Body
Hi Mick,
"When you refer to the primary resonant tank, you are referring to your antiphase transformer?"
Yes.
"Your original antiphase transformer circuit that was providing extra energy- My understanding is the Class D amp was being used a clean sine wave source so the parametric pumping you are speculating on would be derived from the buss pumping to the power rail on the half bridge or some hetrodyning of the double and half frequency, are these your thoughts?"
Something like that. The amplification of a class D amplifier is proportional to its rail supply voltage. Thus any change here is distorting the signal. The reactive power then is not a perfect sine anymore. When measuring the power over at least 20 cycles to minimize the influence of the power in each direction the measurement could perhaps be influenced by the distortion to show pure reactive while actually being something else. Just a thought!
"DId you try the half bridge mode with cap decoupled outputs prior to moving on to the full bridge?"
I don't know if the decoupling was good enough. The power supply was quite unstable when adjusting close to pure reactive power. One supply rail voltage would climb while the other one declined. This looks like power supply pumping. I even tried with eight small lead acid accumulators which gave the same problem. The accumulators were only 1.2 Ah (12V). Perhaps they have too high impedance for this purpose. It was clearly seen that the voltage started increasing at one supply rail while decreasing at the opposite rail.
"Would decoupling the output after the smoothing caps with some more diodes stop this occurrence in a half bridge topology?"
I think only a two quadrant power supply is able to solve the problem when using only a mono block class D amplifier. The power supply must be able to sink or dissipate the stored energy of the inductor. If searching power supply pumping some solutions minimizes the problem when using two channels where they work in anti-phase. The car class D amplifier is actually working that way which makes it easy to bridge couple when given a mono signal and using the two output phases already put in anti-phase. In stereo mode one of the speakers is connected in anti-phase to be able to output the signal in phase with the other channel.
"If the energy is stochastic resonance as Hector often claims then noise and distortion can be our friend rather than our enemy as is tamed in regular audio circuits."
Then the Russian schematic with the double variable autotransformer could be the way to do it. By the way I actually made such a transformer using a stack of ferrite ring cores. The output load is critical which could be the problem at the moment. A variable resistive load for matching the proper impedance is needed. The variable autotransformer puts in inductance when not loaded correct. Too many variables to keep track on! Perhaps I need to find a DC coupled class A amplifier to convert to an electronic load. They're pretty expensive and very inefficient but probably able to do the job as a power generator as well as an electronic load given the right feedback loop.
There are also AC electronic loads out there but they're very expensive. Two or four quadrant power supplies are yet another expensive possibility.
According to W.B. Smith's theory the excess energy is supplied by the all permeating background field(s) by making separate fields coherent.
In the channeled Tesla info he talks about quarts copper and a magnet for generating free energy.
Regards
Ole
---In [email protected], <mkjekyll@...> wrote :
Ole,
When you refer to the primary resonant tank, you are referring to your antiphase transformer?
Your original antiphase transformer circuit that was providing extra energy- My understanding is the Class D amp was being used a clean sine wave source so the parametric pumping you are speculating on would be derived from the buss pumping to the power rail on the half bridge or some hetrodyning of the double and half frequency, are these your thoughts?
On the output side not accounting for the power supply depending on frequency the clocking pulses could provide another source of parametrics if the steep low pass filter were not doing its job. The transformer you made looks as if it could pass some higher frequencies along with your intended ones. But again your spectrum analyzer looked pretty clean so I would rule that out if you were looking that high.
DId you try the half bridge mode with cap decoupled outputs prior to moving on to the full bridge?
Lots of strange things go on inside the class D topology. Maybe go with a class A for your designing as they are much less likely to add as many unforeseen variables.
I have yet to read about the class E and the allegedly channeled data.
Page 12 is interesting concerning the buss pumping. Would decoupling the output after the smoothing caps with some more diodes stop this occurrence in a half bridge topology?
My understanding is when you changed to full bridge mode the power supply is no longer heating up and you are not seeing excess energy. With balanced topology I don't think the antiphase transformer can ring as well. With balanced the PSRR goes way up and even harmonics way down. If the energy is stochastic resonance as Hector often claims then noise and distortion can be our friend rather than our enemy as is tamed in regular audio circuits.
On 1/6/2018 5:30 PM, onielsen@... mailto:onielsen@... [EVGRAY] wrote:
Hi Mick,
The primary resonant tank is not driven at its resonant frequency. If I remember correct it is driven at double the resonant frequency. Perhaps I should calculate or measure the resonant frequency again.
In a true H-bridge configuration the current going through the freewheeling diodes of the two off state switches are used by the other two on state transistors that doesn't have to take current from the power supply rails. This should prevent the voltage on the power supply rails from increasing according to page 12 here: https://www.infineon.com/dgdl/an-1071.pdf?fileId=5546d462533600a40153559538eb0ff1 https://www.infineon.com/dgdl/an-1071.pdf?fileId=5546d462533600a40153559538eb0ff1
Perhaps this parametrically change in voltage is what made the transformer give off free energy. The amplifier used at least as much power as was output by the transformer even though the power going into the primary LC-tank was only reactive power.
According to this channeled interview with Nikola Tesla (Nikola Tesla: Nobody knows my story. It's Time to tell you EVERYTHING https://www.youtube.com/watch?v=WKKEH7IsjRQ https://www.youtube.com/watch?v=WKKEH7IsjRQ) he recommends the class E amplifier of Nathan O. & Allen D. Sokal (1975) to drive tesla coils if he had lived today.
https://wiki2.org/en/Power_amplifier_classes#Class_E https://wiki2.org/en/Power_amplifier_classes#Class_E
http://www.g-qrp-dl.de/Projekte/80m_E-Endstufe/classEampl.pdf http://www.g-qrp-dl.de/Projekte/80m_E-Endstufe/classEampl.pdf
Regards
Ole
---In [email protected] mailto:[email protected], <mkjekyll@...> mailto:mkjekyll@... wrote :
Hi Ole,
When you bridge the impedance you can drive easily is now doubled however the push pull format is what is probably not letting the resonance ring. The resonant part of the circuit needs to wave freely in the breeze, remove anything clamping it.
I still think you will have a problem accepting the energy back into the amp backwards in your four quadrant concept, just decouple supply with some large caps and diodes and move forwards.
On 1/2/2018 6:47 AM, onielsen@... mailto:onielsen@... [EVGRAY] wrote:
Hi Mick,
"Do PF oil caps have lower ESR with the higher voltage spec models like the lytics seem to?"
I don't know. But being physical bigger for thicker dielectric the foil area is bigger too so I suppose the ESR is lower too. Check the datasheets.
"Can you recommend from experience any self healing high voltage PF correction caps?"
No. Check the different manufacture catalogues.
"...hope to see more of your experiments soon."
They're not working too well at the moment regarding OU. I'm using the class D amplifier in bridge mode (two phased floating outputs) to prevent power supply pumping. Only slightly over unity shows up now.
Regards
Ole
---In [email protected] mailto:[email protected], <mkjekyll@...> mailto:mkjekyll@... wrote :
Ole,
Thanks for that!
Was looking two days back and kept getting empty directories, weird...
Do PF oil caps have lower ESR with the higher voltage spec models like the lytics seem to?
Can you recommend from experience any self healing high voltage PF correction caps?
Have a great 2018, hope to see more of your experiments soon.
.....