Avalon said:

QuoteAfter winding and re-winding coils about zillion times I have concluded that the shields are not necessary. Neither are the air gaps in cores. Those exist for are for a specific reason only: fine-tuning the inductance in order to achieve the desired resonance.

Actually in a flyback converter the gaps in the core allow a fast collapse of the remnant magnetic field. Yes, they do also change the inductance, but that is an unwanted by product of the gap, which also reduces inductance thus requiring more copper and ferrite. You could design an efficient inductor that has no gaps with the same inductance and it will be smaller and use less copper and ferrite material, a one piece toroidal construction for example, but it will not work properly in a flyback converter. Remnant magnetism in such a device will not allow a high duty cycle without some means to reset the core on a cycle by cycle basis. It is possible to design a special class of flyback converter using such a core with a special "reset" winding and associated circuitry.

The theoretical considerations of the magnetics requirements and operation of the standard boost flyback converter as noted on page 5 of the Onsemi 34063 data sheet is well documented in the literature.

http://electronicdesign.com/energy/mind-gap-and-improve-your-low-power-flyback-transformer-design

Then there is also the distributed gap type of core:
http://www.mag-inc.com/products/selecting-a-distributed-air-gap-powder-core-for-flyback-transformers

Quote from: avalon on May 09, 2014, 03:18:46 PM
The schematic is on the page 5 of the Onsemi datasheet. The only difference is that I use a variable resistor on the feedback (22K) and there is the secondary coil.
The coils are 37mH primary and 57 mH secondary.

~A
After noted success

Onsemi datasheet

and your second socket is empty...

so 2 windings on a pot core; how leedskalnin-esque; he claims you can start such a configuration itself and get a continual voltage...

there is just a cap to ground and ground on the secondary?



Does it work on... a flyback?


I guess it's cause the chip's pretty low freqency... better to have a high-ish inductance...

Quote from: avalon on May 09, 2014, 08:31:37 PM
The problem still remains of how to determine the resonant frequency of the core. (Yes, I have tried all known methods and still not happy about them).

My previous post suggests a PC / soundcard based automated test to iterate over a range of frequencies to determine the resonant.
I can write the code.  Can anyone design a circuit?

Quote from: avalon on May 09, 2014, 08:31:37 PM
Just for the record...
I am not looking for power output from this circuit. In fact, I could not care less.
All I want is to go to the bottom of this and find out exactly what makes it 'tick'.

We all know by now that it is some kind of resonance. However, I want to know more. I want to know exactly which one. Then and only then will I be able to construct a power generating circuit reliably.

The coils that I am using do not have copper shields. After winding and re-winding coils about zillion times I have concluded that the shields are not necessary. Neither are the air gaps in cores. Those exist for are for a specific reason only: fine-tuning the inductance in order to achieve the desired resonance.

The problem still remains of how to determine the resonant frequency of the core. (Yes, I have tried all known methods and still not happy about them).

~A

How big is your core? do you have a datasheet for it?

1:   $170.93   150mmx30mm   Core height up to 45 mm for
transformers up to 100 kW

mouser has pot cores...

Quote from: semenihin-77 on May 09, 2014, 04:28:58 PM
Все верно, у меня сгорал диод максимум Вся "соль" чтоб найти консенсус между напряжением емкостью всех конденсаторов и частотой. Если нагрузка постоянная то добивайтесь понижения потребления подбирая частоту резонанса, когда вы будете очень близко- ток будет прыгать от - до +, задача найти баланс, очень тонкий баланс.

Сейчас я ищу резонанс по току, резонансная катушка имеет мало количество витков, а конденсатор больше емкость.
В любом случае вы поняли что надо найти, мои поздравления!

All right, I burned the diode maximum All the "salt" to find a consensus between the voltage capacity of all capacitors and frequency. If the load is constant for lowering consumption picking up the resonance frequency when you are very close, the current will jump from - to +, the task to find a balance, a very delicate balance.

Now I'm looking for current resonance, has little resonance coil number of turns, and large capacitors.
In any case, you realize that it is necessary to find, congratulations!

hi semenihin-77,

May i ask what the diode number you are using in your circuit.
(Я хотел бы спросить, какие диод номер, который вы используете в вашей цепи.)

May i ask is there a need to change 510pf capacitor value on MC34063 pin 3 and pin 4 to a higher value to around 2000pf capacitor.
(Я хотел бы спросить, есть ли необходимость изменить 510пф конденсатор значение на MC34063 контакт 3 и контакт 4 на более высокое значение примерно 2000пф конденсатор.)


---------------------------
I am using freetranslation.com which seem to produce little better grammatical accuracy than google translate when i convert from Russian to English.