COP 17 Heater | Rosemary Ainslie

[Harvey]

Thanks! The bulb and diode combination is a great idea. Because of my lack of high-tech test equipment I have often had to get creative simply to be able to confirm or rule out certain possibilities and conditions. The little grey cells don't work as well as they used to, so this is great. Thanks so much!

FWIW, many years ago my electronics teacher told me to get into the test equipment field. He said that technology can only advance as fast as our ability to measure it. Oh! how I wish circumstances had allowed me to be able to follow his good advice.

-kent

For a quick low frequency (up to 20KHz) scope you can use your sound card line input and this software:

Zelscope: Sound card oscilloscope and spectrum analyzer

I don't know much about it, I have installed it before to evaluate the interface but I have never used it. It won't help you during aperiodic modes as these are above 5 times its speed, but it may be a fun project for you for adjusting the timer for on and off times etc.

Cheers,

[Michael John Nunnerley]

Hi Glen

love the scope shots, nice work

Mike

[Harvey]

Hi .99,

Any progress on the negative excursion simulation?

I would like to see if we can get 17W of dissipation in an inductor while getting a net power reading at the shunt of 1.13W because of the reverse current flow into the negative spike of the inductive collapse. We need that spike and subsequent ringing to go deep and resonate as long as possible. It is currently the only conventional answer to this problem I see so far and I think we must allow ourselves latitude in the part values to accommodate the claimed result.

Do you grasp what I am asking for? It may seem a bit convoluted at first but when you see it, it is quite simple. During the negative excursion at the drain we have a compound current influx from both battery poles into that node. That negative spike (and all subsequent rings) form a sink for the current to flow into from both ends. This means that not only does the current from the positive battery terminal not get registered for this period, but also some of the previously registered current is subtracted by the reverse flow from the battery negative terminal.

When we can show this on the simulation, then we will show where the increased COP exists. Don't expect OU, it is not there - but the COP > 1 will be.

[FuzzyTomCat]

Hi Glen,

You did apply heat sink compound to both sides of the mica right? I find silicone zinc oxide works best for mica applications.

Interesting scope shots. Are your probes AC coupled there? I am interested in the missing null period in the 10µs shot that is evident in all the others. You can see a definite energy reduction for each pulse leading up to the small ring in the 5µs shot. Notice that at the end of the small ring (which is that null period) there is a smaller energy pulse, a slightly wider period and then the highest energy pulse. There is a reason that those 3 pulses decrease in amplitude and result in a rest period instead of retriggering. And the smaller energy spike at the end of the ring seems out of place. It would seem that the self triggering would like to do so during that rest period but something prevents it after which it succeeds at the diminished level just prior to a renewed cycle. So it looks like there should be 5 pulses there and their tops seem to map out a capacitive decay curve.

How wide was your timer pulse before aperiodic oscillation?

Hi Harvey,

Everything is DC coupled, probes at 1 ohm DC - X10

Channel 1 is on the bottom [ A1 26mV ] ground is on battery negative side of 0.24 ohm shunt and probe is at 555 negative rail side of shunt.

Channel 2 is on the top [ coupled ] ground is on battery negative side of 0.25 ohm shunt and probe is at mosfet source side of shunt.

I'm not sure about the time pulse, I set the pots at the presets of Aarons in the "Ainslie-Murakami Negative Dominant Waveform Generator" PDF file -

• Gate resistance: 53 ohms
• On pot resistance: 32.8 ohms
• Off pot resistance: 293.9 ohms
• NE555N power adjustment pot resistance: 193.1 ohms
• NE555N timer duty cycle: 55%
• NE555N timer frequency: 915 kHz
• Actual run frequency: 1.17 MHz

The only thing I adjusted was the Gate just slightly and thats what I got .... It's been 7 1/2 Hours now and "NO" change in voltage, still .02 volts above the voltage when I purchased the battery yesterday afternoon.

Glen

[poynt99]

Hi .99,

Any progress on the negative excursion simulation?

I would like to see if we can get 17W of dissipation in an inductor while getting a net power reading at the shunt of 1.13W because of the reverse current flow into the negative spike of the inductive collapse. We need that spike and subsequent ringing to go deep and resonate as long as possible. It is currently the only conventional answer to this problem I see so far and I think we must allow ourselves latitude in the part values to accommodate the claimed result.

Do you grasp what I am asking for? It may seem a bit convoluted at first but when you see it, it is quite simple. During the negative excursion at the drain we have a compound current influx from both battery poles into that node. That negative spike (and all subsequent rings) form a sink for the current to flow into from both ends. This means that not only does the current from the positive battery terminal not get registered for this period, but also some of the previously registered current is subtracted by the reverse flow from the battery negative terminal.

When we can show this on the simulation, then we will show where the increased COP exists. Don't expect OU, it is not there - but the COP > 1 will be.

I too see very little "activity" with the spec'd inductance of 8.64uH. I show it here for comparison to a suped-up inductor with 100x more inductance (864uH), while the other component values are the same.

Shown are Drain voltage with supply current, shunt voltage with Drain voltage. The Drain will of course not really go negative, but the shunt and Gate (as long as there is a Gate resistance) will (Gate not shown).

I'll post some power dissipation shots next.

.99

[poynt99]

Load and Shunt Power

This may be more confusing than helpful, but here it is anyway.

Denoted voltage reversal of Drain voltage is of course across the load resistor (Drain voltage referenced to ground is still positive and gets clipped by the MOSFET body diode when it attempts to go negative). Voltage and current clearly both change to negative which results in a "positive" power product once again.

These shots are with a 864uH value, not 8.64uH. There is no way powers of 17W are possible with this inductance to resistance ratio. However, moving back to the original value of 8.64uH does get closer, but then there is little if any negative excursion of the Drain voltage.

Let me know Harvey what else if anything you would like to see.

EDIT: Added a plot showing instantaneous battery power clearly showing some power returning to the battery while the MOSFET body diode is forward-conducting and clipping the Drain voltage.

EDIT2: A side note here is that if a flyback diode is added to the simulation, most of the power we saw being returned to the battery during the clipped negative excursion at the Drain is now dissipated in the load resistor rather than returned to the battery. So the standard flyback diode connection does not return energy back to the source, it recirculates the current in the load and increases the power dissipation therein. Just an FYI. Note however that a small spike of current will go back into the battery due to capacitances in the circuit components, but this is miniscule in comparison to the current recirculating in the load.

.99

[poynt99]

Load, Drain, and Battery Power with Quantum Circuit

Close up of trailing edge to look for any returned energy to the battery. There is very little if any due to the small inductance to resistance ratio in the load resistor. In other words, there is very little energy being stored compared to what is being dissipated.

The second shot is zoomed out to see the complete picture of a single pulse.

Third shot includes Drain voltage illustrating that no negative excursion is present because of limited flyback voltage.

No flyback diode is present.

Also note that the pulse width is far in excess of that required to fully energize the 8.64uH inductance in this resistor. As I mentioned several weeks back, to make more efficient use of the input energy, this pulse width should be reduced to about 4.5us (or less), from the current 15.4us setting as prescribed in the article. When the circuit goes into this wild HF oscillation, the pulse width will be reduced by default and this may explain much of the gain seen by operating the circuit this way. Again as I've already mentioned a couple times, I would recommend that someone try driving the MOSFET directly at 420kHz (or whatever the average frequency of the aperiodic oscillation is) and see if a similar increase in efficiency is observed.

.99

[Aaron]

back home

Hi Everyone - just back night before last from an incredible week hiking and sightseeing at Glacier National Park in Montana. INCREDIBLE - a much needed break! Will have to post pics later - WOW - talk about Majestic!

Anyway, reading through posts and catching up on email.

[Aaron]

ne555n

I have also went back to the NE555N timer oscillator because of the repeditive wave forms appear not to be there using the other NE555xx timers, and I think Harvey may have came to the same conclusion looking at his postings.

Hi Glen,

Does this mean the ne555n does have some beneficial properties to it that other 555 chips don't?

It is the only one I have used so far.

[Aaron]

data collection

I have suggested that when Aaron has time, we get 3 sensing resistors in place for each leg of interest. B(+), IRFPG50 Source, Timer circuit ground. Then we can run some aperiodic tests with all three being scoped and see how the current is flowing. The only problem with this test, is that his rig is not producing the 17W we are looking for...yet.

Hi Harvey,

There is some specific data collection I'll be doing with Rosemary and after that I can try some of these things.

[Aaron]

self adjusting

I was seriously considering running a full test on this and was about to start taking notes and video when it dropped out of aperiodic operation - on its own.

If the timing signal is dominant, the waveforms will close in on themselves squashing the oscillations. If the oscillation is dominant with timer having long spaces between, then the oscillation can keep pushing the wave forms apart until the oscillation is dominant.

[Aaron]

heat sink

The Question is the aluminum bar stock Aaron used for the last circuit addition "Ainslie-Murakami Negative Dominant Waveform Generator" I'm working on how was that Mosfet Connected to the Heat Sink bar ??

The mosfet is sitting directly on the sink with white grease in between. And it is screwed down with an allen screw right thru the center hole into the sink. I can remove it and see what the difference is.

My earliest experiments were with no sink whatsoever.

[FuzzyTomCat]

Hi Glen,

Does this mean the ne555n does have some beneficial properties to it that other 555 chips don't?

It is the only one I have used so far.

Hi Aaron,

I'm happy you had a great vacation ..... Montana is a beautiful place

As far as the 555 timer oscillator chips the NTE Electronics - NTE955M and the Texas Instrument - NE555P are nothing like the NE555N, those are the only ones I've tried. The biggest and most substantial difference was the Mosfet to Heat Sink insulator that made a "HUGE" diffrence

Glen

[FuzzyTomCat]

The mosfet is sitting directly on the sink with white grease in between. And it is screwed down with an allen screw right thru the center hole into the sink. I can remove it and see what the difference is.

My earliest experiments were with no sink whatsoever.

Aaron I think that the white grease has compounds in varying quantities of conductive aluminum oxide, boron nitride and zinc oxide possibly and not as conductive as what I used ......

Arctic Silver Incorporated - Arctic Silver 5

Which Contains 99.9% pure silver and possibly helps makes a better antenna affect it appears, but the Manufacture states ....

Not Electrically Conductive:
Arctic Silver 5 was formulated to conduct heat, not electricity.
(While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads. While it is not electrically conductive, the compound is very slightly capacitive and could potentially cause problems if it bridges two close-proximity electrical paths.)

I'm just not sure why ..... possibly the high frequency the Mosfet is Oscillating at in combination with the Mosfet drain ( which goes to the 10 ohm Load Resistor ) is connected internally to the Mosfet back plane where the Heat Sink is attached to ??

Glen

[eternalightwithin]

I'm getting comfused now. Do we want electricaly conduction or isolation between IC and heatsink?

Aaron I think that the white grease has compounds in varying quantities of conductive aluminum oxide, boron nitride and zinc oxide possibly and not as conductive as what I used ......

Arctic Silver Incorporated - Arctic Silver 5

Which Contains 99.9% pure silver and possibly helps makes a better antenna affect it appears, but the Manufacture states ....

Not Electrically Conductive:
Arctic Silver 5 was formulated to conduct heat, not electricity.
(While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads. While it is not electrically conductive, the compound is very slightly capacitive and could potentially cause problems if it bridges two close-proximity electrical paths.)

I'm just not sure why ..... possibly the high frequency the Mosfet is Oscillating at in combination with the Mosfet drain ( which goes to the 10 ohm Load Resistor ) is connected internally to the Mosfet back plane where the Heat Sink is attached to ??

Glen