# Brian's Eternal Flashlight Replication

• 1.4K Views
• Last Post 21 hours ago
Brian posted this 16 October 2021

HI Team

I arrived at this forum with the goal of learning how to make a Eternal Flashlight, but found a wealth of knowledge far beyond expectation. After much reading, watching, coils and experimenting quietly, its is now time to see if I can offer something to this effort.

I decided to start with a "Lariman" build due to its circuit simplicity but still with the necessary components to understand the coils and what is required to tune them to the circuit. I have ordered circuit boards based on the above design and have a number of spare which I am happy to post to any one willing to commit to building and posting their progress.

So I will build one up and wind some coils and post progress soon.

Brian posted this 16 October 2021

Ok - so I built up the circuit based on the standard design, but took some liberty with the coil winding based on what I have come to understand. Please tell me if I am on the right track re the coil design. The circuit shows a 28 turn primary and 66 turn secondary. However it seems that there needs to be a bucking arrangement in the secondary.

I used a C core and wound the primary with 31 turns (CW 2.1 meter / 3.67 mH). My understanding is that this should be about 1/4 the total length of the Bucking Coils (L2 and L3) which worked out to be 90 turns CW (27.2mH) and 63 turns CCW (12.94 mH)  or L3 = 0.707 x L2  as we have seen demonstrated in the Mr Preva experiment as well as other devices.

First question is am I over complicating this coil design or am I on the right track?

If so next I need to work on resonance for the primary. But as it is I note some interesting effects.

With the coils together when I power the circuit with 9V I get very little light on the LEDs. But if I slightly separate the C Cores and reconnect the circuit springs to life and the halves of the C Cores attract each other. Current consumption is about 150mA and the IC gets hot. Playing with the separation drops the current drawn and the LED intensity. If I rotate the primary C Core so it is essentially CCW instead of CW the attraction of the cores is less, consumption drops and the LED brightness drops. There is a fair degree of instability noted also.

It has been noted the L2 should be assisting L1(the primary) I am not sure from what I have found so far if the relationship should be L1 CW L2 CW L3 CCW or L1 CW L2 CCW L3 CW (where L3 is 0.707 x L2)

Next I will try tuning the primary to see if I can find resonance and lower the current draw.

Chris posted this 16 October 2021

Hi Brian,

I hope you dont mind, I moved your excellent work to its own Thread, its too valuable to be lost in a discussion thread!

I see your circuit is quite different from what I had poosted here.

I sometimes get the feeling the Circuits are not right. I also replicated the same sort of Circuit: Chris's Replication of Lari Man's Объект 013Б

I was not successfull in this replication even though I felt it would work...

I will post more soon.

Best Wishes and Thank You for Sharing!

Chris

Brian posted this 17 October 2021

Thanks Chris

Are you able to confirm if the secondary coil should be a partnered coil arrangement? I note that the exact working configuration does take some tuning. I don't understand how excess energy could be created without such a bucking arrangement.

Also of interest is how the 2 C Cores attract each other when the circuit comes to life. Is this confirmation that the secondary is assisting the primary? or is it normal for a transformer wound on C Cores?

Regards

Brian

• Liked by
Chris posted this 17 October 2021

All the answers you seek are in those pages! If you dont read the pages, youre wasting your time.

Best Wishes,

Chris

• Liked by
Brian posted this 17 October 2021

Thanks Chris

I have read that thread, but it seems I have missed a concept. I will re read.

On a positive note, I have the current draw down to 11mA with good brightness on the LEDs

I have tried several core types, C-Core, Pot Core and Toroid. To date the toroid is performing the best. All have a primary coil of around 7mH, with the secondary wound as a partnered coil with a 0.707 relationship between the 2 coil windings on the secondary.

Brian posted this 18 October 2021

A quick update.

I revisited various coils with the focus to get resonance on the primary. I managed to fry 4 x 34063 ICs in the process. Good thing they are cheap and I ordered 20 at the time as I had in the back of my mind this might happen on that circuit.

Need to give a bit more thought for the coils.

Brian posted this 18 October 2021

Ahhh - it seems I might have been over thinking it. As Chris noted in the thread attached some time 2 coils can do the job of 3

Time to wind some more coils

experiments / partnered-output-coils

Twenty years later, we are a lot more advanced and know that we must Break Symmetry, and move to Asymmetry. In other words, the use of Three Actions:

1. An Input Pulse.
2. An Output Induction Phase.
3. A Secondary Induction Phase.

Three Phases all up, making an Asymmetrical System!

We can make this happen with Two Coils, or Three Coils, each Coil having its own phase, or one Input Coil becoming an Output Coil also, making up the three Phases.

Using two Coils and three Phases is what Aukla commonly used in his lantern's:, we covered this in Self Powering Above Unity Machines

### When the Mosfet is Off:

Don't forget, Amplifying or magnifying both Voltage and Current is easy, we have covered this many times! We only need combine the two methods, and find Magnetic Resonance, and this bursts into life!

Brian posted this 21 October 2021

Hey Team,

Well I can report that I have found at least 2 dozen ways NOT to wind a coil to make this circuit work

I have tried the following configurations

Primary 28 Turns CW, BC1 42 Turns CW, BC2 28 Turns CCW on Toroid, C Core, E Core and Pot Core

Primary 28 Turns CW, Secondary 66 Turns CCW on Toroid, C Core, E Core and Pot Core

I have also reversed the legs on all combinations to see the effect and tuned with a variable air cap 430pF

I also wound the primary with more and less turns and doubled the turns on the secondary on some to get the voltage up.

My greatest success was with the Toroid and the E Core both wound Primary 28 Turns CW, BC1 42 Turns CW, BC2 28 Turns CCW. Seems this winding in general performed better. The current draw on both were 11mA.

I ask your thought on which wave form is the better of these

Or

The first is E Core with 14Vp-p and the second is Toroid with 20Vp-p

Chris posted this 21 October 2021

Hi Brian,

I believe the waveform was shown, should it not follow this shape:

At least for this circuit? Oh, yes, I see there are a few Circuit Variants, I saw this one:

and this one, which I used:

@Brian, is yours different again? Can you post please, so we have an idea on your direction.

@Other Readers, please correct me if I am wrong or have my Wires Crossed, bad acronym sorry!

Best Wishes,

Chris

Brian posted this 21 October 2021

Thanks Chris

This is the circuit wizard file I built which is similar to the second. But looking at the 2 above the seem to be the same just different layouts. I must be missing seeing some point of difference.

I did have some tests similar to above waveform around the 280Khz but they failed to produce enough voltage to light the LEDs.

Thanks I will go back and revisit them.

One further question - for D1 I am using a 1N4004 - do you think it is suitable or would you suggest something different?

Chris posted this 21 October 2021

Hey Brian,

Yes, that is the same circuit I posted here, its the same Circuit as the original Объект 013Б, I posted a copy of the original files here.

I posted my failure on this circuit here. I was not able to make the Circuit, that I replicated, work as was shown!

I believe the basic layout works, I believe the machine did run as was shown, but believe the Circuit to be fake. Not real or not accurate!

In the future, I will show this circuit running by itself.

This basic circuit is based on Andrey Melnichenko's GLED.

What would I do to the current circuit to make this work?

### This:

Of course, Q1's tunning would be critical!

Best Wishes,

Chris

Jagau posted this 21 October 2021

Hello Chris and Brian
I think the key to this circuit is to make it work in  selfrunning mode and with an asynmetric transformer.
I will post here this circuit here with an oscillator which adjusts his frequency to the coils and to the spread capacitance of the coils.
I am building such a coil with E core. Oscillator is ready and functionnal.
Jagau

Chris posted this 22 October 2021

Hey Jagau,

Yes, I agree! Magnetic Resonance is key, and the Input must be just enough to bring the Interactions of the coils into Magnetic Resonance. At peak Voltage is where we will have Magnetic Resonance.

A great deal of tuning is required for these Circuits, and if one has good SMPS knowledge, then I think that would make this all the much easier for a full circuit design, I posted a thread Here on this very topic.

I wish I did a University Degree on SMPS's and then maybe I would not have so much trouble making these small circuits work!

The Coils are the easy part!

Using various methods they are easy to make do what's required!

Best Wishes,

Chris

Brian posted this 22 October 2021

Hey Thanks Jagau and Chris

This was a really circuit was a really good exercise. Even though I was way off track I did learn a good lesson in tuning the different coils. One turn can be the difference between the circuit coming to life and DOA. It has bought home that the solution could be that next thing or next thought or maybe adding that other value cap. Persistence seems to be key. The more I persistence the better I get the feel and knowledge and better decisions are made.

I am going to put this to one side for a little and move on where I was heading with this which was Andrey Melnichenko's GLED

I have spare boards so if there are members that would like to commit to the build I am happy to send.

This circuit design is from chris-s-replication-of-andrey-melnichenkos-gled post

Thanks Chris

Jagau posted this 22 October 2021

So I redid the proposed circuit with my autonomous oscillator at very low consumption and the circuit operates at about 620Khz as the author suggests and substantially the same waveforms, but as in Akula's circuits the problem of recharging of the primary capacitor does not occur with this type of circuit.

Jagau

Brian posted this 23 October 2021

Thanks Jagau that is good information.

Are you using the 28Turn / 66Turn coil winding design or  do you have the secondary with partnered coils of 42Turn and 28Turn as it appears to be when he unwinds his coil?

Kind regards

Brian

Jagau posted this 23 October 2021

Hi Brian

I used 28T / 66T

the 66 is a POC of 33 CW and 33 CCW

Jagau

Brian posted this 23 October 2021

Team

In regards to the GLED Build what is the recommendation for the FET and Diode?

It looks like Chris used a KT805 in one design yet I see 2N3055A stenciled on the board. I was thinking of using a IRF3205.

For the diode will a 1N5817 be ok? I see a lot of these circuits make use of Schottky diodes or should I just stay with a fast diode like a UF4007?

Brian

Brian posted this 23 October 2021

Just looking at the coil winding now - from reviewing the videos it looks to me that L1 is about 26 arm lengths on the unwind, so lets say 26 meters and L2 is 23 meters with the inner and outer copper strip connected either side of the L1 winding.

I would think this will not produce the impendence recorded on the schematics.

Any thoughts? Else I think I will start with what is observed rather than written.

Regards

Brian

• Liked by
Vidura posted this 23 October 2021

Hi Brian, the kt805 and 2n3055 are bipolar transistor, not MOSFETs. There is a substantial difference in switching characteristics and a certain feedback from the coils seems to be easier to achieve with bipolar transistors ( double pulse). Schottky diodes are superior for higher frequencies in general. Vidura

• Liked by
Chris posted this 23 October 2021

Hmmm, Brian, I see your PCB's are missing the names on My PCB Design:

I wonder whats going on here?

Chris

• Liked by
Brian posted this 24 October 2021

Hey Chris - I used your supplied files unedited - it seems only the white font came out on the screen print.

I also just noticed yesterday the circuit diagram is wrong (around the LEDs) but the circuit boards seems to come out correctly.

I am new to circuit wizard and the gerber files. It took me a couple of goes to get the naming correct with JLC, but I don't think that has caused the missing text.

It is not my intention to remove credit as I respect the enormous effort you have made and still do make with this site and your work.

I do apologize for this error. And will correct the writing (change to white) and reorder with next order I place on JLC as it will be only \$4 to get a new batch with this correction which I will send to any members interested.

Kind Regards

Brfian

Brian posted this 24 October 2021

Thanks Vidura

The circuit shows a MOSFET but it is very good to know the better performance with the use of a bipolar transistor.

I will run with the Schottky diode also.

Kind Regards

Brian

Brian posted this 26 October 2021

Hi Team

I have finished the circuit and wound the Pot core with copper strip.

I did a short test wind and assembled and measured to get a rough idea of the lengths required. From this i noted a couple of points.

Watching the coil unwinding video it shoes he did about 26 arm lengths on the primary and 23 arm lengths on the secondary, which is a long way away from the 22mH and 57Mh shown on the schematic. So I decided to go the path as shown in the Chris's replication of Andrey Melnichenko's GLED group.

My first completed winding with copper strips across the primary resulted in 34.7mH Primary / 72.8mH Secondary resulting in a 88Hz  waveform on the gate tunable to about 167Hz. Calculations show that if I lower the Inductance of the secondary this frequency should drop which is not the 269Hz shown in the successful replication.

You will note also that I have no resonance waveforms on the primary. I used a 560nF timing cap here instead of a 510nF as that was all I could get my hands on. This will result in a lower frequency I believe so I will need to change that.

Next steps

• Change the timing cap
• Take off some windings to get closer to the design

Questions

• Is it confirmed that 22mH/54mH is the working design?
• Do we have to tune the capacitance also by trimming the copper strip?
• Will it work with a IRF3205 or do i need to wait for the KT805 to turn up?

Kind Regards

Brian

Chris posted this 26 October 2021

Hi Brian,

This video is a start to further understanding:

Best Wishes,

Chris

• Liked by
Brian posted this 28 October 2021

Thanks Chris

What I am not understanding is that it is clear that he has the 2 copper strips across the primary. He unwinds 23 arm lengths of the secondary and then 26 arm lengths of the primary (which would not create the inductance ratios he shows on the schematic). Additionally there is another layer of tape he does not take off and at the very end of the video when he puts the unwound coil down (which has the inner copper still on it) there appears to be another layer of tape and a further inner winding.

This inner winding could be the partnered coil of the secondary (but I don't see where he disconnects this)

The length of wire he has would also result in far more turns required to the recorded 26 or 54 mH

Am I on the right track??

• Liked by
Chris posted this 28 October 2021

Hi Brian,

Electrical Energy must be Generated. That is, Voltage is Sepperated from the Copper Atom, and the flow of Current Supported, This is opposing Magnetic Fields. Bucking Magnetic Fields.

Faradays Law, if used Symmetrically, will always be Below Unity! If used Asymmetrically, then Above Unity results can be achieved.

The Circuits we have been given by other researchers are often confusing and hard to replicate! We know this already!

Here is the video Transcript:

``````00:03
so well, that's what happened, I'll show it
00:06
The most interesting
00:21
this one is not to love
003
this now does not work at all not yet
00:40
caught up completely all the signals were checked
00:43
pieces working principle forming
00:47
low frequency resonance in or seconds
00:50
that is, here is our faces of cultures weekend and
00:55
this capacitor
00:59
that is what we have here right here
01:04
there she is
01:16
so we have a series here
01:26
we see the bursts of impulses
010
low-frequency component
01:46
so what, actually
01:49
looking for this thing
02:18
f*cking erase the monsters again I will
02:21
rewind
02:23
the general resonance of ferrite does not match for
02:28
the formation of a low-frequency sinusoid is not
021
I can raise the power
02:52
then you have them for the last time
02:54
will
02:55
saint he would understand
03:00
[music]
04:08
now we are looking out for this
05:16
now this
06:45
this is how you work``````

So, one finds the resonance Akula shows and also explains, then he says:

I can raise the power

How does one do this?

Best Wishes,

Chris

• Liked by
Brian posted this 29 October 2021

Thanks Chris

I hear what you are teaching regarding Asymmetry in bucking magnetic fields, and the need to achieve resonance for efficiencies in the machine. What is confusing is this understanding, yet watching the video of unwinding the coil appears to show 23 arms length on the 54mH coil and 26 arms length on the 22mH (which is later labeled 26mH).

Are you able to clarify how this meets need for asymmetry (as it does work), but windings appear very close to symmetrical.

I do apologize, I feel I am missing something very basic here.

I will wind a couple of different versions, focus on finding resonance and see what reveals itself.

Kind Regards

Brian

Chris posted this 29 October 2021

Hi Brian,

Electromagnetic Induction as it stands in Science today is entirely Symmetrical!

### Conventional Transformer

Primary Turns M.M.F is Equal and opposite to the Secondary M.M.F. This means Ampere Turns are equal and opposite, we see Symmetry as each Force Cancels itsself:1 + -1 = 0.

### Conventional Generator

Shaft Torque is proportional to the Magnetic Field Reluctance on the Shaft due to the Stator Coils. Again M.M.F is the force and Ampere Turns are equal and opposite, we see Symmetry as each Force Cancels itsself:1 + -1 = 0.

Shaft Torque: τ

### τ = r F sin θ

Where:

• τ = torque
• F = force
• θ = angle between F and the lever arm

If the best Generator in the world is 95% efficient, then Shaft torque, Windage Loss and other losses make 100% then we get 95% Electrical Energy, a Symmetrical System can never go above unity!

### Asymmetry or Broken Symmetry

Symmetry of M.M.F + -M.M.F = 0, Asymmetry of M.M.F + -M.M.F + M.M.F = 1, we now see an open door, where all of Science has lapsed entirely this other half of Natural Science of Electromagnetism.

Applying this to the Conventional Transformer is simply using all of the -M.M.F and canceling this entirely so it does not create a negative effect on the Input... Sounds simple, it is once one understands this!

Andrey Melnichenko gave us this analogy:

It is easy to confuse simple things! It is easy to over complicate this! Its important one thinks simple about this and applies the right steps, if one can use this following analogy:

and then apply the Balance and Imbalance of Force, then this becomes very easy very quickly!

The Secondary of any Conventional Transformer has an M.M.F that is entirely Wasted and always has been!

Using smart Switching techniques and different Inductances we can achieve the same Asymmetry, or M.M.F Imbalance!

Best Wishes,

Chris

Jagau posted this 29 October 2021

Hi Brian
Don't worry, researching isn't easy. We are all there. When you watch the inventors' videos, everything looks easy, but these inventors have found something that they will not say in all the details. It is up to us to do the rest.

What Chris teaches is a proven and working method.
The parthened output coil, P.O.C,, are part of the secret which seems to be common to several inventors.
The way in which they are used will be your little secret to discover.

Good research.

Jagau

Chris posted this 29 October 2021

Jagau Thank You My Friend!

All here should know, I still have a bit of trouble making a machine work! I can not Predict a Machine, there is some guess work! I have to feel my way forward. This is why I cant say do X, Y and Z and you will have a working machine!

We have a: Working and Proven Technology, yes we do! Many here have achieved the Goal! I am sure all that have achieved this goal will also admit that there was some work to getting a machine to work properly!

I can guide all wanting to learn to the end goal, but can not do it for the inquisitive experimenter! All answers I have, are in the pages here on this forum! Many have succeeded as a result of this easy path I have laid out for others!

CaptainLoz is one of many:

Tinman used the very same concepts:

Others here also have achieved great success! To achieve success, one must follow the path with passion and dedication! Stick to the basics as directed! Do not deviate!

The same outcome can be reached by many methods or ideas. This phrase refers to the road system of the Roman Empire, in which Rome was positioned in the center, with every road attached to it. All roads lead to Rome, so you can approach the puzzle any way you like, as long as you solve it.

Once one has done this once, it becomes very easy, but still some work is required, its not a 5 minute job to get a machine working.

Starting with an open mind and the will to learn is half the battle! Here, at aboveunity.com you have a support team that will guide you the best we can! Else where, this does not exist!

A war was won, and will be won again: 3 × 7 × 37, the AWB flag is flown here with pride! I use this set of Numbers often: |   Chris| I choose Life!

Best Wishes,

Chris

Brian posted this 5 weeks ago

Thankyou Jagau for your encouragement and Chris for you information motivation from other successes.

My thinking there should be a POC coil on the secondary for this to work, but the video of the coil unwinding does not agree and shows what appears to be a symmetrical coil of about 23 arms length each (it is 26 and 23 or if I approximate it in meters it could be around 20 meters each - but again no way to verify other than a estimate of how long each pull is) This symmetrical arrangement does not agree with the schematic or Chris's proven teachings. The other point causing me some concerns is that it appears to me no apparent capacitor is in a resonant circuit with either of the coils. The 510nF is a timing cap for the MC34063 and the 100uF could have a effect between the two coils, but I am unsure how to calculate this. So is the resonance achieved by the capacitance strip and internal capacitance of the coils.

Another point of note in all of my configurations Q1 does not seem to have any effect if I remove it from the socket, even though it is switching.

For the record which may be of use to others these are the configurations and thoughts I have progressed.

My base here is to accept the core unwinding video as accurate and attempt the replication. All are wound on a 45mm pot core. Yellow trace is on CN1. Blue trace is on CN3.

Test 1 - 20m / 20m with copper strips connected either side of the primary.

On a positive here the frequency is in the ball park and CN3 waveform looks like video, but the voltage need to be increased.

Test 2. 20m / 28.26m (20 /.707) with copper strips connected either side of the primary.

To increase the voltage

Voltage up a little, frequency still ok, but still no resonance.

Test 3. 14.14m/28.28m (in attempt to increase voltage further, had to reduce primary as could not fit any more wire on core) with copper strips connected either side of the primary.

Not much of a change here - expected more voltage

Test 4. 7.07m/28.28m ( a further attempt to increase voltage while trialing to see what effect 1/4 length produce) with copper strips connected either side of the primary.

This is probably the least successful as all other tests were run with 4 volts as the supply, needed to increase to 6 volts. A very slow waveform of 22.4Hz - not the direction required.

At this stage not really sure what to try next??

I feel I am a long way from where I need to be. I will continue with this project as others have succeeded with this circuit, so there is a way.

Kind regards

Brian

• Liked by
Chris posted this 5 weeks ago

Hi Brian,

Sortta getting the feeling youre trying to be a bit difficult and trying to avoid, at all cost, all the information I have given!

While replicating Other Researchers Circuits I have drawn up, they are just that, someone elses!

Your Scope shots have no meaning, they tell us nothing, what are they showing?

I am going to put you on Moderation and you can PM me and we will discuss when you have actually read and start following what I have given!

When you want to become a contributing member and follow the work we have all proven, then youre welcome to start posting again.

Those that wish to succeed will put the work in, and then, succeed! It all starts with Research and then Experiment, confirming the Research! My Pages have all answers needed by any good Researcher!

Best Wishes,

Chris

YoElMiCrO posted this 5 weeks ago

Hello everyone.

First of all apologize for not being active
for a long time, work issues because
these are difficult times.

Chris the capacitor value is wrong, it is not 510nF is 510pF
since it determines the maximum time for Ton.
Ton=(C/0.00004).
The MC34063 is a hysterical DC/DC converter,
this means that you can vary the two times Ton/Toff and by
consequently their frequency of work.
The maximum duty it is capable of generating is approximately
85%, ton duration is controlled by the control loop and current
that senses between the pins 6/7 through the series resistor.
This current is determined by the voltage drop between these terminals and

That is why their waveforms have nothing to do with the replications made.
Brian, it is the capacitor of 100uF that when it is at its maximum negative if
it is placed in parallel to the input capacitor this recharges it.
It is more complex than what I expose here, but it is a good approximation.

Thank you.

YoElMiCrO.

Chris posted this 5 weeks ago

Thank you YoElMiCrO!

Nice Catch, I will correct my Circuits first up in the morning!

Best Wishes,

Chris

Brian posted this 4 weeks ago

Big thankyou YoElMiCrO for finding my mistake. I note alternate circuits use a 330pF cap for timing the MC34063.

I have replaced with what I have available atm which is a 560pF. This has made a significant difference as you would expect

Because of my fiddling as per above the closest coil I have to the 22mH/54mH spec measure currently 20.2mH/55.3mH with inner and outer copper strip connected across L1. I note also that the  Following is its result. Yellow trace is CN1. Blue trace is CN3.

I now have some resonance appearing L1 and only required the drive voltage to be 2.5 Volts which is down form 4V.

Now heading in the right direction at least.

Kind Regards

Brian

Chris posted this 4 weeks ago

Hello Brian,

Great Work Thank You for Sharing!

The key is: Don't give Up!

Get the right Amplitude and the right Frequency, and the Resonance Required, and the rest will fall into place! With little steps, make progress until this becomes an: Eternal Flashlight

This is, of course, part of Research. Replicating takes some work and sometimes solving possible errors made on the part of others is also required, unfortunately!

Thank You also to YoElMiCrO!

It is somewhat disappointing that it has taken nearly 2 years to pickup on this Error I made! I am sorry to everyone for this mistake!

Your Coils, try not to rewind, just add / remove turns as required. Its much easier that way! Try complete Coil swap between each Coil Terminal Blocks, and single Coil polarity changes to find the best performance. Changing values of components a little bit at a time makes for Steps in the right Direction.

Best Wishes and, after our chat, welcome back onboard!

Chris

P.S: I believe your Inductance is currently to low, small increases in Inductance will give you better waveform, in my opinion.

Brian posted this 4 weeks ago

Hi Team

With the recent correction I have been working on small increases in inductance as per Chris's suggestion.

Thanks Chris for posting on your thread and example of the waveform to work towards.

The following is L1 26.8mH / L2 58.7mH on a 45mm OD PotCore. Copper strip either side of L1

There is additional noise on Blue trace when I connect the probe to CN1, so frequency shown is not correct, if you calculate via period it is actually 172Hz as per the second shot below

Yellow trace is CN1. Blue trace is CN3.

Same Core as above but Yellow trace is CN2. Blue trace is CN3.

The timing and amplitude of CN2 and CN3 is not too dissimilar to Chris's results, but still yet to achieve good resonance.

Further increases does improve CN3 waveform.

L1 29.9mH / L2 59.4mH. Yellow trace is CN1. Blue trace is CN3.

Looks like I need to add more to L1.

During the process I have also tried cores without the copper strips which seemed to have little effect on the waveform, but may be important once I get to the sweet spot.

Kind Regards

Brian

Chris posted this 4 weeks ago

Hi Brian,

Thanks Chris for posting on your thread and example of the waveform to work towards.

Little more work:

I did not say anything else, no claims or example made here, so please do not jump to conclusions. All I did was show my Waveform.

I am happy to try and help, but I do not wish to be taken out of context every time I post something. I get that you most probably meant nothing of it, but want to be clear, I am not making any claim or example here.

Best Wishes,

Chris

P.S: Thanks for sharing! Excellent Progress I see!

Brian posted this 4 weeks ago

I do apologize Chris

Kind Regards

Brian

Chris posted this 4 weeks ago

Hey Brian,

I am going to speculate, I can not prove this, yet I have seen results, most of the time.

1/4 is 0.25, now the MC34063 IC has a timing inside the chip. again all this is speculation, but I believe, the chip switches at Time t, so Coil L1 must match the 1/4 Wavelength, and L2 is a Harmonic of the Fundamental. We have covered Antenna Theory already here.

Lets look at Akula's figures:

So, lets ask, what's the Output Impedance? We can only guess on statistics, lets use this chart:

Back to back, we have 30 Ohms approximately right?

Lets use the Aboveunity.com Member Calculator to see the Coil characteristics:

So, turns ratio, are 1 : 1.414, so we could say, 1 / 1.414 = 0.707 and 1 - 0.707 = 0.29278, so not quite our 1/4 wavelength... Which is 0.25. Confused?

I really wish someone would chime in with more experience than me to help and correct me when I am wrong!

There is some guess work involved, that's all there is to it! I am no expert, that's why I said what I did, I am no expert, I have to guess on things. I cannot set any example if I do not have all the answers, I can only try to help.

I am sorry, my lack of Electronic Skills, and lack of answers gives me, and all of us, a disadvantage, but I am fairly confident that's part of the answer.

Can anyone point out the issues I have with my Waveform?

Best Wishes,

Chris

Brian posted this 4 weeks ago

Hey Chris

Amazing that you bring that up. After re reading some of you material, re winding with 1/4 wavelength in mind was my next step.

I arrived at this a little more simply, but I do like your considered approach better

My conclusion came from your information on how to think about magnetic resonance with reference to wire length.

Kind Regards

Brian

Jagau posted this 4 weeks ago

Hello Chris
What does it mean by

the MC34063 IC has timing inside the chip

It is the capacitor on pin 3 that determines the timing of the MC34063 it seems to me?

Jagau

Jagau posted this 4 weeks ago

Hi Chris

For your waveform of an LED where your red arrow is located we are in the linear region of the curve when we know that an LED is not linear it is only a linear region that interests us in fact that which one can foresee a certain linearity.

For the impedance we must not forget that 'we only speak with sinusoidal shapes and different frequencies to calculate the impedance I am sure you know that. For the calculation of the LED it is a resistor (R) and not an impedance) Z).

I'm sure you knew that, maybe I misunderstood your question please explain?

Jagau

Chris posted this 4 weeks ago

Hey Jagau and all,

Yes, technically I am not correct, I have misspoken, so I need to put that bit better!

Pointed out the timing here and various other places, the MC34063 is a PWM Converter Chip, on the IC Schematic, you can see:

We can see that Pin 3 is the Timing Capacitor.

So, my statement:

now the MC34063 IC has a timing inside the chip.

Perhaps should have been: "Pin 3 has a Timing Capacitor that controls the IC Timing or PWM in combination with other features"

I hope this is better and more precise?

It is the case that Readers will need to read between the lines a little, to understand some of the comments I make! I am not always entirely fluent with 100% of the statements I make, although I do make effort to be. If I am tired, then sometimes some things are not always 100% correct, so please read between the lines somewhat!

Best Wishes,

Chris

Chris posted this 4 weeks ago

Hi Chris

For your waveform of an LED where your red arrow is located we are in the linear region of the curve when we know that an LED is not linear it is only a linear region that interests us in fact that which one can foresee a certain linearity.

For the impedance we must not forget that 'we only speak with sinusoidal shapes and different frequencies to calculate the impedance I am sure you know that. For the calculation of the LED it is a resistor (R) and not an impedance) Z).

I'm sure you knew that, maybe I misunderstood your question please explain?

Jagau

Yes, Jagau, I did not mean that Waveform, that's from Manufacturer website, not my waveform.

I meant the waveform from the oscilloscope screen shot I posted, none of the others. That's why I said: "Can anyone point out the issues I have with my Waveform?"

I meant this one:

That's mine.

Best Wishes,

Chris

Jagau posted this 4 weeks ago

Thank for the clarification

Jagau

Brian posted this 4 weeks ago

Hey Chris, I am not sure of the relevance, but the frequency at around 95Hz is a fair way short of the 269Hz shown in Akula's video.

Could this be corrected by reducing the timing cap?

Kind Regards

Brian

Chris posted this 4 weeks ago

Hey Brian,

Experiment is always the best practice to learn, so I suggest do what you feel is best!

If it was me, I would reduce Inductance first, changing the the Cap will effect the rest of the Circuit. I think the Circuit is pretty close to complete. You have all the right Values of Components?

There is a relationship between the coils as pointed out. Focus on that when winding or modifying  the Coils.

An _ _ _ _ _ _ _ must be tuned for optimum Frequency Response!

Best Wishes,

Chris

Chris posted this 4 weeks ago

An _ _ _ _ _ _ _ must be tuned for optimum Frequency Response!

An Antenna must be tuned for optimum Frequency Response!

Coils are actually Antennas that have been coiled for convenience! - Don Smith!

The answers are clear if one follows the correct path! This is a matter of Free Will, do you choose to learn from those before us?

Best Wishes,

Chris

Brian posted this 4 weeks ago

Hey Team

Thought it might be time for a update.

Over the past several days the experiments have followed the line of tuning L2 to a length proportional to L1

I measured the length of L1 including all tails to get the total aerial length (so to speak) and treated L1 as 1/4 wavelength. L1 is about 26mH. As a note the inductance of L1 changes slightly with different windings on L2 (open circuit). Also copper strips are connected either side of L1 and are located on the very inner and very outer of the windings on a 45mm Potcore. Supply voltage 3V.

The measurements of each experiment are as follows:

L1 24.2mH (1/4 wave) / L2 302mH (4/4 wave)

L1 26.8mH (1/4 wave) / L2 187.4mH (3/4 wave)

L1 23mH (1/4 wave) / L2 71mH (1/2 wave)

Unfortunately for whatever reason the pic of the full wave L2 did not save, but the result was unsuitable.

The Above is L2 3/4 wavelength wire

The above is L2 1/2 wavelength wire. (the bursts of resonance are more tightly packed than either 4/4 or 3/4 wavelength winding experiments - I think this might also be showing a limitation of my scope when storing snap shots)

Also to note my circuit is using a 330pF timing cap currently as I did not have a 510pF and I note that several other circuits using the MC34063 are using this value. (a 510pF cap is on its way). I also used a variable air gap cap to explore the effect of changing the 510pF timing cap from 330pF to 840pF. There were no improvements to note with the variance, but this may be down to length of wires and the air gap cap itself.

My feeling is the 1/4 wave - 1/2 wave L1 - L2 relationship was working better on this core as additionally I could drop the input voltage to 2.2 and still brightly light the LEDS and measure 2.8V at the 2200uF cap.

In conclusion I feel that finding the correct L1 inductance and winding L2 to double L1's wire length may be a step in the right direction. Finding the sweet spot value for L1 is the next task.

Chris I am interested to know the answer to your question "Can anyone point out the issues I have with my Waveform?"

Kind Regards

Brian

Brian posted this 3 weeks ago

A quick update.

Rewound L1 to 27mH and remeasured length.

Rewound L2 to double L1 length including tails - 98.1mH

Have lost those tightly packed resonant areas on the waveform but have a increase in voltage on CN1 and CN3

I think I will wait until I have the 510pF timing cap to install to ensure the circuit is an exact replica.

Chris posted this 3 weeks ago

Hey Brian,

You said:

Rewound L2 to double L1 length

This gives a Turns Ratio of: 1 : 2 Right?

I calculated for you:

So, turns ratio, are 1 : 1.414

Which can be factored, and worked out using the Greatest Common Factor, for complete turns. Which are?

Anyone?

I have said, attention to detail, 5 more minutes focusing on Theory can save Hours on the bench Winding Coils. Failure comes from Incomplete understanding, mostly. There are inaccuracies in my post Here, but I do give you a great deal information! That's all I can do to try and help!

Best Wishes,

Chris

Brian posted this 3 weeks ago

Thanks for the clarification - yes I agree I did not understand your instructions

So, turns ratio, are 1 : 1.414, so we could say, 1 / 1.414 = 0.707 and 1 - 0.707 = 0.29278, so not quite our 1/4 wavelength... Which is 0.25. Confused?

I mistook this calculation as proof of a 1/4 wave to fullwave relationship of L1 / L2 lengths.

Thankyou for that clarity.

Kind regards

Brian

Jagau posted this 3 weeks ago

Hi Brian
For the lamp I worked a lot with a first Shark circuit. It is easier to understand and experience with . I have had great success with this arrangement and even today I am experimenting with this circuit which is composed of two BJTs in SCR mode and a third PNP transistor which resets C1 when the switching comes its turn.

Jagau

Brian posted this 3 weeks ago

HI Team

Sorry for that little side diversion. On a positive note I have shown the results of the relationship of L1 / L2 with half, three quarter and full wave windings.

So back to where I left off several days ago.

My best result was a L1 29.9mH - L2 59.4mH which via the calculator works out as 1 : 1.41, very close to what Chris calculated above.

I have adjusted now to as close as I can get to to 27/54  L1/L2 proposed values.

Below is L1 27.1mH - L2 54.2mH

It seems from these results there is better resonance when L1 is higher.

My question now is - with this circuit how do we arrive at what we should aim L1 and L2 to be? If we are relying on the copper strips to create the capacitance for the resonant LC circuit at L1, I am not sure how to measure this, as if I put my meter on it I get a negative result on the meter??? Which makes no sense to me.

Is there a solution to how to calculate the inductance of L1 for this circuit.

It seems from this if L2 is wound with 1 : 1.414 turns ratio then we are in the ball park.

Kind Regards

Brian

Chris posted this 3 weeks ago

Nice work and Thank You for Sharing Brian!

Its not hard to see you're heading in the right direction! IC Current is something to look at, did you look at the links to the calculators I posted Here? Of course if the Current Draw is too much, the Circuit wont self run! A factor to look at!

Best Wishes,

Chris

Brian posted this 3 weeks ago

Hey Chris in regards to current draw seems minimizing Q1 on time seems important here. I don't understand or observe the pumping effect of this part of the circuit and it does not seem to have a positive effect with my tests. If I take Q1 out the current drops yet the wave forms remain the same. Made me think I had something wrong with this part of the circuit yet I can see Q1 switch with a probe on the Drain. I can save 20mA without Q1 and CD4069.

I understand that there is a purpose for this even though I have not been able to observer it.

In regards to progress with the coils, I note my best results with L1 37.1mH - L2 74.9mH which supports the 1:1.414 turns relationship. Current draw of the circuit at 2.2V was 22mA or (10mA with Q1 out)

In general lower frequencies seem to have lower current draw requirements which also result in lower voltages differentials at CN1 and CN3 which seem to match the calculators prediction for current if Vin and Vout are less different.

L1 37.1mH - L2 74.9mH above

Kind regards

Brian

Brian posted this 3 weeks ago

Hi Team

Spent some more time on this project slowly increasing the L1 inductance while maintaining the L2 1:1.414 ratio.

Unfortunately I have not been able to improve performance and lower current draw any further. Current L1/L2 IS 41mH/82.1mH.

A consistent observation is that less current is drawn if tuned to a lower frequency. Also taking Q1 out of its socket has no effect on CN1, CN3 waveforms but reduces current draw about 20mA.

There is something I am missing regarding bringing in the function of Q1 but at this stage I am not sure how to activate it.

My last increase to 41mH / 82.1mH seems to support that increases further is not an advantage. Below are both the same core with differnet tuning to demonstrate the lower current at lower frequencies.

188mA draw at 124Hz

58mA draw at 57.5Hz

Testing indicates that L1 around 29mH has to date produced the best performance. I believe a better understand of Q1 function and how to bring it into effect is required to advance further.

Kind Regards

Brian

Brian posted this 3 weeks ago

Thanks for the info above Jagau.

I will take your suggestion to explore that build to gain further understanding.

It is a little hard to tell if it is wound on a toroid ferrite core? I would expect not.

Kind Regards

Brian

Jagau posted this 3 weeks ago

Hi Brian

For this model I don't think there is a core.
These are rather a coil wound horizontally with another vertical coil (90 degree)
This is an interesting arrangement to experiment with that becomes an asymmetric transformer.

Several inventors have succeeded with this type of arrangement. This is what we are going to find out.

Jagau

Brian posted this 2 weeks ago

Hi Team

Spent several days working with the build Jagau  referred.

It has been very interesting studying the performance of these coils, but as yet I have not been able get the circuit to oscillate.

The Coil has 4 turns on the circumference of a 90mm diameter PVC core (storm water pipe) In a larger wire (blue), then the secondary (red) and sense (black) coils wound over perpendicular to the primary. Makes for a nice looking machine.

On connection with the circuit in the above posted configuration I am unable as yet to produce oscillation. I am thinking the bias conditions on the BJT's are not quite right so need to spend some further time investigating this.

I have connected a signal gen and swept to view its performance.

I noted in a video I viewed on this device that Akula made a reference of 50 - 150Hz, but this seemed too low to me to have any affect in this type of coil and applying these frequencies basically resulted in nothing on the secondary. But above 1Khz we do get results and above 1Mhz they are significant. Interestingly over 100khz a resonant pulse begins to appear which consists of the applied frequency within a wave vibrating much slower. ie less that 500Hz.

Blue trace is signal seen across the primary with a 3Vp-p 100Khz applied. Yellow Trace is signal of Secondary loaded with a 10K resistor.

Next step in this project is to work out why the circuit will not switch in its default configuration.

This has proven to be a bit of a diversion from the original goal, but is interesting to experiment with these alternate transformer designs.

Kind Regards

Brian

Jagau posted this 2 weeks ago

Hey brian
I like the efforts you make and I will give you a very simple little circuit to achieve. As soon as you connect the battery it will oscillate on its own (selfrunner) as long as there is juice in the battery, moreover with a small modification it will automatically adjust to the resonance frequency of the coils that are connected to it, I hope you noticed the resemblance to that Akula,

I wish you good experiences
Jagau

Brian posted this 2 weeks ago

Hey Jagau,

Thank you very much for that circuit. I will get to that in the next day or so.

Given my basic electronic skills I would propose that the inductor in this circuit be replaced with L3 as this would then switch at resonance of the coils. But L1 needs to be powered, so would L1 be in series with the 40V power? Still this does not seem right??

In regards to Akula's circuit I don't see with this default configuration how any transistor can be biased to operate unless there is a significant resonant kick the instant the battery is connected, which seems unreasonable to expect as I have observed that these coils need a much higher frequency to operate. So a single 3V power on spike would not be sufficient. A little work to be done here also.

Kind Regards

Brian

Brian posted this 1 weeks ago

Hi Team

Have spent some time with your circuit now Jagau which has provided some unexpected insights.

Basically I built your circuit with some substitute TO220 transistors (TIP31 and TIP32) which I had in my parts. Wound a 401uH Inductor on a C-core and powered with a 1.5V power supply.

The immediate result was nothing. My first thought was to order some of the designated transistors, but I decided to have a bit of a fiddle first. As it turned out if I shorted the collector / emitter on the PNP transistor to bias on the NPN and get some current flowing in the inductor, it gave the kick it needed to start oscillation. Well this was interesting.

So I then tried my Air Transformer from Akula Lamp.

Nothing again even with the kick.

The inductance on L2 of this  transformer is 14uH, a fair way from the design parameters. But by chance I noticed with a probe on the collector of the NPN transistor, I could short the E/C of the PNP and kick the circuit to resonance. So this gave me the idea of adding some capacitance across the Collector / Emitter of the NPN transistor. Turns out 10nF here was all that was needed. It can now run L1 (7uH) and L2 (14uH) of the air coil as long as a short / kick is provided, but will self start without the short / kick with about 200uH and above inductors connected. All scope shots are taken with the extra 10nF cap connected.

This exercise has provided some new investigation paths to progress work on Akula’s Lamp version.

Great suggestion Jagau!! – Thank you very much.

Kind Regards

Brian

Jagau posted this 1 weeks ago

You learn fast Brian bravo.
I use a 200nf for mine, but if with 10nf it does keep it like that.
I like to share with those who do not let go and experiment, we are here for that.
Getting ahead and experimenting is what is interesting.

Try this one it takes off when you adjust the 5K pot you can adjust the current you need,you will love it

thanks for sharing
Jagau

Brian posted this 1 weeks ago

Thanks so much Jagau.

Sometimes I feel as if I am getting distracted form the original goal, but this stuff in invaluable on that journey.

Kind regards

Brian

Jagau posted this 1 weeks ago

Alright, if you consider that a distraction don't use it sorry for distracting you
Jagau

Brian posted this 1 weeks ago

Sorry Jagau - It was meant as a compliment and a thankyou for sharing very valuable information.

My meaning was that on a journey we set out in one direction not knowing what we need to learn. Having generous people, such as yourself, offer insights that the novice at their point in the journey is yet to understand its importance is a true gift.

I have built your latest circuit above and note how efficiently tunes to the inductor of the circuit over a very wide range. It seems to work much better on inductors with ferrite cores than air cores (or maybe it is down to how I have only that one air coil atm to test with). I had been experimenting with various configurations to explore if it settles on the natural resonance of the inductor. I have not yet concluded on the answer of that yet.

Please accept my apology. I do truly appreciate your advice and in no way meant any disrespect. The comment about me getting distracted is a observation and admission that at times we do not know what we need to know. If we are really fortunate, caring people will present this knowledge if we are open to receive it.

I hope my persistence and progress will regain your trust and thank you again for your support.

Kind Regards

Brian

Jagau posted this 1 weeks ago

ok everything is back to normal.

Another little trick, try it with 3 coils (same number of turns each) including 2 in POC on the last circuit I sent you

and harvest on the third only,

but watch your fingers it will be high.

Jagau

Brian posted this 6 days ago

Thank you Jagau for your understanding.

How many turns do you suggest.

I have started with 50 each on a C Core with all coils being separate (IE not wound over the other)

Tried to tune one POC coil to resonance with a Cap and then harvest off the other POC with the common diode, cap, LED arrangement used.

My thinking is POC coil in resonance assists the primary. The result is about 5ma current draw by the circuit and LEDs well lit. But trying to drive the circuit off the power in the cap is not successful.

Is this the correct approach?

Kind Regards

Brian

Jagau posted this 6 days ago

Hi Brian
With the last circuit that I proposed to you you don't have to worry about the resonance, the way I built it it adjusts itself to the resonant frequency, it is a self runner oscillator (SRO) as long as 'there is juice in the battery.
It is very important that the 3 coils have the same number of turns, with 100 you will have a lot of success.
You did not add a capacitor you know the coils are also capacitors.
Just adjust the 5k pot to get the desired output voltages, do not complicate the circuit as I designed it and have fun with it.

Come share with us.

Jagau

Fighter posted this 6 days ago

Come share with us.

Hi Jagau,

You're moving too fast

It took more than a week after ordering because the courier companies were overwhelmed with this Black Friday craziness.

I'm trying to catch up, hopefully I'll have some time in weekend.

Regards,

Fighter

Brian posted this 6 days ago

It is very important that the 3 coils have the same number of turns, with 100 you will have a lot of success.

Thanks Jagau - I was concerned that I may not have had enough turns.

I was not sure how to arrange the partnered coils for optimal output. I did try a diode as per below, but this killed the resonance.

I have also tried the POC coils in series and flipped, also with the load arranged as below

Seems next step is to increase turns and continue to experiment.

I do really like the circuit, it is very adaptable and efficient.

Thank you again

Kind regards

Brian

Chris posted this 6 days ago

Well done Brian, I see youre paying attention and learning quickly!

Keep this up and I will put you forward for The Elite Builders Club!

In this configuration, Resonance is Peak Voltage, Current and thus Magnetic Field, for a given Frequency and Duty Cycle. We have more to share on this soon!

Best Wishes,

Chris

Brian posted this 6 days ago

Thanks Chris

The support and guidance here is amazing. I mentioned previously I get sidetracked, but not in a bad way. The amount of information provided here, and various adaptions of theory temps me to try little side projects, which can tend to blend in with one another. Every day I discover something new and important on your site and add it to my must do list. I appreciate the collective knowledge born from much work and persistence, and all the team with the wealth of this knowledge offering encouragement back to a proven method.

Kind Regards

Brian

• Liked by
Brian posted this 3 days ago

Hi team

A short update

Continuing the experiment with the Circuit supplied by Jagau I have rewound the coils as 3 separate coils of 100 turns on bucking POC configuration on a C-core. Also applied a load to a one of the POC coils as suggested.

I also introduced a 0.1 Ohm resistor in series with L3 to measure current flowing in the load.

Yellow trace is across 0.1 ohm resistor, Blue is on Collector of Q2

The measured DC current drawn by the circuit is 158mA (via multimeter).

I am a little concerned that I have a earth loop issue measuring the current as a average current of 230mA calculated form the Vavg of 23mV does not seem to practically hold true as this should be enough current to self run the circuit. It does not.

LEDs disconnected to reduce current consumed by the load has no additional effect.

Also I tuned the circuit with a 2.5V supply and with one LED in the load the cap then charged to 2.54V. With this matched voltage I connected to across supply and removed supply. Still it seems the current was not enough to sustain.

I need to consider alternate methods to self power and recheck my measurements and calibrations.

Kind Regards

Brian

Brian posted this yesterday

Hi Team

A quick update

As suspected I did have a earth loop creating the measurement error.

To eliminate I am running the scope off a battery for these measurements. It is also showing some limitations of measurement sensitivity of my cheap scope. I am using a controlled DC power supply for the supply as it has a handy current display as well as voltage control.

Still it seems I am not getting enough current to self run. So considering options I could reduce the secondary turns or increase the primary. The resultant drop in voltage should still be plenty. I respect that Jagau did advise to have the same number of turns on all coils. Please forgive this little side experiment Jagau - I will increase the primary as this is the easiest for a quick test.

Kind Regards

Brian

Brian posted this 23 hours ago

Hi Team

Well doubling the turns on the primary did not provide the result I was looking for. Is this due to the now different rise/fall time of the relationship between primary and POC coils?

It feels to me that I am not getting enough "smashing" of the magnetic waves in the POC to create the excess energy required. Do I need to increase turns on the POC coils now to match the primary? If so will this result in higher excess energy? That is using 200 turns on all coils instead of 100 turns.

Is the ferrite C-core I am using suitable or should another core be used. Also is the the 0.35mm copper enameled wire suitable or should it be thicker? I see various devices from many builders such as Akula or Andrey Melnichenkos using what appears to be similar cores and wire gauge.

I accept also that the approach I am taking to self power may be faulty thinking. Basically I am matching the voltage on the cap to the input voltage by the load LED arrangement across the L3 coil, then connecting to the power input, retuning a little as the load is now changed, then removing the power source.

At this stage I am not really sure what is the next step to reach this threads goal of an eternal flashlight.

Kind regards

Brian

Jagau posted this 22 hours ago

Hello Brian

On page 49 and up it's going to get even hotter.

Jagau

Brian posted this 21 hours ago

Thank you Jagau

Kind Regards

Brian

Members Online:
Since 03.08.17 - Widget: 27.11.18
Recomended Protocol:
Our Above Unity Machines:
Start Here:

More than anything else, your contributions to this forum are most important! We are trying to actively get all visitors involved, but we do only have a few main contributors, which are very much appreciated! If you would like to see more pages with more detailed experiments and answers, perhaps a contribution of another type maybe possible:

PayPal De-Platformed me!

They REFUSE to tell me why!

We now use Wise!

Donate

The content I am sharing is not only unique, but is changing the world as we know it! Please Support Us!

Perhaps a Bank Transfer?

• Bank: Bank of Queensland
• Name: Christopher R Sykes
• BSB: 124-001
• Account: 21580359

Thank You So Much!

Weeks High Earners:
The great Nikola Tesla:

Ere many generations pass, our machinery will be driven by a power obtainable at any point of the universe. This idea is not novel. Men have been led to it long ago go by instinct or reason. It has been expressed in many ways, and in many places, in the history of old and new. We find it in the delightful myth of Antheus, who drives power from the earth; we find it among the subtle speculations of one of your splendid mathematicians, and in many hints and statements of thinkers of the present time. Throughout space there is energy. Is this energy static or kinetic? If static, our hopes are in vain; if kinetic - and this we know it is for certain - then it is a mere question of time when men will succeed in attaching their machinery to the very wheelwork of nature.

Experiments With Alternate Currents Of High Potential And High Frequency (February 1892).