Claimed OU circuit of Rosemary Ainslie

[MileHigh]

.99 and others:

A recurring theme for me recently due to the questionable data from the DSO has been "Back up!  Back up!  Back up!"  By that I mean that you go back to basics and use very simple measurement techniques or do some other related experiments to get your feet on a solid foundation first.  You have to make some measurements that all check out fine, something to give you confidence to move forwards.

Well I just had a great idea for doing a related experiment to find that solid foundation that would be common ground for everyone.

Let's start with the 3.7% duty cycle 2.4 KHz waveform.  That "sort of looks like" the following:  A square wave that is 3.7% ON and 3.7% OFF, covering 7.4% of the 2.4 KHz available time slot.  (100/7.4) = 13.5.   Therefore a 50% duty cycle square wave of about (2.4 KHz x 13.5) = 32.4 KHz would sort of resemble the 3.7% duty cycle 2.4 KHz waveform.

We are now going to make an assumption:  Measuring the power consumption for the components in the circuit would also be difficult for the DSO if the 555 output was a 32.4 KHz 50% duty cycle square wave.

Now here is the interesting part:  We can vary the frequency of the 50% duty cycle square wave, and we know that the power measurements should always be essentially the same.  We also know that the lower the square wave frequency, the less and less the load resistor inductive component, and the stray capacitive and inductive effects, should have on the circuit.  In other words, the spikes become less and less of an issue as the square wave frequency lowers.

So here is the test:  Make measurements where the 555 output is a 50% duty cycle square wave at say....  50 Hz.   Yes, a lousy 50 cycles per second.  At this frequency the load resistor will look much more like an ordinary resistor with no inductance at all.  The spikes will still be there, but they will be negligible in comparison to the resistive burn energy that will be happening every time the MOSFET is switched on.  I think .99 just mentioned that when the MOSFET is on you sustain about a 4-volt drop across it.  So the load resistor voltage is about 20 volts less the shunt voltage.  You can easily calculate what the power in the load resistor for a 50% duty cycle should be on paper, it's a no brainer.  You just have to switch the MOSFET on for a few seconds and measure the voltage drops and calculate the current and you are ready to go.

So you do a run with a 50 Hz 50% duty cycle square wave and make all of your DSO recordings and crunch the data.  The DSO will be sampling at a much lower rate because of your much slower time base but of course you will still be able to get thousands of samples per cycle.

We are going to assume that the DSO data crunching and the on-paper calculations and the real world thermal profiling are all going to be in accord here.  They simply have to be in accord at this very low frequency.  If they aren't then something has to be very very wrong somewhere.

Now the next step is to start to up your square wave frequency.  100 Hz, 200 Hz, 500 Hz, 2KHz, etc, etc.  We know that as the frequency increases, the MOSFET switch starts to see more high frequency "smack" spikes as well as more body diode conduction cycles.  Therefore you should start to see a trend in the data that shows slightly less power dissipated in the load resistor and slightly more power in the MOSFET.

At a certain point as you keep upping the square wave frequency you should start to see anomalies in the trending of the power data for each component.  I am making this assumption based on the multiple instances of strange load resistor power data that we have seen so far.  This is telling you that something somewhere is causing something to go awry with your measurements.  .99 speculated as to some of the possible causes in a previous posting.

For example if you start to see your load resistor power start to shoot way up (as evidenced in the data seen so far) but you measure the temperature of the load resistor but nothing special is happening - then you are 100% certain that you have pushed the DSO past it's limits - or - (and this one is much more likely) you have to be very highly skilled in understanding how to use your equipment properly under these conditions and have to have a really deep understanding of transient responses in circuits, and how the geometry of the wires and the layout comes into play.

Anyway, supposing that someone does this investigation and finds out that the results start to go awry when the square wave frequency is above 10 KHz.  Since the real 3.7% 2.4 KHz excitation frequency is akin to a 32.4 KHz square wave, that explains why the DSO measurements seem to be out of whack.

You can see the "Back up!" mentality at play here.  You are trying to answer a basic question:  How accurate are my DSO readings for the Ainsley circuit as I sweep my excitation frequency higher and higher?

We all jumped into the DSO measurements assuming that it would be a piece of cake to make the measurements on this circuit.  We have seen strange data, so it is time to take a step back.

MileHigh

[MileHigh]

Glen and Rosemary:

I am going to put on my "fuzz" hat - like a cop handing out speeding tickets, sorry.   I am going to hunt down and rebut some of your points as a reality check for both of you and for the readers.

Glen:

Here is some data from a test run using a new 10 ohm Mosfet gate potentiometer to try to bring a better percentage accuracy to the required 5.8 to 5.3 ohms that seems to make this circuit run much more efficient

More efficient than what?  It is a serious question, what are you implying?

The word "efficiency" is bandied about by everyone in the forums in the same vein as "resonance."  Think about this:  Any variation of the Ainsley circuit you build is essentially 100% efficient at producing heat, but the wattage draw from the battery will be different from circuit to circuit.

And the results were interesting to say the least ..... and the final Image and Data dump on the 100ns had gains that hasn't ever been seen before, and if possible to maintain these values would be incredible ......

I see very high drain spikes in the last image for sure, but what do you mean by "gains?"  Again, that is a serious question for you.  This is typical for the forums, make a statement like that and everyone agrees with you but what are you really talking about, gains relative to what?

I am assuming that you are implying energy gains somehow, and not a voltage gain in the spike.  Simply because there is nothing "incredible" about the voltage spike.  If you are talking about the voltage gain in the spike and are not implying energy gains, then I will ask you the same question, what is so amazing about this?  What is this information telling you?

I am giving you a hard time here because I can imagine the chorus of cheers if you record an even higher voltage spike.  The chorus of cheers would imply that everyone thinks that this is "near proof" that the circuit is showing some kind of energy gain.  The problem is that right now there is no proof at all that a higher voltage spike on the MOSFET drain pin equates to energy gains at all.  I feel compelled to say this, to being everyone back to reality.

I also have a challenge for you:  Now that you have posted the thermal profiling, you are in a position to analyze your data.  Perhaps .99 could give you the spreadsheet that he used to crunch your data and you could play with it and try loading in other data captures that you have done in the past.  You can plot your thermal profile data on a piece of graph paper with a pencil and ruler, or you could plot it in Excel, which really shouldn't be too hard.

You have generated a lot of good data and posted it, but the real issue and the real challenge at hand is to analyze it.  I hope that you take the plunge and don't be shy and post some questions if you need help getting up the learning curve.

Rosemary:

Just noted the harmonics in that last posting of yours.

Well I saw those postings myself and I didn't see any harmonics.  Is it possible that what you are calling "harmonics" are something else?  You should be very careful with that word.  Normally harmonics are much lower in amplitude than the fundamental frequency of the waveform and are nearly invisible to the naked eye.  You simply can't see them.

For example, a sine wave of 100 Hz added with a 200 Hz first harmonic that's at -20 dB will still look pretty much like an ordinary sine wave to the naked eye, you simply can't see the harmonic.

Sorry for being a bad boy but there is good food for thought here.

MileHigh

[fuzzytomcat]

My Ainslie Circuit Test Plan. This is essentially what I am following.

It may not be perfect, but at least I wrote and posted one. 

Notice in the photo on page 4 the wave forms on the scope. This is what is responsible for the skewed negative power in the MOSFET when the scope multiplies those two wave forms together. That represents a -2250W (~150ns) power spike and over one cycle amounts to about -0.84W.

The investigation continues...

.99

Hi .99

Nice PDF ........ it's to bad your not using a replication of a "Quantum" 10 ohm "load resistor" thats outlined in the original October 2002 issue

http://www.feelthevibe.com/free_energy/rosemary_ainslie/transient_energy.pdf 

My prototype is similar to it's size and configuration with much better results because of the obvious differences in micro Henri's .....

I see your also not using any battery's at all for your testing and data collection ..... not much of a "replication" ......

I also noticed in your set-up photograph included in the PDF you have the "load resistor" pointed directly at your "groundloop" PCB circuit ....... this could be possibly the worse place in the world for it to be, with getting strange and outright weird effects that come from the ends of the 10 ohm load resistor during operation ( point it at a compass or play with some magnets ) ..... just a observation 

Fuzzy 

Edit - added battery note

[poynt99]

Those aren't harmonics in the posted multi-cycle wave forms.

What you are seeing there Rose (and Harvey) is the difference between what is really there and the scope's ability to display and acquire what is really there. You may think of it as a beat frequency of sorts.

There are no harmonics there. Reduce the time base. It is unnecessary (and undesirable) to use such a long time base here; all those missing spikes etc., really are missing from the data.

It's as MH mentioned. You really need to have a good handle on the use of these scopes and techniques because the quality of the acquired data depends on it. Grounding is also a huge factor, as I hope to illustrate shortly.

.99

[Hoppy]

.99 and others:

A recurring theme for me recently due to the questionable data from the DSO has been "Back up!  Back up!  Back up!"  By that I mean that you go back to basics and use very simple measurement techniques or do some other related experiments to get your feet on a solid foundation first.  You have to make some measurements that all check out fine, something to give you confidence to move forwards.

Well I just had a great idea for doing a related experiment to find that solid foundation that would be common ground for everyone.

Let's start with the 3.7% duty cycle 2.4 KHz waveform.  That "sort of looks like" the following:  A square wave that is 3.7% ON and 3.7% OFF, covering 7.4% of the 2.4 KHz available time slot.  (100/7.4) = 13.5.   Therefore a 50% duty cycle square wave of about (2.4 KHz x 13.5) = 32.4 KHz would sort of resemble the 3.7% duty cycle 2.4 KHz waveform.

We are now going to make an assumption:  Measuring the power consumption for the components in the circuit would also be difficult for the DSO if the 555 output was a 32.4 KHz 50% duty cycle square wave.

Now here is the interesting part:  We can vary the frequency of the 50% duty cycle square wave, and we know that the power measurements should always be essentially the same.  We also know that the lower the square wave frequency, the less and less the load resistor inductive component, and the stray capacitive and inductive effects, should have on the circuit.  In other words, the spikes become less and less of an issue as the square wave frequency lowers.

So here is the test:  Make measurements where the 555 output is a 50% duty cycle square wave at say....  50 Hz.   Yes, a lousy 50 cycles per second.  At this frequency the load resistor will look much more like an ordinary resistor with no inductance at all.  The spikes will still be there, but they will be negligible in comparison to the resistive burn energy that will be happening every time the MOSFET is switched on.  I think .99 just mentioned that when the MOSFET is on you sustain about a 4-volt drop across it.  So the load resistor voltage is about 20 volts less the shunt voltage.  You can easily calculate what the power in the load resistor for a 50% duty cycle should be on paper, it's a no brainer.  You just have to switch the MOSFET on for a few seconds and measure the voltage drops and calculate the current and you are ready to go.

So you do a run with a 50 Hz 50% duty cycle square wave and make all of your DSO recordings and crunch the data.  The DSO will be sampling at a much lower rate because of your much slower time base but of course you will still be able to get thousands of samples per cycle.

We are going to assume that the DSO data crunching and the on-paper calculations and the real world thermal profiling are all going to be in accord here.  They simply have to be in accord at this very low frequency.  If they aren't then something has to be very very wrong somewhere.

Now the next step is to start to up your square wave frequency.  100 Hz, 200 Hz, 500 Hz, 2KHz, etc, etc.  We know that as the frequency increases, the MOSFET switch starts to see more high frequency "smack" spikes as well as more body diode conduction cycles.  Therefore you should start to see a trend in the data that shows slightly less power dissipated in the load resistor and slightly more power in the MOSFET.

At a certain point as you keep upping the square wave frequency you should start to see anomalies in the trending of the power data for each component.  I am making this assumption based on the multiple instances of strange load resistor power data that we have seen so far.  This is telling you that something somewhere is causing something to go awry with your measurements.  .99 speculated as to some of the possible causes in a previous posting.

For example if you start to see your load resistor power start to shoot way up (as evidenced in the data seen so far) but you measure the temperature of the load resistor but nothing special is happening - then you are 100% certain that you have pushed the DSO past it's limits - or - (and this one is much more likely) you have to be very highly skilled in understanding how to use your equipment properly under these conditions and have to have a really deep understanding of transient responses in circuits, and how the geometry of the wires and the layout comes into play.

Anyway, supposing that someone does this investigation and finds out that the results start to go awry when the square wave frequency is above 10 KHz.  Since the real 3.7% 2.4 KHz excitation frequency is akin to a 32.4 KHz square wave, that explains why the DSO measurements seem to be out of whack.

You can see the "Back up!" mentality at play here.  You are trying to answer a basic question:  How accurate are my DSO readings for the Ainsley circuit as I sweep my excitation frequency higher and higher?

We all jumped into the DSO measurements assuming that it would be a piece of cake to make the measurements on this circuit.  We have seen strange data, so it is time to take a step back.

MileHigh

I'm not able to pin down the cause of these anomalous readings but bearing in mind we have a current switching direction, I think MH may have touched on the cause because I think its possible that we have a 'standing wave' developing on the mosfet drain / source conductors, which is giving rise to a varying voltage along the length of the conductors which would result in various results depending on the positioning of the DSO probes. Kirchoff's laws are reputed to be non-applicable to pulsed inductive circuits. 

Hoppy