Rosemary Ainslie circuit demonstration on Saturday March 12th 2011

[poynt99]

The correct schematic as per the actual build. I combined all the CSR (shunt) resistors into one on my diagram.

.99

[poynt99]

Have any of you wondered why Q2-Q5 are mounted on large heat sinks while Q1 is only on a small U-channel heat sink?

Well if you have, I can tell you it's because Q1 probably doesn't get very hot. Q2-Q5 however probably do. Why? Because Q2-Q5 are sourcing most of the current to the load.

The important realization here, is that as a result, very little of the net current in the load is flowing through the CSR. 

.99

[Rosemary Ainslie]

Have any of you wondered why Q2-Q5 are mounted on large heat sinks while Q1 is only on a small U-channel heat sink?

Well if you have, I can tell you it's because Q1 probably doesn't get very hot. Q2-Q5 however probably do. Why? Because Q2-Q5 are sourcing most of the current to the load.

The important realization here, is that as a result, very little of the net current in the load is flowing through the CSR. 

.99

Guys,

I want to know how upwards of 48 amps flowing over more than 80 percent of the duty cycle - can possibly represent 'very little' of anything at all.  The more so as the voltage behind that current is also upwards of 48 volts.  During the oscillation phase there is a really, heavy duty current flow.

The real question is this.  What in that switch transposition enables this?  My own take is this.  The gate is either presenting a positive signal to the drain or a negative signal to the source.  Effectively what standard settings on that MOSFET enable, is the positive to the drain or a zero or a negative, to the drain.  That's it.  So then?  Does that mean that there has, perhaps, been a negative potential on the circuit that was simply not allowed to fully generate?  This because our transistors were designed on an ASSUMPTION?  And that assumption being that the negative voltage component was simply stored energy?  And, somehow this stored energy is released in those really high spikes? 

The puzzle is this.  Parasitic oscillations are known to cross zero and result in that complex sine wave that is sort of snubbed out at each transition as it manifests.  It's the fact that it crosses zero at all that perhaps should have given us a clue.  Some early indication of this negative potential difference on the circuit itself.  I suspect that the doping in those transistors is designed precisely to prevent this.  Or certainly to prevent it from persisting as it does here with this transposition. 

Either way.  What is clear as daylight is that there is no restriction to the flow from the battery and no restriction to the flow from what has been seen as counter electromotive force.  And IF that CEMF is the result of 'stored' energy - then how is it that it can exceed the energy from the supply in the first instance.  No-one has difficulty - for some reason - in explaining this when it's a spike.  But there is some greater effort needed to resolve this under classical paradigms IF and when we find that continuous oscillation - or that continuous resonance. 

It seems that there is NO STOPPING IT - provided that the signal at the gate stays negative.  And that is definitely NOT acceptable in any standard or mainstream school of thought.  And please note.  The MOSFETS ALL stay relatively cool.  They are NOT unduly stressed under normal conditions.  They only become stressed when we apply too much signal at the gate.  Then it seems to want to default into a kind of 'runaway' condition - as I think Happy referred to it.

Regards,
Rosemary

[Rosemary Ainslie]

And Guys - another thought.  Consider too that these parasitic oscillations are only evident when those MOSFETs are put in parallel.  What actually has happened there?  Could it be that all those extra body diodes now come into play?  That there are then enough of them to better enable that returning flow?  I'm sure this is part of it.  But I absolutely do not have the answers.  I've been reading up on transistors but we really need an expert.  Perhaps someone here can comment.

And Dear God - Please let it not be Poynty.  I'm heartily sick of his negativity.  First it was a misreading of the voltage across the battery.  Then it was inductance on the wire.  Then it was gross undersampling. All of it gross and unadulterated nonsense. 

Again - with regards,
Rosemary

[nul-points]

hi Rosemary

the important thing at this 'poynt' is not to get distracted

it seems that there are now at least two follow-ups suggested by recent developments:

A) with the circuit as tested, go on to make whatever developments are necessary to enable continuous operation**  (eg. try those suggestions for DC negative gate drive, etc)


B) with a circuit modified to place all MOSFETS truly in Gate/Drain/Source parallel, establish whether it's still possible to achieve parasitic oscillation - and if so, does this version provide significantly different energy conversion results to your March 12 test?


**
i realise that establishing proof of your 'new conduction model' is your own personal driver in performing this experiment, but i believe that you also have an altruistic desire to see people benefit from the 'clean green' energy which this circuit might address: eg. off-grid water heating

somebody recently made the valid point that if your circuit is be used in this way, then it must be made to operate continuously without danger of overheating or of 'slipping out' of the preferred drive conditions

unfortunately, i think that until you successfully enable the continuous safe operation of your circuit then nobody will be prepared to either to give any credence to your conduction model or to try and apply the 'technology'

it would be a shame to have come so far, in the face of so much flak, and not to bring this 'Herculean' task to a successful completion


i hope that you will take this post in the constructive spirit in which it is intended

np


http://docsfreelunch.blogspot.com