Kapanadze Cousin - DALLY FREE ENERGY

[NickZ]

  Yes, I know that now. I bid on a digital scope also, (never used second hand digital scope), but I lost that bid. 
It's no problem, I'll get the hang of it. But, I wish that I'd known about the math previously, it would have made a difference. 

  There are a couple of ways to tune the Mazilli set up, but it may not be as easy or as effective as turning some trim pots for the control of the PWM and frequency on the drivers.

  At least I got the Mazilli working normally again, and with the Kacher on 24v, I can light a 20 watt 220v bulb on the grenade's 168 turns output coil, and even a CFL off of the grenade,  as well. This is without any connection of the 28 turns yoke coil to the grenade. That bulb previously would not light at all when running the Kacher on only 12v.

[TinselKoala]

1. The capacitor in the Mazilli circuit is no more of an "energy recycler" than the coil itself is. The capacitor and the yoke coil form a resonating tank circuit in which the energy supplied by the power supply is "sloshing" back and forth between the magnetic field of the yoke coil and the electric field of the capacitor. The frequency of the resonance ("sloshing") is determined by the capacitance and inductance working together. You can attain the same resonant frequency by using many different values of L and C but there is probably some value where the losses inherent in the "sloshing" process are minimized. In other words, a small capacitance and a large inductance can give the exact same resonant frequency as a large capacitance and a small inductance, but perhaps a "medium" capacitance matched with a "medium" inductance value gives the best performance. The Mazilli is "tuned" to a desired frequency by adding or removing turns from the yoke coil, and/or by adding or decreasing capacitance of this main capacitor. You could also tune it by varying the gap between the core pieces, or even by using some "bias" winding on the core to vary its saturation level, hence varying the inductance.
Use the Resonant Frequency Calculator here:
http://www.1728.org/resfreq.htm
where you can enter any two of the three values F, L and C to find the unknown value. You will see that using this calculator and the known values of your capacitor and the frequency of the oscillation, you can "measure" the inductance of your yoke coil. Once that is known, you can then use that value, along with various capacitor values, to calculate the resulting frequency of oscillation of the tank circuit. And you can then confirm that resulting frequency using the scope.

2. The higher in frequency, the more important your _layout_ and physical construction becomes. For a good performing Mazilli/Royer oscillator, you will want to use a tight and symmetrical layout, with high-current conductors between the mosfet Drains, the capacitor and the yoke coil. Sockets (like 3-position terminal blocks) for the mosfets is a good idea. Use ultrafast, over-rated rectifiers for the diodes, like UF4007 or similar.  See the photo below. That one is tuned to oscillate a custom 5+5 turn primary on a flyback transformer core at about 30 kHz, near the resonant frequency of the flyback's secondary. Hence the 6, 1.0 uF 250V poly film capacitors in the series-parallel arrangement giving 1.5 uF at 500 V. Smaller ceramic caps can be put in parallel with the main cap bank if necessary for "trimming" the resulting resonant frequency.

3. To read the frequency of a repeating signal on the Analog scope, you simply divide the number of cycles by the time it takes for those cycles. On a scope like the Tek 2205 this is done like this.

  a. First set the horizontal time base so that you are showing as many full cycles on the screen as you can comfortably distinguish. This is one time when you don't want just two or three cycles showing-- get as many as you can actually count on the screen.

  b. Use the center 8 graticule marks, don't use the left and right edges. Analog scopes can distort a little at the edges, so you want to use the center portion of the screen for this measurement.

  c. Use the horizontal position control to set a peak exactly on the second graticule marker from the left edge of the graticule (The edge itself is the first marker). Now count the peaks from this peak (#0) all the way over to the second-from-right edge graticule marker. This will be 8 full divisions on the graticule of the 2205. Estimate the partial cycle at the rightmost. So let's say you count 19.3 peaks (full cycles) in those 8 divisions.

  d. Now multiply the timebase setting by 8 to find the duration of 8 divisions on the screen. Let's say you are set to 1 us/div.... obviously then the 8 divisions are representing 8 microseconds, or 0.000008 second. (kilo = 3 places to right of decimal, micro = 6 places, nano = 9 places, etc.)

  e. Now simply divide. You have 19.3 peaks / 0.000008 seconds = 2,412,500 cycles per second, or Hz. But since you only know your "peaks count" to three significant digits, you have to round your answer to three sig digs to avoid false precision, so you can report a frequency of 2.41 MHz for your measurement.

It takes longer to describe than to actually do it on the scope. Once you've done a few it will be second-nature. For greater precision, set the timebase to give as many cycles on the screen as you can comfortably count. If your peaks are irregular or multiple per cycle, then you can use the zero-crossings instead of peaks if those are easier to read. But the Mazilli Drains should be giving you a nearly perfect sine wave, so it's pretty easy just using peaks.

It is _much better_ to learn to do this on an Analog scope, before going to DSOs with their "numbers in boxes". The DSO's measurements will not always be accurate or even close, and digital scope users should develop the habit of actually reading the trace using graticule markers and channel/timebase settings to confirm and back up what the "box numbers" are telling you.  I have encountered DSO users who did not know how to read the traces, and who therefore have made serious errors in scoposcopy by relying exclusively on the "numbers in boxes" that the DSO spits out.


ETA: When doing quantitative measurements on the Analog scope, make sure your "cal" knobs on the vertical scale and timebase knobs are turned all the way to the right, in the "cal" detent if there is one.

[TinselKoala]

Got it.
Did you ever try to pick up the standing wave node with a neon bulb and a fiber-fed phototransistor ?  That would be cool!

That's an interesting idea and would probably work quite well. I use neons quite a bit but I've never used them as direct sensors of operating frequency. I'll have to try that when I get a chance. I do have one video where I show that the nodes of the standing wave can be visualized on a gutted CFL bulb:
http://www.youtube.com/watch?v=TKtjGLbf2Kg

[NickZ]

   Verpies:
   Thanks for that video link to the scope precautions.  I'll make sure that I have both the scope and PS going to the same AC house ground connection. And I'll try to use only batteries for running the circuits, instead of grid connection input sources.
And, I'll try not to connect that black negative crock clip to hot sources, as well. That's very good advice on that video link, as mistakes are easy to make, and POW!!!  Like he says...

   TK:  Thank you for taking the time to explain the process of doing the math on the scope. I'll be doing all that, very soon.
           
   Funny thing, last night I was able to light a 50 watt bulb, on the yoke's 28 turns coil, only using 12v, 2 amps. And, also a 100w bulb off of the grenade's 168 turns coil, but only dimly. The yoke's 3 turns coil (6 turns on mine) is only connected to the grenade's inductor 12, 12 turns coil. This was done with the Kacher running on 24v.
  I've been able to tune the yoke/grenade's running frequencies, similar to how Geofusion showed in his last video, but, by moving ferite cores inside the grenade's former tube, (pvc pipe). As well as by using magnetite inside the yoke core.  Works the same as tuning the pots on the TL494 circuits. Plus I also use different tuning caps, like the 0.47uf etz, and adjust coil turns counts, as well.  Although I don't have the higher voltage 2000v WIMA 0.47uf caps yet, but I'll have to get some of those, also.  As using only the 275v O.47uf caps, as I am, may be cutting off or limiting more of the "effect", that's generated by the Kacher's superpositioning.

  Edit:  I use neons all the time as HV sensors, as they will light on a lower voltage than the CFL bulbs need to light on, and also draw less input from the source, to light. Therefore the small bulb neon have less of a negative effect on the Kacher's output operation, when using them as signal sensors. Yet, they will allow one to know when the HV output from the Kacher circuit is operating properly, or when it's not, and to what degree, without needing to use any other bulbs on the circuits, or a scope, to see what's going on.

   

[magpwr]

   Verpies:
   Thanks for that video link to the scope precautions.  I'll make sure that I have both the scope and PS going to the same AC house ground connection. And try to use batteries for running the circuits, instead of grid connection input sources. And, I'll try not to connect that black negative crock clip to hot sources, as well. That's very good advice on that video link, as mistakes are easy to make, and POW!!!  Like he says...

   TK:  Thank you for taking the time to explain the process of doing the math on the scope. I'll be doing all that, very soon.
           
   Funny thing, last night I was able to light a 50 watt bulb, on the yoke's 28 turns coil, only using 12v, 2 amps. And, also a 100w bulb off of the grenade's 168 turns coil, but only dimly. The yoke's 3 turns coil (6 turns on mine) is only connected to the grenade's inductor 12, 12 turns coil. This was also done with the Kacher running on 24v.
  I am able to tune the yoke/grenade's running frequencies, as Geofusion showed in his last video, by moving ferrite cores inside the grenade's former tube, (pvc pipe). As well as by using magnetite inside the ferrite yoke core.  Works the same as tuning the pots on the TL494 circuits. Plus I also use different tuning caps, like the 0.47uf etz, as well.  Although I don't have the higher voltage 2000v WIMA 0.47uf caps yet, but I'll have to get some of those, also.  As using only the 275v O.47uf caps, as I am, may be cutting off or limiting more of the "effect", that's generated by the Kacher's superpositioning.

hi Nickz,

Nice to know you finally got a new scope after all these years in the forum.

Mazilli is easy circuit to work with but tuning is a pain since it's output frequency is dependent on the L/C which means it is harder to alter frequency merely with add/remove capacitors or with ferrite to alter inductance.

PWM circuits based on TL494 or SG3525 or 3825 was already posted in this forum.It is easy to assemble on a breadboard and very easy to vary frequency.

But i am not here to teach basics.

----------------------------------------------------------
I am on working on few FE related projects at the moment-

Nelson circuit as provided by him in this forum but was further modified to run with less than 2mA using 9volts and still produce high voltage.

Finalized design once more  and about to start PCB creation related to Don smith china replication video using 7.2kv using 6 x SIC Mosfet in series to switch via magnetically isolated signal driver(10Mhz rated) and 24volts power module all 1.5kv electrically isolated individually.

Gave up on Akula new device due to insufficient input and gathering parts for the older outdoor device.
(For the experienced -Study how grid-tie inverter works to lower mains power consumption.Synchronized sine wave.But i am still trying to figure out how 3x frequency can be in "phase" with  frequency eg:~55khz)