PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2017-11-19, 08:53:01
News: Check out the Benches; a place for people to moderate their own thread and document their builds and data.
If you would like your own Bench, please PM an Admin.
Most Benches are visible only to members.

Pages: [1] 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ... 29
Author Topic: The Rosemary Ainslie Circuit  (Read 265622 times)

Group: Administrator
Hero Member
*****

Posts: 2610
It's not as complicated as it may seem...
Recently, after much work in testing, data-gathering, and writing, Rose and her team have put together a paper which encompasses the intent of this work. This was all in an effort to prove her thesis regarding the true operating nature of a switched inductive/resistive circuit in terms of the COP that can be obtained when specific variables in load resistor dimensions and drive frequencies are met.

The essence of her claims is that having met these certain parameters, the switched load resistor will exhibit more output energy in heat when compared to the total energy supplied to the circuit, and a further sub-claim that indeed this COP>1 is at least partially achieved by the active recharge of the source wet lead acid battery.

The paper has been submitted by Rose and her team to the IEEE and is here for public review.

.99
   
Group: Guest
.99:

After reading the paper the first thing that comes to mind is "Where's the beef?"  The Ainsley team needs to present power-out vs. power-in numbers to validate the claim as a first step, and they are not there.

MileHigh
   
Sr. Member
****

Posts: 284
.99:

After reading the paper the first thing that comes to mind is "Where's the beef?"  The Ainsley team needs to present power-out vs. power-in numbers to validate the claim as a first step, and they are not there.

MileHigh

Yes, the AT team have left out the most important and juicy bit. Maybe they cannot find a way to convincingly show that the system is operating OU using their test results.

Hoppy
   

Group: Administrator
Hero Member
*****

Posts: 2610
It's not as complicated as it may seem...
IMHO, the paper does not contain convincing evidence to prove any claims of overunity. My own tests have shown that "conventional" testing methods are the main reason the test results appear inconsistent, and questionable.

Any response from the IEEE will be interesting, however I believe that their conclusions (if any) will not be all that different from the ones expressed here so far.

I have developed a detailed analysis of the problem areas and also provided practical solutions to overcome these problems. Although the AT implemented only a small portion of these solutions, there was a marked improvement in the accuracy of the results.

I will post my analysis within the next few days.

.99
   
Group: Guest
The paper mentions that a necessary part of tuning requires putting the circuit into a chaotic state whereby crosstalk from power switching causes gate excitation and so the resulting waveform is aperiodic. This makes for a great wall of obfuscation to hide behind and makes comprehensive analysis through finite sample analysis impossible.

I won't commit myself either way at the moment, but I will need more than this paper offers to convince me 100% that COP of this system is special.

I will say though that the paper was well written and presented and makes a more convincing argument than alot of what I read in that volumous OU thread.
   
Group: Elite
Hero Member
******

Posts: 3036
It's turtles all the way down
Quote
The paper mentions that a necessary part of tuning requires putting the circuit into a chaotic state whereby crosstalk from power switching causes gate excitation and so the resulting waveform is aperiodic. This makes for a great wall of obfuscation to hide behind and makes comprehensive analysis through finite sample analysis impossible.

I totally agree with this and therefore have offered an alternative test method that does not require finite sample analysis. This in no way should be meant to take away anything from the noteable work Poynt has done in this area. Kudos to Poynt for his Herculean effort.

But for those who do not have access to expensive data loggers, exotic scopes and wish to play with the Ainslie circuit, I would highly recommend a simple way of testing the circuit by using the attached diagram. A simple differential thermocouple setup will show discrepancies in temperature between the DUT and the control.

Since both resistors will be pulsed identically, there is no need for absolute temperature measurement, any differences between the DUT and control will be a differential in temperature. The control resistor should be a carbon composition or other non-inductive type with about the same thermal mass and surface area as the DUT.

Also a potentiometer can be placed in the gate of the DUT firing FET and adjusted for the so called "chaotic" condition.

I will later post the circuit for the differential thermocouple technique which can be used with a cheap DVM.
« Last Edit: 2009-12-15, 20:58:48 by ION »


---------------------------
Just because it has a patent application or is patented does not always mean it really works.
   

Group: Administrator
Hero Member
*****

Posts: 2610
It's not as complicated as it may seem...
Fraser,

Actually, the "aperiodic mode of oscillation" is a contentious issue, as imo use of the term is really a misnomer in this case. It would more aptly be called a "quasi-aperiodic mode" of oscillation, or QAM oscillation for short. I have obtained this QAM oscillation in my own testing of the circuit, and it is purely attributed to the non-periodic re-triggering of the 555 timer circuit used to drive the MOSFET. The resulting wave form output has a non-periodic component, but I would not classify it as chaotic.

The 555 can be configured as a crude voltage controlled oscillator (VCO) and the signals on pins 4, 5, and 6 will influence its timing parameters. Without going into a great deal of analysis, the gist of what occurs is that voltage transients on the 555-3 (output) and ground reference couple into the 555 circuitry and cause not only re-triggering of the output wave form at pin 3, but it can be adjusted to do so in such a manner that resembles the output of a swept VCO. The resulting pulse train is a pattern of between 2 to 6 pulses, each one with a decreasing pulse width and off-time between pulses, then the pattern repeats (i.e it is periodic). The very same result could be achieved by sweeping a 555 VCO circuit with a saw-tooth control voltage of appropriate amplitude and frequency

In any event, the switching wave form produced by this QAM oscillation being periodic, can easily be sampled and quantified using standard data acquisition at the appropriate sampling rate.

It should be noted however, that my observations are that the AT's data was not acquired with the circuit operating in the QAM oscillation mode. This can be easily verified by studying the wave forms as posted at EF.

.99
   
Group: Elite
Hero Member
******

Posts: 3036
It's turtles all the way down
Here's a better drawing of the Differential Temperature Test Technique (DTTT)

Any DVM with a reliable mV scale can be used for the meter. I recommend type "T" thermocouple wire cut from the same roll. Thermocouples are cheap, rugged and low impedance devices that are relatively noise immune, or can be easily filtered.

Alternately a Fluke 52 dual thermocouple meter or equivalent testing device can be used. Thermocouples are not absolutely necessary, a simple thermistor null bridge can be easily made with a small battery and a few resistors. Thermistors have some self heating issues which can cause errors unless low currents are used in the bridge.

Thermocouple and meter connections should be made to an isothermal block (I'll expand on that next post)

Anomalous energy production will show up as a difference in temperature between the two resistors, which can be easily detected with this setup.

This will only identify existence of anomalous energy production and not meant to make quantitative measurement of the energy production.

Quantitative energy production can be determined with DC equivalence test in which R2 is fed from DC power supply and power is nulled against DUT.

I tend to prefer null balance techniques because they quickly demonstrate any results (if they exist) without needing to log or interpret reams of data.
« Last Edit: 2009-12-16, 18:09:14 by ION »


---------------------------
Just because it has a patent application or is patented does not always mean it really works.
   

Group: Administrator
Hero Member
*****

Posts: 2610
It's not as complicated as it may seem...
Hi ION.

I think this is a great idea.

I would only have two suggestions/concerns, from a devil's advocate point of view:

1) Using a buffer between the 555 and the MOSFET may preclude us from obtaining the quasi-aperiodic mode (QAM) of oscillation, as it is reliant on feedback from the capacitively-coupled Drain spike to the 555 pin 3.

2) It could be argued that M1 and M2 need to be closely matched, which is not too difficult to do with the right test setup.

Cheers,
.99
   

Hero Member
*****

Posts: 2888
tExB=qr
what does the signal to the mosfet look like?   Just curious.
   

Group: Administrator
Hero Member
*****

Posts: 2610
It's not as complicated as it may seem...
Grumpy,

In "normal" mode, the 555-3 drive signal looks fairly square, with some spikiness, and sloping, but in QAM oscillation, there are extra distinct spikes on the 555-3 which couple into it's circuitry causing re-triggering of the 555.

When I get back to testing the RA circuit, I will show some nice scope shots of what I obtained for QAM oscillation.

.99
   
Group: Elite
Hero Member
******

Posts: 3036
It's turtles all the way down
Hi Poynt, Grumpy

My last post was an elementary schematic based on similar testing I have performed in the past, but not on "Ainslie" device. The buffers are optional, I just threw them in. One or both may be removed. There are many permutations of this basic method, e.g. Q2 could be slaved off Q1.

Below is an example of how to set up a proper differential thermocouple using type "T" thermoelements (others can be used).

Also a simple R-C filter can be easily added if noise pickup is experienced.
« Last Edit: 2009-12-17, 12:42:48 by ION »


---------------------------
Just because it has a patent application or is patented does not always mean it really works.
   

Hero Member
*****

Posts: 2888
tExB=qr
a simple disruptive discharge coil can melt tungsten with only a few watts of energy

How is this accomplished?

   

Group: Administrator
Hero Member
*****

Posts: 2610
It's not as complicated as it may seem...
Squeeze that few watts of power into a 10ps time frame.

.99
   

Hero Member
*****

Posts: 2888
tExB=qr
Squeeze that few watts of power into a 10ps time frame.

.99

is the Rosie circuit any different?

Also, when one of you have Rosie's Circuit replicated, try putting the inductive resistor in a vacuum (like a little box around it and use a hand vacuum pump) or place it in oil as see if this effects the heat output.
   

Group: Administrator
Hero Member
*****

Posts: 2610
It's not as complicated as it may seem...
They would be two different animals imo.

.99
   
Group: Guest
guys - I would have thought this subject has been exhausted.  In any event let me at least advise you that there's a schedule of results detailed on test 13 as being the one used for full description on the paper.  It's at the end of the paper because the IEEE want their schedules and diagrams submitted that way.  they then postion them as required.

the other thing is this.  there is some requirement that - in the event that Poynty concludes that the test results were erroneous - presumably as a result of his own failed attempts at test replication - then I have sworn under oath that I will retract the original claim.  I really can't oblige here Poynty.  If this IS required then I need to bow out of this forum.  My concerns are that you may not be able to replicate or worse still - may not want to.

then.  I have no idea how to find the threads on this forum.  I got to this one courtest Poynty's link on his email.  How does one find ones way around here?

and - finally - on a personal note.  I'm as blind as a bat.  I can just manage the print size of energetic forum.  am more comfortable with overunity.com. But this one entirely defeats me.  I'm typing blind - in the hopes that this is half way readable and ask that something be done here.  else I'll neither to able to read the posts with any degree of confidence - nor answer them.  white against light blue is another major stress for the focus.  But it's my problem not your's.  It's just I won't be able to get here with ease.  It now occurs to me that maybe that's why I can't find the threads.  they're marked in white on blue?  maybe. 

golly.  this typing is exhausting me.  can't see a thing.
   

Group: Administrator
Hero Member
*****

Posts: 2610
It's not as complicated as it may seem...
Rose,

Sorry for the troubles you are having here at the new site. I have checked our fonts against the fonts used at OU, and they are the same type and size. So I am uncertain wherein lies the problem. MileHigh has also expressed concern that the font is small. However, I do agree that our buttons are still difficult to see and read, and should be larger. I hope we will have that changed soon.

Regarding your paper and claims, they were in existence before this forum opened, and also since you have submitted your paper to the IEEE, I am confident they will come to the same conclusions that most of us here have already. Therefore, it will suffice to let the verdict fall on the IEEE, if you are in agreement with that.

I think the basic disagreement between your team and the rest of us is whether the data presented in the paper does or does not support the claim of COP>1. I have asked several individuals whom I know and respect about the paper and the test results in general, and all agree that there is nothing to substantiate a claim of COP>1, and I am in agreement with that, as I have been from the moment I realized the test results (both yours and my own) were faulty.

What will your position be if the IEEE comes back with either no comment or a simple rejection of the paper?

.99
   
Group: Guest
Hi Poynty.  I will not accept a 'verdict' by the IEEE.  They've got the responsibility of bringing new 'discoveries' to the public academic forum via their reviewed journals.  They are not expected to comment on experimental results.  IF however, there are any apparent contradictions in terms of the arguments in the paper against the documented experimental evidence - then that may be addressed by the reviewers prior to publication.  The onus is on the wider forum of all the journal readers to discuss the findings and either replicate, question, or refute the findings.  The IEEE have never been entitled to such sweeping judgements on the contributions on behalf of an enormous body of experts available in their readers.  Golly.  I fondly believe that the academics themselves would be angered at such gross violation of the their rights to information.  We need to advise them.  All of them.   ???  (Hope I've chosen an appropriate emoticon)
   

Group: Administrator
Hero Member
*****

Posts: 2610
It's not as complicated as it may seem...
guys - I would have thought this subject has been exhausted.  In any event let me at least advise you that there's a schedule of results detailed on test 13 as being the one used for full description on the paper.  It's at the end of the paper because the IEEE want their schedules and diagrams submitted that way.  they then postion them as required.

Exhausted? IMO, no.

Quote
the other thing is this.  there is some requirement that - in the event that Poynty concludes that the test results were erroneous - presumably as a result of his own failed attempts at test replication...

I think many, including myself are still waiting for a clear explanation of what a "successful" replication would be.

I believe I have for the most part, obtained very similar results to Fuzzy's, it terms of the wave forms.

.99
   
Group: Guest
Hi Poynty.  Your comments regarding the replication.  I'll detail my argument again.  You need to show that the energy being dissipated at the load exceeds the energy delivered by the battery.  If you do this, then you definitely have a gain.  The extent of the gain is open to discussion.  The waveforms are pretty chaotic - but when they're dominated by that harmonic they seem to flirt with overunity numbers.  Without the harmonic we seem to get values at some co-efficient of performance that varies between >4 and >9 as measured.  

The reason I'd like to 'move on' is that you guys are still debating that there was any gain at all.  Poynt - you cannot seriously propose that - because you couldn't replicate - the experiment is somehow wrong?  Surely?  

In any event - Fuzzy's 'live show' is on offer.  If you have questions you just need to ask and my guess is he'll be able to give you better indications directly off the experimental apparatus.  He's got it all there.
   
Group: Guest
Rosemary:

I am under the impression that what you and Glen call the "harmonic" is the sweeping up and down waveforms that you see embedded in the main waveform.  If that is the case then you are incorrect.  All that you are seeing is what are called "aliasing effects"

The standard display that Glen shows may or may not be under-sampled by the DSO's A/D conversion system since you are asking it to look at a fairly large time span.  This concern was voiced by myself and Poynt many times.  Then that data is re-sampled again for purposes of displaying the data on the screen.  Here is where the DSO has to make big compromises because it has to take a large amount of data and squeeze it down to a much smaller amount of data to display it on the screen at a fixed and limited pixel resolution.

The net result of all of this data manipulation to make it displayable on the DSO screen are the aliasing effects, sometimes called "aliasing" or "aliasing artifacts."  So what you call "harmonics" are just pretty patterns on the screen that actually don't exist in real life.

With respect to real harmonics, every single waveform you are looking at consists of thousands and thousands of harmonics, it all depends on how precise you want to be about it.  You can't "see" harmonics normally unless you are looking at very simple examples.  Just the fact that the waveforms are irregular is telling you right away that the waveform is made up of thousands of harmonics all added together.

To beat an old horse to death:  You have presented zero data to back up your claim of COP > 1.  For myself personally, all of the discussions about the aperiodic oscillations giving you greater COP are a red herring.  The is no substance to that argument.

For me, we are still back at Square One, you have not presented and credible data showing thermal power out being greater than electrical power in.  I see no numbers.  Yet you claim that you do have data.  This is where there is a total disconnect and I simply don't get it, I feel that you are being irrational with respect to this.

MileHigh

   
Group: Guest
Hello MileHigh

As I understand it aliasing is the result of waveforms that are not 'captured' by the oscilloscope - either the range setting is incorrect, meaning the voltage is too big for the full waveform to be shown on the screen, in which case the value is truncated and therefore incorrect, or the frequency is such that the scope is unable to see the required data.  Neither apply in this case.

The harmonic, as I understand it - is a resonating number that 'kicks in' and holds a signature form that is repeatably evident - hence its term as an harmonic.  But I'm no expert and would prefer to stay out of this discussion.  It is better held with Harvey or Fuzzy.

Then - I keep saying this.  We used #13 sample capture as the definitive example in our tests.  This is only because the results were sufficient to prove the point.  But they were, by no means, the optimal values.  These are better evidenced on earlier and later tests and Fuzzy frankly did not have the time to optimise this.  At the time we were anticipating the need to return the scope on completion of the paper or on the 1st of December whichever was the sooner.  So there was a need for speed.

The results of test 13 were given in a schedule - and schedules and figures had to be appended at the end of the document to accommodate the editing requirements of the IEEE.  I can't remember the number of that schedule - but if you look for it - it's at the end of the paper.  Sorry that this is not more prominent.  But I can assure you it's there.  You will notice that we did not even bother with the heat measured to be dissipated from other circuit components such as the Fet or the Shunt.  The heat from the fet was possibly as much as the heat dissipated at the load - but it would have required more detailed analysis and Fuzzy did not have extended use of the power supply needed to establish this.  In any event there was an average of 4.5 watts or thereby dissipated (from memory) at the load and this was enough for proof of concept and it allowed for generous margins for error - required for the crude heat measurements applied.

   
Group: Guest
And by the way.  As I understand it, the DPO takes a digital recording of the waveforms and translates this into the waveform graphs.  This is displayed but never in real time - just very nearly real time.  But the representations of the waveforms on each channel co-incide with the representations so it will display any shifts if these are applicable.  Best accuracy on his TDS3054C DPO is at 700 samples over a single screen shot of 10 000 samples.  I have not been able to ascertain the number in the sample range Fuzzy shows on the live broadcast.  The point here is that the greater the number of samples for analysis the more accurate the values.  I suppose there's an upper and a lower limit to this.  Just not sure what they are.
   
Group: Guest
Hi Rosemary,

Thank you for the season's greetings on OU.  I was going to reply to you, part of my excuse for being late is that I go through phases where getting to sleep is a problem.  So for the past few days I have been very very tired and restricted myself to the forum topics.  I have been kicked off of OU, so please accept my best wishes for the season and the new year from here!

Going back to your paper, I looked at the results of test 13 at the end of the document.  One more time Rosemary, I have to say it:  There is no thermal power out vs. electrical power in data at all to be found.  There is the chart for test 13 that Glen filled out by hand with the recorded temperatures, and a few DSO screen shots, that's it.  I just don't know what to say but to simply repeat that your paper shows no data supporting your claim of over unity.  What is it going to take for you and the rest of the team and the supporters to acknowledge this simple fact?  It truly boggles my mind that you can all be congratulating each other for a job well done when the job has not even been finished.  You have to produce numbers that are backed up by your test data, it is as simple as that.  You can skate around the issue all you want, but that fundamental fact is unavoidable.

I also want to give you some context for trashing Aaron's "treatise" that he posted on your thread on the EF that I then copied over to OU and bashed.  It hurts my ears and brain to read stuff like that especially knowing that "young and impressionable" newcomers to electronics are reading it.  In engineering parlance, it can be referred to as "obscene talk."  It is analogous to two truckers talking about explicit sex and using foul language in front of a bunch of six-year-old boys and girls.  So from my perspective it truly is offensive to read that kind of junk talk and once in a while I respond to it forcefully, just like I would respond forcefully to a bunch of truckers talking dirty in front of a group of young children.  I don't know if you can understand that but his "witch doctor" approach to electronics almost gives me a headache sometimes.  And he truly can't utter more than six sentences about electronics and energy without making a mistake.  I was not exaggerating or intentionally being mean when I said that.  I was watching every step of the way when he was doing his testing of your circuit and my comment stems from that experience, it was pretty awful.

MileHigh
   
Pages: [1] 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ... 29
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2017-11-19, 08:53:01