Beyond Unity - Jagau's successful ZPM replication

Jagau posted this 09 December 2022 - Last Edited 03 April 2023

My replication of Figther's ZPM. As my two coils were already mounted on an AMCC320 I was able to replicate quickly.
In the first image the 2 coils mounted on the AMCC320

In the second image when the power supply is turned on at 24 volts DC he sees an input power of 4.32 watts on 24 volts and 180 ma when my TL494 oscillator is supplied to both coils through an IGBT pulsed at 38.2% on the high side. Channel1 is IGBT pulse

Image of the oscilloscope with the current probe on the PS

You can notice that the oscilloscope show on a my TEK current probe (green) 292mv =292 ma (rectangular form) and 38.2% DTC

Last image the 118 volt 4 Watts lamp lit with 34 ma x118.8v = therefore 4.0392 watts of power produce at the output on a large capacitor and 4 diodes in parallel in order to produce an almost pure DC which can be read by the 2 DDMs

Calculation of input power in DCM mode with reactive component.

As indicated by the power supply we have an instantaneous pulse power of 4.32 watts, this is the instantaneous power not the average power over time.

V and I is what PS provided a DM as wrong lecture for input due to high frequency so to calc PIn we take well known formula for inductor.

Irms on scope is 0.292 Amp, DC voltage is 24volts

Pin= I*V*sqrt(Duty_Cycle)/sqrt(3)

Pin= 0.292 Amp x 24 Vdc x sqrt 0.382 / sqrt 3 = 2.533 watts

So for Pout of 4.032 watts i need only 2.533 watts at input, Still a pretty good COP
Jagau

#Permalink
10
0

beyond-unity zpm

Liked by
and 6 others

bigmotherwhale posted this 12 December 2022

The period (time) of the cycle and the lenght of the magnetic pathway.

Propagation delay is equal to d / s where d is the distance and s is the wave propagation speed.

Liked by
and 1 others

Munny posted this 12 December 2022

Quoting:bigmotherwhale

could you try altering the capacity of one of the coils with a capacitor and measuring again or try pumping the core at this freqency?

Curious BMW, wouldn't we want as much of the ringdown as we can get? I can understand closing up the blank time--that's just hardware sitting there waiting for the next pulse. It was always my impression we wanted to lengthen the ringdown for as long as we could get it, usually by means of removing as much resistance as possible or anything that increases the Q-factor. There must be more to this I'm not familar with.

Liked by

bigmotherwhale posted this 13 December 2022

Im not saying its good or bad, im trying to find the reason for it, is it an LC , related to the speed of propagation in the core material ,ferroresonance or magnetostriction phenomenon?

Liked by
and 1 others

Jagau posted this 13 December 2022

Curiosity is a very important tool in research like we do. No problem with that BMW

In another order of idea, with regard to the ZPM the type of load is very very important.

Having tested different configurations, it should be non-linear, as an incandescent lamp can and the combination of nominal operating voltage and power is proportional to what you will get at the output.

Jagau

Liked by
and 2 others

bigmotherwhale posted this 13 December 2022

I just did my own mini replication as a metglas core i ordered cheap from china second hand arrived today, I can see now why it was so cheap, its tiny! Its an AMCC 6.3.

I built a circuit arround what i think was right, i grabbed whatever i had at hand, coils were wound 15 turns on one side and arrouund 30 ish on the other, i didnt pay too much attention as it was pre used copper wire and only for a rough idea as to whats going on.

I tried direct output and output through a diode but without a capacitor as well as them connected directly.

At arround 40khz and 80khz and somewhat at 125khz I have a dip in input current draw, its nothign too much but its noticeable and makes my PSU spack out a bit even through a very large DC link cap, The range is so limited beacuse my mosfet drivers are not rated for this at all.

The only thing i have to report of interest is that the 240v flyback protection LED bulb on the mosfet flashes periodically at the same time as the 12v bulbs I have on the output, im using this bulb as an overvoltage protection device. they should in theory be wired through a diode to the same direction across the coil so how this is happening i dont know. I will have a look at the circuit properly and see if im missing something or this is a hairpin effect of some sort.

Liked by
and 1 others

Jagau posted this 13 December 2022

Ok BMW
I've never tried a 6.3 but you could ask Atti, I think it's more or less the same size as yours and for him it worked for his demo.
Second thing, can you roll more turns of thread for your coils? For the different configs that I tried, the best ones were from 100 turns each coil.
Because at this low number of revolutions the frequency of operation will be very high, but all the same you see him it worked for Atti, Don't give up, it's a start
thank you for sharing
Jagau

Liked by
and 1 others

Jagau posted this 14 December 2022 - Last Edited 14 December 2022

Hi all

On the following site there is an online calculator that can make your life easier, just enter the first 3 values, voltage, inductance and current, you won't have to do any calculations and that's what I explained above.

https://www.daycounter.com/Calculators/Inductor-Current-Power-Calculator.phtml
These are maximums or minimums that you can reach so you can go more or less according to these forecasts.

if we take example of my last demonstration
24V, 9mh, 0.292amp

the site indicates the i can reach up to max power drawn at thei nput by this inductor 2.861 watts with DTC max of 50% as they suggest

as I used 38.2% DTC and 1 Khz and I got 2.5 watts as I demonstrated, so I could rework my circuit to achieve less input consumption.

We are talking about the consumption at the input of my circuit

Jagau

Liked by
and 1 others

Munny posted this 14 December 2022

I like where you are going with this, calculate the power ceiling on the input based on the fixed parameters. Then you know there is no possible way you could draw more than X watts. That makes any COP dead simple to know if you're going in the right direction when tuning.

I like it! I like it a lot, though the site won't let me give you two thumbs up. 😜

Liked by

Admin

Fighter posted this 14 December 2022

Quoting:Munny

I like it! I like it a lot, though the site won't let me give you two thumbs up. 😜

Sorry Munny but we didn't implemented the multiple thumbs up feature. 😀

We apologize for the inconvenience ...

Regards,

Fighter

"If you want to find the secrets of the universe, think in terms of energy, frequency and vibration."
		Nikola Tesla

Liked by

Jagau posted this 14 December 2022

Yes that's what coilcraft a great coil manufacturer uses to know the maximum can be reached, they code their inductors with the maximum but nothing prevents us from using them below their saturation point and that's the same thing for the DTC it is the range available before saturation which is useful, it is the maximum not to be exceeded.

Jagau

Liked by