BEMF harvesting key to OU?

[Itsu]

I was asked to start a replication on a circuit that uses BEMF harvesting to determine if it could show any abnormal behaviour.

After some PM's forth and back, the specific circuit to be used, or at least started with, is the one shown in picture 1 below.

As this is based on the works of user "GMeast" and his video is copyright protected, i went in contact with him and got permission
to use this circuit, but he asked me not to copy anything over from the video which i will respect.
His video which is a great piece of work can be found on his youtube channel and here:  https://www.youtube.com/watch?v=CYW_TYIBkgA

In further PM communication with GMeast he stressed that for the used resistor/inductor it is important that the Heater Element (inductor) be a nickel/iron alloy
and he will try to find the specific partnumber for this "standard power resistor ... wire wound/hollow ceramic core".
He mentioned it was 10 Ohm, but no indication as of yet about the inductance.

The used frequency should be above 300KHz and the duty cycle shows to be around 30% (from his video).

A first very crude attempt (including chaotic connected coloured clipleads  ) to setup the circuit can be seen in picture 2 below.
I used a 25W wirewound 220 Ohm power resistor which measures 223 Ohm / 23uH, but should be replaced by the above mentioned 10 Ohm ??H inductor/resistor.
The MOSFET used is an IRFP260N, the diode (Schottky) an IDH12SG60C.
Frequency of the FG was set to 100KHz at 30% duty cycle.

The screenshot in picture 3 shows:

yellow trace (drain/source voltage)
blue trace   (gate signal)
green trace  (return lead to battery current using my ac/dc current probe set to 100mA/div, so we have 132mA pp)

It reveals that we lack inductance in RL as we hardly see any ringing on MOSFET shutdown (like in GMeast his video where according to him the extra energy could come from)
It also shows that probably my FG is unable to hard open the MOSFET, so i need a MOSFET driver here.

I will be tidying up the circuit and use fixed connections etc., the right resistor/inductor and power both the FG and the MOSFET driver
from the same 2 battery stack so we have all power consumption from one source.

A problem i see and which also was mentioned by GMeast is how to accurately measure/compare the electrical power input and the generated heat
in the resistor/inductor, MOSFET, diode etc.

Any comments are welcome.

Regards Itsu

[ion]

While gmeast's data appears to be accurate, in fact it may not be.

e.g. using a thermocouple meter near a high frequency / high current pulsed source can possibly skew the readings. A shielded thermocouple would be required or testing to show that there is no erroneous measurement produced by induced noise, both series and common mode into the thermocouple.

Also his statement that the diode may allow the bemf to charge the battery is incorrect. The diode just keeps the bemf current circulating in the load, it does not return to the battery.

This smacks of R. Anslie.

Poynt has spent a great deal of effort on this type of circuit and is uniquely qualified to comment in greater detail regarding possible measurement errors.

I'm not saying there is not a gain, just saying all possible measurement errors must be routed out before making claims. Testing of supposed gain must be done several different ways before this conclusion is reached.

Regards, ION

[Itsu]

ION,

Also his statement that the diode may allow the bemf to charge the battery is incorrect. The diode just keeps the bemf current circulating in the load, it does not return to the battery.

Concerning this, could you please elaborate, as the idea may be to use the collapsing magnetic field to generate a bemf which only way out is via D1 to the battery (point of lowest resistance).
Why would the bemf current keep on circulating in the load (which on its own would not be bad as it adds to the heat generated in this load).

Thanks,    regards Itsu

[ion]

ION,


Concerning this, could you please elaborate, as the idea may be to use the collapsing magnetic field to generate a bemf which only way out is via D1 to the battery (point of lowest resistance).
Why would the bemf current keep on circulating in the load (which on its own would not be bad as it adds to the heat generated in this load).

Thanks,    regards Itsu

The diode provides the path for the current to keep flowing in the same direction in the inductor. Without the diode, the voltage across the inductor will rise to a very high level, because the inductor is now attempting to be a current source. This can sometimes lead to deterioration or breakdown of the driving semiconductor device due to excessive voltage.

http://en.wikipedia.org/wiki/Flyback_diode

Minus diode losses all of the energy is returned to the inductor (or resistive inductor)

I have covered looping of bemf on my bench here: http://www.overunityresearch.com/index.php?topic=604.0

While this work applied to the blocking oscillator, it can be applied to any switched inductor.

The common flyback power supply, now probably at least 80 years old or more exploits bemf to capture the inductors stored energy and provide a useful output. To my knowledge no one has ever produced more power output than was supplied to charge the inductor.

There is, however an area ripe for research with the common flyback supply, especially a looped version as I have pointed out on my bench. This would be where the inductor is of a special design such that NMR, NAR , thermo-electric or other atomic level effects allow for a tiny bit of extra energy to be absorbed and transferred by the inductor upon it's release of stored energy, enough to overcome circuit losses and provide a self running system.

A common inductor does not seem to have these properties, if they can even be shown to exist. Thus far they seem to be in the realm of many hypotheses.

Regards, ION

[forest]

Imho BEMF can be used to obtain OU only if clever using Lenz law. It is my personal conviction that induction phenomena explanation is not correct. Proof me wrong but where there is a direct proof about transformator converting magnetic energy of primary side into electric energy of secondary side , if no electrons from primary side are flowing into secondary ? Or where is the direct proof that a dynamo is converting mechanical energy into electrical ?

[ion]

Imho BEMF can be used to obtain OU only if clever using Lenz law.

Apparently, no one is clever enough to demonstrate this today, right now, and especially not in a practical self running system.

Debates about the operation of transformers or dynamo's can be quite time consuming. When the dust of the debate has settled and there is a seeming winner this person now has the burden to either prove without doubt a new workable hypothesis for their operation and/or design experiments to prove it conclusively on the bench.

Agreed, the present day explanation of many devices requires models that seem magical at times.

[forest]

Yes, without working prototype of OU generator all theories are possible 
My theory is simple and according to it there is no OU in BEMF, but BEMF can be used to direct already available ambient energy.
I found confirmation of my theory in comments and patents of various inventors from the past, also by Steven Mark.
It is so simple , that it hurts , that nobody assume it as real ! Earth magnetic field is not static field !

[ion]

Yes, without working prototype of OU generator all theories are possible 
My theory is simple and according to it there is no OU in BEMF, but BEMF can be used to direct already available ambient energy.
I found confirmation of my theory in comments and patents of various inventors from the past, also by Steven Mark.
It is so simple , that it hurts , that nobody assume it as real ! Earth magnetic field is not static field !

Sounds interesting. Please, kindly further explain your theory, and if possible show your proofs by bench experiments here.

[Itsu]

The diode provides the path for the current to keep flowing in the same direction in the inductor. Without the diode, the voltage across the inductor will rise to a very high level, because the inductor is now attempting to be a current source. This can sometimes lead to deterioration or breakdown of the driving semiconductor device due to excessive voltage.

http://en.wikipedia.org/wiki/Flyback_diode

Minus diode losses all of the energy is returned to the inductor (or resistive inductor)

I have covered looping of bemf on my bench here: http://www.overunityresearch.com/index.php?topic=604.0

While this work applied to the blocking oscillator, it can be applied to any switched inductor.

The common flyback power supply, now probably at least 80 years old or more exploits bemf to capture the inductors stored energy and provide a useful output. To my knowledge no one has ever produced more power output than was supplied to charge the inductor.

There is, however an area ripe for research with the common flyback supply, especially a looped version as I have pointed out on my bench. This would be where the inductor is of a special design such that NMR, NAR , thermo-electric or other atomic level effects allow for a tiny bit of extra energy to be absorbed and transferred by the inductor upon it's release of stored energy, enough to overcome circuit losses and provide a self running system.

A common inductor does not seem to have these properties, if they can even be shown to exist. Thus far they seem to be in the realm of many hypotheses.

Regards, ION

Thanks ION,

Great info on the flyback diode, and in the linked thread.
I will check if i can use anything from it.

By the way, GMeast has responded that he did suspend the hf/hv pulsing during the thermocouple measurements, so your concerns for skewing the data because of that are not valid.

Regards itsu

[ion]

From Itsu:

By the way, GMeast has responded that he did suspend the hf/hv pulsing during the thermocouple measurements, so your concerns for skewing the data because of that are not valid.

I was concerned also about direct inductive eddy current heating of the thermocouple junction itself , not only interference with the measuring electronics. This would take a while to settle out and if readings were taken immediately after shutting off the switcher (as they should be) there could be a residual heat measured.

There is difficulty of taking heat measurements in a hf switched environment (or shortly thereafter). A mercury thermometer will also register eddy current heating. Even a non-contact infra red readout at some distance will have a degree of uncertainty.

The best method might be calorimetric where a body of water is brought up to a certain temperature, then the thermal measuring device can be inserted to measure the difference above ambient temperature rise.

A lot of care must be used in designing the experiment and the results must be verified using several differing methods before claims can be made.

[Smudge]

You BEMF guys might be interested in my paper attached.  It considers a pair of bucking coils and takes account of the magnetic propagation delay, so the best experiment yielding maximum delay would use coils on each end of a ferrite rod.  Then apply a square voltage pulse to the coils and note that the BEMF on switch off at the back end of the square wave could yield OU.  The waveform deduced, taking account of the delay, bears a striking resemblance to Osamu Ide's published waveform.  The clockwise traverse of the flux v current loop offers an anomalous energy source.

Smudge

[Smudge]

And while thinking about coils wound on ferrite rods, here is another thought about propagation delay and pole flow.  Can we separate voltage induction due to dPhi/dt from that due to v X B?  This paper suggests we can.

Smudge

[ion]

Smudge: Thanks for your thoughts and the papers.

Itsu: regarding the use of nickel iron wire, you may enjoy this thread: http://www.overunityresearch.com/index.php?topic=1505.0

It is a short, adventure filled seven pages.

[Itsu]

ION,  nice read, so be aware of what alloy one is using.



Ok, i cleaned up my circuit and used a LM555 timer to generate the square wave signal and feeding it into an ucc37321 MOSFET driver, the IRFP260N MOSFET is now cleanly switching on/off.

The whole circuit is being driven by the same battery stack, 24V to the MOSFET / resistor, 12V to the logic.
I have a switch in the diode lead, so i can remove or insert the diode at will.

When running at 100KHz (300KHz also runs ok, but ringing is worse) i see a ringing on the 555 output pin and on the return current to the battery when the diode is IN the circuit.

When opening the switch to remove the diode from the circuit, this ringing mostly dissappears.

The gate and drain signals of the MOSFET are/stay clean.

Before continuing, i would like to know why i have this ringing on the 555 output pin and return lead current and if possible eliminate it.

Any idea?

Screenshot 1 is with diode IN the circuit (normal operating),
Screenshot 2 is without diode.

Yellow is the gate signal
Blue   is the drain signal
purple is the 555 output pin signal
green  is the return lead to the battery current signal (set at 100mA/div.)


Video here:  https://www.youtube.com/watch?v=cieDhgPxk5g&feature=youtu.be


Thanks,  regards Itsu   

[TinMan]

TOPIC-->BEMF harvesting key to OU?

Until recent event's i would not have thought this possable. I will refer to it as inductive kickback rather than BEMF.

Nice work Itsu-i truly believe there is truth in this now.

Brad

[Itsu]

TOPIC-->BEMF harvesting key to OU?

Until recent event's i would not have thought this possable. I will refer to it as inductive kickback rather than BEMF.

Nice work Itsu-i truly believe there is truth in this now.

Brad

Thanks Brad,

any details on why you are NOW thinking that?  Are you doing similar experiments?

Thanks,   regards Itsu

[TinMan]

Thanks Brad,

any details on why you are NOW thinking that?  Are you doing similar experiments?

Thanks,   regards Itsu

Have PM'd you

[Smudge]

You BEMF guys might be interested in my paper attached.  It considers a pair of bucking coils and takes account of the magnetic propagation delay, so the best experiment yielding maximum delay would use coils on each end of a ferrite rod.  Then apply a square voltage pulse to the coils and note that the BEMF on switch off at the back end of the square wave could yield OU.  The waveform deduced, taking account of the delay, bears a striking resemblance to Osamu Ide's published waveform.  The clockwise traverse of the flux v current loop offers an anomalous energy source.

Smudge

If an inductor consisting of a pair of bucking coils in series wound on a core, with separation between the coils so that there is a time delay along the core, can create a clockwise flux v. current loop, then it should give out more energy in its flyback than is put in when energizing the inductor.  That is the premise of my paper.  Of importance in this is the magnetic delay along a core, we want it to be as large as possible.  I have experimented with creating magnetic delay by having a series of coils along the core each one shorted by a capacitor.   This truly creates a magnetic delay line, albeit lossy, and I went as far as building a transformer on a toroidal core.  Somewhere I have the photo and the scope waveforms and I'll try to find them.  Interesting thing is it still worked as a transformer when the delay was 180 degrees, i.e the flux in the secondary was in antiphase to that in the primary, the flux was not continuous along the core.  You won't find that in the text books!  Essentially it had a primary occupying only a small part of the ring core diametrically opposite to a similarly wound secondary.  Between these two windings were the other coils each shorted with a capacitor.

Meanwhile here is a paper that deals with a magnetic delay line and gives formula so that others can experiment.  Maybe you flyback circuit experts would be interested in trying something like this.

Smudge

[ion]

Itsu:

I reviewed your video and it appears you may have the diode in / out switch positions reversed.

The no ringing should be when the diode is in the circuit and full ringing when the diode is out.

This is how these circuits have always worked for me, as the diode usually clamps the ringing,  but I could be wrong.

Regarding why the 555 output is ringing there are a few possibilities I can think of.

One is parasitic capacitance to the gate from the drain.

Two is insufficient bypassing of the driving IC's

Three is ground loop currents getting into the scope leads.

I suggest also a "star" grounding technique where the battery minus is the common point for all grounds, signal and current. A separate ground should be used back to this point from the high current FET switching (source).

Scope grounds can act as pickup loops especially when more than one scope ground is used (creates a ground pickup loop)

Properly probing this stuff can be tricky.

Smudge: Thanks for your magnetic delay line paper.

From tinman:

Have PM'd you

Tinman: May I ask if I may be pm'd also?

[Smudge]

OK, I found an old paper that shows the transformer.

[Itsu]

Itsu:

I reviewed your video and it appears you may have the diode in / out switch positions reversed.

The no ringing should be when the diode is in the circuit and full ringing when the diode is out.

This is how these circuits have always worked for me, as the diode usually clamps the ringing,  but I could be wrong.

Regarding why the 555 output is ringing there are a few possibilities I can think of.

One is parasitic capacitance to the gate from the drain.

Two is insufficient bypassing of the driving IC's

Three is ground loop currents getting into the scope leads.

I suggest also a "star" grounding technique where the battery minus is the common point for all grounds, signal and current. A separate ground should be used back to this point from the high current FET switching (source).

Scope grounds can act as pickup loops especially when more than one scope ground is used (creates a ground pickup loop)

Properly probing this stuff can be tricky.

ION,

i double checked on the switch, but its how i said it in the first place, i put a wire jumper across the switch, but the ringing is definitely there then (so when the diode is IN the circuit).

I now used a short piece (5 cm) of coax cable to link the output from the 555 to the input of the MOSFET driver, and now this signal is clean.
Guess it picked up the noise somewhere in between (5 cm).

The only ringing (with the diode in) is now seen (picked up?) by my current probe.

I have the "star grounding" system, but not the separate MOSFET ground back to the battery, so i could try that.

Thanks, regards Itsu

[Itsu]

I made some improvements and moved closer to the real environment, like upping the frequency to 300KHz and
using a 10 Ohm resistor (made of a stack of 4 resistors each 10 Ohm in a parallel/series configuration)
The inductance is measured to be 3uH, so very low.

We are drawing much more current and the resistors getting hot quickly.

I shortend all the leads where possible and this resulted in much more stable signals and with the separate ground for the source of
the MOSFET a more normal behaviour with the diode in or out of the circuit was seen (as mentioned by ION).

Current draw is 2A pp through the resistor.

Two screenshots, first with diode in the circuit, second without the diode.
I had put the current controller to 2A/div., so not anymore as can be seen in the video at 1a/div.

Yellow is gate signal
Blue   is drain signal
purple is 555 timer output signal
green  is the current through the 10 Ohm resistor (controller set at 2A/div.).

Video here: https://www.youtube.com/watch?v=4sg1sOIV1T0&feature=youtu.be    

Regards Itsu

[Smudge]

I have opened up a topic under EM Theory which will deal with magnetic delay lines, since they have important features beyond the bucking coils mentioned on this thread.

Smudge

[Itsu]

Meanwhile i have put an inductor in series with the stack of resistors to simulate the higher inductance.
Its a 4 turn coil wound on a nanoperm toroid measuring 780uH.

Now the current through the resistor/inductor is much lower so it can stay "on" all the time without heating up, but i need the correct value resistor/inductor before proceeding.

Regards Itsu

[Tommey Reed]

Hi all,

Its funny how some people are now changing their opinion on BEMF.