[Prev][Next][Index][Thread]
Re: Tesla Coil Stuff
[The following text is in the "ISO-8859-1" character set]
[Your display is set for the "US-ASCII" character set]
[Some characters may be displayed incorrectly]
There are several discussions going on here at the Tesla List that
ultimately affect all of us Coilers at one time or another. I have written
up the following paper to try and explain such things as:
* Sparks appear on the SIDE of my coil
* My Coil works for ten seconds and then goes wimpy
* I get Sparks at top winding instead of at Ball on Top
* But I just had this sucker TUNED yesterday $^%#-at-*&%!!!
* Mine doesn't work AT ALL!
Sometimes it helps to understand the underlying CAUSES for all these
things. And that means learning a little bit of Theory. But don't worry. I
won't use any math or equations at all. I just want to share with all of
you something about What Is Really Going On with these Coils With An
Attitude Problem.
When you have a spark coming off of the SIDE of your coil, it is not tuned
properly. By SIDE, I mean anywhere up the column EXCEPT at the TOP of the
winding. When you get sparks any where ELSE along your column, it indicates
that the greatest electrical pressure (voltage) occurs at THAT particular
spot. And it is happening THERE because that is where the Here And Now
State of your system has set up the Primary and Secondary Resonances. You
WANT the Resonaces to be THE SAME. In addition, you WANT to get the
1/4 wave resonance at the Secondary to coincide with the top of the column.
Literally EVERYTHING in the Tesla coil system can affect this resonance.
The transformer voltage affects the resonance because it partially
determines how fast the main capacitor(s) charge, and how High they charge.
The transformer is affected by the power it receives from the power mains.
If the power company reduces the mains voltage by 10 volts, you will
definitely notice it. Changing your LINE CORD can affect the power. The
thicker it is, the less power it wastes in the form of heat. Believe me,
even the PLUG can affect resonance. A lousy AC switch can build up arcing
damage over time and add enough extra resistance to the circuit to rob you
of several percent of your total power. Even if you are using a good
quality Contactor as your power switch, it can stand an occasional contact
cleaning.
If the transformer itself gets hot, that slightly changes the output
impedance (resistance), but very slightly. Are you using power resistors in
your filter circuit? As they warm up they can change by about 20%. Each.
Filter coils don't normally change value while they are running, but the
act of adding or removing such coils from your Tesla circuit definitely
affects the tuning. Caps added for Filtering affect the Tesla circuit only
slightly as long as they are kept small in value.
By the way, ALL electrical connections in the Tesla circuit affect the
Resonance and the Q of the circuit. The WEAKEST link in the system will
generally affect the Entire machine. Use wire that does not have an
adequate surface area, and NO MATTER HOW THICK IT IS, it will not
adequately conduct the high frequency currents. Any connection that is not
soldered can be a problem. Remember that at high frequencies, currents flow
over the Surface. A non-soldered connection will be seen as a low
resistance connection ONLY if there is adequate surface area at the
connection point. That is why a strap is so much better than something like
a (yuck!) alligator clip. ALL WIRING IN THE HIGH FREQUENCY SECTION NEEDS TO
HAVE THE LOWEST **RF _AND_ DC** RESISTANCE POSSIBLE.
The Capacitor affects the circuit in several ways. It is the Energy Storage
Device. The MAXIMUM energy that you can couple into your system will always
be LESS than the energy stored in the capacitor. That's because the
Capaitor itself has a few energy wasting mechanisms. Any corona in a
capacitor is WASTED Energy. And in a High Voltage capacitor it can be
fairly significant! The charge and discharge of the capacitor causes
heating of the dielectric (the insulating plastic/glass/mica/whatever)as
the molecules change position due to the electrostatic forces. This varies
with the material, and is also a function of the frequency. If the
dielectric is really "lossy", then the heat developed can cause it to
become soft or even molten. Zap! One hole in the insulation and your
capacitor is trashed.
The thicker the insulation in a capacitor, the LONGER it takes to dump the
energy. But you need enough thickness to make sure the insulation does not
Break Down. A capacitor made of a series of lower voltage capacitors
connected in series to increase the overall voltage rating will usually
have a much higher Q factor than a single High Voltage Capacitor of the
Same Rating. That is because lower voltage ratings mean a reduction in the
thickness of the dielectric (insulation between plates). Thus Two .1 mfd
10KV capacitors in series is generally better than a single .05 mfd 20KV
capacitor. They both end up with the same RATING as to voltage and
capacitance, but their ability to DUMP Energy RAPIDLY is quite different.
If you are making a homemade plastic capacitor and have a means to measure
capacitance, try this experiment: After you have rolled the capacitor up,
and before you stick into its container, measure its capacitance. Now,
while you are measuring its capacitance, push down on the center of the
capacitor with both hands about six inches apart. SURPRISE!!! Did you see
how MUCH the capacitance changed? It Increased Greatly when you pressed
down, because you decreased the distance between the two plates (There is a
LOT of air space in there!) When the Capacitors Charge and (ESPECIALLY)
when they Discharge, the Electrostatic charges present will make the
capacitor plates vibrate back and forth. Changing their Capacitance.
Upsetting your Fine Tuning! From Moment To Moment, no less! And you thought
your coil was perfect! 'Taint No Such Thing. All we can ever do is try to
minimize the effects Once We Are Aware Of Them. Do you see now, why you
sometimes have to do a lot of re-tuning when you MOVE the Capacitors in
your system?
The Spark Gap GREATLY affects the tuning. If you have a single spark GAP
that has a large separation, then you get Fewer Pulses, but BIGGER pulses.
These Big Pulses will ONLY give you spectacular effects if the Tesla Coil
Secondary is resonating at the SAME frequency that the Tank Coil is RINGING
at. That is why it is HARDEST to tune the coil for Single Pulse Operation.
(But, boy is it Worth it when you see that Multi-Meter Monster Crash into
Existence)
If you bring the Gaps closer together, they break down sooner. This reduces
the energy in the circuit, but makes for MORE SPARKS across the gap in a
given period of time. This tends to dump less energy, but it also produces
this energy over a BROADER SPECTRUM. That is why many Tesla coils will
"work, but not too good" with this arrangement of the gaps close together.
It just so happens that SOME of the energy is at the right frequency. Note
that it is ALWAYS the case here that MOST of the energy is NOT at the Right
Frequency, even though Some of it IS! That is ONE reason why NO Tesla Coil
works Really Well with the Spark Gaps TOO Close.
The QUENCHING at the Spark Gap influences the Q of the circuit, because the
Hot Plasma tends to want to continue conducting AFTER the initiating spark
jumps the gap. What happens then is complex, but a simple explanation would
be that energy that SHOULD have been transferred from the primary circuit
to the secondary circuit ends up looping back around through the primary
POWER SUPPLY LOOP. This needlessly wastes transformer/capacitor energy, and
so the NEXT spark is delayed, and all you have to show for your energy loss
is an awfully hot spark gap ELECTRODE.
And the heat from the ELECTRODE warms the air, promoting premature
discharge the NEXT time around, and the vicious cycle continues. (That is
why your Tesla Coil can run Beautifully for ten seconds and then slump into
a stupor) Whenever your spark gap FLAMES in ANY WAY, that is Bad News for
the overall system. So how can you fix THIS up? One of the Best Solutions
is the time-honored Rotary Spark Gap.
The Rotary Spark Gap has a point in its operation where the rotating
electrode set is far apart. This gives the capacitor a chance to charge up
to a reasonable value. As the rotating electrode approaches, it reaches a
point where a spark will strike. The neat thing is that because the
electrodes were initially Pretty Darn Far Apart, the spark that you get is
Really Long, and that means Mucho Energy. You also get less loss through
brush discharge, because the electrodes are close together only part of the
time. The Ideal Rotary Gap would have the electrodes first APPROACH, SPARK
(Just Once, if we could have our Wish Come True...), and then SEPARATE. The
Separation QUENCHES the arc and prevents After-Flame. This reduces Heat,
since the Heat was originally caused by the After-Flames. And, to cool
things off even further, the rotating electrode only has a portion of
itself arcing at any moment, and the whirling around in the air cools it
off a bit as well. Some Rotary Gaps have FINS to increase the effect even
further. You STILL have to watch the heating of the Stationary part of the
Rotary Spark Gap, a point some designs don't address.
Yeah, but what if you don't have the inclination, drive, Bucks, or Parts to
build a Rotary Gap?
Build a Non-Rotary Gap that addresses the same problems, but in a different
way.
Two Basic Solutions here: You can build a SINGLE Gap Spark Gap or a
Multiple Gap Spark Gap.
The Single Gap Spark Gap is what many of us started off with. It works, but
only after a fashion. It has ALL kinds of problems. Brush discharges rob it
of energy BEFORE the First Strike. When it strikes, it doesn't want to
quench at all, hardly, so you have to resort to things like blasting it
with a high speed stream of compressed air. The only Single Gap Spark Gap I
ever read about that looked like it had a chance of doing anything good was
a design by Tesla in which two MASSIVE blocks of metal were ground flat and
then inscribed with deep circular channels. The two blocks would then be
set facing one another at a very small distance. Tesla designed this
monster to handle Tesla Coils that had input voltages of just a few KV, but
very high POWER ratings, like 100KW. As the Sparks heated up the air
between the gaps, it set up a movement of air in the channels that blew
out the After Arcs. That and the fact that the MASSIVE metal SUCKED UP all
the available heat and therby extinguished the flame. The book I read (it
was published about 1900), didn't give any further details, but the picture
looked like Tesla had added an oil cooling system to the Massive Electrodes
to KEEP them COOL for extended periods of time. The only other thing I
remember about this Spark Gap was that Tesla like the way it "Sang".
For many Coilers the answer to the Spark Gap Problem is a Series Spark Gap
with Forced Air Cooling. The Richard Quick Spark Gap is a good example. The
idea here is to break the spark up into a Series of Small Spark Gaps, and
to keep the gaps as Cool as Possible. Using Hollow Copper Tubing of a fair
sized diameter allows for a reasonable surface area for the sparks. This
enhances the ability of the Gap to handle the large RF and PULSE currents
without excessive damage. Curved outer surfaces cut down on corona
discharge, and the large internal and external surface area of the tubing
allows for maximum cooling while offering minimal resistance to air flow.
Because the assembly is in a tubular structure, air flow can be confined
and directed quite readily. If you don't mind, I would like to make a
comment here about some of the implementations of this type of Spark Gap.
Many people are not providing adequate AIRFLOW to these devices. Use a Fan
at BOTH ends, with ONE End set up to blow air IN, and the OTHER to Suck Air
OUT. This arrangement is MUCH better than a Single Fan, and EASY to
implement. Use fans that provide LARGE AIRFLOW. I have seen pictures of
some units where the fans were obviously way too small. If you can get your
hands on a High Speed BLOWER, that is even better. Cooling is a function of
Total Air Flow Per Unit of Time, Air Temperature, Electrode Surface Area,
and the Heat Transfer Characteristics of the Electrode. In short, INCREASE
the Volume of Airflow. Cooler Air Is Better (But we'll just use what's
available in the room anyhow). Make the Electrodes LARGE, TUBULAR, With the
Walls AS THICK AS POSSIBLE, and make sure there is ROOM around them on ALL
SIDES so that the MAXIMUM Surface Area is Exposed to the SPARKS and the
AIRFLOW. Use a material that transfers heat well. For example, Copper is
MUCH better than Aluminum. Understand all the factors that affect the Spark
Gap, and then see what you can do to Make Existing Designs Even Better.
When we talk about the Primary Coil, we often forget that EVERYTHING IN THE
LOOP THAT FORMS THE OSCILLATORY SECTION IS PART OF THE PRIMARY. Bend a
wire, and that is a fractionary turn added to the impedance of the primary
circuit. KEEP WIRES AS SHORT AND AS STRAIGHT AS POSSIBLE. Everything in the
Oscillatory Section (Capacitor, Spark Gap, Wiring, Primary)carries RF.
Remember that the Primary Coil is FED by the other parts! THEY need the
SAME Great Care that we lavish upon the Primary! The Energy from the
Capacitor is DELIVERED to the Primary THROUGH a Series Circuit that
includes the Spark Gap and ALL the WIRING that connects them. Since this is
all RF being conducted, we should apply what we know about RF circuits.
Keep conductors Short, Straight as Possible, and with a Large SURFACE Area.
I always shake my head a little when I see a Wonderful Spiral Primary
Lovingly and Painstakingly formed out of 3/8 inch copper tubing, and then
notice that parts of the Oscillatory Section are wired using #12 Copper
wire. It Works, YES, but the performance is ultimately LIMITED and Less
Than It COULD Be. Try this experiment: locate the longest wire you have in
your oscillatory section. connect one end of this wire to a #12 copper wire
of slightly greater length. drive a nail into the end of a broomstick and
attach the other end of the wire you just attached to this nail. Power up
your coil (Low power mode is fine). While it is running, use the broomstick
to touch the free end of the wire/nail to the point where the original wire
goes. If you can see a spark, then you can probably improve on your wiring.
The Primary will create the Magnetic Fields that will Coerce the Secondary
into Oscillation. We want to create A STRONG Magnetic Field that RAPIDLY
Builds and Decays. We want the Magnetic Field to Couple as much of the
Secondary as possible. The Q of a Tesla Primary Coil is a measure of the
effectiveness of the coil in producing its magnetic effects. Since DC
resistance limits current and thus the magnetic field, Q is often expressed
as the Ratio of the AC Resistance at a given frequency to its DC
Resistance. So, we want to keep the DC Resistance LOW, and the AC
Resistance High. We keep the DC Resistance Low by using the best conductor
we can get. Gold would be COOL, but awfully Expen$ive, so we settle for
COPPER. (Some people Silver Plate their Copper. This Works too, since Tesla
currents exhibit the Skin Effect). When we talk about the AC Resistance, it
is important to remember that we are talking about the resistance due to
Magnetic Effects ONLY. We want to INCREASE the Resistance due to Magnetic
Effects, but DECREASE any Resistance due to the Skin Effect (which is NOT a
Magnetic Effect). For those versed in electrical terminolgy, we want to
Increase the Reactance of the Circuit, and Decrease its Resistance.
When you form a wire into a coil, it increases its Inductance. The more
turns, the more Inductance. The Closer the Turns are to One Another, the
greater the Inductance. The Magnetic Field around a Coil is a function of
the shape of the coil and the current flowing through it. We have already
discussed how to maximize the current flow. Now we want to Optimize the
Shape of the Magnetic Field produced so that it encompasses As Much of the
Secondary As Possible.
Ideally the Primary's Magnetic Field should encompass the Entire Secondary.
There are two ways to do this. The way that immediately comes to mind is to
make the Primary WINDINGS completely encompass the Secondary. This normally
doesn't work too well, because the Secondary has this Really Nasty Tendency
to hurl Lightning Bolts at the Primary. This normally burns insulation off
of the Secondary, resulting in an unneccesary trip to the Garbage Can or
Local Dump, depending on the size of your Secondary. We DON'T REALLY WANT A
UNIFORM MAGNETIC FIELD ANYWAY!
Keep THIS in mind: the Secondary needs to have a VERY LARGE ELECTRICAL
CURRENT at the BOTTOM of the Coil, and a VERY LARGE ELECTRICAL PRESSURE
(Voltage) at the TOP of the Coil. Since the CURRENT is a function of the
Magnetic Field Strength, it becomes Evident that the Magnetic Field we need
to produce needs to be Strongest at the Bottom of the Secondary. A FLAT
SPIRAL seems to Fit The Bill. Banking it up at an angle (The Saucer
Primary) is a Variation that is useful for Smaller Coils where the Primary
Q is Lower due to the smaller surface area employed. On Larger Coils the
Saucer Primary probably creates more problems than it solves. It INVITES
Strikes from the CENTER of the Secondary when the System is Out Of Tune.
Because it is raised up, and the OUTER part of the Saucer Primary is Most
Likely BEYOND the Shielded Area provided by a Toroid, it then needs a GUARD
ELECTRODE. And if that isn't an invitation for Corona and Close Strikes,
then What IS?
Reputable Coilers have reported that Really Good Results come when they
make the Diameter of the Primary About Equal to the Height of the Secondary
Winding. This makes sense, and you can actually see the effect quite
clearly in certain photos of these machines. When a photo is taken with the
lights off and the Spark Gaps shielded, you can clearly see a corona field
between the Primary and Secondary that looks like a nice inverted parabola.
Where the Voltage Difference is GREATEST is Between the Top WINDING of the
Secondary and the OUTER Winding of the Primary. When such a Coil is run in
Pulse Mode they can easily throw Discharges Longer than the Length of the
Secondary.
Let's assume that you want to build a 6 foot High Secondary. That would
mean a 6 foot Diameter Primary. Let's say that your Monster Secondary Coil
is 1 foot in Diameter. We want to leave room between the Bottom of the
Secondary and the Inside of the Primary, so that we can use a Separate RF
Ground. (And let's assume we are using FLOATING Center-Tapped Neon
Transformers to boot). Let's allow 3 inches between the Outside of the
Secondary Column and the Inside of the Primary Coil. That makes the
Beginning, Inside Diameter of the Primary Spiral 18 inches, and its
Outside, Final Diameter 72 inches. Let's say that we want 15 Turns in this
Spiral. First, take the difference between the Inside and Outside Diameters
of the Primary Spiral in inches. 72-18=54 (OK, so I lied... there IS some
math in this treatise. But SIMPLE MATH!). We then have to divide this
answer by 2 to get the number of inches difference on Just One Side, Not
Both Altogether. 54/2=27. OK, so we have 27 inches in which to create
15-1=14 Equally Spaced Divisions. (Think About It... when you divide
something into Two Pieces you have Three Lines. The First, the Middle, and
the Last. Lines-1=# of divisions). 27/14=1.92 (mental pause). OK, so we
stare at that for a moment and say, "What the Heck, Let's call it 2 inches
and Get On With Life!" This is reasonable, because you WILL be Tuning this
thing anyhow, so MINOR variations here have literally NO EFFECT.
14 Equally Spaced Divisions at 2 inches per division gives us 28 inches
instead of our original 27 inches. If I Use 3/8 inch diameter copper tubing
I can space each turn by 1 and 5/8 inch. 14 Divisions times 1 5/8 is 22.75
inches of total spacing. You are definitely NOT going to get arcing from
one winding to the next on THIS baby! Each adjacent winding has a breakdown
of well over 10KV WORST CASE. This means that if you use 10 taps, it can
handle a total of 100KV across the Primary. Allowing for all kinds of
inductive shenanigans, we are still Quite Safe with dumping the output of a
Capacitor charged up to about 30KV into this Primary.
Connections to the Center of the Primary can be Permanent. The outer
connection needs to be made adjustable so you can Tune it. Try to use wide
copper straps for the Adjustable Connection. The idea is to make a Large
Surface Area Connection. A Compression Fitting of some kind made out of
Copper Flashing would be Good. Make sure that the Connection is as FIRM as
you can get it.
Avoid ALL Ferromagnetic materials such as Iron and Steel. If it is
attracted to a permanent magnet, avoid it like the Plague. Any
Magnetizeable material such as Iron or Steel will tend to attract the
Magnetic Field in their direction. Not only will this draw energy out of
your system, but it may heat the metal up so high through eddy currents
that you have a fire hazard. (This is epecially the case near the Primary.
What you have here is a Glorified Induction Heating System!) Soldering the
Copper Tubing is OK. Copper is OK. Aluminum is Tolerable. AVOID ANY METAL
AT ALL IF IT IS CLOSED INTO A RING SHAPE. It will Couple the Magnetic Field
to itself, robbing the Secondary of Energy, and causing the ring of metal
to get VERY HOT! The Toroid is OK, because where it is the Magnetic Field
is very weak.
Note that there is an interaction that takes place when you Tune the
Primary. You not only change the FREQUENCY of Operation, but you also
change the Shape (Size, Extent, Curvature) of the Primary's Magnetic Field.
That is one reason why Tuning a Tesla Coil can sometimes be so Frustrating.
Improving ONE Aspect can cause ANOTHER Aspect to Degrade.
The Secondary has already been mentioned a couple of times. There are some
additional things you should know. The Length of the Wire in the Secondary
should Ideally be the same Length as 1/4 the Wavelength of the Resonant
Frequency Desired. Thus the Height of the Secondary Coil is a function of
the Length of the Wire, the Width of the wire (including insulation), and
the Outside Diameter of the Coil Form. (Oh, see how all the pieces Interact
at every Bend in the Road!)
The Secondary has an Inductance and a Capacitance, and
in MANY Respects behaves like a Transmission Line. The Capacitance is not a
Lumped (Single) Capacitance, but a Distributed Capacitance. That means that
the Coil Acts Like a Whole Bunch of Windings. Each having its Own
Inductance, and a Whole Bunch of Capacitors distributed at Every Turn of
the Secondary. When you Throw a Toroid or a Sphere or Anything Else At All
onto the TOP of the Secondary, THAT acts like a Significant Capacitance at
the END of the Transmission Line. Even COMING NEAR a SECONDARY or its
Terminal can significantly change the behaviour and Tuning of the System.
That is another reason why we have to RETUNE any time we move a Tesla Coil
from One Place to Another.
Secondaries should be CLOSE WOUND. That means the insulation of one turn
should just touch the insulation of the next turn. This implies that you
use VERY Good Insulation. Double Formvar Copper Magnet Wire is about the
best stuff that is readily available. The Form should be PVC or better
Plastic Pipe. Clean it thoroughly, Dry it, and give it several coats of
Shellac, Varnish, or a Clear Epoxy Paint. There are many different Brand
Names. Look at past postings for details and Personal Preferences. The key
thing is to apply SEVERAL Thin Undercoats of (let's call it) Shellac. Put
each coat on VERY THIN and let each coat dry before applying the next. If
you have a coil winder built, you might want to use it to rotate the coil
form while the stuff is drying so that it doesn't run and form Beads and
Lumps. Three coats would be a minimum. I use about Six Thin Coats. But
then, I'm Me, not You. I like to keep this undercoat smooth, because it
forms the Base upon which the Wire will be wound.
When you wind the wire, leave several inches free at the top and bottom of
the coil for mounting purposes. I Glue round Plexiglass Plugs inside the
top and bottom with Copper or Brass type Bolts that connect to the wires
and then are use to both anchor the Column to the Insulating Base or
Adjustable Platform (If I desire to fine tune the coupling), and make a
good firm electrical connection the the RF Ground. The Top Bolt is used to
attach Toroids and such.
It's getting late here and I have to finish this up for now, but a quick
note about Sparking from the Top winding and Toroids. A Toroid adds
Capacitance to the Top of the Secondary and therefore affects the Tuning.
It also prevents discharges off of the top of the coil Winding IF the top
of the coil winding is within its Zone of Protection. You see, the Toroid
produces an electrostatic field that is greatest at the periphery. You will
hardly EVER see a discharge originating from the Inner Side of the Toroid,
because it is actually the INSIDE of a Charged Body, and so experiences
little or NO Electrostatic charge. If the Toroid Extends Outward Far Enough
and its lower portion extends BELOW the the Top of the Coil Winding. then
the Coil Winding at the Top is ALSO Shielded by the Toroid. Note that no
such protection is provided by a BALL.
A Fragment of a thought before I REALLY call it quits tonight:
Sometimes the Secondary is resonating at a MULTIPLE of the Primary
Resonance. When this happens the Secondary Sparks are especially
transitory. You can get the sucker to work Pretty Good, but Tuning is far
too critical, and you get mostly whiskers of sparks with an occassional Big
Guy that snakes around for a second or two, then disappears. DRASTIC
Retuning required in this case.
All for Now!
If all else fails, try something else.
Fr. Thomas McGahee