This is it !

[gmeast]

calculations

I just watched your video...It's an impressive device for phase adjustment on the fly! I'm assuming you're working towards power in/power out calculations with battery charging?

Hi skaght,

Thanks for your comment. The thing was a bit of a chore. At present it has only 12-bit resolution and really needs 16-bit res for higher frequencies. I'm not sure I know how to figure power in such a complex system yet. Power out is difficult because it's connected to itself ... which is the object of this investigation ... that is to show a 'real' self-supporting, self-charging configuration using one battery and one battery only ... no 'swapping'.

This could be the 'Holy Grail'. People tend to shun claims for systems like Bedini's where the proof and evidence are indirect and subjective.

I have yet to see the evidence I seek with this setup, however, I'm hoping Bob Boyce will chime in. I don't know if I'm even in the right range of frequencies. I seem to recall that Bob said 'power density is a function of frequency' with this thing, so I'm operating in the dark. Plus I am aware that the chokes I'm using are items that must be 'tuned to the task' .... but how?

Later ... thanks again.

[gmeast]

B-Hex

Greg dont forget the B-Hex yahoo forum where Johan (who rep Bob's Hex) frequents, they will be VERY interested in this Bro

Ash

Thanks Ash,

I'll see if they'll let me on. The version of my thing that goes to PCB will be a 16-bit version for higher res at higher frequencies.

Thanks for your help. Later

[gmeast]

interesting oscillation

Here's an interesting shot at 30kHz and all 3 coils firing at one time. I don't really know what it means though ... any guesses?

http://02d1852.netsolhost.com/radian...k_simult01.png

Later

[gmeast]

more scope caps

Hi all,

Below are scope captures to compare. The two captures are of the PPC/FET board exciting the toroid at 50kHz with a PW of 800nsec and about 120deg phase angle.

compare_diodes-common-choke.png is probed at the location where all of the diodes common together on the way to the recovery choke.

compare_coils-common-choke.png is probed at the location where all of the Toroid Common primary coils join on the way to the B+ choke.

The two captures are basically the same ... but not. I don't know how to account for the differences actually or the significance of either over the other.

http://02d1852.netsolhost.com/radian...mmon-choke.png

http://02d1852.netsolhost.com/radian...mmon-choke.png

Still, I thought they were interesting.

Later

[gmeast]

testing progress

Hi all,

Ok, for the last couple of weeks, I have managed to keep running and changing the convenient variables such as Frequency, Phase Angle and Pulse Width. I have determined that equally-space phases or 120 degrees is NOT necessarily optimum ... though I don't know exactly what is.

Now, it is doubtful that I am charging because over the course of the past 5 weeks my battery has gone from 12.66VDC down to 12.06VDC. Even though, during that time interval I have had shorts, burnouts and even deliberately induced conditions that would draw excess current, those events were few and far between compared to the length of steady-state running time, so If I was charging it was not by enough.

My best running used phase angles like the one in the following scope capture. This kind of relationship seemed to draw the least amount of current and drain the battery slower:

http://02d1852.netsolhost.com/radian..._20110424a.png.

I'm now going to attempt to calculate the power consumption based on computation ... coil resistance, PW, etc. and see if there is even a suggestion of charging.

Bob had told me that the function of the chokes in the charging configuration was to control the 'direction' of LE flow. Those are two more variables that can be tested in concert with the variables mentioned above perhaps by selecting and holding constant the most promising Phase Angle, Frequency and Pulse Width ... and I have no idea on where to start with the chokes.

Well, there ya go. I only hope Bob contacts me again with some more information.

Later

[gmeast]

now some arithmetic

So, If I determine the equivalent power of a non-inductive load like resistor, what do I do? I know that pulsing an unloaded toroidal transformer with short 'square' pulses is NOT anything like pulsing the same transformer with a sine wave. If unloaded, the transformer primary(s) returns to the generator what is passed to it during half a cycle, and does so with practically no loss. But a transformer does NOT return such a favor back to the generator if the pulses are short and square ... and especially if you are doing it with solid state switches like FETS and other Transistor types. I suppose complimentary switches would come close to allowing it, but other things creep into the picture in the form of losses that can't be avoided ... voltage drops across devices, etc. I guess that's why Tesla stuff of the day stood a chance of working ... he used commutators, brushes, slip rings, switches ... plus, he made sparks .... sounds like Joseph Newman.

I'm going to pretend that nothing gets back to the battery ... a pessimistic assumption. So I'll apply some averages over the course of my recent testing. I'm going to use a frequency of 50kHz. I'll use a Pulse Width of 800nsec or 0.8usec. I'll also use an average battery voltage of 12.2VDC and I measured my coil resistance to be around 0.6Ohms. Now I can derive a duty cycle from that of 0.0000008 / 0.00002 = 0.04 or 4%. A straight-up E=IR calculation gives me 12.2V / 0.6Ohms = 20.333 Amps and the 4% applied to that gives me 0.8133 Amps but I have three coils so 0.8133 X 3 = 2.44mp equivalent continuous.

This doesn't help much does it? I guess I could take a 75 Watt, 5-Ohm power resistor and hook it across the battery. I know it would drag my battery down in much less time than 5 weeks. Even if I started at 12.6VDC on the battery, it would drag down to 12.06VDC in less time than 5 weeks.

Quite a bit seems to be getting to the battery and from what I have seen on the oscilloscope, it can't all be coming back from the diodes ... some has to be LE. I'll try and get the amp-hour rating for the battery. It doesn't seem to be in plain sight on the battery itself.

So, any suggestions on where to go from here?

Later

[gmeast]

PPC resonance hit

@7imix. This one's for you-

OK ... I decided to pursue the resonance thing a little further before giving up. In addition to the several resonant frequencies for the T650-52 Toroid, there are also frequencies where it will more readily 'accept' the energy of a particular pulse ... it's Amplitude and Pulse Width. This may also be tied to some characteristics of the MOSFET.

This turned out to be a very important characteristic of the setup and these powder-metal cores. At the frequencies I was pulsing, the Toroid somehow would not allow as short a pulse as the controller was giving the FET gate.

So I increased the frequency and reduced the configuration to 2 phases at a 180 degree phase angle separation.

I've included four scope captures of that run:

This first picture, resonance3.png, shows the gate signals from the PPC controller ... green and yellow. You can see a frequency of 511kHZ for each PPC signal and a PW of 0.235usec. for each giving a combined frequency of just over 1MHz.
http://02d1852.netsolhost.com/radian...resonance3.png

This second picture, resonance2.png, shows one of the PPC gate driver signals (green) and a resonant wave in the form of a pretty good sine wave (yellow). Note that the sine wave frequency is at the combined frequency of the two PPC driver signals. This is in line with the principles of resonance.
http://02d1852.netsolhost.com/radian...resonance2.png

This third picture shows the resonant sine wave (yellow) and the B+ voltage (green) ("B+" = "Battery +"). The resonant wave reaches 60VDC and has a P-P of almost 80V. The reason I included this picture is to give a glimpse of a possible charging affect here. My scope doesn't function as a very good analyzer but if you look at the areas enclosed by the sine wave above and below B+, it seems like the area above B+ might be greater than the area below B+. My battery is staying fixed at this voltage, so at least the 'overhead' is being met.
http://02d1852.netsolhost.com/radian...resonance4.png

This last picture is a single phase (only) used to pulse the Toroid, but at twice the frequency as a single phase (above) ... and things are the same as things should be ... same sine wave.
http://02d1852.netsolhost.com/radian...resonance5.png

So ... I guess one resonant frequency of my Toroid is just over 1Mhz.

One more point ... the amplitude really doesn't change at all when I fire all 3 coils simultaneously ... another proof that this is a resonant frequency of a Toroid with very little loss.

Later
Edit/Delete Message

[gmeast]

LE From A Poly-Phase Toroid

Hi all,

Before I finish with this version of the 12-bit PPC, I wanted to try and get a resonant waveform using all 3 phases. Well I almost got there. I just don't have the fine adjustment I really need. The 16-bit machine should give me that in conjunction with a multi-turn pot for fine tuning the frequency.

The following scope capture certainly tells the story. Look at the waveform data in the picture. Everything is almost right ... the Toroid's resonant frequency is still at just over 1MHz and each phase's frequency is just shy of the 333kHz for perfect 3-Phase operation ... it's 292kHz which probably accounts for that 4th 'phantom' bump that I'm sure will vanish at close to the 1/3 frequency. I think it's interesting the way that '4th' wave was simply synthesized ..... mmmmm ..... what does it really mean, huh?

Here's the capture:
http://02d1852.netsolhost.com/radian...resonance6.png

Later

[gmeast]

LE From A Poly-Phase Toroid

Hi guys,

OK, I fiddled a little more and found a near-perfect 3-phase resonance setting for the PPC. This does NOT seem to be charging (maintaining) as well as the post with the "phantom" wave in it, but it's awfully nice looking. I think this is just a lucky shot as far as the symmetry goes.

I say that because 1/3 frequency for 1.03MHz is 0.3433MHz and I ended up pulsing at 0.342MHz. It takes all 3 phase coils firing to maintain it though, so it must be 'sitting' at either the lower or upper limit(s) of something.

Here it is:
http://02d1852.netsolhost.com/radian..._resonance.png
Enjoy. Later

[gmeast]

progress

Hi all,

Well I thought I'd post a progress report. As those of you who had been following this thread earlier, you notice that Bob Boyce had a brief attendance here ... albeit NOT to assist initially. But in the background, Bob PM'd me on 3 separate occasions with valuable information relating to watkykjy's video(s) showing his controller used for self-charging capabilities, details and suggestions about the choke(s) used, placement of components and other valuable goodies ... for which I am grateful and feel privileged to have been given.

I followed his instructions carefully and have had some interesting results and gained some insight as to the nature of LE. Yet, I have NOT seen what could be regarded as self-charging performance.

On the power-side of the arrangement (FETs and Gate Driver), he said to use exactly the same (output) components as in the 3GPWM's parts list, but that's sort of hard to determine since he refuses to release a schematic of that board revision. So I did the best I could with regards to that aspect of things.

Recently I re-examined the PMs he sent me and uncovered a key piece of information. Apparently, Bob had written some 'code', for the AT-Mega processor used on the Hex Controller, specifically addressing the self-charging use-mode of his controller, but refused to share that code with watkykjy and other 'selected' individuals, so watkykjy wrote his own (simplified) code that was just barely able to produce the effect and did involve pulsing phantom phases like the '1-3-5' channel sequence you saw in his video(s). I'm now convinced that there is much more to the 'pulsing' sequence than shown in watkykjy's video(s).


Later

[gmeast]

next steps

Hi all,

In an attempt to actually pull this off, AND because I have not heard further from Bob, I am assuming that there is a 'Pulsing Sequence' that needs to occur and that it is NOT simply 3-Phase operation of the Toroid/Controller. I am assuming that it is something more complex ... yet still repeating such as a firing order like 1, 2, 3 ... 1, 2, 1 ... 2, 3, 1 ... 1, 2, 1 ... then repeating (or something) ............. I don't know. Several have suggested the sequence used for Hydroxy production. Whatever, a Micro could be handy for this. But since I've done so much with my PPC, and understand what it can do, I can accomplish the same with the PPC. Now a 'Random' sequence would be almost impossible to achieve with the PPC, but not with a Micro assuming the correct algorithm, but I don't see that as being very useful.

To use the PPC on discrete sequences only requires a 'Phase Channel' for each phase coil in each sequence. So if you predict you will have 4 very specific sequences, each involving a phase 3 times in a sequence, like the example above, you would set 12 Phase Channels on the Counter Data Bus. Now this gets complicated, because now the MOSFET Gate Driver ICs must listen to multiple input signals ... the Driver IC is shared between several inputs. So that would require a 'pull-down' or 'turn-off' resistor on the input pin of the Driver IC and each Phase Channel's output being 'Diode-Connected' to the Driver IC. This all makes the use of a Micro very attractive. But at the same time, the individual Phase Channels are cheap and just 'plug' onto the a data bus.

I'm shutting down testing to do several things. 1 is to switch to MUR410 diodes, 2 is to test the 'shared inputs' issue and 3 is to somehow get the Counter Board, Phase Board and maybe also the FET Board to PCB. All that takes $$$ I don't have, so it will be a while from this point.

Later

[gmeast]

Progress Report

Hi all,

I'm happy to report that my approach to controlling the Driver IC using 'shared' inputs is successful. I use a 1k resistor to pull the UCC27322's input pin LOW and all of the PPC signals that are to be shared by that pin are diode-connected. I used a 1N914's in the test ... no biggie! Wooooppppie !

Later

[gmeast]

driver shared signal

Hi all,

I'm happy to report that my approach to controlling the Driver IC using 'shared' inputs is successful. I use a 1k resistor to pull the UCC27322's input pin LOW and all of the PPC signals that are to be shared by that pin are diode-connected. I used a 1N914's in the test ... no biggie! Wooooppppie !

Later

Here's a video of the shared signal on a MOSFET gate driver IC

YouTube - Poly-Phase Controller Shared Signal Input

[sucahyo]

In an attempt to actually pull this off, AND because I have not heard further from Bob, I am assuming that there is a 'Pulsing Sequence' that needs to occur and that it is NOT simply 3-Phase operation of the Toroid/Controller. I am assuming that it is something more complex ... yet still repeating such as a firing order like 1, 2, 3 ... 1, 2, 1 ... 2, 3, 1 ... 1, 2, 1 ... then repeating (or something) ............. I don't know. Several have suggested the sequence used for Hydroxy production.

I think the sequence in hydroxy application relate to frequency. Where the first is double of the second, the second is double of the third.

[gmeast]

frequency

I think the sequence in hydroxy application relate to frequency. Where the first is double of the second, the second is double of the third.

Hi sucahyo,

Yes, I recall that an example is 10.?kHz, 21.?kHz, 42.?kHz or so was used. In the D9.pdf is goes further to say something like "... the frequency depends upon the characteristics of your cell ...", etc., and was using one of the example for calculation purposes beginning with the Toroid's secondary and primary turns ratios and so forth ... it's quite comprehensive.

I have no idea if any of THAT is pertinent to the self-charging task, but the battery has got water in it even though we're not trying to electrolyze it ... or are we ? ooooohhhhh

Later