No announcement yet.

An energy harvesting project

  • Filter
  • Time
  • Show
Clear All
new posts

  • It’s good to know your method anyway as I will most likely need to tweak my ‘CoP measurement’ method.
    'Consciousness came First'


    • Hey fellas, just wondering if you measure input vs output on amp metres on your builds?

      Dave Wing


      • Hi Dave,

        Input yes but not output normally since most meters will not respond fully to the pulses in the output line. Output energy has to be measured by its net charging effect on the battery and which requires a clear methodology.

        It’s still worth putting a meter in the output line, analogue, digital and clamp, to see how each behaves. For example I put a clamp meter on the output line and it read negative while the input read the right normal value

        'Consciousness came First'


        • I hardly used meters on the output in the past, but now I use them all the time. The meter was the easy way to determine that my SG also had a flow of generating current charging the secondary battery, which is additive to the coil collapse and boosts charging output amperage.

          I say this because I see in your build you appear to be using high inductance coils. Have you noticed anything extra going into your secondary batteries?

          Dave Wing


          • Whatever extra is going in is not easily measured, hence the need to use the battery's response to measure any 'extra' energy input. Given the very low efficiency of the pulse-generating process itself, the fact that in several cases I got a CoP of around 1.1 strongly suggests that there is indeed something 'extra' going in.
            'Consciousness came First'


            • Hi Rodolphe,

              I will reply to your post using categories as before:


              While I found that the extra ‘magnetic’ mass in my setup contributed to CoP values, I have not undertaken enough tests to ascribe a degree of certainty to how much of an increase occurs. Repeatability is not my focus at this stage. Rather, by changing one variable at a time I have built up a picture of what factors are most important and which are less significant.

              The same can be said of my comment about OU with the SSG, or more accurately, my SSG. I have had OU results but not consistently even with the same parameters. It’s value to me has been more in the area of comparing the use of a trigger coil and using the power transistors and R45 core.

              By SSG here I mean a single bifilar coil with a welding rod core and the well-publicized SG mode circuit (monopole circuit). However, I have adapted that to run in Generator mode as well and most of my tests were done in Gen mode.

              When I refer to an SG circuit then I mean the type described in the three SG e-books by Peter and Aaron.

              During all my v4 tests (my own take on a rotor and solid-state version of an Aspden/Adams type generator) using the PWM, my rotor and its magnets were stationary in the middle of all the coils and it was noted that removing it dropped performance a significant amount.

              I will clarify my coil arrangements and also those of the impending v5:

              In my SSG there is one coil with a power winding and a trigger coil. In my v5, which is the v4 developed and informed by my ‘retro’ work with the SSG (BD1 circuit), I have wound the main coil with four power windings and one trigger coil. The other four out of five spaces for coils will use the v4 coils as additional recovery coils. With the five switched channels available, four will be used with the MC and the remaining single channel will operate with the four recovery coils in parallel.

              This is all shown in the layout image together with notes on the various charging options. I would expect the most effective charging to be seen with the HV (from the MC) concurrent with cap dump pulses derived from the recovery coils via the cap dump unit.

              Click image for larger version  Name:	V5 Generator Layout.jpg Views:	0 Size:	457.9 KB ID:	74371

              As indicated, one can use just one channel of the v5 such that it will be in effect the BD-1 and with the revisions.

              SOLID STATE

              My findings so far strongly indicate that using a PWM to drive a FET, while a very good way to drive one, does not produce as good results as a trigger coil and a power transistor. While it is possible to drive the MJL21194 with the PWM it is not ideal as these transistors are more current oriented while FETs are voltage devices. There must be more to the performance differences than that as the coils can in principle be switched on and off by either method, but it would appear something else is happening when the trigger coil is used that does not happen when a PWM is used. Perhaps some form of system resonance or maybe it’s because you need a degree of current with the electrostatic pulse and the MJL and 2N provide pulses that do that (their ‘h’ shape) while FETs tend to give the sharp spikes with less current component, and so a much smaller ‘h’ shape. That being so then one needs to drive the device with the appropriate trigger, which for the MJL is a current-based arrangement provided by the trigger coil.

              I still believe that a solid-state option is both possible and desirable and I have included the option for a PWM input on the v5. With the formal research project starting early next year (details further down), then by then I may or may not have a solid-state version to use for it.

              If any of your contacts have some views on this then yes, please ask them.

              CoP MEASUREMENT

              It is good to hear about the different ways that CoP is measured and I’m very open to refining my method in advance of the research study. I have used a fixed charging time so far as that is in keeping with only changing one thing at a time since charging time itself also affects the final CoP value.

              As I say in my manual, the part of the charging profile you work on will affect the value due to the extrapolation used although here I’m hardly extrapolating at all since the input and output energies are almost matched in the two stages of the test.

              Also, it is easier to conduct a test with a fixed charging time since one doesn’t need to keep checking on where the voltage is in order to stop a test. The voltage at switch-off is going to drop anyway to a degree depending on the interval before the discharge stage.

              When it comes to the research study my method needs to be easily repeatable and stable, and of course, set in advance via the Pre-Registration.


              I see what you are saying here but on the assumption that there is at least a little radiant component to my charging then that can at least start the ball rolling in that direction. My understanding of conditioning is that if you start with hot charging then that does not reduce the amount of ‘cold’ charging conditioning required anyway so there is little advantage to doing it to somehow prepare the battery. I will be consulting others on this issue.

              OUTPUT PULSES

              My current thinking is that in SG mode, with the output Neg connected to the input plus, the output pulse is having an effect on the supply side which may account for the double pulse. In CE/Gen that connection is not there and so there is no feedback to support a second pulse.

              As SG mode on the BD1 has not worked properly I can’t comment on its effect on my PSU. With the v5 that should be testable.

              V4 EXPERIMENTS

              Regarding the supply current flow, having just wound my new MC (4+1), I thought I would do a test using the v4 board on a small 7Ah battery.

              I connected up just one of its four coils and also one of the recovery coils and the current was about 1.2A. This resulted in a faster voltage rise in the battery and a shape similar to what you have obtained with your v4. But again, after 2 hours I still only reached a 12.8V peak, falling back slowly. I then did a discharge and derived a CoP of 0.35.

              Here is the graph:

              Click image for larger version  Name:	Charging curve.jpg Views:	0 Size:	276.5 KB ID:	74372

              I think there is more that is similar now between our v4s than before, but it’s still not clear how and why you can get a 12Ah battery up to 15V after only a few hours. I’m less concerned by that than I was but it is annoying.

              RESEARCH STUDY

              The more formal research study will start early next year and be conducted transparently using the Open Science Framework (OSF). The project can be seen here:

              A link to see how the OSF operates is here:

              Details of this can be seen on my updated website at:

              The aim here is to get this whole topic into a mainstream scientific journal, and if achieved it would seem to be the first time ever. This would get the ball rolling for further planned studies and, with the world’s various challenges that lie ahead, this topic cannot be ignored in building multi-faceted solutions and contributing to the energy aspect. With careful work, some long-standing hurdles and ‘boggle thresholds’ can be overcome. Well, that’s the mission - going in by the tradesman’s entrance instead of banging on the front door!

              V4 DIFFERENCES

              The only real difference between our systems that I can identify is regarding the ‘magnetic mass’. The presence of 4 other coils with ferrite cores (originally) in close proximity to the rotor magnets, is going to intensify some aspect of the resulting pulses, even if not the peak kV - that being limited by the avalanche rating of the active device.

              The pulses naturally carry a charge component and perhaps that is significantly enhanced by the extra magnetic mass. If you want any of my design drawings then I can send them over.

              The first gallery on my site shows the early assembly and setup that is still at the heart of the v4/5. This is at:

              That lot will do for now

              Last edited by JulesP; 09-12-2023, 12:16 PM.
              'Consciousness came First'


              • Hi Gary, Julian,

                I didn’t dare to post here for a while after reading Gary’s opening sentence of post #118… Just kidding; As Julian responded already; it was in reply to attached document of his post #113.

                I watched Peter Lindemann’s presentation “John Bedini’s Self-Recharging Monopole Motor”.

                And have the following questions and comments, the numbers refer to approximate time stamps, the sentences are paraphrased by me, for the exact sentence see the movie/presentation:

                09.35 – Standard SG
                -> So Classic SG / Radiant mode (not common ground/generator mode)

                09.49 / 10.35 - Circuit boards, two motor drive coils.
                -> Initially I thought it looked like an 8-transistor PCB, but if they actually are 2 boards (placed together so it gives the impression it is one), driving the two coils, I assume those are the 4-transistor boards:

                12.15 – input battery + output battery + large storage system
                -> Large Storage system = large battery? Will this not start off-gassing if it is constantly connected in parallel to the output battery? Or because its size (capacity) id so big compared to the power of the coils, it will not get so easily to that point as long as the output battery gets swapped/used quick enough?

                12.45 – Switch input and output battery
                -> Topic that has come up a couple of times already: In other movies John said this was not possible in classic SG /radiant mode, that the batteries would run down if you’d do that in this mode. So is the difference here that there is a large storage system in parallel and that it IS possible in this specific configuration?

                16.48 – Keep it at 0.25A
                -> This is input power to PCB? If so, my interpretation in the context in which PL says it is; JB did not want to have too much output power (spikes) to the output battery.
                But if I look at the setup; those 2x 4-transistor boards will have combined certainly more than 0.25A of input. Although in theory the resistors of the trigger circuit can be chooses so it would be less than 0.25A, but then I miss the point of having two PCBs.

                24.53 / 26.25 – Voltage grounds out / but it was not
                -> what does he mean by that? Just the fact that you cannot measure the HV peaks by putting a volt-meter on the battery terminals?

                29.0 – If you haven't seen it run in your home, it is because the batteries are not able to elicit the process. Keep doing it and doing it and the process will show up.
                ->In general after doing multiple runs, day after day, I would not see an significant increase in the COP with my Bedini SG in common ground/generator mode. In general what did yield a higher COP was after the following: After the charge cycle was done, I would discharge 1Ah out of it. Then the next morning it would charge the OUTPUT battery again with the TeslaGenX charger, discharge again 1Ah out of it and only then start a new charge cycle with my Bedini SG.
                In that those cases the COP could be quite high… incidentally above 1…. It gave me always the idea that it was more what that TeslaGenX charger did to the battery that made the COP high, rather than the machine I built -OR (maybe put more correctly) – That it was the solid state cap dump system that provokes an over unity effect in the battery, which my machine in common ground / generator mode was not able to… But with just doing charge/discharge cycles with my Bedini SG, I would not get COPs over 1, far from in general.
                Would be interesting to talk with PL about this.

                Best regards,


                • Answer to my own comments/questions 12.45 from previous post:

                  Post nr #216 from this thread:

                  Another thing that popped into my mind regarding my previous post 09.49 / 10.35:
                  I did not understand why PL would use 2x 4-Transistor boards while 1x 8transistor boards are also available...
                  The advantage of using these two boards (or better said -> 2 coils) is that you have double the core material.
                  Last edited by pearldragon; 09-22-2023, 06:10 AM.


                  • Hi Rodolphe,

                    Taking each issue in turn:

                    Using a PSU:

                    Yes, I agree that it can’t deliver current in the same way as a battery and also can’t be used in SG mode but I’m not using one to get the best results but rather for convenience. When I’m looking for the best performance then I will use a supply battery.

                    For a battery supply to receive some radiant assist then the system will need to be in SG mode but that doesn’t appear to give the best results, generator or common earth seem to. As with most other areas, suck it and see.

                    Interesting note that current measurements should be spot checks and not continuous. This does then require one to be around the test area a whole ot more.

                    Battery Setup:

                    Gary offers some useful advice on how to set up the batteries for a test run. Starting with both batteries fully charged and only being able to use radiant charging can take quite a while and so far I’m never sure when full charge has been reached as I have not been able to take a battery to 15.2V. The c20 rate seems low but workable although I’m inclined towards a c10 or C15 rate.

                    Battery Swapping:

                    Useful confirmation that swapping of the supply and charging batteries is ok and that it is using a hot charger that undoes the conditioning process and is to be avoided.

                    Measurements on leads:

                    Gary’s use of a sniffer coil is a good idea and why I have tried using a clamp meter on the input and output leads. Of course, this does not give me the scope detail but with its Hall sensor and coils it does detect the magnetic fields. The strange thing is that the clamp meter on the input leads reads a current of say 0.6A while the meter on the output cable reads -0.2A!

                    This implies the magnetic field is in the opposite orientation or has a changed direction. If you changed the orientation that you put the clamp meter on the input cable it makes no difference as the detector is reversible, I think, so the negative reading indicates something else. Magnetic energy entering the wire instead of radiating out as normal?

                    I would like to build a ‘sniffer’ coil for a more detailed picture of events so any details would be appreciated.


                    Gary also proposes using time as the main factor with the CBA and that is how I work. Using a voltage is too vague and imprecise for an accurate Ah or energy calculation. Voltage is a consequence of the chemical state and you can't readily measure that; but you can measure the energy or Ah in and out.

                    Extra battery:

                    I wondered the same that if the parallel REAC type battery is always being charged then it will reach full charge. But then if it has a large capacity and also when it reverse charges up the newly swapped battery, perhaps these keep it from reaching full charge. Also, it perhaps can start in its role from being nearly depleted. I will be trying out a REAC in the near future.

                    Low Input Currents:

                    A figure of 0.25A for input current is about what I found with the BD1 and any higher and it erodes the CoP value since the denominator is increasing in the CoP energy calculation.

                    This is seemingly a problem with the v5 I am testing which has four windings in its MC. The current is four times higher but the charging effect is not up by the same factor. Maybe using a REAC will improve matters.

                    Here are some V5 charging graphs, the upper with HV from the 4 winding MC and the lower from the cap dump unit.

                    Click image for larger version

Name:	V5 Charging 1.jpg
Views:	343
Size:	287.1 KB
ID:	74381

                    Click image for larger version

Name:	V5 Charging 2.jpg
Views:	177
Size:	350.2 KB
ID:	74382

                    After doing most of my tests with a small 7Ah AGM battery, I am now starting with a 110Ah deep cycle battery. Everything takes a lot longer though!

                    Voltage Grounding:

                    When the spikes meet the low impedance environment of the battery then they are effectively earthed so you won’t measure the spikes at that location. However, their impact on the electrode and electrolyte has not ceased and is having repercussions within the electrochemistry. This can result in the ‘over-potentialising’ that has been spoken of.

                    Consistent OU:

                    It may be that the TeslaGenx charger is indeed doing something that your SG is not able to do but whether it is the cap dump system is not clear. I have used a cap dump system and it gives nice-looking graphs with lots of spikes on it but the CoP is not really any better. So what is different about one cap dump from another?

                    The answer is probably the mix of voltage and charge delivered in a particular pulsed format. While PL's demo cart may have no ‘secret sauce’ I reckon the waveforms delivered for best results are indeed a very carefully chosen arrangement of impulses that have specific effects. That is IP and so is not shared widely. That’s the way of the world.

                    We each need to find our own ‘secret sauce’. If I find it then I will indeed share it so long as I’m not infringing someone else's ‘secret sauce’.

                    'Consciousness came First'


                    • Hi Julian

                      I still intend to respond a bit more elaborate at your post #126 and #129. But here are some things I wanted to respond immediately to:

                      #126 CONDITIONING
                      If you’d have a TeslaGenX charger I’d recommend you to charge with that one only.
                      If you have a new battery, the normal ‘breaking in’ (or one that has been on the shelf for a while) you can do till 14.4V with a normal charger in absence of a Radiant charger… then at least you know that the battery is accepting voltages to 14.4V and if none of your machines can get it to at least that level, then there I would start looking. But again, if you can do away with a normal charger: better; it was just a suggestion how to do some checking with the tools you have.

                      #129 Discharging:I only discussed with him how I explained it to you, which he mentions in #118. But there are more ways that lead to Rome (or Overunity ).

                      #129 Measurements on leads / Sniffer coil:
                      From post #216 onwards on this thread Gary and I discuss the sniffer coil:
                      If you’ve more questions let me know, but it’s basically buying a small spool of coper, put it 90@ orientated to your wire and hook the spool of copper up to your scope.

                      #129 Low Input Currents:
                      If I look at the first image of your graph there, it looks exactly like my graphs. To me it just seems you would need to let it run for waaaaay longer… that dimple you see, it is the same in my graphs, only after that the voltage starts rising slowly again.

                      #129 Consistent OU:

                      There are a lot of ways and parameters to change with cap dump systems, you probably know more about them then me. I tried to contact TeslaGenX a couple of times already to, although I’ve talked extensively with them in the past also via Skype, at the moment no response….

                      Best regards.

                      Click image for larger version

Name:	Sniffer Coil.PNG
Views:	166
Size:	26.5 KB
ID:	74384


                      • Hi Rodolphe,

                        Perhaps I should consider getting a TeslaGenx charger as a reference device to do some output tests on and to charge up my supply battery instead of using a regular hot charger or waiting many hours using my system. Which model would suit for charging up my 110Ah deep cycle battery, and smaller ones when required? But I could only really afford the TX-2A12.

                        Would such a charger take a battery up to 15V like my SSG and v5(SG) are supposed to do?

                        SNIFFER COIL
                        Would a simple inductor serve the purpose? I have a bag of various inductors as here:

                        LOW INPUT CURRENTS
                        If I could throttle the input current that would be useful but so far the only control I have is to increase the trigger coil adjustment resistance, and so make that circuit operate less efficiently, or to disconnect some of the MC windings, e.g. four down to two.

                        CONSISTENT OU
                        They are probably cautious of revealing some ‘secret sauce’

                        Last edited by JulesP; 09-26-2023, 10:36 PM.
                        'Consciousness came First'


                        • Hi Julian,

                          SNIFFER COIL
                          Would a simple inductor serve the purpose? I have a bag of various inductors as here:

                          Making a sniffer coil is really very simple. I doubt that most commercially made inductors would work for this. Here's a link to a post where Rodolphe and I were discussing this with a photo and explanation showing what I use.

                          Measurements on leads:

                          Gary’s use of a sniffer coil is a good idea and why I have tried using a clamp meter on the input and output leads. Of course, this does not give me the scope detail but with its Hall sensor and coils it does detect the magnetic fields. The strange thing is that the clamp meter on the input leads reads a current of say 0.6A while the meter on the output cable reads -0.2A!

                          This implies the magnetic field is in the opposite orientation or has a changed direction. If you changed the orientation that you put the clamp meter on the input cable it makes no difference as the detector is reversible, I think, so the negative reading indicates something else. Magnetic energy entering the wire instead of radiating out as normal?

                          I would like to build a ‘sniffer’ coil for a more detailed picture of events so any details would be appreciated.

                          I've found that clamp meters being digital don't give accurate results. And yes, the magnetic fields around the input and output leads are opposite orientation from each other.

                          Gary Hammond,
                          Last edited by Gary Hammond; 09-27-2023, 09:46 AM.


                          • Thanks Gary, I will explore how to make one.

                            Does the magnetic field being in the opposite direction imply a magnetic field is being drawn in to the wire rather than radiating out as normal?

                            'Consciousness came First'


                            • Hi Julian,

                              Originally posted by JulesP View Post
                              Thanks Gary, I will explore how to make one.

                              Does the magnetic field being in the opposite direction imply a magnetic field is being drawn in to the wire rather than radiating out as normal?

                              No. I think it means that current is flowing in the opposite direction. From the run battery the supply current (electron flow) is traveling from negative to positive. And the battery being charged has electron current flowing into the negative terminal and out of the positive terminal. A current direction reversal is necessary to charge a battery. Of course, with this system there is high voltage spikes and very little current doing the work of recharging the receiving battery.

                              All the leads have both an electric field and an associated magnetic field around each wire. I think the "sniffer coil" produces a voltage when the magnetic field cuts across the windings. This voltage can then be displayed on the o-scope. It is very interesting to use the "sniffer coil" at various places around the machine while it is running and watch the traces on the scope.

                              Gary Hammond,
                              Last edited by Gary Hammond; 09-27-2023, 02:46 PM. Reason: spelling error


                              • Thanks, Gary,

                                I will build a 'sniffer' coil and see what it reveals.

                                I also wanted to ask for your suggestions on reducing the supply current in an SSG/SG-type device.

                                Results I have observed indicate that the supply current is one of the primary factors in CoP and that in my going from a bifilar coil with my SSG to a four-winding main coil in my SG (v5) device, the current is up by a factor of 4 but the pulse effect on the battery is not.

                                So I am looking to turn down the supply current a bit to find the optimum balance. Of course, the trigger circuit has its own pot to tune the trigger timing, and some have used this pot to adjust the supply current (which would surely change the optimum timing setting for running), but there is normally no such equivalent pot for adjusting the coil current.

                                If I wanted to include such an adjustment would a simple pot in series with each channel be suitable to take the total supply current down from around 0.9A to around 0.25-0.5A, or perhaps elsewhere in the circuit to affect all channels equally?

                                My concern is that dropping nearly half an amp at around 12V means dissipating 6W so does that mean that the only sensible way to drop the current is to reduce the coil voltage? This can be done with my Buck Converter but that has other consequences.

                                So I’m just asking what is the best method to reduce the total supply current, which in my case is made up of my 4 recovery coils in parallel (being switched by one transistor) and the 4 windings in the main coil each with their own switched channel.

                                'Consciousness came First'