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

Re: Bifilar coil



Original poster: Harvey Norris <harvich@xxxxxxxxx>



--- Tesla list <tesla@xxxxxxxxxx> wrote:

> Original poster: Illicium Verum
> <sebas@xxxxxxxxxxxxxxxx>
>
>
> Hello,
>
> I'm wondering if someone has ever experimented with
> a bifilar coil
> and what his or her results were?
Tightly wound pancake coils when adjacent on their
planes have a high mutual inductance. Flattened wire
can add to internal capacity which lowers the spirals
natural resonant frequency.
>
> This coil uses two wires laid next to each other on
> a form but with
> the end of the first one connected to the beginning
> of the second one.
>
>
> Tesla patent 512340
>
<http://www.keelynet.com/tesla/00512340.htm>http://www.keelynet.com/tesla/00512340.htm
>
>
> Best regards,
>
> Sebastiaan
The ramifications of Tesla patent explorations here
may not be evident to most. Peculiarly Tesla does not
take the next step in making his comparisons here. By
having capacitance between layers, instead of
capacitance between horizontal windings a (
multilayered square conductor)primary could also be
structured to have internal capacity.

Some confusion exists with regard to the term bifilar.
The actual word bifilar doesnt appear anywhere in the
patent, but people often refer to this as Tesla's
bifilar patent. Radio Shack sells a spiral coiled set
of speaker wires, known as Mega Cable, that has about
1/8 th inch flattened wire in the spiral. Significant
Reductions in what should be the resonant frequency of
the 100 ft of wire have been recorded by scoping the
spiral(s) and giving it a (spatial)high frequency
influence. With only a single set comprising two
layers of 50 ft, only one option gives magnetic unity,
the returned layer method, shown in the tesla patent.
However in this method, the 1/2[CV^2] internal
capacity has 50% of the entire (line
connected)potential placed across its wire length
appearing between the layers and the voltage
differential between layers is fairly constant. Even
though for this particular case only the edges of the
cable layers are involved with the higher voltage
reference points, its influence can be measured by
employing 4 layers (2 adjacent Mega Cable sets) and
measuring changes in resonant frequency by varying
routing paths through the 4 layers. A special option
now exists for making the voltage reference points
between layers into a NON-LINEAR BIFILAR arrangement
where zig zag spiral paths are employed. By employing
a double negative two oppositely wound spirals can be
wound together for magnetic unity, producing for dual
layers a highest and lowest voltage distribution
between layering reference points. This of course is
the most sensible wiring routing to begin with, as the
extra amount of wire to make a connection from outside
to inside points of the adjacently layered coils is
avoided, and simultaneous inner and outer connection
points are instead employed on oppositely wound spiral
layers. Since the V term is squared the increases in
stored energy seems concievable by this method. In
making measurements for four layers @ 200 ft total, a
25 pf  Heath scope recorded 330,000 hz for the
returned layer method vs 250,000 hz opposite direction
coil spiral sets having identical side connection
points. This value is well below, ( about 5 fold ) at
what the resonant frequency would be as a straight
wire. Complicating the issue is the scopes own
internal capacity, and how much this slows the
recorded vibration. Another larger wind 20 *30 square
coil arranged in spiral zig zag appers to resonate 13
times lower then straight line value.

At first the inductance of these spiral sets would
seem way to high to be of value as a tesla coil
primary. Conversely however the C(int) value may
appear in series with the C(ext) value and L we choose
to resonate with according to equation. If in fact
C(int) and C(ext) were equal the total acting capacity
should be halved, if they were to act as being in
series. Thus in the schemes of matching L and C(ext)
in the arc gap it would seem that more capacity can be
used then what we would be predicted by equation.
Another facet would be the use of a marx type gap.
Again the charging of caps in parallel, in this case
where each cap ends though a spiral length and then
the discharge of these caps in series means the acting
capacity on the figure 8 LC loops on discharge is
actually half that of the caps themselves.

The inductance of 4 layers of these spiral sets is
~2.3 mh. I use 3 of these sets for a 3 phase maximum
energy transfer resonance attached to a car alternator
480 hz arrangement. This is not specifically connected
to the subject of tesla coils however. The sets
display an acting q factor of 5 and 8.5 between the
phases. They have an impedance of 7 ohms @ 480 hz. An
interesting aspect of the voltage rise circuit is that
we can then attach a pole pig or unregulated high
voltage transformer supply as a load inside this METR
circuit. The arc obtained from the transformer
secondary in a conventional TC primary, where the arc
shunts the supply as a short circuit in this situation
we can plot out whether the supply can meet the
demand, as the extraordinary aspect to the (open
loaded)METR resonances is that they can pull as much
amperage from the generator as what is recorded when
the generator is shorted. Thus depending on the
appearance of the circuit prior to TC capacity
charging by secondary transformer attachment, what we
wind up with is an output very sensitive to its load
as to what voltage appears, but unloaded it is a 8.5
fold resonant voltage rise circuit on the transformer
primary itself. Thus a 20 volt alternator can input
170 volts to a 30 fold step up transformer becoming
5100 volts unloaded potential. From there the amount
of primary amperage required based on the demand of
the used C (TC primary) value at 480 hz can be
calculated and compared to known short values. In this
particular case the primary will be current limited to
7*1.7 ohms or ~12 ohms, probably far from a practical
value.
However in terms of generator supply and demand, some
other unusual aspects may come into play. Here we
speculate that the 480 hz resonant circuit can supply
2 amps when shorted when driven at 24 volts stator
input based on a 12 ohm impedance regulation. However
the actual current obtained from the stator lines is q
times less then the actual shorted current
measurement. And during an (interphasal METR) short
the appearance of the supply voltage must also
increase, where the initial conditions of the (open)
METR circuit are such that the load drops the supply
voltage down to half of its open circuit condition. In
this situation the measurement of the internal phasing
short also changes the load from one of series
resonance to one of a tank resonance, whose impedance
is generally predicted to be q squared higher then
that of its maximum current producing ability.
Thus given a 24 volt stator the circuit can supply a
8.5 fold voltage rise to the unloaded primary of the
tranformer @ 204 volt input producing 6120 volts from
30 fold step up transformer. The secondary, having a 2
amp primary regulation should only deliver 66 ma on
short.

These values are similar to what are obtainable with
NST's. However at 8 times the frequency compared to 60
hz, we should have 8 times the energy transfer between
primary L and C so that a  1nf cap ay 480 hz does the
same work as a 8 nf cap at 60 hz. Finding the
impedance @ 480 hz for 1 nf where X(C) = 1 /2piFC,  ~
331,000 ohms thus the 6100 volt supply would have a
conduction of 6100/331,000 = .018 A, well worth the
limits obtainable with the system.



But it should someday provide for study of a quenched
arc gap, which essentially means the supply voltage is
withdrawn q fold after arc gap firing.
HDN