Electrical / Electronic Devices > Tariel Kapanadze's Devices

Kapanadze / Dally

<< < (2/4) > >>

exnihiloest:
Neither the TL494 nor the transistor КТ926 allows high frequencies. The highest frequency components are likely around some hundreds Khz. This is too low to get resonant modes in the coax, because at these frequencies the coax length is much shorter than a quaterwave. Imho one can discard line effects and can consider the coax coil as an ordinary coil of a double-wire (the inside one and the shield).
As the current goes in opposite directions in the shield and in the central conductor, the magnetic field from the coax is canceled. The coupling between the coax and the other coils can be only capacitive, especially near the not connected end of the coax. Although the shield is connected to the ground, at the opposite end the potential is not null due to the inductive impedance of the "coax shield coil" that is in series with the ground.

ion:

--- Quote from: exnihiloest on 2012-09-19, 09:03:50 ---Neither the TL494 nor the transistor КТ926 allows high frequencies. The highest frequency components are likely around some hundreds Khz. This is too low to get resonant modes in the coax, because at these frequencies the coax length is much shorter than a quaterwave. Imho one can discard line effects and can consider the coax coil as an ordinary coil of a double-wire (the inside one and the shield).
As the current goes in opposite directions in the shield and in the central conductor, the magnetic field from the coax is canceled. The coupling between the coax and the other coils can be only capacitive, especially near the not connected end of the coax. Although the shield is connected to the ground, at the opposite end the potential is not null due to the inductive impedance of the "coax shield coil" that is in series with the ground.

--- End quote ---

Both devices you mention may be operated in avalanche mode i.e. extremely sharp turn on pulse from a high voltage supply. While the repetition rate may be in a few hundred kilohertz max due to recharge rate, the frequency components from the fast risetime pulse due to avalanche breakdown may easily be in the tens of megahertz if not even higher. Do not confuse the maximum frequency from the data sheet for this large die device (KT926) or SCR T7 with the turn on time when operated avalanche mode. The recharge rate of the pulse discharge capacitors sets the maximum repetition rate possible. It is possible KT926 is operated as a simple switch since the applied voltage is 150 and the max c-e is 200 V.

As I mentioned before, the SCR is operated in pure avalanche mode with a debounce  switch circuit used to turn the circuit on and gate the SCR to avalanche mode.(unless whoever drew the schematic forgot the feedback line for the gate circuit, thus turning it into a one shot instead of oscillator). It then operates independently at a frequency determined by the recharge rate of the capacitor C11 in conjunction with inductor DR+R.

In Dally1, note that "R", the SCR gate resistor must be experimentally determined and will differ for each SCR used. It is probably a high value, the idea to bring the gate close to turn on threshold of avalanche mode, not normal SCR mode.

In the case of KT926, if operated in avalanche mode, it is gated by 74HCT100 oscillator. The inductor and capacitor set the recharge rate, which is faster than required by 74HCT100 gating.

The idea is to produce (comb effect of) very fast risetime pulses with relatively low repetition rate into the coax.

These items are merely a replacement for the small spark gap used by Kapanadze. As such they do not actually carry full power to the load by induction, but may just be a catalyst towards NMR effects, if we are to believe that theory.

Standing waves of electrostatic potential may show themselves at intervals along the skin of the coax when the coax is impulse excited at its resonant frequency.

exnihiloest:
@Ion

I was only speaking of this setup:
http://realstrannik.ru/media/kunena/attachments/630/Dally2012.JPG

I don't confuse the working frequency and the frequency components of the pulse.
The diode КД203 and the transistor КТ926 are power components. I didn't found their datasheet but I saw their case. They are not designed for HF.
So their inside parallel capacity should be rather high and prevent high frequencies (2-3 Mhz max?). Moreover the inductance of TR1 probably increases the impedance. 

What maximum frequency do you think you have in the coax? For me, no more than 2 or 3 Mhz at a significant level; probably less.

ion:

--- Quote from: exnihiloest on 2012-09-20, 15:40:20 ---@Ion

I was only speaking of this setup:
http://realstrannik.ru/media/kunena/attachments/630/Dally2012.JPG

I don't confuse the working frequency and the frequency components of the pulse.
The diode КД203 and the transistor КТ926 are power components. I didn't found their datasheet but I saw their case. They are not designed for HF.
So their inside parallel capacity should be rather high and prevent high frequencies (2-3 Mhz max?). Moreover the inductance of TR1 probably increases the impedance.  

What maximum frequency do you think you have in the coax? For me, no more than 2 or 3 Mhz at a significant level; probably less.

--- End quote ---

As I understand it, avalanche mode is a different mechanism that does not rely on current injection into the base to effect turn on, rather, the entire die is shock excited, and an entrained  electron flow is produced. True, the die capacitance will limit repetitive pulsing at a high frequency, so there is a recovery time involved.
I agree that such a device would only be usable as a low frequency oscillator when connected normally using the base as a feedback or drive element of the oscillator.

It is true that HF transistors (and SCR's) use a different construction technique that allows more efficient propagation of the gate signal across the die.

It is probably the fast risetime that produces harmonics which excite the line at it's resonant frequency not the actual repetition rate.

Tr1's inductance will be a function of the load it sees (plus any leakage inductance). We need to determine this, but it should not be a huge factor in pulse coupling if properly engineered.

I have not tested any 5.9 meter lengths of RG58A/U coax yet, but I am sure coax as used in the TV and LAN industry can support 100 MHz  or more without extreme loss.

exnihiloest:

--- Quote from: ION on 2012-09-21, 04:00:03 ---...
Tr1's inductance will be a function of the load it sees (plus any leakage inductance).
...

--- End quote ---

In a 100W HF transmitter powered under 12v, the push-pull power stage is followed by a transformer whose the primary has only one turn of a big cylindrical conductor inside a double core, and the core is constituted of several sections of different ferrites in order to cover a wide band (1.6-30 Mhz).
I'm very pessimistic about the possibility of Tr1 to pass high frequencies at a signicant power if it is built as an ordinary transformer that we can find in switching power supplies.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

Go to full version