 1 inch of PVC uncovered by wire at the
bottom and
> 2 inches at the top. How much space between the top of the secondary
> and my toroid?

The use of carpet tape is a new one and if it's clear, odds are that
it's not a problem, but I still wouldn't bet the farm on it.  I
personally would have put a few more turns, out to at least 21", but
nothing is cast in stone here.  The lowermost extent of the toroid
should generally be no more than 1-2 inches above the top turn.

> Starting at the wall plug, I want to put an EMI filter between the
> outlet and the variac. My first stupid question. My variac says that
> side is common to line and load. Does this mean that the variac is NOT
> acting as an isolation transformer?

Absolutely - it is not isolated.  You should look at the wiring; the
neutral is common to input and output.

> Do I ground the variac case to my house ground?

Yes, and that's the only thing, with the possible exception of an EMI
filter case, that goes go to house ground.

> From the variac, I plan to go to the NST. Where do I ground the
> NST, house ground or to the RF ground?

RF ground, although some may differ.

> From the NST, I want to wire to a
> safety gap with the center grounded to my RF ground. Should I provide
> bypass cap between the NST and safety gap? Which caps are suitable for
> this? Do I want to put in RF chokes between the safety gap and main
> capacitor?

Use an R-C filter.  Chokes used to be thought to be effective, but more
recent research shows that they do more harm than good.  See my web site
at http://www.laushaus.com/tesla/protection.htm for the filter circuit.

> Any recommendations on which of the caps to use that I already
> have? The software recommends approximately .006 mF cap for the
> The Java software recommends using a larger cap than is recommended by
> program. Here is where I think I am a little fuzzy.

The software is computing the mains-resonant value - the value that
would resonate at 60 Hz with your NST's secondary.  This was at one time
thought to be the value that would extract the most power from the NST.
This value will work, but it turns out that there's nothing magical or
particularly sweet about it.  I've found that a value of 2 times the
mains resonant value (when using static gaps) worked best for me.  I
would try using your two .03uF caps in series for .015uF, or your .01uF
cap, if it is in fact TC-worthy.

> How is the value of the
> primary cap determined? What is the impact of using a larger than
> cap? Does the size of the cap, the distance between gaps in the spark
> the inductance of the primary coil and the size of the toroid all
> this calculation? Won't any of these factors change the resonant

The cap size is generally chosen by sizing it according to the power
supply - by choosing some multiple (1.5 to 2.0) of the mains-resonant
value.  The secondary and top load sizes are also chosen based on the
power supply size, although there is no simple or agreed upon formula
here.  Mostly just experience and what "looks right".  Then the primary
inductance is chosen to bring the primary and loaded secondary
frequencies to the same value.  The main gap width is strictly a
function of the power supply voltage, and has no bearing on frequency of

> Next, I want to wire to the series spark gap. How do I determine the
> of each individual gap? At what total gap should I start at? Will too
> of a spark gap cause excessive load on the NST? I had trouble getting
> NST and I paid too much for it. I want to do everything I can, within
> reason, to protect it!!!

The ONLY way to set a static gap is to wire it across the NST at maximum
variac, and set it to it just barely fire.  The caps must be out of the
circuit for this adjustment.  For a ballpark figure you can use 0.25"
total gap width for a 15kV power supply, but the exact figure will be a
function of the gap geometry, airflow, air pressure, electrode
temperature, texture, number of gap segments...   Many variables.  Too
wide of a gap will permit excessive voltage on the caps and NST - FAR
higher than the NST secondary rating - and this is why so many new
builders fry their NST's.  Performance will always be much better with a
wider gap, and it will be very short lived.

> What type of wire should I be using to wire from the NST to the bypass
> to the safety gap to the cap to the series SG and then to the primary
> I was planning to form the primary with a 6 inch diameter. One inch
> the inner winding and the base of the secondary coil. Is this
> spacing? I plan to space .25 inch between windings. I calculate that I
> get between 13 and 15 windings. To attach to the primary, I plan to
> my wire to an ATC fuse holder and then snap the fuse holder onto the
> tubing wherever I like.

The wiring from the NST to the safety gap to the main gap carries only a
small current, so any size will do.  The loop consisting of the primary
coil, main gap, and tank cap should be short, direct, and heavy.  For
use with a 15/30 NST, probably anything 14gauge or heavier will do.

> I am going to great lengths to build a substantial RF ground
> separate from my electrical service and water pipes. The ground here
> very good for grounding, red clay with a high iron content. I plan to
> at least 5 eight foot ground rods with 4 of them forming a square 5
feet to
> the side and cabled to each other and then cabled each to a central
> which will be my attachment point for RF ground. This attachment point
> be approximately 10 feet from the TC location in the front of the
garage. I
> plan to use two 4 AWG car jumper cables, 4 strands total to attach the
> to ground.

Your ground will be far better than most ;-)

> I plan to attach the NST case, the safety spark gap, and the base of
> secondary all to RF ground. Can someone please explain to me
everything I
> need to know and am afraid to ask about a counterpoise ground?

With a ground that good I don't think you need a counterpoise.  I've
never used one.

> Any and all comments will be most gratefully accepted. If I seem to be
> particularly off base in any area, please point it out. I am striving
> learn as much as I possibly can while doing this, instead of just
> someone's recipe. Oh yes, I almost forgot. I was contemplating using
the HV
> diodes I have to rectify the NST output to pulsed DC. Would this help
> ease the burden on the primary cap?

I would skip the diodes.  The stress on the caps comes not from the 60
Hz stuff, but from the tank frequency oscillations.

> If anyone would like to communicate with me offline about this, please
> My email is pbbrodie@xxxxxxxxxxxxx
> Think positive!

Regards, Gary Lau