PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2018-10-19, 07:50:01
News: Registration with the OUR forum is now by invitation only.

Pages: 1 2 3 4 5 6 7 [8] 9
Author Topic: Dr. Stiffler returns with a new device: SFM  (Read 12925 times)

Hero Member
*****

Posts: 2495


Buy me some coffee
Still not quite ready to give up on this at the moment so here is another test of the biased Stiffler circuit but with a different twist.  The difference is that the biased cap is now the input buffer C2 and not the output C1 as in the previous tests.

In reference to the schematic below, a DC buffer cap C2 was added that basically charges to the supply voltage from V2.  To guarantee this, C1 is shorted prior to each scope trigger.  This insures that when the signal generator starts, it starts at zero volts so no fault is initiated in the sg.  Also, a 1K precision non-inductive resistor R1 is added and used as a load in the final output calculations.  The circuit operation is not as simple as it might appear so I'll try to explain it the best I can.

The first scope pix shows the input power of 7.129mw supplied by the signal generator which equals an input energy level of 7.129e-3 * 100e-3 = 712.9uJ .  This is a product of the input voltage times the input current over time.

The second scope pix shows the start and finish voltages of 1.823 and 15.86 respectively across C1 which is an output.  The energy is then (15.86^2-1.823^2) * 3.94e-6/2 = 488.9uJ .

The third pix shows the voltage across R1 is both -85.5mv mean and 1.46v rms on the Math channel over the 100ms time period.  The rms output energy across R1 is (1.46^2/1e-3)*.1 = 213.2uJ .  The negative mean voltage across R1 actually indicates energy is fed back to the 20v power supply V2.  This negative energy is (-0.0855/1000) * 20 *.1 = -171uJ .  A test was run but not documented here that used a charged capacitor C3 to replace the V2 power supply and indeed the finish voltage on C3 was higher than the start voltage confirming that energy was being fed back to the dc supply. 

A not so obvious function is that energy is drawn from C2 during operation and the fourth scope pix measures this.  The start voltage across C2 is 19.17v and the finish voltage is 12.92v.  Therefore the energy consumed  in C2 is (19.17^2-12.92^2) * 0.22e-6/2 = 22.1uJ .  This energy is considered as an input to the circuit function.   

The energy totals are as follows: Input = 712.9uJ + 22.1uJ +(-171uJ) = 564uJ and output = 488.9uJ + 213.2uJ = 702.1uJ .  The apparent COP = 702.1/488.9 = 1.24 .  If one considers the negative energy as an output,  then the COP = 488.9 + 213.2uJ +171uJ/712.9uJ + 22.1uJ = 873.1/735 = 1.19 .

Regards,
Pm

Pm

I am lost as to how L1 in your circuit has 330uF of capacitance.
Im guessing that should read uH ?.


Brad


---------------------------
Never let your schooling get in the way of your education.
   
Hero Member
*****

Posts: 704
Pm

I am lost as to how L1 in your circuit has 330uF of capacitance.
Im guessing that should read uH ?.


Brad

Brad,

LOL, yes you are correct.  Thanks for pointing that out and I will make the change.

Regards,
Pm
   
Hero Member
*****

Posts: 2853
Slider will probably update here too

another Vid from Stiffler ,and apparently Ground is on the Menu [rather have a cheeseburger :'}

Mark
Quote from Stefan's forum

The latest video has piqued interest:
https://www.youtube.com/watch?v=66KTJGlbKmM

Dr. Stiffler is using a 12V SLA, a SEC circuit, a couple of L3's, a 100ohm resistor, the diode loop and the Cree board.

What I like is the description underneath:
Quote

  stiffler quote
  Doing final check on the parts to be used in the loop distance test. There will be no single wire, test will be through the earth only.



Outdoors through the earth...exactly what many have been stuck on, including myself most frustratingly for years.
   

Hero Member
*****

Posts: 1296
On the other hand, we used to harvest earthworms for fish bait, by sticking the electrodes of an old army surplus telephone dynamo into the soil and cranking away.  The earthworms come to the surface, for yards around.

https://www.ebay.com/itm/German-Army-60V-20Hz-Hand-crank-Kurbelinduktor-Field-Phone-Generator-N-I-B-/362370750825



---------------------------
"The easiest person to fool is yourself" -- Richard Feynman
   
Hero Member
*****

Posts: 1331
... .-.. .. -.. . .-.
TK  O0
A proper belly laugh on seeing that and yet it's completely what should be investigated. A high voltage localised source, doing the job very effectively.

Such a thing makes a bunch of sense for literal field tests with receiving circuits or lamps. No batteries, no earthing worries from a mains feed. Walk anywhere with it and simply hand crank.
It's the Trevor Bayliss wireless energy machine !


---------------------------
ʎɐqǝ from pɹɐoqʎǝʞ a ʎnq ɹǝʌǝu
   
Hero Member
*****

Posts: 704
I don't want to derail this thread but honestly, I'm not sure where Dr. Stiffler is leading this parade so I'm focused on the potential ability of a so-called single wire device to produce OU.  I will continue to do so and unless someone has a problem and speaks up, I will continue on.

Here is a variation of the Stiffler/Avramenko device as seen in the schematic below.  The open wire connected to D3 and D4 is 61cm in length and hangs off the edge of the bench.  The length of this wire is critical to obtaining a COP>1 along with the operating frequency and component layout.  Diodes D1-D4 form a bridge from the input to space that allows the charging of C1.  The measurements are relatively straight forward as there is no bias, etc.

I've included a spreadsheet that gives the COPs for sweep periods from 10ms -70ms in 10ms steps.  For those not able to read Excel, the max COP =1.173 at 10ms and the min is .991 at 70ms but doesn't vary linearly.

The two scope pix show the sampled input wattage and the output voltage across C1 measurement profiles at 40ms.  Prior to any measurement cycle, C1 is fully discharged and the current probe is zeroed.

I've also included a pix of the layout.

Regards,
Pm

   
Hero Member
*****

Posts: 704
All the previous test were made using a sine wave from a signal generator.  This test uses an Ixys IXDD609P1 mosfet driver operating at 25vdc supplying a ground referenced square wave to L1.  The only other change is the operating frequency used was 2.37MHz.

The following is taken from the spreadsheet for this test and I hope the formatting holds.

Sweep Period (ms)   Input Power (mW)   Input Energy (uJ)    C1 Voltage   Output Energy (uJ)      COP
               
20                                    2.368               4.74E-05                 10.01           1.97E-04             4.17
30                                    5.4                       1.62E-04                 14.84             4.34E-04             2.68
40                                    9.066               3.63E-04                 19.25             7.30E-04             2.01
50                                  12.51               6.26E-04                 22.93             1.04E-03             1.66
60                                  15.88               9.53E-04                 27.31             1.47E-03             1.54
70                                  19.00               1.33E-03                 30.92             1.88E-03             1.42
80                                  21.81               1.74E-03                 33.71             2.24E-03             1.28
90                                  24.16               2.17E-03                 36.2             2.58E-03             1.19
100                                  26.31               2.63E-03                 38.03             2.85E-03             1.08

The scope pix shows a 50k expanded 2us period of a measurement cycle so one can see the resolution is reasonably accurate at 100M samples/second on the horizontal sweep.

It appears that this device shows promise using square wave drive which is far easier to implement compared to sine wave IMO.

Regards,
Pm

Edit: I see the formatting is messed up so I've attached the spreadsheet.
   

Hero Member
*****

Posts: 1944
Good work PM

Have you thought that the diodes are forming a varactor where the capacitance is being varied by the voltage in the output cap? Just a thought!

Regards

Mike 8)


---------------------------
"All truth passes through three stages. First, it is ridiculed, second it is violently opposed, and third, it is accepted as self-evident."
Arthur Schopenhauer, Philosopher, 1788-1860

As a general rule, the most successful person in life is the person that has the best information.
   
Hero Member
*****

Posts: 704
Good work PM

Have you thought that the diodes are forming a varactor where the capacitance is being varied by the voltage in the output cap? Just a thought!

Regards

Mike 8)

Mike,

I hadn't thought of that but yes I believe it is possible.  I'm presently using 1N4148s for the diodes and haven't really tried anything else at this point but I will. 

I also plan to do some spectrum analysis while changing the length of the "antenna" to see if I can see anything relative as I suspect the possibility of the antenna wire forming a transmission line with the permittivity of space resulting in a slowing of velocity.  This would explain the "tuning" results with such a short length of wire.  I'm also going to try to sweep the signal generator at a rate equal to the decrease in capacitance of the circuit in an attempt to maintain a constant transfer function.

The question remains that if the results are truly OU, then what is the source?  IMO the device is "open" and connected to the aether so this would be my best guess at this point.

Regards,
Pm

Edit: BTW, I do have some varactor diodes around here someplace!
   

Hero Member
*****

Posts: 1944
Mike,

I hadn't thought of that but yes I believe it is possible.  I'm presently using 1N4148s for the diodes and haven't really tried anything else at this point but I will. 

I also plan to do some spectrum analysis while changing the length of the "antenna" to see if I can see anything relative as I suspect the possibility of the antenna wire forming a transmission line with the permittivity of space resulting in a slowing of velocity.  This would explain the "tuning" results with such a short length of wire.  I'm also going to try to sweep the signal generator at a rate equal to the decrease in capacitance of the circuit in an attempt to maintain a constant transfer function.

The question remains that if the results are truly OU, then what is the source?  IMO the device is "open" and connected to the aether so this would be my best guess at this point.

Regards,
Pm

Edit: BTW, I do have some varactor diodes around here someplace!

PM

The diodes cathode to cathode as are, with the output cap changing voltage to the combined dielectric of the 4148's (-//-) supplying the bias, is what is happening I think.

In the past I have used two diodes like this for making a VCO, works really well ^-^

Regards

Mike 8)

A sort of auto tune with the aether maybe, moving according to what is in that local area :-\


---------------------------
"All truth passes through three stages. First, it is ridiculed, second it is violently opposed, and third, it is accepted as self-evident."
Arthur Schopenhauer, Philosopher, 1788-1860

As a general rule, the most successful person in life is the person that has the best information.
   
Hero Member
*****

Posts: 704
PM

The diodes cathode to cathode as are, with the output cap changing voltage to the combined dielectric of the 4148's (-//-) supplying the bias, is what is happening I think.

In the past I have used two diodes like this for making a VCO, works really well ^-^

Regards

Mike 8)

A sort of auto tune with the aether maybe, moving according to what is in that local area :-\

Mike,

Interesting!  I was involved in some experimentation years ago on parametric devices using varactors and found that mosfets with the gate connected to the source made better non-linear sources than varactors or other diodes by a magnitude or more.  Another test variation to add.

In the meantime, here is another test result from the same circuitry but with a slight frequency adjustment.  The apparent gain at 20ms is infinite with the other readings being higher than the previous.

Regards,
Pm

Edit: Correction
   
Hero Member
*****

Posts: 704
These test variations use three paralleled diodes for each bridge element to raise the capacitance of each leg and C1 is changed to 1.98uf as seen in the schematic below.  The tests were run at 2.2MHz and 2.3 MHz for comparison.  The test protocol has also changed in that eight samples are averaged with C1 discharged between each sample as this seems to yield more consistent measurement results.

Input and output scope pix measurements at 50ms are attached for examples plus the data tables for each frequency.

Regards,
Pm

 
   
Hero Member
*****

Posts: 2301
These test variations use three paralleled diodes for each bridge element to raise the capacitance of each leg and C1 is changed to 1.98uf as seen in the schematic below.  The tests were run at 2.2MHz and 2.3 MHz for comparison.  The test protocol has also changed in that eight samples are averaged with C1 discharged between each sample as this seems to yield more consistent measurement results.

Input and output scope pix measurements at 50ms are attached for examples plus the data tables for each frequency.

Regards,
Pm

 

Wow = intriguing results, PM?   COP's of around 4 and 5...?? 
All I can say is, keep plugging!
I had about given up on Stiffler, but your "variation" is looking promising --
with actual measurements of Pout and Pin which is VERY good progress.

I now have a new FeelTech FY3200S signal generator... 2 MHz easily and tunable of course, 20V p-p...  Do you think this would
work with your circuit?
   
Hero Member
*****

Posts: 704
Wow = intriguing results, PM?   COP's of around 4 and 5...?? 
All I can say is, keep plugging!
I had about given up on Stiffler, but your "variation" is looking promising --
with actual measurements of Pout and Pin which is VERY good progress.

I now have a new FeelTech FY3200S signal generator... 2 MHz easily and tunable of course, 20V p-p...  Do you think this would
work with your circuit?

I'm not really familiar with that generator but it sounds like it would work.  You could lower the circuits optimum frequency if needed by increasing L1 but I would try to keep the frequency as high as reasonable with your measurement tools.  I plan to increase L1 and drop the frequency until no OU is seen and if the kHz range can be reached, the measurements would be greatly simplified and more accurate.  Interestingly, a spectrum analysis shows that the 4th and 6th harmonics IIRC are stronger than the fundamental with the pulse drive!?

The problem is in the scope used for the measurements.  I'm hoping to figure out a protocol and circuit that will allow lower performance scopes to be used so others can replicate if they desire. 

Since the OU appears to be greater at the start of the cycle and diminishes as the cycle continues, my next step is to try sweeping the applied frequency to maintain the voltage/current phase relationship thru the entire cycle for a more constant apparent OU as the capacitance of the circuit decreases.

Pm 
   
Group: Elite
Hero Member
******

Posts: 3227
It's turtles all the way down
Going one step further, with a COP>5 it should be simple to loop this device. The oscillator circuit (SE or PP) can be driven directly off the capacitor and HF transformer coupled back to the input for reasonably high conversion efficiency and best impedance matching.


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   
Hero Member
*****

Posts: 704
Going one step further, with a COP>5 it should be simple to loop this device. The oscillator circuit (SE or PP) can be driven directly off the capacitor and HF transformer coupled back to the input for reasonably high conversion efficiency and best impedance matching.

That's a good idea ION as the C1 voltage at 2.2MHz reaches ~15v but the energy is low at 223uJ and if cycled every 10ms would yield a potential power of only 22mw.  That is if the OU is real!  However, with that amount of gain, the supply voltage should continually increase producing increasing potential power levels.  Sound familiar?

Pm
   

Hero Member
*****

Posts: 1944
Going one step further, with a COP>5 it should be simple to loop this device. The oscillator circuit (SE or PP) can be driven directly off the capacitor and HF transformer coupled back to the input for reasonably high conversion efficiency and best impedance matching.

Coupling off the cap I think would change the system totally, I have tried that many times, it just does not work, whatever the cop is. As far as I am concerned, the ou power input has to be from within the circuit and part of it, output to input just detunes every time.

Regards

Mike


---------------------------
"All truth passes through three stages. First, it is ridiculed, second it is violently opposed, and third, it is accepted as self-evident."
Arthur Schopenhauer, Philosopher, 1788-1860

As a general rule, the most successful person in life is the person that has the best information.
   

Hero Member
*****

Posts: 1944
As an addition to my above post, any electronic circuit, especially in the RF area, will never be able to be looped directly from output to input without detuning the +cop created in the first place.

Now if you look to the return currents and voltages  within the system (standing waves etc), then I think there is a chance, no, I think it can be done. The + cop has to come from somewhere, and that somewhere has to be from what is around us or a physical or chemical change due to what the circuit is doing.

If anyone knows of something else, let me know.

Regards

Mike 8)


---------------------------
"All truth passes through three stages. First, it is ridiculed, second it is violently opposed, and third, it is accepted as self-evident."
Arthur Schopenhauer, Philosopher, 1788-1860

As a general rule, the most successful person in life is the person that has the best information.
   
Hero Member
*****

Posts: 704
I was finally able to get a gated sweep to drive the Ixys driver to the Rev7 circuitry and after some optimization this is the result.  With these parameters I couldn't believe the results so I ran the test several times and took the worst case data to post here.  The sweep frequency used was from 2.3MHz to 2.7MHz and again the reason for this was to attempt to maintain a constant voltage/current phase relationship that was giving the high COPs in the starting charge periods as the circuit capacitance decreased over time.  The circuit used is the Rev7 as previously posted with the only change being the sweep.

The data table is shown below along with the sample scope shots showing the 50ms measurements.

I think I will now attempt to lower the operating frequency to see the effect on the COPs to find a bottom limit.

Pm

   

Hero Member
*****

Posts: 1944
Nice results PM O0

Seems that some kind of auto frequency (tracking) is needed in relation to the capacitance change/ voltage rise of C1.

Is the floating ground still the short wire hanging off your bench?

Regards

Mike 8)


---------------------------
"All truth passes through three stages. First, it is ridiculed, second it is violently opposed, and third, it is accepted as self-evident."
Arthur Schopenhauer, Philosopher, 1788-1860

As a general rule, the most successful person in life is the person that has the best information.
   
Hero Member
*****

Posts: 704
Nice results PM O0

Seems that some kind of auto frequency (tracking) is needed in relation to the capacitance change/ voltage rise of C1.

Is the floating ground still the short wire hanging off your bench?

Regards

Mike 8)

Mike,

Thanks.  Frequency tracking is definitely in order and I failed to mention that the sweep was linear but will try log.  Yes, still the same 56cm wire hanging off the bench.  I'm now curious to see how using mosfets for the diode with the large non-linear drain to source capacitance will work but need to lower the frequency first.

Regards,
Pm
   
Hero Member
*****

Posts: 2301
  Parzman, at 40 ms sweep, you record a cop of 20.8 which is awesome   O0

My question regards the data at 10, 20 and 30 ms sweep - which show negative Pin (and Uin).  What is the meaning of this negative input??

I'm also very curious to see what happens as you lower the frequency.  Thanks again!
   
Group: Elite
Hero Member
******

Posts: 3227
It's turtles all the way down
Coupling off the cap I think would change the system totally, I have tried that many times, it just does not work, whatever the cop is. As far as I am concerned, the ou power input has to be from within the circuit and part of it, output to input just detunes every time.

Regards

Mike

Dear Mike

Your observation is correct for a basic tank circuit, however the capacitor in this case is a progressive peak charging accumulator that is not directly involved in the tank resonance, but is inserted into it. In that respect, the accumulator capacitor is not really a factor in the overall resonance of the circuit, this resonance being a function of the self resonance of the coil and the tank formed by the coil, external capacitor and diode capacitance.

The only direct involvement I can think of is the charging voltage on the capacitor back biasing the diodes, thus altering their capacitance, this would be the parametric element, but I could be wrong about this, as back bias may be far less effective than forward bias in changing capacitance.

If the circuit is redrawn with the  stray capacitance of the dangling clip lead returned to the ground of the sig gen, we have a series resonant circuit formed by the stray capacitance of the clip lead and the coil, and the variable capacitance of the diodes, this in addition to the coil self resonance due to it's own parasitic capacitance. The peak detector is inserted in this loop and it's capacitor does not directly contribute to the resonance of the circuit, it's value is too low for the frequencies involved.

What suggestions do you have for looping the circuit, as I thought the idea I suggested had some merit based on the topology.

Maybe if I post a schematic what I had in mind would be clearer.

regards

P.S. when performing these experiments it is a good idea to elevate the coil with some porcelain standoffs, silk thread or nylon line for consistency of capacitance to the air, as lying on a bench can lead to variabilities in capacitance. I would for consistency also suggest using a tunable "gimmick capacitor"  back to the gen ground instead of flying cliplead. A sturdy breadboard with fixed positions for all components should produce greater consistency in measurement.


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   

Hero Member
*****

Posts: 1944
Dear Mike

Your observation is correct for a basic tank circuit, however the capacitor in this case is a progressive peak charging accumulator that is not directly involved in the tank resonance, but is inserted into it. In that respect, the accumulator capacitor is not really a factor in the overall resonance of the circuit, this resonance being a function of the self resonance of the coil and the tank formed by the external capacitor and diode capacitance.

If the circuit is redrawn with the  stray capacitance of the dangling cliplead returned to the ground of the sig gen, we have a series resonant circuit formed by the stray capacitance of the cliplead and the coil, and the variable capacitance of the diodes. The peak detector is inserted in this loop and it's capacitor does not contribute to the resonance of the circuit, it's value is too low for the frequencies involved.

What suggestions do you have for looping the circuit, as I thought the idea I suggested had merit based on the topology.

Maybe if I post a schematic what I had in mind would be clearer.

regards

A schematic would be good ION. :)

PM stated that the sweep is linear, looking at the change as the C1 charges and applies bias to the diodes, I think any change is going to disrupt things IMO :-\

I have more work to do on this but, I found that if any "natural" return charge is available without taking from the final output, things don't detune and show no OU.
This is the biggest stumbling block >:(

Soon back on the bench, though it is bloody hot here, 37ºc and storms at night, very Caribbean, even the sea temp is 30ºc which is warmer than my pool, 28ºc :D

Regards

Mike 8)


---------------------------
"All truth passes through three stages. First, it is ridiculed, second it is violently opposed, and third, it is accepted as self-evident."
Arthur Schopenhauer, Philosopher, 1788-1860

As a general rule, the most successful person in life is the person that has the best information.
   
Hero Member
*****

Posts: 704
  Parzman, at 40 ms sweep, you record a cop of 20.8 which is awesome   O0

My question regards the data at 10, 20 and 30 ms sweep - which show negative Pin (and Uin).  What is the meaning of this negative input??

I'm also very curious to see what happens as you lower the frequency.  Thanks again!

Assuming that the measurements are accurate (which is always a question), the negative input IMO means that the excited circuit is receiving energy from some outside source that exceeds the circuit's basic energy requirement.  IOW, if the input energy is a positive value and less than the output energy, the COP is a positive value >1.  If the input energy measures as a negative value, then we have a negative COP which would be infinite.

My initial tests at a lower frequency of 400-600kHz look promising but a more complete analysis is in order to be able to optimize the results.

Pm
   
Pages: 1 2 3 4 5 6 7 [8] 9
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2018-10-19, 07:50:01