PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2024-05-18, 20:19:33
News: Forum TIP:
The SHOUT BOX deletes messages after 3 hours. It is NOT meant to have lengthy conversations in. Use the Chat feature instead.

Pages: 1 2 [3] 4
Author Topic: Fighters "Romanian ZPM (Zero Point Module) replication.  (Read 7726 times)

Group: Moderator
Hero Member
*****

Posts: 4102
Hi Itsu,

It is very strange because the moderator Fighter wrote here https://www.beyondunity.org/thread/fighter-s-zpm-zero-point-module-replication-by-itsu/?p=18&order=all#comment-af7905b6-b1de-427a-8059-af6a015c6228 : "Actually, you know what ? Prepare your bags, you'll go back to your buddies within 24 hours. We have no obligation to keep this crap you produce on our site. It's time to end this bad joke."
So the 24 hour is not over yet, he wrote it 9 hours earlier yet.

All you have done is you attempted to replicate the ZPM "effect"  i.e. to achieve a significant drop in the input current at a certain input frequency while the output power manifests in a brightly lit incandescent lamp and unfortunately your lamp remained faint when your input current dropped but your COP was < 1.   

I would like to suggest for Jagau to use a simple solution which would put an end to the debate on which of the following two methods is correct for the input power measurement:
- using the math formula he adheres to as described here: https://www.beyondunity.org/thread/energy-in-a-pwm/
- or using the Math function of a good quality digital oscilloscope to multiply the instantaneous DC input voltage and current you use (besides you use a non inductive current sensing resistor to double check),

so a very simple solution would be to apply a step down isolated DC-DC converter and loop the 118-120 V AC or rectified DC output back to the 24 VDC input supply.

The formula Jagau uses for calculating the input power gives 2.5 W for his ZPM circuit and the output power he measures (correctly) at the output of the circuit is 4.03 W, a COP of 1.59, indeed a pretty good COP value. His power supply shows 4.32 W taken out by his ZPM circuit (24 V, 0.18 A). 

This isolated DC-DC converter here has an efficiency above 80% https://www.ebay.com/itm/133998157732 and can receive AC input between 100 V to 380 V and its isolated output gives a regulated 24 VDC at 0.42 A rated load current. The tech specifications for the converter is included by the seller.

Such converter can safely receive the output voltage coming from Jagau's circuit which presently feeds a 118V 4W rated incandescent bulb https://www.beyondunity.org/thread/jagau-s-successful-zpm-replication/ 
Jagau rectified and filtered the AC output (frequency 1 kHz) voltage of his circuit and the converter I refer to can accept either the AC 1 kHz output voltage or the rectified DC output of his circuit without any problem.

I understand that Jagau firmly believes in his input power calculations but you have spent a lot of efforts and time attempting the replication of the ZPM circuit and you also firmly believe in your measurements.

So a solution would be to loop the output back to the input in Jagau's circuit.

Gyula


Hi Gyula,

yes, i thought i still had some time to prepare the promised video (see above) about the standing waves, but i was blocked posting much earlier.

Apparently only successful replications are allowed there and when you fail you are banned.
I know there are many more failed replications done there, but these members were much smarter than me (not difficult) and they already knew what would happen if they came forward with it.

You can suggest all you like for Jagau, but he won't be reading here, so your suggestions are useless, no matter how simple and practical they are.

The formula used by Jagau does IMO not take the input power into account, but only the pulsed part across the inductor, and thus discarding part of the device needed for creating the output in the first place.
I am not a math wizard, but i understand this formula uses the frequency / duty cycle twice almost halving the input power, so the COP would be easily > 1.

Itsu
   
Group: Tech Wizard
Hero Member
*****

Posts: 1125


You can suggest all you like for Jagau, but he won't be reading here, so your suggestions are useless, no matter how simple and practical they are.



Well, I sent him a PM to draw his attention to my post here.  We shall see what he does or not.

Gyula
   

Group: Moderator
Hero Member
*****

Posts: 4102
Gyula,

OK, good,

Jagau is a good researcher and i have no problem with him, i worked fine with him for a while on his Melnichenko replication, but this drive for using only the pulse power for calculating the COP is something i do not agree with and seems to have troubled our relation instead of just "agreeing to disagree".

I do agree with Vidura his latest comment that it is useless to find standing waves while the "effect" is not present.
But the only one which has the "effect" present is Fighter who has no time to experiment.
I did show that the used Fluke DMM (or any) meter can cause a wrong indication in both a positive and negative way, so are not to be trusted in this scenario.
 
Itsu
« Last Edit: 2022-12-14, 10:58:30 by Itsu »
   

Group: Moderator
Hero Member
*****

Posts: 4102
Back to basic.

I went back to basic with my ZPM replication as suggested.

The picture shows the present setup stripped from all unneeded items:



We have the 24V switch mode PS with display for voltage and current, with some filtering and with grounding connected to the minus lead.
We have the core with both coils in opposing (checked again)
We have an IRFP260 MOSFET directly driven by the FG (5V driven from mains, not grounded)
We have 3x 12V / 5W bulbs parallel together with 2x 12V / 21W bulbs parallel each on different length of wire to detect standing waves.
We have short, thick wires.

Some initial testing shows some special effects like:

# a strong resonance at 63kHz,
# decrease of the current on the display of the PS dipping to 100mA minimum,
# Unfortunately we also see the bulbs go off during that dip but,
# we do see the both 12V / 21W bulbs dim differently (one earlier than the other) pointing to some sort of standing waves.
 
I will make some further measurements later today and shoot a video showing these effects.

Regards Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4102

Since that i now am attacked and accused of lying and making personal attacks to members via PM or in threads on BeyondUnity.com, i am forced to respond to defend myself.

I never made any personal attacks on members there, not in threads nor in PM's, prove it, show us.

I made copies of all PM's i send and of all PM's i received (i always do) so i know exactly who wrote what and who responded what.

I did send some members a PM asking for a second opinion on the used formula, and they gave me their thoughts about the matter, with which is nothing wrong IMO.

I was promised that in due time, a second opinion / analysis of the used formula / procedure will be given in a public thread.

I still have good hope that this promised analysis will be given in such thread, in time, as those are the words of decent truth wordy and honest members.

Thanks,  Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4102

As promised here, a video of some testing i did today,

I show the difference in brightness of the 2x 21W bulbs parallel, but it turns out that no standing wave are in play here IMO.
But i don't know what is causing this either.

Also shown is the resonance peak around 60kHz manifest itself on the scope (voltage becomes sine wave, current minimizes) and on the PS (dip in current), but also as mentioned yesterday the bulbs go off  :(

 
It looks to me like normal behavior of a parallel LCR circuit going into resonance and experience high impedance and thus minimum current.

So one can argue that i do not have the "effect" still, and i cannot disagree with that because that requires that the bulbs stay on brightly during the current dip, but how to find that "effect" is still the question for me.

2 effects that were noted in several replications including mine even without having the "effect" (the DMM on the input voltage varies upwards (24V to 36 or even 44V) and the difference in brightness of 2 similar bulbs on different length cable) seems to be not caused by standing waves or any sub-harmonics IMO.
 
Video here:  https://youtu.be/sao3w7qSIx0

Regards Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4102

Some people who have seen my earlier screenshots of the input voltage and current may have noticed the difference in signal shape.

Each Power Supply that i have used (now 5 in total, including switch mode and conventional) plus the use of batteries with or without a filter produces different signal shapes.

This leads me to conclude that the used power source becomes a part of the active circuit (ZPM) it feeds.

The different filters and / or buffer capacitors in such power source partly determines the workings of the ZPM including the (parallel?) resonance peaks frequency.

This makes it extra difficult to exactly replicate the ZPM as it often is unknown how a PS is laid out internally.

Itsu 

   

Group: Moderator
Hero Member
*****

Posts: 4102

Here the same situation as yesterday, only i have removed the incandescent bulbs and hooked up only a 12V 3W AC led lamp.

It shows the same behavior as with the other bulbs as that at resonance (around 58kHz) the current decreases, but also the led lamp goes dimming.

Short video here:  https://youtu.be/wN3efpxjJxs

Playing with some ceramic capacitors across the drain - source of the MOSFET (200pF, 1nF, 2nF, 4nF and 100nF) does not show much difference on behavior, only with 100nF the dip / decrease is much less.

Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4102

It turns out that the used power source in Fighter his ZPM is not a switch mode PS.

The QJ6005E is being sold as a "Linear PS" in several ads on the internet and this link shows a picture of the bulky PS it contains (scroll to the bottom):  https://www.ebay.com/itm/263185946677

As the power source seems to be an integral component of the ZPM device, i will concentrate on using a Linear PS as the source to further narrow down the possible different components used in my replication.

Itsu
   
Group: Experimentalist
Hero Member
*****

Posts: 1987
...
This leads me to conclude that the used power source becomes a part of the active circuit (ZPM) it feeds.
...

Hi Itsu,

This may be due to the internal DC resistance of the power supply but often it is due to AC impedance because varying currents produced by the powered device can superimpose the DC. A good precaution is to always decouple the power supply by paralleling the DC input of the circuit with a low impedance capacitor at the signal frequencies (eg several thousand µF for frequencies of order of 50 Hz, or tens µF or nF the higher the frequency).
The disadvantage with large capacities is that you are no longer well protected by the current limitation of the power supply, which can only intervene with a delay.


---------------------------
"Open your mind, but not like a trash bin"
   

Group: Moderator
Hero Member
*****

Posts: 4102

Hi F6,

yes, i agree, but both the internal DC resistance and/or the AC impedance are different for each used PS which makes a replication very tricky.

And the question is if we NEED such superimposed DC for the "effect" to occur, so eliminating it by the use of low impedance capacitors in parallel to the DC input might be not the right idea.

The "effect" being, resonance in the circuit around 600kHz, having a dip (to nearly 0mA) in current measured by the internal current meter on the PS, while the 2x 12V 55W (100W) bulbs stays on brightly.

Thanks,  Itsu
   
Group: Experimentalist
Hero Member
*****

Posts: 1688
Hi F6,

yes, i agree, but both the internal DC resistance and/or the AC impedance are different for each used PS which makes a replication very tricky.

And the question is if we NEED such superimposed DC for the "effect" to occur, so eliminating it by the use of low impedance capacitors in parallel to the DC input might be not the right idea.

The "effect" being, resonance in the circuit around 600kHz, having a dip (to nearly 0mA) in current measured by the internal current meter on the PS, while the 2x 12V 55W (100W) bulbs stays on brightly.

Thanks,  Itsu

Itsu,

I've been loosely following your effort in this replication and if the condition for OU is as stated above, then mystery solved!  To really validate if such a reading from the internal PS current meter can be used in calculations to show overunity, the entire power supply circuitry would need to be probed and measured Before and After the current meter and then and only then could a determination be made as to validity of the claim.

I mean if one wishes to ignore the potential of such errors, I can easily demonstrate infinite OU from my own experience and circuitry.  It all depends on the reference points used for the measurements.

I've always suspected that the "filters" used between the PS and the ZPM circuit was the cause of the apparent OU at resonance.

Regards,
Pm
   

Group: Moderator
Hero Member
*****

Posts: 4102
Hi Partzman,

thanks for your comment, i do agree that an internal PS current meter is not ideal to use as an indication for a pulsed load, especially as the below picture shows on the right side
near the yellow arrow, very probably the shunt used for this internal current meter, intended to measure DC loads.

The coil like shape of this shunt will have some inductance and with 600kHz pulsed load will have some reactance and thus will deviate from its intended DC operation spec.

However, there were different current measurements methods used, but they were classified as undetermined because the results could not be matched with what the PS internal current meter showed or
deemed unrealistic because they supposed to show more current, the PS could deliver. (5A @ 24V = 120W, mind you it is a 60V PS) etc.

Methods used were: using DMM's to measure voltage across a csr and directly as a current meter, using a "kill o watt" meter at the PS 240V side and a watt meter like shown here:
https://nl.aliexpress.com/item/32274797082.html   which states DC Watt meter.

It might be that some methods did show considerable current still flowing while the internal PS meter showed near zero current, but they were put aside as unreliable / undetermined which could be true due to the fact the DMM's are not designed for 600kHz operation.

But the fact of the matter is that I too (and other replicators) can show this decrease of input current to near zero (see my above video’s) both on the used PS current meter, as via the scope (current) probes as well with a "kill o watt" meter, which makes this decrease real at resonance, but i and other replicators problem is that the bulbs go out too at that current decrease resonance point.

Could it be a so-called sub resonance inside this specific PS current meter circuit, showing minimum current there while full current passes through to the device?

Regards Itsu
   
Group: Experimentalist
Hero Member
*****

Posts: 1688
Hi Partzman,

thanks for your comment, i do agree that an internal PS current meter is not ideal to use as an indication for a pulsed load, especially as the below picture shows on the right side
near the yellow arrow, very probably the shunt used for this internal current meter, intended to measure DC loads.

The coil like shape of this shunt will have some inductance and with 600kHz pulsed load will have some reactance and thus will deviate from its intended DC operation spec.

However, there were different current measurements methods used, but they were classified as undetermined because the results could not be matched with what the PS internal current meter showed or
deemed unrealistic because they supposed to show more current, the PS could deliver. (5A @ 24V = 120W, mind you it is a 60V PS) etc.

Methods used were: using DMM's to measure voltage across a csr and directly as a current meter, using a "kill o watt" meter at the PS 240V side and a watt meter like shown here:
https://nl.aliexpress.com/item/32274797082.html   which states DC Watt meter.

It might be that some methods did show considerable current still flowing while the internal PS meter showed near zero current, but they were put aside as unreliable / undetermined which could be true due to the fact the DMM's are not designed for 600kHz operation.

But the fact of the matter is that I too (and other replicators) can show this decrease of input current to near zero (see my above video’s) both on the used PS current meter, as via the scope (current) probes as well with a "kill o watt" meter, which makes this decrease real at resonance, but i and other replicators problem is that the bulbs go out too at that current decrease resonance point.

Could it be a so-called sub resonance inside this specific PS current meter circuit, showing minimum current there while full current passes through to the device?

Regards Itsu

Hi Itsu,

I would have to think that your highlighted comment above is the case!  What else could it be?  Under the right circumstances, this particular PS meter arrangement in resonance with all other connected circuitry shows a current null while an obvious current flows to the load.  All replications by yourself and others have apparently failed to produce Fighter's results due to the uniqueness of his test setup.  To me, in order to claim OU in this premise, a complete independent verification of Fighter's device would be necessary to validate the OU claim.

Regards,
Pm
   

Group: Moderator
Hero Member
*****

Posts: 4102

PM,

i tend to agree with you.

I am trying to find out what these "right circumstances" are, so i am looking into parallel resonance circuits, but up till now they behave as what i see on my bench, when in resonance, there is a dip
in current on the meters, which also kills the current through the bulbs, and they go out.

By the way, with  "a so-called sub resonance" i more mean something like a "separate resonance" in that current shunt sensing circuit.

Thanks for your insights.

Itsu
   
Full Member
***

Posts: 114
Hi itsu
I saw that the replication is a bit stuck.If you are still interested, have a look at the replication from Atti in his thread :
https://www.beyondunity.org/thread/atti-s-successful-zpm-replication/?p=2
You can see there in his videos very clearly the standing wave pattern of the hot area which lit the neon, and how it shifts to another pwhenart of the circuit when the ground is connected to the centre tap. This is typical for resonators with distributed parameters. He stated that the effect of input reduction was present while the voltage got high. Of course it could be argued that it is not the original layout as by Fighter, I think the additional taps allow a better impedance matching. Note that he also got the effects while powered from a battery, with a filter I think.
Just as a suggestion, you could try to make some substantial changes in your layout, until you can observe the above mentioned effects. The neon as indicator works fine, and maybe with additional taps or including an additional winding it is possible to bring the voltage up, and the standing waves to manifest. Regards.
   

Group: Moderator
Hero Member
*****

Posts: 4102

Hi Vidura,

I did see Atti's videos as i am still following him (and others if any).

The neon is interesting, as i can only light one if the bulbs are disconnected, and then only the coils ends and center.

The bulbs seem to have a too big of a load on the signals to light up the neon, and / or my coils are not big enough.

I have received my AMCC 320 core and start winding the new coils on it at 150 : 300 turns.

I can use the old coils / core to extend with some wire like Atti has to see if i can show an effect as he has.

I don't think he uses the filter with the battery, so that's interesting too, but i do not consider this a one to one replication of Fighters ZPM as Atti himself mentioned this also, but it does show similar effects.


Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4102


I received my AMCC 320 Hitachi core and put 2 coils on the far ends in a 150 - 300 turns relation following this layout:



The coils specifications are:

L1                                        L2

150 turns 0.8mm                    300 turns 0.8mm
Inductance 268mH @ 10kHz     Inductance 1094mH @ 10kHz
Resistance 1 Ohm                   Resistance 2.1 Ohm

Coupling coefficient 0.97
Series aiding inductance 2260mH @ 10kHz
Series opposing inductance 290mH @ 10kHz (this is what we are using).



To compare with old AMCC 200 core:

L1                                       L2

100 turns 0.8mm                    200 turns 0.8mm
Inductance 89mH @ 10kHz       Inductance 277mH @ 10kHz
Resistance 0.9 Ohm                Resistance 1.5 Ohm

Coupling coefficient 0.96
Series aiding inductance 655mH @ 10kHz
Series opposing inductance 63mH @ 10kHz

Here is a picture of the both cores side by side:



Here some VNA plots of the new AMCC 320 core including the series inductance (upper left) confirming the 290mH (292.4mH) opposing induction at 10kHz.
Also seen is what seems a self resonance point at 121kHz (upper right) as the impedance peaks to 114 KOhm there.



I will insert the new core in the circuit and start making some measurements.

Regards Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4102

Some initial tests showed this is a nasty beast, with 57W of bulbs as a load (3x 12V 5W plus 2x 12V 21W) and 12V input from a battery i have already 560V of spikes going back into the input (battery).

No wonder, diodes and capacitors in the input get hot and / or fail after a while.

In this situation i have the Fluke meter again raise to 36V or so due to these interfering pulses, and also the both 12V 21W bulbs each on a different length of cable go off at different points depending on how
high the frequency is.

A 24V linear (not switch mode) PS dropped its 24V with increasing frequency till 4V or so (not due to overload), causing the bulbs to go out with a small range of frequency just before it were to bulbs / voltage
is fluctuating slowly (few Hertz).

These are IMO all reactions of the used equipment on these nasty spikes coming back from the ZPM.

A quick sweep of 10kHz to 5MHz did not show the effect we are after, being the bulbs fully on with minimum current on an input current meter.

Need to find a way to safely make some further measurements.

Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4102

Here a video showing the present setup with the AMCC 320 core and the 300 / 150 turn coils including their north / south pole detection.

An 10kHz to 1MHz frequency sweep is made showing how the bulbs react and what voltage / current they present to the 24V PS including again the effects like a Fluke DMM showing more voltage (almost twice) then really is there and the unequal dimming of the 21W bulbs.

The effect we are after (decreasing current to almost 0 while the 57W bulbs stay on brightly) was not seen yet.

https://youtu.be/5orC-UaIEIU

Regards Itsu
   

Hero Member
*****

Posts: 545
Interesting results Istu, thanks.

Since inductance is roughly proportional to the square of the number of turns, it might be worthwhile to perform a test with a winding ratio that has an even multiple inductance, not an even multiple turns ratio.

For the AMCC200:
AMCC200 - 300 turns = 1094mH
estimated ~212 turns to hit 547mH

For the AMCC320:
AMCC320 - 200 turns = 177mH
estimated ~142 turns to hit 88.5mH


For a back-of-envelope test, instead of rewinding you could try placing a ceramic/neo magnet on top of the higher-turn coil to reduce the permeability of the core on that side, and by extension its inductance.


---------------------------
"An overly-skeptical scientist might hastily conclude by scooping and analyzing a thousand buckets of ocean water that the ocean has no fish in it."
   

Group: Moderator
Hero Member
*****

Posts: 4102

Hi Hak,

I think you mixed up the cores and coils, but i see what you mean.

However, this circuit uses the both coils in series (series opposing), so i think they act as one single coil with a resulting inductance at 10kHz of 290mH (much lower at higher frequencies, see the VNA plot).

Placing a stack of ceramic magnets on top of the higher-turn coil (or anywhere else on the core / coils) shows that this resulting inductance of 290mH @ 10kHz does not change much (few mH), even with a stack
of more powerful neo magnets, the inductance does not change.

Sliding the core halves slightly up or down or left or right has more of an influence on the inductance than the magnets.


Thanks,   Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4102

Still doing some tests with my AMCC320 core.

The 150 : 300 turns coils in series opposing on it need to get into resonance to get the "effect", so i tried to find out what parallel capacitance brings this core/coil into resonance.

I used the core coil only, so no load (bulbs) or MOSFET attached, just some parallel capacitors.

I excite the coil using my FG with a single turn around one of the coils (loosely coupled).

Scope uses a 3pF probe, which was taken into account.

The following caps showed the following results:

Capacitor            resonance frequency        calculated  inductance (https://www.1728.org/resfreq.htm)

450nF                   462Hz                          263mH
100nF                 1000Hz                          253mH
45nF                   1470Hz                          260mH
3nF                     5130Hz                          320mH
23pF                  14130Hz                        5H
3pF (probe only)  14230Hz                       41H

The graph below shows the result.

So this shows me that the resonance shown by Fighter his setup of around 600kHz cannot be caused by any parallel capacitance with his coil.


Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4102
To compare how a simple air coil responds to a vna sweep, i have this simple air coil measured:





The resulting "series inductance" plot (10kHz to 10MHz) shows a steady increase in inductance till its self-resonance point, then continue as a capacitor.




Itsu
   
Group: Experimentalist
Hero Member
*****

Posts: 1987
Itsu, Thanks for reminding me of all the things you can do with a nanoVNA. I use it in radio for antennas, and I always forget to use it when I'm tinkering around the FE! :(
But as I'm in the process of redesigning my current gradient device, now I'll use it.
« Last Edit: 2023-01-07, 17:54:07 by F6FLT »


---------------------------
"Open your mind, but not like a trash bin"
   
Pages: 1 2 [3] 4
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2024-05-18, 20:19:33