Kapanadze Cousin - DALLY FREE ENERGY

[d3x0r]

I do not understand how this saturable core transformer circuit works.

The article, that explains the operation of this DSRD driver with saturable core transformer, is in Russian language.
You will have to ask Сергей or Vasiliy to explain how it works, based on that Russian article.

I have never succeeded in getting any saturable core transformer DSRD driver to work.
I only had success with this transformerless DSRD driver.

P.S.
Adding a ferrite core to an inductor increases its inductance but decreases its speed because ferrite introduces a lag.

Yes that's what I thought... I think I want to delay the current a while...

Well; neither the coil or capacitor is in Fig.22 of Belkin book that describes this pulse generator...
(Shapers power nanosecond and picosecond pulses on a semiconductor unit basis)[/size]


Content: The purpose and principles of the pulses with nano-and picosecond fronts. Circuitry formers using DSRDs, Dzlp and saturable inductors. study industrial diodes as DSRDs iDzlp. techniques to improve the quality parameters of pulses. Broadband dividers current and voltage measuring elements as shapers. Examples of specific schemes shapers. Conclusion. Institute of Nuclear Physics of the USSR Academy of Sciences, Preprint 91-51 , Novosibirsk 90, USSR. Mr. 36str. Abstract: In the paper, a review of methods of shaping. Proposed new circuit design techniques that consider practical scheme using serial diodes.
-----------------


wish there was an entire translation of that book somewhere...


went back and read Dally's original postings, and now some more things make sense; unfortunatly have to find the spirit of the work in some cases...


DSRD; why?
Why isn't this just a simple diode, that a high voltage pulse is fed in on the - side so it doesn't go through, and when it reaches the end, the signal will still be high, and go through that diode... eventually smoothing out and/or consumed by voltage drops in the diode...


wikipedia on DSRD


... Short pulse of current is applied in the forward direction of the DSRD ... So I even have my diodes backwards?  or is it working on the BEMF spike ?  Mine is, but should it?[/size]



Immediately when the voltage drop happens, there is a spike on the output side, and then a graceful surge after... hmm I really wish I had 4 scope points in parallel (again, if I had that I'd wish for 8 )

maybe a better direction is to tune L2 closer to L1... but also, as the impedance on W3-TR2 changes, the optimal frequency changes... so I guess I need at least a fake load? Oh if I tune it for like 1/4 a wave, then a wider range of high freqencies will resonate the lower frequency?

[Hoppy]

Do you realize that a 50V pulse on the collector of Q5 (e.g. KT926) can still provide 1kV nanopulse at the DSR Diode when it abruptly stops conducting in reverse?
If not, see here.

Yes, I'm not doubting this.

Regards
Hoppy

[Hoppy]

Shottky diodes do not work well as DSR Diodes, because they do not have semiconductor to semiconductor junctions, instead they have metal to semiconductor junctions and those junctions have very sharp borders.
DSR Diodes require semiconductor-semiconductor junctions (PN or PIN)

SiC semiconductor-semiconductor diodes (PN or PIN) could work if they were manufactured differently (more diffused), but I do not know of any that exhibit the DSR effect. There was a guy that manufactured his own experimental SiC DSR Diodes on purpose, but they were not Shottky diodes - the were diffussed PN junctions AFAIR.  I can dig up his name if you want.

The ideal DSR diode structure is similar to a PIN diode that has blurred border between the P-I junction and between I-N. 
In other words, for best DSR effect,  the semiconductor junctions have to be diffused (not sharp).

It so happens that the old high power silicon rectifiers were poorly made which resulted in blurred border between their semiconductor junctions and now they unintentionally work well as DSRDs.

The are very few DSR Diodes available that were tailored for this effect purposely - all of them are custom experimental devices.

P.S.
DSR Diodes must be able to conduct very high current in reverse for apx. 20-70ns - that's the essence of the DSR effect !

Thanks Verpies. I now have a much better understanding to the nature of DSR diodes.

Regards
Hoppy

[verpies]

DSRD; why?
Why isn't this just a simple diode, that a high voltage pulse is fed in on the - side so it doesn't go through, and when it reaches the end, the signal will still be high, and go through that diode... eventually smoothing out and/or consumed by voltage drops in the diode...

Because DSR Diodes are one of the the fastest high power opening switches (SAS) known to man.
Conversely, Sprytrons and Krytrons are the fastest high power closing switches in existence.
Transistor cannot open as quickly as DSRDs.
Transistors can open in 10s of nanoseconds and only withstand 10s Amps at some 100s Volts.
DSR Diodes can open in picoseconds while kiloAmps and kiloVolts are flowing through them in reverse !!!

In a circuit, the DSR diode first conducts forward for several 100s of ns and next it conducts in reverse for several 10s of ns.
When it abruptly stops conducting in reverse, the inductive flyback spike forms the kV pulse.
So, yes - inductance is necessary to get the DSRD to form a kV pulse.

The funny thing is that those regular old rectifying diodes can be used as DSR diodes, bacause of "sloppy"manufacuring practices in old times.

wikipedia on DSRD

I wrote that Wikipedia article

... Short pulse of current is applied in the forward direction of the DSRD ... So I even have my diodes backwards?  or is it working on the BEMF spike ?  Mine is, but should it?[/size]

For the DSR effect to occur, the diode must conduct in both directions.  First forward and than in reverse. The abrupt opening happens only when the diode conducts high current in reverse.  This abrupt opening forms a huge flyback ("BEMF") spike at the output.

P.S.
For the Russians:
In English, the word "opens" means "stops conducting"
In Russian, it is the opposite.

[mihai.isteniuc]

Hello 2 all,

It's been a while since my last post. In all this time I was hardly testing parts of the Dally's device and I can tell you that both Сергей Ð'. and verpies are right. I will not go into details of their posts cause it will take too much time. But for both of them and for all of you thank you for your testing and researching. You have done my job easy showing to me the right and the wrong directions with your info.

First of all I have nanopulses. Good news. Bad news is that they are not the right ones and for producing them I use too much energy. I know now what I did wrong and I hope soon I will correct the errors.

THE MISTAKE (a huge one!!! at least for me). The pulses coming from the nanopulse generator (the low power part - are too long).  The first LA3 is the generator itself. The second LA3 is a sharpener if I can call it like this. In point A (see the schematics the pulses look like the ones in channel 1 of my oscilloscope (disregard the frequency at this point). In point B they should look like the ones in channel 2.

Dally with measurement points: http://postimage.org/image/5qxmh9uvd/
Pulses in points A and B  http://postimage.org/image/zbg8rflhl/ (probe settings: 1x)

For me it was a connection error that finally burn my TL494 converter, because I was forcing it with huge loads (BTW it's repaired and alive again, and also burn out at least 4-5 MOSFET transistors - replace for KT926 - still waiting for the KT to arrive).

I have ask you to confirm previously if the duty cycle is somewhere 40-60. You said yes. WRONG. The real duty cycle should be very low somewhere maybe 10-90 or something, as you can see. And it's logic too. Let's consider the propagation delay between point B and D let's say 25ns. After the nanopulse it 's generated in point D, well you're done. Your goal has been achieved. If your transistor KT926 will stay in conduction more then 25ns you will only consume energy for nothing.

The measurments for DSRD effect where done in point C and ground. If the DSRD diode it's not connected there are only normal regular pulses. when the diode it's connected (in my case it's a romanian model D25A10 rated 25 Amps, 1KV) the pulses appear. They can be generated as low as 50V supply line (instead of 150V). In the image below the only detail you are not seeing it's the probe with 10x settings). I have only wounded the primary with 6 turns on the ferrite ring. I don't know what's happend in secondary and beyond. It's irrelevant as long the first part it's not working properly.

My nanopulses: http://postimage.org/image/myctli65j/

To further sustain my claims see the attached doc. At page 4 at the bottom of the page you will find this:

DISCUSSION

During the series of the experiments have been got the following results. Minimum
rise time was 1.6 nanosecond. As high maximum peak voltage as 550 Volts on the 50-Ohm
loading has been achieved under 30 Volts power supply voltage. Power consumption was
less than 6 Watts with 20 kHz PRR. Connection of two identical diodes in parallel
improved impulse shape without visible rise time degradation.
Maximum PRR volume was 25 kHz and it is determined by output pulse transformer
overheating. Note that peak pulse power decreasing did not allow increasing of the PRR
value. Evidently higher PRR using proposed schematic can be achieved by utilization of
other transformer with improved electrical parameters.

So the nanopulser needs somewhere around 6 Watts to work. Let's say 10W. I hope I will soon confirm this. And U2.

Mihai

As usual if I'm wrong in any way don't be a stranger ...