That motor won't help either. I'll explain.

If you use an inductive load on the tesla switch, you have a BEMF effect that will reduce the collected current. Transformers are the worst. Motors are bad but they have less of an effect as only one side of convetional motor makes the BEMF. Here's why..

I probably need to diagram this..but

In regular motor you have 2 poles present from permanent magnets. Your winding on the rotor create the other 2 poles for attraction or replusion or both in one pass. You could have 1 of three combinations in motor.

Replusion motor like pulse motor create the highest amount of BEMF do to the inductive spiking. The simplest form of this is a bifiliar pulse motor with the scondary windings disconnected from a power source and hooked to bridge rectifier.
The BEMF left over after the coil is charged creates a false ground. If this type of load is put on a ONE WIRE coil that energy is still present and has to be disapated by the incoming energy. This creates a ground scenerio. In which your loosing energy.
Pushing away from Multi poled motor acts the same.

An Attraction/Replusion motor Pulls the coil into place then switches the polarity and pushes it away. This also creates an equal amount of BEMF, BUT... Some BEMF does not show up under a Tesla switch. The BEMF produced will actually flow out of the system without disapating our current. But you still get a fair amount and it still causes this ground scenerio.

A pure Attraction Motor performs the best. The coils in the rotor are always charged opposite of the magnet. And so do to the state of the communtator they can never align but are always trying. Pulling in....
This produces the lowest amount of BEMF. Only the coil that is trying to line up to the north magnet creates BEMF that will not flow. The BEMF created on the south poled magnet does not show up in negative way. It actually adds voltage to the system rasing the potential in the coil.

I should probably add that that this is how I see it working with the aid of Oscope. I also measure the battery voltage and line amperage to help determined the more efficient of the setups.

The best possible Tesla switch/motor combanation I have been able to think of and test is a standard tesla switch driven with relays and the motor is permanent attraction but with the same pole on both sides of the motor.

I used 2 south pole magnets for stators and 2 coils independant of each other, both pulling into the magnets both charged to the north. The other thing that matters is how you wined the coils.If the switch is running the load from the ground side, You want a counter clockwise winding. The power from the ground side of the batterries goes into the start of the coil... charges the coil...then exits from the end of the coil back into the power recovery side.
I've only been able to test it once, It caused a good charging effect in the batterries. And the batterries maintained a room temperature. But the motor failed early and I haven't built another one yet.

Right now I am trying to work on semi solid state switching with high power transistors. This has turned to a whole new mess that kept me scratching my head for a couple of months now. But I got it worked and soon as its built and stable and tested, I'll post it. Then I'll rebuild the motor.

A big thing to understand in the Tesla Switch is the fact that it makes any load an open loop system. But you have to watch what and when enviromental energy flows into the system. Negative energy and Positive energy from the electron cannot co exist on the wire at the same time. They cancel each other out. But the positive energy from the HOLE current doesn't seem to have this effect when negative energy enters a system. They seem interact well. Maybe because they are all really the same thing, I don't know. You got to find that combination of Load and system that allow for the enviromental energy to stay in the system with out disapating the positive energy your using to do work with. Or ideally do work with the negative energy and gain more everytime you do. but....

Hope you guys follow what I am saying..

Cheers
Matt

Hi, I was wondering if you have had any contact lately with Jack? I was following his progress for a while and then he started to have problems typing messages and then just up and quit posting. Has not logged into OU for some time as well. He had a sweet piece of work for sure.

thaelin

Nope not a word from Jack. It has me wondering whats going on there. But his concept is pure genious. Who would have thought it would take less energy to direct the flow of magnetic flux as it does to create the magnetic flux in the first place. It definetly has me thinking that even if we get the best energy system that the motors are so inefficient that it would just be a waste. But then again this motor has a lot of merit to it and it should be expanded on. Like I said I tried to post about it on this forum and didn't really get any hits at all. No one had any real comments on it or for that fact any improvements to mention. I mean the motor is super simplistic it needs more to it to make it worth the investment to build. The way I see it it would cost about 2-300 to build a prototype with all the best stuff like silicon steel and neo52 mags but no one has tried to replicate this on here.... Hmmm I wonder why?
Well this has nothing to do with the Tesla switch so If you would like to comment or even add to my post please do.
Anyone know the inventor of...

All right, I admit. I was a wrong about not being able to explain more efficient charging, when charging in series with classical models.

However, that does not mean this principle is useless at all. After all, we can conclude it *is* more efficient to charge capacitors, as well as batteries, in series.

Having said that, the other half of the story, charging with radiant energy, will also be more efficient when charging in series. Therefore, I believe this principle will prove to be an important key into turning devices harnessing radiant energy that are on the edge of COP 1.0 into COP > 1.0 devices.

As I explained in previous posts on this thread, radiant energy travels as waves outside of our circuits, which means we have to consider our circuit as a sort of wave guide when thinking about the phenomenon assiciated with radiant energy spikes. As I explained before, diodes or transistors are not fast enough to switch or otherwise control the flow of radiant energy, beside that they operate on the energy flow *inside* the circuit only...

Therefore, the answers to controlling/designing radiant systems lie in the spacial design of the circuit. By using different lengths of wires, one can control the relative timing of when the radiant spike arrives at a certain point in a circuit.

From this line of thinking, I came up with a circuit design that might be used as a basis to charging capacitors in series and discharging them in parallel.

The scetch shows a bunch of capacitors in series, with some resistors in between. When the transistor is open (non-conducting), the capacitors are in series. When the transistor closes, drawing current, the capacitors are in parallel due to the diodes. Of course, the transistor will have to be open well before the moment the radiant spike leaves at the coil.

The question of course is: do we gain anything with this, or will the additional loss resistors spoil the whole game?

Attached Images

Scetch_parallel_charge.pdf (236.2 KB, 53 views)

What about shunting the bemf of the motor back to charge a cap then recycle it into the motor so that it doesn't even affect the tesla switch? It could even be a part of the motor cicuit since some motors need a driver cicuit anyways it would make what comes out of the switch exactly that and non reflective as bemf from the motor get reused.

The problem there is how to redirect it. About the time it shows up, it wants to head towards the higher potential.
I have several pulse motors that are open loop and do switch off anything that shows up after the initial pulse including the energy I put in. But in this case none of those motors perform as expected. You would expect them to perform as a normal power source. But it doesn't.

Like I said to generate any power on the shaft with it you have to generate a BEMF that will flow out of the system and ad additional power. This can be done and it should perform as would be expected from a high torque motor.
You just need a few more bells and whistles and some simple rulles to follow.

Mind ya, this just where I am heading. There very well maybe a way to go about it as you describe. I might be wrong, I have been before. So if your testing don't just take my word for it. Try it.

My stuff is creeping along right now cause I'm building a home. Towards the end of summer I am going full swing. I'll be able to get more answers together then.

Cheers
Matt

Why not use any form example bemf from bedini type Fast switching Diodes Redirects the flow to a cap which also has a fast switching diode as well to add to the normal charge after the bridge rectifier. Kinda like a one way loop back system. Like I said the motor will need a controller like any motor on a scooter would have from my reference. Hence this motor would have one too. It actually could be used like a sudo regenerative braking cicuit on like a scooter controller or any good pwm motor controller would have. I bet it could be done with some experimenting.
Oh on another note I lost two batteries before i got to test the switch you provided me go figure. So that is on hold atm. Right now I am looking into improving the motor design to get way more out then in like this hilden-brand motor.
But I shall not go too much into details friend since this is about the switch and not the motor.

Thats what I'm telling ya. It won't work the way you expect it too.
With the Tesla Switch setup, the potentials are always exposed, or the amount of offtime is so minimal that the extra energies travel to the source battery or to the opposite source battery before they travel to any other point. Even a cap inline will only have a detrimental effect. The positive energy in the cap will be reduced from the negative coming in.

Its hard to explain, you have to see it to understand.

Matt

I think you cant see it because I am not explaining it right. The recapture cicuit would be after the bridge rectifier and the looping circuit would also be after the bridge. Not going back to the batteries but back to the motor. Most motors that are used for scooters are pwm motors and so is the motor I am looking at. The input of the motor is only 150 ma just enough to guide the magnets field from blue to red. With the combined pull of stacked mags i figure I could get 300 lbs pull from each side. But I guess I'll have to put it together as a prototype with less powerful mags and experiment from there. With the current off it would recycle the field flux in the ends like a leedskanin type setup until current is supplied then it would shift the flux to the central part combining the flux of both sides saturating the core and spining it to correct orientation- up down. This would be a pulsed motor. Timing would be a factor and getting that timing right between pulses and recapture would be tough but not impossible.

You cannot collect the bemf. This force in brushless motors occurs during the powered phase there is no actual current flowing backwards just an increased resistance to flow. If you were to attempt to capture it you would have to turn off the power from the tesla swtich. Because that resistance is created by voltage you will not be able to get the capacitors fully charged before the switch reverses current.

Solution is to have a non-inductive motor that minimizes bemf. Sort of like a starter motor. A static magnetic or em field and a armature with brush contacts for the rotor coils.

I have been researching the Tesla Switch for some time now. I don't know if I can shed any light on the subject or not, but perhaps I can rekindle some interest in the project.

I started off very simply in attempting to understand the original circuit as it appeared on John Bedini's website. I started off with a single battery and a dpdt switch (see schematic) which I flipped by hand, in effect reversing the battery polarity with each switch. I was trying to understand the concept of Tesla's OSC and how the large capacitor's act as energy sinks, and how the energy flows between the sinks (the plates of the two separate capacitors) whenever the potential on the opposite plates of the capacitors is changed. This helped me understand the concept of "one-wire" energy flows, and how it can be shuttled around at will, and even through a load if one desires.

Further, I read Tom Bearden's document regarding Bedini's method of forming a negative resistor in a lead-acid battery. As I understand it, an oversimplified explanation of the theory is this: In order to create a negative resistor in a battery "you wish the ion current in the battery to be about 180° out of phase with the electron current in the load". Tom says that you need to switch the high voltage spike across the battery in 5 nanoseconds or less, and 1 nanosecond is even better. This will overpotentialize the battery briefly (20 to 40 nanoseconds tops) and the overpotientialization can be quickly redirected into an external load. If you can time the next overpotentialization correctly, the battery will continue to charge, and the load will also continue to be powered. This requires "microwave switching techniques".

I see this as an indication that the timing is the most critical thing. As Tom Bearden states in the same article, Bill Nelson (a Microwave switching engineer) "reasoned that the motor was just a load, and all the action was in the battery as controlled by the switcher. Bedini confirmed that this was correct." Bill Nelson's replication used an ordinary lamp for the load and also kept it's own batteries charged.

Tom Bearden points out that "getting everything timed just right, is still a significant undertaking". To me, the timing is where I need to focus my attention.

I've read this thread with interest, and noted another important (at least to me) thing. There was a mention made regarding the "David Bowling continuous charging device thread" wherein a Tesla Switch was built and tested, but subsequently abandoned because after 4 test-cycles the batteries were taking far too long to charge. In Tom Bearden's document on Negative Resistor effects in Batteries, Tom makes it clear that a battery that has been charged with negative energy ("conditioned", if you will) will manifest EXACTLY that effect when placed on a normal charger. I take that to mean that they (the experimenters) were very, very close to getting the timing just right.

I've been playing around with simple variations to the circuit I've attached. I've discovered that there is a lot more going on in even a simple circuit like this than they ever teach you in school. One unexpected thing I found was that if I connected a 10:1 transformer across the AC side of the bridge rectifier, the motor speeds up, as if to indicate that there is some ringing going on between the capacitor plates, not to mention that I can get rectifiable high voltage out of the secondary winding.

In attempting to understand the solid-state version I have come to appreciate that the transistors must somehow come to have their own forward bias voltages applied in a way that is completely electrically-independent from the rest of the circuit. I figure this concurs with Tom Bearden who states that you always have to pay for the switching. Bedini's circuit seems to use the independent transformers to alternately forward bias the transistors into an ON state, and then reverse bias them into an OFF state.

This leads me to wonder if perhaps there is a clever way to utilize the AC shuttling between the capacitor plates to self-resonate a specially wound transformer that has 6 separate (very low voltage) output coils that could be properly phased to drive the sets of switching transistors. The rise and fall time of each transistor would, I imagine, be the critical limiting factor to applying the overpotentialization within a 5 nanosecond or less time span. Things such as transistor over-saturation would matter greatly under conditions like this.

Intuitively I think that the exact 50% flip-flop effect is critical to circuit operation, but that doesn't, I think, mean that the transistors themselves are conducting 50% of the time. There must be a very brief lag as forward-bias voltage rises when the transistors are still off, and this would allow the overpotentialization to "relax" into the load before the next cycle sling-shots the lead-ions once again.

There is so much more to learn, and discover, and figure out. But I'm trying...

Attached Images

tCad1_test_001.jpg (30.3 KB, 65 views)

Remember Transistors have come along ways since that paper was wrote. If you settle on the fact that it has to cost, you may not be looking hard enough for alternative switching combinations. Low power, high speed SSR's , and mosfet combinations hold more promise every day.

I am not sure I agree, the type of load is as important as the rest of the setup. Loads need to be configured differently based on the output wave form. Transformers may hold the key to an "any load solution", but to date I have seen nothing but loss from conventional step up or step down transformers.

As far as timing goes it is the most critical part. But based on the load type resonating the load and the switching with each other is the key..

Cheers, its nice to have you here.

Matt

Hello Matthew. It's a pleasure to make your acquaintance.

My plan is to build a mechanically-switched device first. I'd rather like to replicate the one I see in the video you have posted on you-tube. I have some specific questions about it.

I'd like to try calculate the approximate switching frequency of your unit. What speed does the motor run at? What is the circumference of the wooden spools that make up the switching mechanism? It looks like there are 4 switches made per revolution, making 2 complete switching cycles per rotation, but what is the duty cycle? Were you striving for a particular value when you made it? Is it your opinion that a different switching rate (faster or slower) would have any effect? Do you think the internal resistance of the motor windings is very critical?

I have many other questions, but I'm sure a lot of them will be answered when I get my own device working so I can play with it.

Any information, and insight you have would be welcome and most appreciated.

Thanks for sharing your time and experience with me.

-kent

The motor was running at or about 1200 rpms. Yur right about 4 switches per revolution. Duty cycle may have turned out a little off but was supposed to be 50. Since the communtator was hand built it may have been flawed here and there.

If I were planning another machanical I would do it different. But it worked.

I am trying for a solid state but I got alot going right now. Just moved and I am building a house.


Cheers
Matt

Matthew,
Thanks for the info.

My buddy just finished renovating a 100+ year old house. Then he tore down the old dilapidated barn and now he's building a nice big workshop. The plan is to build up a Tesla switch device and work on a Kromrey converter, too. Anyway, I've helped him an awful lot over the past year, so I can empathize with the lack of time and energy which conflicts with the eagerness to get going again. So I once more thank you for your time and trouble.

I'll keep lurking here (and periodically asking questions) until I have something slightly newsworthy to report.

There's an interesting article about the Tesla switch at keelynet:

The Wiseman Theory of Energy Conservation

http://www.textfiles.com/bbs/KEELYNE...GY/flexflo.asc

Attached Images

	proof1.jpg (28.7 KB, 70 views)
	flexwat1.jpg (33.6 KB, 56 views)
	flexflo1.jpg (17.3 KB, 45 views)
	flexflo2.jpg (27.5 KB, 46 views)

Greetings! Thank you kindly for the link. In all my travels I've never come across that document before. Thanks sincerely.

The experiments mentioned therein are exactly the kinds of things I have been doing on my workbench. If only I could work as fast as my imagination does...

Hi have you got a link to the diagrams, as the attathments are not very good quality and are unreadible when printed out

I found the pictures here:

http://www.anomalies.net/archive/Kee...s/FLEXFLO1.GIF
http://www.anomalies.net/archive/Kee...s/FLEXFLO2.GIF
http://www.anomalies.net/archive/Kee...s/FLEXWAT1.GIF
http://www.anomalies.net/archive/Kee...ams/PROOF1.GIF


Alternately, you can get the same four pictures from these two zip files:
http://www.keelynet.com/expcirc/flexflo.zip
http://www.keelynet.com/expcirc/proof.zip

I used the schematic provided in the mueller report to arrange my transistors. i am using a bipolar motor with 6 independant triggers to run the switching system. As soon as i turn it on the transistors started to heat up which the current that was going to my motor power coil, the transistors never heated up before. I traced back through the schematic and found out why they did that. The two transistors that are closest to the bridge create a hot short when they turn on. I disconnected these two and left the 4 that create the alternating current going to the bridge and it functions fine now. Still have to load test it though.

how long do diodes take to oppose electron flow?

Surely that's nanoseconds, at least for some type of diodes?? My scope should come in the next few days, I'll look into it...

Love and light

You can fire the transistors nearest to the ground with ground energy from the CHARGING BATTERY. Eliminating your heat / ground. If memory serves me right.
If your using normal NPN transistors. Somthing along the lines of a TIP3055.

If you probe through transistor setup's you can find several ways of turning on the transistor without any ground present. Just build a one way setup with no switching. Add transistors at the key points. (IE the series connection in the batteries) Then start probing for firing. The series for example will fire from teh hot side of the series connection.

I have schematic but I will be couple of more days before I can post it. I am away now.

Keep us up to date.

Cheers
Matt

I have mocked up a simulation of a solid state version using LTSpice. I seem to have worked out some important bugs using the simulator. Now to take it to the next stage, an actual working device... So many things to keep straight in my mind. And so many unknowns still.

I can provide the spice file if anyone is interested. For now, here's the circuit diagram I plan to use. Will it work? Time will tell, though I expect progress will be slow, as in a month or more.

I split it into two sections for clarity. All 8 coil windings are on the SAME toroid core.

(Oops. I originally posted the Switching Circuit incorrectly. M4 and M6 *were* upside down. The schematic below has now been corrected. I also simplified the oscillator section.)

Attached Images

	TS_pt1.jpg (169.6 KB, 38 views)
	TS_pt2.jpg (102.3 KB, 36 views)

I tried replicated the Tesla Switch using
four 8.4V 150mAh Nickel-Metal Hydride 9V rechargeable batteries
and two double-poll-double-throw relays.

I was able to get the current to flow back and forth
and run motors, incandescent light bulbs, and LEDs for
what seemed to me to be LONGER than had I run
them with just 4 batteries.
But, I found that the relays were too slow and that the
pulse timing was KEY.
If you have the pulse on too long, the current flows
into the load and not your batteries.
If too short, the load isn't able to see enough current.
So a balance must be struck -- it must be TUNED.
And tuning is manual and changes with temperature, the load, etc.
One could imagine using a microcontroller to auto-tune
in a fuzzy logic way with hedging.
So my point here, I guess, is that yes this can be made
to work very nicely, but the tuning probably is a bit problematic
unless its "smart".

I think led acid batteries are better here. There is something
very magic about the way the led-acid batteries charge.
The internal chemistry flow has its own resonant frequency
that possibly can be exploited to get chemical current flow
to occur in the charging direction while using a small bit
of power in your loads.

The interesting thing about the Tesla Switch is that if your
load is a low-resistance load, that is more ideal actually.
So the notion of INCREASING your load (making the resistance less)
is a GOOD thing.

I think the Tesla Switch idea is amazing. Brilliant.
Even if its not OU ... its very efficient ... with COP approaching 1
if you tune well, have low resistance loads that don't produce heat
and friction.

So the "free" energy can come from solar or wind ... and the Tesla Switch
approach is a more efficient way to use your harvest of energy.

Apparently, the duty cycle needs to be accurately 50%
and the speed of operation 200 cycles per second or
more.re, the better the results until things get dangerous.

An inductive load is important. I can't remember why.
Probably the L is needed. check this useful doc:
http://www.free-energy-info.co.uk/Chapeter5.pdf
Paul-R

Hi Kent,


I see a couple of problems with your simulation. If I understand correctly you are planning to wind all the coils on the same toroid. How do you plan to control which output coils turn on then? The way I see it if you turn on one control coil that will turn on all the output coils. I think you are going to have to use two toroids. The other problem I see is how are you going to control your on time of the mosfets? The way a transformer works is that during the rise time of current on the primary you get a current flowing in the secondary. As soon as the current on the primary reaches its peak then current stops flowing in the secondary. This is only going to give you a few milliseconds of on time for the mosfets and therefore I don't think you can get the 50/50 duty cycle you need for the proper operation of a tesla switch. A circuit very similar to this was posted a few months ago and I don't think that person ever posted back with any results on whether they got it to work or not. Maybe they will read this and let us know how it turned out. Good luck, citfta

Hi citfta,

You're right. There is only going to be a brief on-pulse during each cycle. I believe that is precisely how the circuit diagram in the Mueller report is designed to function. Look very carefully at the SG3524 oscillator section and the way it drives the transformers. Everytime the squarewave rises, or falls, you get a small voltage kick on all the secondaries. The voltage is either positive or negative depending on 1) whether the pulse in the primary is rising or falling, and 2) the phasing of the secondary winding. If the circuit switched no faster than 20cps, there is no way the secondary windings can hold a power-on state for 50mS.

In the Mueller report diagram, all the transformers are marked such that the primaries are all in the same phase relationship, while the secondaries have 3 phase dots going to the transistor bases. and the other 3 go to the emitters. So the secondaries are wired with 3 going positive while the other 3 go negative. These positive and negative pusles either drive the transistors on, or further into the cut-off region of operation. So while all the secondary coils are outputting pulses, only the transistors with positive going pulses will be forward biased and turn on. The others will stay off.

As for the 50-50 flip-flop, from all that I've read and researched, and from all my "playing" on the workbench, the *instant of change* is when the batteries, capacitors, and load "see" the disequilibrium that has been introduced into the circuit. The rest of the time the circuit is trying to relax and restore equilibrium - through the load and "into" the charging batteries - which it almost does just before the next change occurs to flip it all back again. This "relaxing" can continue even after the transistors have turned off. The "perfect 50%-50% flip-flop" mentioned by Bedini may not necessarily refer to a perfect 50%-50% on-time, but only to a perfectly even time difference between switched states. At least, that's the understanding and premise I've based my circuit design upon.

What I *AM* concerned with is not having the sets of 3 mosfets turned on simultaneously. If one is faster to turn off, or slower to turn on than the others, then it's possible that a complete circuit might never be momentarily completed, and nothing will happen. There are a few other concerns I have, too, such as batteries preferring to be only charged OR only discharged, but gradually becoming useless when constantly being switched quickly between modes, but it's too early to tell. I need to get it built so I can observe and experiment with it.

Are you using transformers to replicate the Mueller Diagram? I know it has them. Do you think they will have positive outcome?? I just wondering so please let me know.

You don't need this to make the effect work. Transistor and mosfets have come along way since that report was published. High powered can be turned on with as little as 3 volt. And can carry between 1 - 200+- volts at 100 amps. You get a small loss (or restriction of volts and amps) each time you run the power throught the mosfet but you just use a higher potential. Say 32 volt to 16 to acheive 13.5volt.

Or use 2 volt cells and add as many as you want. "Enersys Cyclon 2volt 2.5 amp hour". Stack them up. They only cost 8 dollers a peice.
Mosfets...Heres a good one http://ixdev.ixys.com/DataSheet/8779...5958adb7c9.pdf.
The only downfall is it has to be turned off. Maybe they are all that way. I don't know.
A couple big diodes and a big bridge.
Oh ya an Astable multivibrator, play with it a bit and you got an oscilator.
Some solid state relays so you can fire and route the correct power to the transistor and you can get it going. NO GROUND AT ALL!!!

Somebody asked is it overuntity. Depends on the baseline. How much time do you power the load convetional and how much time can you power the load with the switch? And whats your definition of overunity.
It not rocket science but there some tricks.
Motors don't run right on it. They only put out half power. Perm Mag motors have one pole doing just the opposite of what its supposed to do, therefore reducing torque. Alot. Thats easy to solve though. Brandt must have...
Transformers don't run along time extra on it, because of the BEMF made when you shut the transformer off. Why? The power output winding needs to be reversed so the BEMF come out the same polarity as the energy from the battery. So you can collect it or allow it to serialize inside the tranformer so your input is reduced. But that takes good timing.
Inverters fall under that. Except for transformerless and thats just another ball of wax.
Light bulbs with low resistance work well but also eats of power. The output of light in general should if you think about it. Light Photons are not a byproduct of the electron (Or whatever energy we have here) the carbon filament converts the energy into light through heat. Heat is a loss... period.
But you still get more light for your money.

Maybe you just wanna find out for yourself?

It works, and real well if set up correctly. And I am sure if you look hard enough you can find a good load for it, but you'll have to build that as well.

I'll post my schematic this week when I get home.

Still I wanna know why you are duplicating the old stuff when you could just skip most of it, so if anybody can answer that would be great.

Cheers
Matt

People like to replicate what is documented first because it is there already for them to see and makes a good starting point. But when you have two different diagrams for the same device only one of which can function properly because the other has a short circuit in the schematic. You have to go to the drawing board and figure out what's what, where can i tap energy off of the system, and what losses can i eliminate.

Through experimentation and careful comparisons of the different diagrams i was able to find out that by rearranging the 6 switches and using a common positive you don't need the diodes at all. That eliminates several volts in losses.