Rosemary Ainslie’s COP = 17 Circuit

In January 1999, Rosemary Ainslie, a scientist in South Africa, made a patent application for a device
which has seventeen times more output power than it’s own input power. I, personally, was never in
any doubt that Rosemary Ainslie would encounter fanatical opposition to her scientific paper, because
“scientists” (for want of a better word) are unwilling to accept her simple system with vastly more power
output than the power input.

Rosemary Ainslie has produced a pulsing oscillator system which has been measured at a performance
of COP = 17, that is, the output power is seventeen times greater than the circuit’s input power. As far
as I am aware, this has not yet been replicated by other people. The Panacea-bocaf.org people are
working to produce an independent implementation of the circuit. At this point in time, the heater has
been built to a prototype testing scale for laboratory examination and measurement and not been
produced in the kilowatt range, which, hopefully, will come at a later date. Rosemary arranged her
circuit to operate a heating element because scientists like her find it convenient to make reliable power
comparisons based on heat measurements.

Panacea-bocaf have produced a 250-page document describing the research, the testing, the theory,
etc. and that can be downloaded free using this link:
http://www.panaceatech.org/Rosemary%20Ainslie%20COP17%20Heater%20Technology.pdf

As that document contains the details which scientists need to see for serious testing and development,
it may be a little technical for some people. This is Rosemary’s circuit:

1
In very broad outline, the circuit produces the same very short, very sharp voltage pulses that are the
basis for so many "free-energy" devices. The circuit used looks very simple but in spite of that, the way
that it operates is not at all simple. To a quick glance, the circuit looks like a standard 555 timer chip
circuit, used in many existing applications. However, if the circuit is operated as a 555 pulsing circuit,
then the output power will defnitely not be greater than the input power (“COP<1”).

Looking more closely, we notice that the link between the output of the 555 chip on pin 3 and the input
gate pin of the Field-Effect Transistor, is unusual as it is not the usual voltage divider between pin 3 and
the 0-volts ground line. Instead, the gate is directly coupled to the 555 chip output by a single, low-
resistance wire-wound multi-turn varaible resistor.

Many NE555 chips struggles to reach 50,000 cycles per second. To get the circuit into it's COP>1
operation, the resistor marked "GATE" is adjusted very slowly to find the point at which the circuit
becomes unstable, over-rides the normal operation of the 555 chip and starts oscillating at the resonant
frequency of the overall circuit, forcing the 555 chip to become a feedback component. The circuit then
produces the sharp, short voltages spikes at more than ten times the operating speed of the 555 chip
and pulsing the 10-ohm heating element marked "LOAD" at about 500,000 pulses per second. I’m told
that some 555 chips can operate that fast but many 555 chips definitely can’t.

The timing elements of the chip should be producing a much lower frequency, as indeed it does before
the "GATE" resistor adjustment causes the circuit to break out of its normal design-mode operation and
start the high-speed spike generating, resonant performance which draws in additional energy from the
local environment

Most circuits which draw energy in from the local environment, generally need to be tuned to achieve
resonant operation. It is also found that a waveform rich in harmonics is needed to produce the best
results. For example, Ronald Classen recently produced an analysis of the operation of Bob Boyce's
electrolyser toroid pulsing. Bob's circuit generates three separate waveforms, one at about 42.8 kHz,
and two harmonics, one at around 21.4 kHz and the other at about 10.7 kHz. Ron examined the
operation with the two harmonics slaved exactly to the master frequency and then again with the two
harmonics free-running and not quite synchronised, so that a random pattern of harmonic pulses were
generated. Surprisingly, he found that the random arrangement gave much higher gains than the
"precision" circuit.

The same sort of situation is found here in the Ainslie circuit as very precise adjustment of the "Gate"
preset resistor "R1" has a major effect on the circuit performance while the other two, R4 and R7, are
used to adjust the frequency of the pulses and the ratio of "On" time to "Off" time. Like almost every
other circuit which produces a greater power output than the input power required to make it operate,
very careful adjustment is needed. The characteristics of the "Load" heating element "R3" are also very
important. With some configurations, there is no excess power generated, while with others there is a
2
very marked increase in power and the prototype apparatus produced power outputs in excess of four
times the input power.

A quick glance at the circuit diagram makes it appear that there is no significant connection between
the NE555 timer chip and the IRFPG50 FET transistor. This is not the case as the arrangement as
shown generates transients which modify the oscillation of the NE555 chip. This is presumably due to
the nature of the current draw by the gate of the FET or through induced currents caused by the pulsing
of the inductive load heater coil "R3". We tend to think of FET transistors as having next to no current
flowing into the gate, but the IRFPG50 FET can draw up to a massive 6 amps for the Gate to Source
current flow. The NE555N chip supplying that gate current (with no current-limiting resistor between the
two devices) can supply a maximum of only 200 mA (or possibly 300 mA at a push) which is only 5% of
the possible current draw by the FET. The circuit of the NE555N chip is:

From this it appears that the direct coupling of the output could allow some modification of the chip
timing and waveform if the output current draw is well above the design value, the internal resistors
preventing destruction of the chip and reducing the effect so that it just modifies the functioning of the
chip.

This is also suggested by the fact that the adjustment of the "Grid" variable resistor, which controls the
NE555N current draw, is the most critical adjustment of the circuit. Supporting that idea is the fact that
the required chip operation does not take place if the "Grid" resistor setting is too high or too low.
Presumably, the setting has to be an exact amount so that the specific physical NE555N chip operation
is altered to make it generate waveforms not envisaged by the chip designers. The physical separation
of the "Load" resistor and the circuit board may also be important as there is almost certainly a
magnetic feedback element as well.

I would love to tell you that the circuit operates in the way that the circuit diagram would suggest, with
3
the timing and Mark-Space values controlled as expected by the 555 chip designers. However, that is
definitely not the case. If the 24V battery is disconnected, then the NE555 chip section of the circuit
performs exactly as expected by the chip manufacturer. If the "R1" GATE resistor is at the correct
setting and the 24V battery is then connected, the result is that the normal running of the NE555 chip is
overridden and the circuit immediately switches into a completely different type of operation. The Mark-
Space ratio is forced into an approximately 55% setting and the pulsing rate is bounced to over 500
kHz. This is the waveform generated by the circuit:

You will notice that this waveform has repeating pairs of pulses, neither of which is a square wave. The
overall circuit is clearly not operating as an NE555 chip circuit any longer but is oscillating in an
unexpected way. This high radio-frequency pulsing produces electromagnetic waves which radiate out
from the load resistor, an effect which is seen on a nearby television set. This is not really surprising,
as the circuit should really be presented like this:

This is because the 10 ohm "resistor R3" is actually a coil of wire. The specification for this component
shows that it has a length of 150 mm (6"), a diameter of 32 mm (1.25") and is an air-core coil, wound
with 48 turns of resistance wire with a 1 mm gap between each turn. The lack of a core, allows the coil
to oscillate at this high frequency, and any coil driven at that frequency radiates radio waves.

It is almost certain that these electromagnetic waves are inducing voltages in the wiring surrounding the
NE555 chip circuit, causing it to run wildly outside its design. The wire-wound adjustment resistors are
little coils which have the potential for picking up transmitted waves. This pickup mechanism is strongly
supported by the fact that only an NE555N chip will operate in this way and three other makes of 555
chip which were tested, failed to produce this runaway action. The higher runaway frequency is
important for achieving power gain. Don Smith states that the extra power being drawn into a circuit is
proportional to the square of the pulse frequency. If this is correct, then moving the pulse rate up to
over 500,000 per second will have a major energy effect and explain why tuning the circuit into this
high-speed mode is important.

The practical method of tuning the circuit into its self-oscillating non-symmetrical, power-gaining mode
is by monitoring the voltage of the "V1" 24V battery. When the circuit is out of tune, the battery voltage
gets pulled down quite noticeably. When the circuit is tuned correctly, there is a slight increase in the
battery voltage. If the circuit has been built as described, using an NE555N timer chip and a high
inductance load "resistor" coil, then tuning the circuit is performed as follows: Connect a digital
voltmeter across the 24-volt power supply and note the exact reading. Set the "ON" preset resistor to
its minimum value of zero ohms. Set the "OFF" preset resistor to its maximum value of 10K ohms.
4
These resistors are generally left at these settings throughout.

The "GATE" resistor is now adjusted very carefully, watching the voltmeter reading. As the circuit
comes to its best possible tuning, the battery voltage will rise. Pick the resistor setting which gives the
highest battery reading. The rise in battery voltage is caused by the inflow of external energy. Some of
this flows through the "LOAD" causing heating effects which can be 17 times greater than would
normally be expected. Part of the inflowing energy flows back into the power supply, and that flow
lowers the current draw from the 24V battery, which in turn, allows it to show a higher voltage reading.
This mechanism is exactly the same as described by Tom Beardon when explaining the operation of
John Bedini's battery-charging pulse circuits - part into the load and part back into the power supply.

Although it is not mentioned in the Parts List, it is very important to mount the FET transistor on a heat
sink as the current flowing through it causes it to heat up. Also important is to use a mica gasket
between the FET and the heat sink. A mica gasket is a thin layer of mica which electrically insulates
the FET from the heat sink while still acting as an extremely good conductor of the FET heat to the heat
sink. This is necessary because the "Drain" pin of the FET is connected electrically to the metal
mounting strip of the FET and if the FET is not insulated from the heat sink, then the heat sink acts as a
radio aerial and radiates an embarassingly large level of radio waves. The heat sink can be a simple
sheet of aluminium, or it can be a commercial finned design of which there are many from which to
choose. A suggested physical layout for this circuit is given towards the end of this document, and can
be used if you are inclined so to do.

This is a circuit which cries out for replication and investigation by both experienced and inexperienced
experimenters. There are no expensive components in the circuit and the circuitry could hardly be any
more simple than it is. If this circuit can be scaled up to operate as a household heater it would mean
that electrical heating costs could be reduced to a tiny fraction of what they are at the present time.
That sort of cost reduction would make a major difference to a very large number of people, which
makes this circuit very interesting indeed.

The operating methods which are used in this style of circuitry are described in considerable detail in
the patent application WO 99/38247 which has been filed for this system. Reading those descriptions
can be helpful, so here is a digest of part of that patent:

Patent: WO 99/38247 Date: 22nd January 1999 Inventor: Rosemary A. Ainslie

HARNESSING A BACK EMF

ABSTRACT
A method of achieving high efficiency of energy usage which includes passing current through an
inductor, causing the current to be repeatedly interrupted, thereby generating a back EMF in the
inductor and thereafter, harnessing the back EMF so generated, to supply energy to an energy-
receiving or processing device. The frequency of interruptions should be 40 Hz or more and is
achievable by rectifying the current. The invention extends to apparatus for harnessing such back EMF
and energy generating means comprising an inductor and a current interruptor connected to an energy-
receiving device.

FIELD OF THE INVENTION

The invention relates to a method of harnessing back EMF for use in powering a load or replenishing a
depletable energy source and extends to apparatus used in performing the method.

BACKGROUND OF THE INVENTION

Conventional switching circuits are well known in electrical energy conversion technology, and switch
mode systems have been employed to enhance energy utilisation efficiencies. The concept of
absorbing electrical energy released by the collapse of auto-electronic emissions from a discharge tube
is disclosed in US 5,449,989. This document discloses a circuit which includes an output port
5
connected to a current sink which is able to absorb at least a substantial portion of such emissions.
The current sink may be an electric motor or a secondary battery.

The concept of applying a back EMF in electrical circuitry is also known. For example, in US 5,521,476
there is disclosed a control circuit for a disc drive motor, in which back EMF blocking circuitry is
employed to prevent dissipation of a back EMF through a power supply. By contrast, publication WO
9,613,892 discloses the use of a back EMF to trigger a response in a control system for a mechanical
system, so that driving pulses are generated to accomplish a desired displacement motion.

In the present invention, to achieve high energy efficiencies, greater than unity in relation to a
conventional test circuit, a back EMF which is generated in an inductor, is harnessed so as to return
energy associated with the EMF, to a depletable energy source which is supplying such a circuit, or to a
load included in the same primary circuit as the energy source. It is envisaged that a wide range of
electrical supply sources will derive benefit from the invention disclosed below.

A first aspect of the invention is a method of harnessing back EMF in an electrical circuit in order to
increase the efficiency of energy usage to 90% or more, (compared to a Resistor-Temperature v
Wattage calibration circuit). This is done by arranging the circuit so that it contains an inductor and an
energy-receiving device configured so that the current flowing through the inductor generates a back
EMF whose energy is used to provide both additional energy to the circuit itself, and the back EMF
energy to the energy-receiving device.

In a preferred form of the method, the back EMF is generated by interrupting the current flow through
the inductor, ideally, interrupting and restoring the current flow repeatedly and rectifying the current.
The rate of interruptions should be at least 40 times per second and preferably 50 or more times per
second. The duty cycle of the interruption should be at least 50% and ideally be 75%. That is, the
current flow through the inductor is "On" for 50% to 95% of the time and "Off" for 50% to 5% of the time.

In a further preferred form of the invention, a back EMF is generated which is large enough to cause the
comparative energy efficiency to be at least unity. This can be achieved by setting and controlling a
suitable value for a variable selected from one or more of:
The frequency of interruptions from the wave rectifier;
The duty cycle;
The thickness of the wiring in the circuit;
The efficiency of the inductor core,
the value being set in accordance with the operational requirements of the desired application.

In another preferred form of the invention, the energy-receiving device is either an energy-requiring
load, and/or an energy storage device, ideally a replenishable source of either DC or AC electrical
energy. Ideally, the method also includes providing at least one inductive load associated with each
receiving device. The inductor may be a transformer or other suitable inductive device.

A second aspect of the invention is a method of restoring electrical energy to a source, which is done
by providing a closed circuit containing a source of electrical energy which passes current through the
inductor, creating an extruded magnetic field around the inductor, which field is then collapsed, creating
a back EMF which is then fed to the source with an energy usage efficiency factor of 1 or more when
compared to a Resistor Temperature Versus Wattage Calibration Circuit.

This feedback of energy can be to an energy-requiring load or to an energy storage device.

In a further preferred form of the invention, the bias-changing mechanism is a wave rectifier and the
method of use is to make the wave rectifier output interrupt the electric current.

Ideally, the inductor used should have a solid core which is capable of inducing a magnetic moment
associated with a collapsing magnetic field.

The method used in this invention includes selecting a value for:

The frequency of interruptions from the wave rectifier;
The duty cycle;
6
 The thickness of the wiring in the circuit;
The efficiency of the inductor core,
so that the magnitude of the back EMF generated when the magnetic field collapses, is in a
predetermined range which suits the requirements of the energy-receiving device and its intended use.

In one preferred form of the invention, the inductor is a transformer with a primary winding large enough
to create sufficient voltage from the back EMF, to feed power back into the circuit. If the current feeding
the inductor is AC, then the current interruptor can be a diode or a triac.

A further aspect of the invention is an apparatus comprising an inductor having a core suitable for the
generation of back EMF from collapsing magnetic fields, and an electrical circuit containing that
inductor, a replenishable energy source, and energy-receiving device and means for changing orbital
bias of a magnetic field set up in use and associated with the inductor, both it and the source with
variable frequency and variable Mark-Space ratio, being configured to operate the inductor, and
arranged so that the magnetic field of the inductor is made to collapse and be restored repeatedly,
thereby generating electrical energy, the circuit being capable of conducting the energy and providing it
to the energy-receiving device.

A BRIEF DESCRIPTION OF THE DRAWINGS

Fig.1 illustrates schematically, a circuit to which the invention may be applied.

7
Fig2. illustrates an electrical generator which may be used with this invention.

Fig.3A illustrates a control circuit which is described in Example 1 below, and Fig.3B illustrates a test
circuit, the performance of which is compared with the circuit shown in Fig.3A.

8
Fig.4 illustrates the test circuit described in Example 2 below.

Fig.5 illustrates the circuit referred to in Example 3 below.

DETAILED DESCRIPTION OF THE INVENTION

By connecting an inductor in a load-bearing circuit and causing back EMF to be established in the
9
inductor, there may be created a voltage of sufficient magnitude to restore energy to the circuit's source
of power and so reduce its rate of depletion. It is not suggested that the load would consume less
energy, but that additional energy from the back EMF can be supplied to either the load or the power
source supplying that load.

The circuit can be supplied with either DC or AC power and while the inductor may be any suitable
inductor, the use of a transformer is preferred. An alternative is a winding or a choke, preferably
containing a core capable of inducing a magnetic moment associated with a collapsing magnetic field -
typically an iron core, but it could be any suitable liquid or gaseous medium or combination with or
without additional solids.

To generate back EMF, the current flowing through the inductor needs to be interrupted repeatedly
which can be done by any suitable means. A preferred method is the use of a variable-duty cycle
chopper. If the current is AC, then the interruption can be cause by using a wave rectifier such as a
diode or a triac. If the current is DC, then the current interruption is achieved by the use of an oscillator,
MOSFET or an equivalent means whereby a fluctuating magnetic field in the inductor can be created.

The method of recycling energy which is the substance of this invention has the following steps:
(1) Setting up a circuit containing an inductor which has an extruded magnetic field and which is
arranged in such a way as to allow electrical energy to be passed both to and from the inductor, and
(2) Changing the orbital bias of the magnetic field around the inductor, causing the collapse of the
magnetic field and the creation of the back EMF current.

These two steps are repeated in rapid succession and when the current flowing through the inductor is
interrupted, an alternative circuit is provided in order to direct the back EMF current to the desired
destination. Preferred inductor core materials are iron and other ferrous materials.

The circuit does not need to be complicated but it needs to be able to either interrupt or reverse the
current through the inductor as already described. The invention will now be described in greater detail
by referring to the diagrams:

In Fig.1, the wave rectifying diode 14 is placed in series with a primary inductor 12, and acts as an
interruptor of the electrical current supplied to the circuit. If a sine wave or square wave waveform is
applied to the circuit through points 16 and 18, a pulsing DC waveform is created in the winding of
inductor 12. The interruption of each waveform cycle in the inductor winding 12, induces a fluctuating
magnetic field in the iron core 20 inside the inductor.

It is thought that the back EMF causes a reverse waveform in the inductor winding 12 which is a full
sine wave in the case of an alternating current powered circuit, or a full square wave if the circuit is
powered by DC pulses. The inductor 12, may be connected with a load (not shown) in series or in
parallel at any of the points marked 18.

Depending on the frequency of the interruptions, the duty cycle, the thickness of the wiring and the
efficiency of the core, the voltage across the inductor 12 may be conducted through a closed circuit to
be used in powering the load or returned to the power source. It is desirable, though not essential, that
the frequency of interruptions should be not less than 40 Hz although 500 Hz or higher is more
appropriate for some applications.
10
An example of a suitable closed circuit employing such a system is a battery powering a lamp. A
transformer may be connected in series with the lamp along with a current chopper which has a
variable duty cycle. The output from the transformer can be routed through a diode, a high value
resistor and a capacitor all in series. Here, when the chopper service is on, the current flows through
the load and transformer. Repeated opening and closing of the current-chopper system causes the
generation of electric current in the transformer secondary and that current is passed back to the
battery, exceeding the current draw.

Fig.2 shows another variation of the circuit where a primary winding 30, having for example, 220 turns
around a cylindrical core 32, made of a ferrous metal such as iron or an iron alloy, is associated with a
secondary winding 34 of about the same number of turns. The secondary winding is wound around the
core adjacent to, or on top of the primary winding 30, producing a magnetic coupling between the
windings, enhanced by the core 32. The circuit input 36 is connected to an AC source 38, typically a
220V 50Hz mains supply. The circuit output is taken from the secondary winding 40. A diode 42 is
connected in series with the primary winding 30, causing the full-wave AC input to become a pulsating
input to that primary winding.

On each positive-going half cycle, the primary winding induces a corresponding current in the
secondary winding 34. However, when, due to the blocking effect of diode 42, the magnetic field
resulting from the current in the primary winding 30 collapses, the resulting back EMF in the primary
winding induces a corresponding negative-going waveform in the secondary winding 34. Hence the
output 40 from the secondary winding is a full-wave AC waveform.

Although this description is for a circuit with one inductor, it is clear that additional inductors could be
used to achieve even greater enhancements in system performance. For example, two or more
primary windings could be wired in parallel where just one is shown in Fig.2 above, each providing a
separate, independent full-wave AC output. Alternatively, more than one secondary winding can be
placed on the transformer core, utilising the magnetic coupling of the core.

Example 1: Two tests were conducted on two wire-wound, 10-watt resistors manufactured by Philips.
The resistors have identical surface areas. The object of the test was to compare the rate of current
draw of a standard "Resistor temperature Versus Wattage Calibration Circuit" (the "control") indicated in
Fig.3A:

11
to a test using a switching device and an inductor as indicated in Fig.3B. The same battery was used
in both tests. The control test shown in Fig.3A, had a thermocoupled 68 ohm resistor 40, and a
sensing resistor 42, placed in series with the battery terminals 44. All measurements were made after
the temperature of resistor 42 had reached its maximum value of 95OC. The current was measured as
being 196 mA and as the battery voltage was 12.28 volts that represents a power level of 2.406 watts.

The test circuit shown in Fig.3B, had a MOSFET switching circuit transistor 56 driven by a square wave
signal (shown in green) whose Mark/Space ratio was adjusted until the load resistor 42 reached its
highest value of 93OC and all quoted measurements were made after that time. The pulsing signal was
running at 5kHz with an "On" time of 22.87% and an "Off" time of 77.13% of the time. The current flow
was measured as 171.8 mA which represents a power input of 2.109 watts. The room temperature
remained the same during the entire test period.

Allowing for a 5% error in the measurements, this result shows an energy output which is 8.6% greater
than the power input, or COP=1.086.

Example 2: The following tests were conducted to prove that subject to specific circuit configurations,
an inductor is able to enhance energy efficiency to levels beyond the standard capabilities of an
electrical power supply source. The tests also indicate that if a resistor is placed in series with a power
supply and an inductor as shown in the Test Circuit, then the correct wattage analysis of that power
may be calculated as the energy source voltage multiplied by the amperage (V x I) and that I2R no
longer holds as a base calculation of the wattage and power generated in this particular system.

12
With reference to Fig.4, the Test Circuit 60 comprised a 47 ohm, 10 watt, Philips wire-wound resistor
62, placed in series with two 6-volt batteries, 64, and 66 connected in parallel. A inductor 68, was
placed in series with load resistor 62. A positively-biased diode 72, was placed in parallel with the
inductor 68 and above an n-channel MOSFET transistor switching device, 74. This wire was then taken
back to the positive terminal of the batteries. The battery voltage was measured at 6.12 volts.

The duty cycle was adjusted to a 50:50 Mark-Space ratio, giving equal times for the On condition and
the Off condition. The load resistor reached a temperature of 30OC and the ambient room temperature
was 22OC. The waveforms for the three sensing resistors SR1, SR2 and SR3 are shown in Fig.4
below the circuit diagram.

The voltage waveform across the SR1 sensing resistor in series with the load resistor 62, is roughly
triangular but followed an exponential rise and fall during the On and Off periods of each cycle. The
voltage did not fall below zero. The peak positive voltage was measured as 0.006 volts which
corresponds to approximately 0.169 watts which is less than would be expected from the temperature
of the load resistor. It would be expected that 0.375 watts would be required to produce the measured
30OC of the load resistor 62.

The voltage waveform across the SR2 sensing resistor placed in series with battery 1, marked 64, was
roughly triangular in form with some exponential curvature as shown. The average current draw from
the battery was measured and calculated to be 0.025 amps, which is a power draw of 0.153 watts..
.
The voltage waveform across the SR3 sensing resistor placed in series with battery 2, marked 66,
showed a waveform with equivalent amounts above and below the zero voltage level. The On voltage
peak was 0.0055 volts and the Off voltage peak was -0.0055 volts (i.e. below zero volts). No power
was being drawn from this battery and in fact, the shape of the two sections of the waveform indicate
that there was actually a slight degree of charging on this battery although this was ignored as being
too small to be significant.

The inescapable conclusion from these tests is that to achieve identical heating of the load resistor, the
standard circuitry required 0.0612 amps while the test circuit required only 0.025 amps. This means
that the pulsing circuit is more than 100% more efficient than the conventional circuit. These
measurements represent a Coefficient Of Performance of 2.45 as the output power is 2.45 times the
input power.

These two examples shown here do not necessarily represent optimised values and further gains may
be attained by using two or more inductors, two or more energy sources or energy storage and its
switching circuitry, and other measures.

Example 3: A further set of tests was conducted to investigate the relationship between power
13
supplied by the battery marked as 82 and power dissipated by a resistor R1 in the circuit of Fig.5.

This is to test the efficiency of the energy conversion as the duty cycle of the FET switch Q1 is adjusted.
This circuit includes an inductor 84, which has equal primary and secondary windings and a 350 VA
rated core. The circuit also contains a positively-biased diode 86 and other components mentioned
below. The tests were conducted with "On" times of 90%, 80%, 70%, 60% and 50% and the results are
shown in this table:

Duty V1-3 Average V1-2 Battery V1-3 RMS V3-4 Load Pload /
Cycle Average Current DC Power rms Current rms Power Pbatt
% mV A V W mV A V W ratio
90 69.5 1.390 12.57 17.46 102.5 2.05 10.02 20.54 1.176
80 38.2 0.764 12.64 9.657 73.1 1.462 7.58 11.08 1.148
70 20.9 0.418 12.69 5.304 51.1 1.022 5.36 5.478 1.033
60 7.9 0.158 12.73 2.011 34.1 0.682 3.19 2.176 1.082
50 1.2 0.024 12.76 0.306 15.9 0.318 0.94 0.299 0.976

**************************
The important thing to note from these figures is the way that the ratio of the output power to the input
power (which is the Coefficient Of Performance or "COP"), shown in the final column, varies with the
Mark-Space ratio shown in the first column. For all On time ratios over 60% in this very simple circuit,
the COP is greater than 1 which conventional science swears is "impossible" in spite of it being
demonstrated over and over again by different people and different styles of apparatus.

Rosemary Ainslie's techniques shown here where the back-EMF pulses are harnessed and used to
perform useful functions, achieve COP values from 4 to 17 in tests performed to date.

John Bedini's captured back-EMF battery-charging circuits have been replicated with high-voltage
battery banks being charged by one 12V battery and yielding COP=11 results.

The pulse motor design of Robert Adams which utilises the back-EMF pulses and other techniques,
reaches COP figures of 8 or higher, depending on the quality of the build and the accuracy of the
adjustments.

Thane C. Heins demonstrates on video http://www.youtube.com/watch?v=RbRPCt1-

WwQ&feature=channel a very simple transformer arrangement which produces COP=3.77 a result
which you can easily check out for yourself.

14
Rosemary's neat technique which produces this energy gain has every appearance of being a more
easily adjusted method of producing the gains of the Tesla Switch which has to have a substantial
inductive load in order to get its COP>1 performance and which is very tricky to adjust.

It should be stressed at this point that it is physically impossible to draw more energy out of a circuit
than the energy flowing into that circuit. Energy can't be destroyed or created and you can't have more
than 100% of anything, anywhere, any time. But Rosemary Ainslie and others have demonstrated very
clearly that carefully designed and operated circuits definitely put out more energy than the user puts
into the circuit. I do not know of any way to prove where that extra energy comes from, but it definitely
comes from somewhere, flowing into the circuit from outside. However, let's not concern ourselves with
trying to discover the source of this extra power and instead, just learn how to capture and use it for our
own benefit.

So, let's recap on how Rosemary's circuitry is set up and used. The initial basic circuit which gives an
energy gain is:

Here, a mains-powered light bulb has two components connected in its normal circuit. The first
component is a diode "D" and the second a transformer "T":

The diode has a very low resistance to current flow in one direction and a very high resistance to
current flow in the other direction. We tend not to think about it, but the ordinary mains current flowing
through a light bulb switches its direction of flow dozens of times per second - actually, sixty times per
second in the USA and fifty times per second in most other parts of the world.

If we put a diode in the circuit as shown in the diagram above, it gets in the way of every second surge
of current through the bulb. This causes the current flow to be in only one direction and there are fifty
or sixty gaps per second in the flow of current through the bulb. This pulsing current flow passes
through the left hand transformer winding (shown in blue in the diagram), called the "primary" winding,
and it generates a voltage and current flow in the other winding of the transformer (shown in black in the
15
diagram and called the "secondary" winding).

The two lines drawn between the two windings indicate that the transformer has some kind of magnetic
core. Having a core in a transformer can be a very mixed blessing. It will work very well if there is no
core material - generally called an "air-core" transformer. Energy gains in a circuit like this, increase
with increased voltage and even more so with increased rate of pulsing (called the signal "frequency").
An air-core coil or transformer will operate at very high frequencies, limited mainly by the wire diameter.
Most powerful transformers are usually supplied with an iron core as that improves their magnetic
coupling at the very low pulse rates used with mains power. That iron core has very limited frequency
performance as it is limited by how fast the iron can alter its direction of magnetisation. It is unlikely that
you would get good performance even at the low frequency of one thousand pulses per second ("1
kHz"). As each of these pulses feeds a little packet of extra energy into the circuit, obviously, you would
like as many as possible per second, so that the energy inflow is very great. You will notice in
Rosemary's patent, that she mentions raising the pulse rate to five hundred per second to increase the
power gain.

However, that does not matter here as we are using a mains electrical supply which is just creeping
along at well under one hundred pulses per second in order to explain the technique in a simple form.
Anyway, the voltage generated in the secondary winding of the transformer is a full-wave voltage
waveform just like the original mains waveform with no gaps in it. This energy in the secondary winding
could be used for a wide range of different purposes. The one shown here is the charging of a battery
or a bank of batteries arranged to work at almost any voltage. Contrary to popular belief, the voltage
used to charge a battery is not particularly important provided it is high enough, but what is very
important is the current flow into the battery, and that needs to be controlled carefully. Ed Gray
demonstrated that charging with a high voltage was a perfectly good method and he used a capacitor to
control the current flow into the battery. Eventually, he gave up doing that and used an ordinary car
alternator to charge the battery as it was difficult to get the capacitor value just right to achieve the
desired current.

Rosemary uses the same technique but adds in a resistor "R" to make sure that the charging current
never becomes excessive. The diode "D" converts the alternating voltage in the transformer winding to
positive pulses, that is, pulses where the voltage rises above zero volts and never falls below zero volts.
This is the sort of voltage which we need for feeding to the positive terminal of a battery.

In passing, while the capacitor "C" does act as a current-limiting device, it may also act as a conversion
device as extra energy flowing into the circuit from outside can be of a somewhat different type to the
electrical current drawn from the mains, and a capacitor is a well-proven method of converting the
incoming energy into the more familiar conventional form.

16
So, looking at the circuit again, the mains is converted to a pulsing 50% Mark-Space ratio current flow
through the primary winding of the transformer "T". When that flow cuts off suddenly, there is an inflow
of energy into the winding from outside the circuit, forming what is called a "back-EMF" brief voltage
pulse in the opposite direction. This fills in the pulse gaps in the secondary winding, giving it a full-wave
waveform in spite of the primary being fed only half of that waveform.

The secondary power has the negative pulses (below zero volts) chopped off by the diode on the
battery side, giving a series of positive pulses at the same frequency as the mains. The capacitor "C"
and the resistor "R" control the current feed to the battery and allow it to charge at a suitable rate.

So, that is the basic circuit - simple and elegant and very effective in use. But, it does not stop there as
that basic idea can be used in various other ways. For example, like this:

This is the same circuit, but two transformer primary windings are connected across each other (called
being wired "in parallel"). The operation is exactly the same as before except that two copies of the
mains waveform are made by the magnetic coupling of the transformer windings. Each is "rectified"
into positive-going pulses and fed to the battery, creating a larger charging current. An alternative
version of this is:

17
In this variation, the transformer is wound with one primary and two secondary windings. The magnetic
coupling of the transformer core generates copies of the mains waveform in both of the secondary
windings. Each are rectified and fed to the battery as before.

If this circuit was being built using standard off-the-shelf transformers, it might be easier to use two
separate transformers connected "in series". This would depend on the application and the windings of
the particular transformers to be used.

The diagrams show up to now have all suggested charging one or more batteries, but this has just been
for the convenience of presenting a simple application. As is pointed out by Rosemary, it is perfectly
possible to drive some other load such as a heater or a motor using these additional power take-off
connections. However, for our continuing description of the circuit options, we will stay with battery
charging. So, using two standard transformers, the circuit would be as shown here:

All of the Ainslie circuits mentioned so far have used the mains, but there is very considerable scope for
circuits and arrangements which do not use the mains at all. Admittedly, a certain amount of electronic
construction work is needed, but the results can be very rewarding. For example, instead of charging a
battery bank, it is quite possible to charge the battery which is driving the circuit itself.

Now, before you start to say that this is an impossibility, please remember the little girl and her small
electric car battery. Her father found that if he left the charging circuit on too long that he needed to put
a bulb in as a load in order to avoid over-charging the battery, and that battery (appears to be) what
18
powers the charging process. In all of these systems, please remember that additional energy flows
into the circuit from the local environment, so charging a battery which is driving the circuit is perfectly
possible. For example, Robert Adams of New Zealand ran his motor for a ten-hour test and the battery
voltage was exactly the same after the test as it was before the test started. If you think that is
spectacular, then consider John Bedini's self-charging motor. John ran that non-stop in his workshop
for more than three years !! So please don't try to tell me that this sort of thing is impossible because
that's what you have been told. Self-charging can definitely be achieved if you know what you are
doing.

Here is an Ainslie self-powering circuit:

Here, the signal generator, which is probably just a simple 555 timer circuit, produces a train of pulses
with a Mark-Space ratio of more than 50% On time. That signal is used to switch a transistor On and
Off in rapid succession. The transistor type is deliberately not shown as it can be an NPN silicon
transistor, an FET type of transistor, a Darlington pair, or one of those fancy new IGBT devices.
Whatever the type chosen, the lamp will be switched on and off so rapidly that it will light up. The
fluctuating current through the transformer "T" will produce an alternating voltage in its secondary
winding and that will pass through the diode, resistor "R" and capacitor "C" to charge the battery in spite
of the fact that the battery is powering the signal generator circuit and the lamp.

Obviously, all of the other options and variations discussed above in connection with a mains-powered
version of the circuitry will apply equally well to a battery-powered version. If running from a battery or a
bank of batteries and high voltage is wanted, then an off-the-shelf inverter can be used to generate the
high voltage used for the mains supply.

If you would like to test the operation of the circuit and the design generally, here is a stripboard layout
which might be used:

19
The preset resistors are high power units looking like this:

These are adjusted with a flat screwdriver inserted into the slot at the end of the shaft although they
could have a knob attached. It takes ten full turns of the shaft to move across the full range of the
resistor. If you are adjusting the Mark-Space ratio and the ratio goes up when you turn the shaft to the
left but want that to happen when you turn the shaft to the right, then just swap over the wires going to
the outermost terminals of the resistor and that will reverse the effect when you turn the shaft. You can
stick the base of the resistor directly to the stripboard using "Impact" Evostick or any similar adhesive
and that will hold it securely but still allow you to prise it off the board at a later date if you should need
to.

There is no need to use knobs as the circuit will be set up, adjusted for best performance and then left
untouched. The circuit can be built using stripboard like this:

20
The view above is of the underside of the board as that shows the copper strips running horizontally
between the holes. The copper strip is quite thin and can easily be broken using the tip of a drill bit or
alternatively, a modelling knife. The spacing of the holes is arranged to match the pin spacing of
standard Integrated Circuits such as the NE555 timer chip used in this circuit. The only place where the
strips need to be broken in the layout above are between the pins of the NE555 chip and if you didn't do
that, then the four pairs of pins would be short-circuited together, preventing the chip from operating. It
is a good idea to use an 8-pin IC socket soldered to the board as that prevents any heat damage to the
NE555N chip during soldering, the IC being plugged in after the soldering has cooled down. It also has
the advantage that if the chip ever gets damaged, then plugging another on in is a very easy thing to
do. After the board is completed, it is also probably worth running a solder layer along the copper strips
which carry some current, that is the plus and minus strips and the strip between pin 3 of the NE555N
and the point where the connection to the variable resistor is made. You will notice that the layout of
the board includes four holes to take mounting bolts. When these are drilled, the strips under the board
need to be cleared away to make sure that no short-circuits can occur when the bolts are in place. The
board mounting is like this:

Suppose we wish to replicate and test this circuit:

We need to remember that this is just the outline for a practical circuit and that it does not show the
normal extra items like and On / Off switch and a fuse or circuit-breaker which are essentials for any
circuit which contains a powerful battery. Please remember that you can't see current flow and if there
is an accidental short-circuit, the first you may know of it is smoke !! That tends to be expensive,
especially if some of the components are pricey and/or hard to get.

If we work with the Ainslie pulsing circuit shown at the beginning of this document, then a physical
layout convenient for experimenting might be:
21
The "heat sink" shown in the diagram above, is just a piece of aluminium bent to raise the centre
section slightly and allow good air circulation and clearance for the FET's locking nut. The FET is
bolted securely to this plate in order to allow the aluminium plate to let the FET run cooler than it
otherwise would. The lamp would be a 12V car type and while many people just solder directly to the
bulb as shown here, there is no reason why a bulb socket should not be used. Car accessory shops
usually have low-cost "reversing lights" which are a small plastic housing, a bulb socket, a bulb and two
pieces of wire already attached to the bulb holder - very convenient, especially since it is very easy to
change over to bulbs of different ratings for different tests and the bulbs themselves are cheap.

This circuit is of course, the same as the driver circuit for the heating element circuit. The green link
wire shown in the diagram above gets replaced with the 30-watt 0.25 ohm resistor and the resistor
should be positioned so that it is in the air, well clear of everything else as it may get hot during
operation in spite of its very low resistance value.

Disclaimer: It must be understood that this document is presented for information purposes
only and it must not be construed as being an encouragement to either build or experiment with
this or any other circuit. The people who have investigated, designed, built or described this
circuitry are in no way liable for any loss or damage caused by your actions, should you decide
to experiment with this or any other circuit. Should you choose to do that, the responsibility for
your actions rests entirely with you alone. This document, while presented in good faith, does
not warrant that all attempted replications of the circuits described in it will definitely perform in
the same way as those which were investigated during the tests which form the basis for this
description.

22
Patrick Kelly
www.free-energy-info.com
www.free-energy-info.co.uk
www.free-energy-devices.com

23