another small breakthrough on our NERD technology.

[Rosemary Ainslie]

Hello PhiChaser,

You've based your argument on the fact that a MOSFET transistor is a capacitor because it shares some material properties with a standard capacitor.  Which is interesting.  However, the only thing that it does not share is the ability to store charge.  It is designed to deliver charge.  If it had the ability to store charge then indeed your argument would hold.  But it is physically IMPOSSIBLE to disconnect a MOSFET and then apply it to the same or to an alternate circuit and expect it to then discharge what was first delivered.  That - essentially - would be required for one to argue its definition a capacitor.  Quite apart from which that capacity in that transistor then needs must be ENORMOUS. 

I see where you're going.  You're proposing that during the time that the battery is applying a positive charge to the gate of Q2 - then it's in the process of storing that charge.  Then the switch at the Gate of Q1 changes to become positive.  And simultaneously the switch at the Gate of Q2 changes to become negative.  And somehow, during this transition then all that energy STORED in Q2 is then discharged. On the face of it - it could perhaps be plausible.  Assuming always that it can even find a path through it's source leg Q2S to the Drain rail or battery positive - as Poynty refers to it.  Again.  Bear in mind that IF this energy is being returned it is still showing a voltage that is greater than zero.  And also.  Bear in mind that in the process of returning this energy it is also discharging about 72 000 amps of stored energy.  That's to account for the battery voltage reducing from 12 volts to 0.5.  Which not only begs a storage capacity somewhat larger than a standard capacitor.  But for some reason - rather confusingly - this returning energy BACK to the Drain rail of the battery - that positive terminal - somehow manages to then REDUCE that battery voltage from 12 volts to 0.5 volts,  Under normal circumstances IF energy is returned - one would expect it to recharge that battery.  And one would also expect the voltage to then be less than zero.

But that slew of improbable events is actually not even relevant.  Because, in point of fact, the signal at the Gate of Q2 does NOT change from a positive to a negative during this oscillation.  It stays negative.  For the duration. So.  I'm not sure that this proposal can be resolved by proposing that the MOSFET is acting as a capacitor.  Unless I've missed something.

Regarding your questions. The drawing of the paralleled MOSFET's is doable.  Poynty has some in his own schematics.  And I am not about to alter that artifact - nor am I interested in doing other experiments.  But there's nothing to prevent you from doing this.  And thank you for this proposal.  It is SO much more palatable than Poynty's rather repetitive dialgue while he tries to duck the issue.  At LEAST it's arguing the case.  Most appreciative PhiChaser.

Kindest regards,
Rosemary

[Rosemary Ainslie]

Thanks Rosemary for replying to my email so soon and directing me to this new thread on your circuit.

I attempted a replication of this circuit today, but ran into some confusion. 
In my replication Q1 gets very hot but Q2 stays totally cool to the touch.   
Yet the circuit clearly says that Q2 should have the 4 MOSFETS and not Q1. 
Even with 4 MOSFETS connected in parallel at Q1 all 4 get very warm. 
Is there an error with the numbering in the circuit?  If not then what is going on here?
The fact that Q1 gets hot makes sense, because it is connected directly between the heater and .25 ohm resistor.
Q2 shouldn't get hot because as you said, it isn't connected to the negative, it is floating.  So how could it get hot?
If Q2 must be made of 4 MOSFETS, then the 4 MOSFETS must be acting as some kind of energy collector. 
I have to have 4 connected in parallel at Q1 or they will fry, but I will also connect 4 in parallel at Q2. 

Question: Do you need heat sinks on any of your MOSFETS?

Hi AbbaRue,

I was hoping you'd come into the discussion - for many reason, not least of which are your skills at replicating.  May I ask if you found that oscillation?  And, by the way, there is no need to parallel those diodes at Q2.  We actually only did that by accident.  And I'm not about to change the circuit.  Not until our papers are published.  Nor do we find any of the resistors ever getting that warm.  That's even on rather high voltage applications.  Those body diodes seem to cope quite adequately. I can't account for your Q2's getting warm.  Unless it's because you've set the offset that the duty cycle is barely on.  We use that setting for our first test as described in our first part of the 2-part paper. 

Please let us know.
Kindest regards,
Rosemary 

EDITED. (twice)
I highlighted the question.  This may answer your concerns about the heating of those MOSFETs.
And by the way - we do use heat sinks.  They're quite substantial.  I'll see if I can find a shot of these somewhere.

[PhiChaser]

Rosemary,

So the charge isn't held in the MOSFETs, I can see that so scratch that off the list of possibilities. I remember reading somewhere that voltage spikes can go reversed bias through a diode. Has this possiblility been considered?
Oscillation means waves meeting waves in some sort of harmony (?) to me (just because I'm a musician maybe?)... 4/4 is a pretty common time signature... (Well, there ARE 4 MOSFETs on one side?)
Have you tried to get the same (similar) results using just 2 MOSFETs for Q2 and adjusting the frequency from your generator to double (or half?)?
And lastly (heh heh, for now) do you think the type of MOS you're using is responsible for your (anomalous) results Rosemary?
I swear I see a mobius loop between Q1 and Q2...

PC

[Rosemary Ainslie]

Rosemary,

So the charge isn't held in the MOSFETs, I can see that so scratch that off the list of possibilities. I remember reading somewhere that voltage spikes can go reversed bias through a diode. Has this possiblility been considered?

Indeed.  If you read our second part of that two part paper (    I'm getting as repetitive as Poynty Point)you'll see that we RELY on these diodes to conduct that energy.

Oscillation means waves meeting waves in some sort of harmony (?) to me (just because I'm a musician maybe?)... 4/4 is a pretty common time signature... (Well, there ARE 4 MOSFETs on one side?)
Have you tried to get the same (similar) results using just 2 MOSFETs for Q2 and adjusting the frequency from your generator to double (or half?)?

Indeed.  One of our collaborators has done many circuits with many different applications.  And he has never paralleled those transistors.

And lastly (heh heh, for now) do you think the type of MOS you're using is responsible for your (anomalous) results Rosemary?

That would be nice.  Then we'd bottle those specs and sell some really unique MOSFETS.  During the nearly 2 years of testing we have replaced 2 FETS.  But we've had some wild voltages that were responsible. 

I swear I see a mobius loop betweem Q1 and Q2...

I'm not sure what gives here PhiChaser.  We went to GREAT LENGTHS to explain in that paper that we only had PARTIAL solutions.  That it needs the input of the expert.  Which, indeed, is why we even wrote that paper.  We need to get it to the academic forum and some dedicated RESEARCH.  The simple fact is we are able to generate a really robust current flow during that oscillation.  And the battery supply is ENTIRELY disconnected so it cannot be considered the power supply source.  Which is also why we refused to do a standard computation of wattage.  How does one argue a negative wattage?  That's absolute nonsense.  No such thing.  Unless our thermodynamic laws are nonsense.  And I'll stake my life on it that they're NOT.  There's nothing wrong with the standard model.  Which means what?  Are we indeed allowing inductive/conductive circuit material to show their potential?  For the first time?  Or have we just got some glitch in that design that is entirely OVERLOOKED.  If the latter - then TRUST ME.  There are minds - considerably better than even our collaborators can bring to the discussion - that have NOT BEEN ABLE TO FIND IT. 

Kindest regards,
Rosemary

PC

[Rosemary Ainslie]

OK Rosemary,

Please identify using two of the six "P" numbers on the diagram, where the scope probe and probe reference were placed for your battery voltage measurement.

P2 AND P4 WITH THE USE OF ONLY ONE BATTERY.  I trust that answers your question Poynty.  But having said that I need to add the caveat that we do NOT use that circuit you've drawn.  IT'S SIMPLY WRONG.

Regards,
Rosemary