"Enhanced Generator" from JPKBook

[JulesP]

Measuring the HV current on the Solid State generator

Hi all,

I am close to starting to make output power measurements as a function of oscillator frequency in my solid state generator. Before I do that I have prepared two devices that will help me make various measurements; one is a adjustable potential divider for safely measuring the HV pulses on a scope and the other is a 1 Ohm resistor to enable me to view the 'hot' current aspect of the HV pulses. The photos shows the latter connected up and a scope trace of the voltage across the resistor.

John Koorn, if you are reading, this I would appreciate some interpretation of the scope trace as you have quite a bit of experience doing this type of measurement. The image is of course frozen but in real time it was jumping all about and not a nice stable waveform.

Regards,

Jules

[John_Koorn]

Hi Jules,

Where is the 1 ohm resistor placed in the circuit?

Regards,

John K.

[JulesP]

Hi Jules,

Where is the 1 ohm resistor placed in the circuit?

Regards,

John K.

As per the attached. Between the bottom of the coils and the battery.

Jules

[John_Koorn]

Hi Jules,

I'm not sure why you're seeing spikes going both positive and negative on the scope. I also can't understand why the level of the pulses is not regular (as in my scope pics).

In the picture of your setup I can see red wires going to your batteries, so it kind of looks like the shunt resistor is in parallel with the batteries (which also look to be in parallel). That looks different to your last picture of where the shunt is in the circuit.

John K.

[JulesP]

HV Current Measurement

Hi Jules,

I'm not sure why you're seeing spikes going both positive and negative on the scope. I also can't understand why the level of the pulses is not regular (as in my scope pics).

In the picture of your setup I can see red wires going to your batteries, so it kind of looks like the shunt resistor is in parallel with the batteries (which also look to be in parallel). That looks different to your last picture of where the shunt is in the circuit.

John K.

Hi John,

Yes there are connections to the other set of batteries but my swapper system, and the large relay you can see on its side, isolate one set at a time to be charged. Also at this moment I only have the one battery in each set connected so am running the system on 12V and not 36V. To check I disconnected the other positive lead to the right hand battery as seen in the picture you referred to and it made no difference. I made a 7 sec video of the scope trace which shows this jumping around and I have no idea what is causing this and why I have both negative and positive spikes.

The video is at: https://www.dropbox.com/s/87u5yw3vte...ement.MOV?dl=0

Any thoughts would be appreciated.

Thanks

Jules

[JulesP]

Solid State Generator Update

Hi all,

Over the last few months I have been working on my solid state generator, which effectively replaces the rotor based triggering of the coils with a square wave generator system that gives me the option of firing coils all at once in parallel or sequentially. After dealing with the inevitable glitches, faulty components etc I now have a system that allows me to test various coil configurations and, in the near future, to replace the 'small' coils with three of my previous rotor based coils using enhanced Ferrite cores to boost the voltage (more about that later). So I wanted to share what I have found so far and raise the inevitable questions that are part of steering the project onwards.

To give an operational overview, the completed generator, as shown in the labelled pic and based on the attached schematic, was designed to see the effect of increasing the HV pulse rate on battery charging. In this setup there are two sets of three batteries giving me 36V to run an inverter at lower currents than using just one 12V. The drive and charging sets are separate and switched every 2 mins by the 'battery switching circuit' and 'main switching relay'. The coils wired up as shown are fired sequentially run by a 4017 chip that sequences through the 7 channels with each clock pulse from either the internal oscillator or the external PWM module, near the main switch, and which has the option of a variable duty cycle. One channel out of the 'trigger circuit' allows for the full pulse rate to be directed to one or more FETs so I can compare the performance firing each coil in turn to get a particular value of HV pulse frequency or firing a set of coils in parallel with one FET to get the same HV frequency. You may wonder why I opted to fire the coils sequentially and the reason was to avoid the very low resistance of 7 x 2 ohm coils in parallel that might result in very high circuit/coil current.

Okay so what have I found so far?

Firstly that ramping up the frequency of the HV pulses does not result in proportionately more energy being deposited into the battery when the pulses are fed directly to the batteries. I have good reason to hope that the situation is different when one uses a capacitive discharge system as a capacitor probably has a much faster reaction time to convert the cold electricity to hot for subsequent direction to the batteries but an SLA battery has to move heavy ions around and my thinking is that once pulses get above say 500Hz, the battery chemistry can't respond fast enough to make the most of the incoming ZPE.

Secondly, in my case firing the coils sequentially was no better or worse than firing them all in parallel with just one FET. Counter intuitively the current remained very low with the coils in parallel and the HV output values very similar. I found that a 20% duty cycle was ample with no obvious benefit of increasing it to 50% or more. I attach a scope trace for the latter situation and where I am using a 45:1 voltage divider to pick up the spikes so their actual voltage is 650-700V. That's not a particularly big value which might partly explain why charging has been only modest and is only just able to compensate for the circuit demand with no external load. While the coils are wound on to Ferrite rods with turns 5 layers deep, there is not the same mass of wire as in my rotary coils which is why I plan to employ them soon and which leads me on to my third main point.

It is well known that the back EMF from a coil inductor is linearly proportional to the relative permeability of the core material, with the other variables of length, cross sectional area and current being the same (see attached equation). Especially if you have used spools to create your coils, you can replace the cores relatively easily (unless they are resined in!). In my rotor coils I used solid 'Bright' steel rods with a Rel. Permeability of 100-200. Using some Nickel Iron Ferrite rods off eBay one can have a value about 1000 and which will dramatically increase your HV pulse voltage. My rotor pulses were about 1,200V at 90Hz and, having switched to Ferrite cores taking me 20 mins, my output jumped to 4,050V at 122Hz. So this is a simple tip to get more bang for your buck. I am awaiting some 'F6 Neosid' Manganese Iron cores from Australia that have a RP value of 1800 and so can expect to produce HV voltages of around 6kV or more. When I have measured the difference I will instal three of these coils in place of my current 7 slim ones and do some more tests and on the assumption that a much high voltage will result in better coherence with the ZPE and draw more energy into the system.

So as the experiments progress questions arise and I would appreciate constructive thoughts on one particular area in particular to help steer future work:

1. The use of capacitive discharge systems seems quite common with Bedini type devices and I understand about the electret phenomenon but Bedini himself observed that battery charging was very effective by feeding cold electricity directly to them instead of indirectly via a capacitor. However, based on what I said above about how a battery might not be able to process high frequency pulses as well as hoped, maybe a capacitor system would be an appropriate option here?

2. I have seen on the forum various capacitive discharge circuits and am interested in the type that has some voltage sensing feature and where you can vary the discharge frequency so as to optimise the point in the capacitor charging curve when the energy should best be released to the battery. Is there a circuit out there that has shown itself stable and reliable?

3. Handling big capacitors gives me the jeebies as I know that one handling mistake could be ones last. So whenever I have had to use them, even low voltage ones, I build in a means to short them out safely (like a toggle switch) so they are not holding on to their charge while I forget about it. With large or super caps what is the best way to discharge them? For example could one have a momentary button to discharge them to the battery after the generator is switched off together with some indicator like an LED to show if it still has some charge in it?

I look forward to your thoughts and, as always, I am happy to share any details of my work including circuits, schematics, pics etc.

Regards,

Julian

[mml]

Julian,
Thank you for sharing.
Concerning the circuit of discharge, I built the circuit whose diagram is in this post #106, for battery 12V:
http://www.energyscienceforum.com/sh...ll=1#post18112

I set the hysteresis with the 2 variable capacitors, on the bench, between 17 and 23 V to charge the 12V battery.
I have never had a breakdown in 4 years of use, this circuit is very reliable.
I added a simple white del simply installed in series with the output pin 6 of the comparator and R1 (1.2k) which feeds the front end of the opto-coupler, as indicated in this post #114:
http://www.energyscienceforum.com/sh...ll=1#post18535

I have not tested the one for 36V which is in post #14:
http://www.energyscienceforum.com/sh...ll=1#post13081

cordially
mml

[JulesP]

Hi mml,

Thank you for bringing all those threads together. I think I need to build some form of cap dump system, if only to compare the performance of feeding HV straight to the 36V battery sets. However, I am very new to this circuit and I hope you won't mind helping me with a few questions based on the 36V one I have reattached.

1. The FWBR on the left suggests that this circuit is being powered from an external ac supply. Can one run it from a 12.7v supply as I do for the other circuits I am using?

2. What does the term inverted and non-inverted refer to in these types of circuits?

3. There is a lot of forum discussion on small modifications to the various circuits, for example to control oscillations at low frequency, stabilisation from false triggering, to improve the sharp switching of the FETs and so on. Is the one attached updated in this regard as I notice it still uses the optocoupler and some have said it impairs the FET switching?

4. What determines the size of the main storage capacitor? I note some have a range of switchable values. Are super caps of 1-2F relevant here?

5. In your circuit do you have a way of showing that the main cap is charged (LED or similar) and do you have a way to discharge it safely?

Kind regards,

Jules

[mml]

Hello Jules,

answers below

1. The FWBR on the left suggests that this circuit is being powered from an external ac supply. Can one run it from a 12.7v supply as I do for the other circuits I am using?

The FWBR (bridge rectifier) is shown only to indicate where to place the two inputs (negative and positive). There is no external power supply, this is not the goal.
The collapsing wire of the coils must be connected to the negative of the Cap Pulser, side of the negative of C1, with a ultra fast diode upstream.

2. What does the term inverted and non-inverted refer to in these types of circuits?

This term was chosen because the device turns cold electricity into conventional electricity. (I think)

3. There is a lot of forum discussion on small modifications to the various circuits, for example to control oscillations at low frequency, stabilisation from false triggering,
to improve the sharp switching of the FETs and so on. Is the one attached updated in this regard as I notice it still uses the optocoupler and some have said it impairs the FET switching?

This map is at the point. I have never noticed any dysfunction.
As I said before, I did not reproduce this card which is designed for 36 V (3 batteries of 12 V in series), but the card according to the diagram of the first link of my previous post.
Nevertheless they only differ by three components: R13, C4 and R12, which are added to the Cape Pulser in 12 V. If you decide to build it,
I advise you to build according to the diagram of 12 V. in order to properly adjust the lower and upper discharge thresholds.
The lower threshold must be set above the full charge voltage of the batteries. For a 12V battery, the full charge is at least 15.8V, but can rise to more than 16.5V depending on
the condition of the battery.
For 3 batteries of 12 V in series, I do not know, but the lower setting will always be based on a slightly higher tension at the voltage of end of charge, otherwise when the batteries will have reached this reference voltage, the comparator of the Cap Pulser will permanently open the mosfets, and there will be no pulsation to charge the batteries. We want to keep the pulsation throughout the charging period.
The upper threshold is about twice the voltage of the battery, or 24 V for a 12 V battery, but I guess for 36 V, it will be 36 V + 12 V, or 48 V. So, we must remember 12 V above the battery voltage.
(Ref. Bedini SG - The Complete Intermediate Handbook, p 46).

4. What determines the size of the main storage capacitor? I note some have a range of switchable values. Are super caps of 1-2F relevant here?

Do not put super caps, they do not have the right properties. The capacitors that must be used are capacitors with low ESR (fast charge and discharge), like photo caps.
These 15000μF chemical capacitors are very suitable for me: https://www.ebay.de/itm/Chimique-Con...item35dc97e3ec
I put 4 on my Cap Pulser, and this is very sufficient to charge 2 batteries of 125 AH. 2 capacitors would also do the trick - their unloading in the batteries will be faster.
It can be charged up to 80 V, so they could also be suitable for a load up to 48 V.

5. In your circuit do you have a way of showing that the main cap is charged (LED or similar) and do you have a way to discharge it safely?

No, it is not useful, because the generator constantly feeds the Cap Pulser, and the voltage of the capacitors varies permanently, according to the rhythm of their charge and their discharge
(for me, it is on average 2 to 3 times per second).
The white led that I mentioned in my previous post is sufficient to indicate the two states of Cap Pulser (charge: the led is off, discharge: the led is on). The led has never grilled,
it's simple and effective.
You can still measure with a meter or oscilloscope if you have one.

cordially
mml

[JulesP]

Hi mml,

Thank you for the detailed replies:

The FWBR (bridge rectifier) is shown only to indicate where to place the two inputs (negative and positive). There is no external power supply, this is not the goal.
The collapsing wire of the coils must be connected to the negative of the Cap Pulser, side of the negative of C1

Ok so the HV pulse from the generator's FET Drain (which reverses polarity and goes HV+) will connect directly to Pos of the Capture Cap. Where do the LM741 and the H11D1 get their power supply from?

I advise you to build according to the diagram of 12 V. in order to properly adjust the lower and upper discharge thresholds.
The lower threshold must be set above the full charge voltage of the batteries. For a 12V battery, the full charge is at least 15.8V, but can rise to more than 16.5V depending on
the condition of the battery.
For 3 batteries of 12 V in series, I do not know, but the lower setting will always be based on a slightly higher tension at the voltage of end of charge, otherwise when the batteries will have reached this reference voltage, the comparator of the Cap Pulser will permanently open the mosfets, and there will be no pulsation to charge the batteries.

The 12V build then is suitable for a 36V setup and perhaps the lower threshold might be set at 40V and the upper at 48V but the lower in particular might need tinkering. Is it R10 that adjusts the lower and R13 the upper? I appreciate what hysteresis means in magnetic terms but not sure in this context (the amount of lag between the charge and discharge?) To clarify these thresholds visually is the attached right and the cap would normally discharge at the lower threshold?

I put 4 on my Cap Pulser, and this is very sufficient to charge 2 batteries of 125 AH.

My small batteries are 7Ah each so three in series would happily be met with just one of those 15000uf caps?

Yes I have a scope so should be able to see the real time charge discharge and make the threshold adjustments to achieve the optimum charging.

Jules

[mml]

Hello Jules,

answers below

Ok so the HV pulse from the generator's FET Drain (which reverses polarity and goes HV+) will connect directly to Pos of the Capture Cap. Where do the LM741 and the H11D1 get their power supply from?

The collapsing current of the coils, when the supply is cut off, is a positive current that must flow to a negative terminal (of a capacitor, or of a battery).

The 12V build then is suitable for a 36V setup and perhaps the lower threshold might be set at 40V and the upper at 48V but the lower in particular might need tinkering. Is it R10 that adjusts the lower and R13 the upper? I appreciate what hysteresis means in magnetic terms but not sure in this context (the amount of lag between the charge and discharge?) To clarify these thresholds visually is the attached right and the cap would normally discharge at the lower threshold?

It seems to me that the roles of R10 and R13 are those. The simplest method of tuning I found was to tune the high and low thresholds by feeding the Cap Pulser with a DC lab supply and placing a voltmeter at the output of the mosfets. Place a resistive load at the output to drop the voltage when cutting the source.
I specify that I have never tried to set the threshold higher than 48 V, but 23 / 24V.
The term "hysteresis" can also be applied to a device that has a control of the level of the ignition voltage different from that of stop. This is the case, for example, of the thermostat of a refrigerator. In Cape Pulser, there is a hysteresis comparator.

My small batteries are 7Ah each so three in series would happily be met with just one of those 15000uf caps?

A single capacitor 15000uf would suffice for 3 X 7 Ah.

cordially
mml

[JulesP]

Hi mml,

From your helpful information I have made my own attached drawing which is part of the build process for me. I hope it and the credits are right.

I would need to feed the discharges via my battery swapper to direct them to the battery stack that is not driving the whole circuit at that moment.

When I have finished testing the HV feed direct to the batteries I will start on this and let you know how it goes.

Regards,

Jules

PS. The 12V zener diode IN5242 seems hard to come by in modest numbers so can I replace that with a BZX55-C12V as I have quite a few of those? The other diodes I can get fine.

(In anticipation of what one earth will happen with Brexit )

[mml]

Hi Jules,

I read a little more my notes and some post of the thread "Bedini Comparator Cap Dump". I found this information concerning a modification to be made between the 2 Cap Pulser 12V and 36V.
For the 36V, D3 is a 15V zener diode;
For 12V, D3 is a 12V zener diode.

You can find the zener diode IN5242 at Farnell. I mainly buy my components at their home.
Link: https://uk.farnell.com/on-semiconduc...5?st=1N5242BTR

In your diagram, reverse the polarity of the white del (between terminal 6 of the LM741 and R1).

cordially
mml

[JulesP]

Thanks. If D3 needs to be a 15V Zener instead of 12V then perhaps I can just use the same as D2 (IN4744A)?

Regards

Jules

[JulesP]

Hi mml,

Having drawn up the circuit in readiness for the build I have come across an issue that seems to be relevant to the solid state design.

When looking at how to arrange the negative side connection ports on the circuit board I realised that if I connect the capture capacitor to the main negative rail of the generator then I am effectively bypassing the cap dump circuit which switches the negative line back to the batteries in order to release the stored charge. I have shown this in the attached pic and where the cap dump circuit is traditionally connected to an independent coil. In my case the HV pulses are being generated in relation to the battery negative/ground (FET Source) and so I can't connect the negative of the capture cap to that as it will discharge straight away.

I wondered if you can think of a simple solution to this?

Regards,

Jules