Kapanadze Cousin - DALLY FREE ENERGY

[NickZ]

   Lost_bro:
   So, if I raise the input voltage from 12v to 24v, will the current draw be reduced by 1/2.  Or not? 
   I'm trying to find a solution to the fet overheating issue, still.  So, if by raising the voltage will not help the overheating, let me know. 
    The total power drawn may be the same, but the overheating may or may not be reduced.

 
   Forest:  Obvious, no.  As I don't rectify my grenade's output to run the AC 120v bulbs.
   I rectify the device's output only for the feed-back loop back to the input side.  Which by the way is now actually helping to raise the bulb's lumin levels, instead of dropping the output when I connect this feed back loop up, like before.
  So, I can say that the (rectified DC) feed back path is working now, and the higher the feed back loop's voltage/current, back to the input side, the brighter the bulbs become, while using the same fixed power input source. 
Not self running yet, but, I'm working on it.  Still a long road ahead...

[lost_bro]

   Lost_bro:
   So, if I raise the input voltage from 12v to 24v, will the current draw be reduced by 1/2.  Or not? 
   I'm trying to find a solution to the fet overheating issue, still.  So, if by raising the voltage will not help the overheating, let me know. 
    The total power drawn may be the same, but the overheating may or may not be reduced.

 
   Forest:  Obvious, no.  As I don't rectify my grenade's output to run the AC 120v bulbs.
   I rectify the device's output only for the feed-back loop back to the input side.  Which by the way is now actually helping to raise the bulb's lumin levels, instead of dropping the output when I connect this feed back loop up, like before.
  So, I can say that the (rectified DC) feed back path is working now, and the higher the feed back loop's voltage/current, back to the input side, the brighter the bulbs become, while using the same fixed power input source. 
Not self running yet, but, I'm working on it.  Still a long road ahead...

Good day NickZ

will the current draw be reduced by 1/2.  Or not?    Yes, for the *same* size load............Total power consumption remains constant unless you have achieved O.U.

fet overheating issue  There are many causes related to MOSFET overheating issues:

Overstaying it's welcome at the *Miller Plateau* ie;  slow Tr on gate signal.........  increase decoupling capacitance on driver, use a more powerful driver. Speed-up/clean-up  Tr (rise time) on gate signal.

Not reaching the *full-on* voltage level necessary, ie; from spec. sheet.

Not turning Off completely; Check with scope

Voltage going *too far* below ground while switching off; check with scope.

Spurious oscillations on GATE due to feed back circuit; Check with scope.

To low of an inductance on primary coil/ to fast of a slew rate.

Too High a Frequency; check specs...... spec RF FETs available.......

Vds overvoltage, ie; usually burns up fast............

Over current conditions, ie; usually burns the FET.

Faulty Source to ground connection, not common.

Insufficient heat sinking.............

This list in not comprehensive,,,,,,,,,,,,,, just wanted to throw out some ideas for ya....


take care, peace
lost_bro


EDIT..... not being sarcastic but,  you cannot do this kind of work without an 0-scope................. believe me.

[magpwr]

Good day MagPwr

Good to hear that you enjoyed the videos.........

Want to take advantage of the moment and make a couple of comments on the schematic that you posted:
As we know it uses Pin#14 so it is TypeII PLL, which is different from TypeI PLL on Pin#2 in that @ resonance, both the signal and comparator input are in Phase = Zero shift.  Of course the draw back to Type II is that it is very much more sensitive to EMI and sloppy board layout than Type I.

I used TypeII PLL in my SSTC design also... All the other TC designs I have seen on the web use Type I exclusively for that very reason (EMI problems).  The type I really requires much more circuitry to implement because the phase shift is variable 180 degrees through the VCO range. (90 degrees @ center VCO). 

Type I will also lock onto *harmonics* or overtones of the principle frequency, Type II will NOT lock on *harmonics*.

I took the liberty to comment a couple of items on the schematic, please see attached *commented* schematic.

Take care, peace
lost_bro

hi lost_bro,

The circuit at pin1 of 4046 is not required it is just visual indicator.I merely connected green led with resistor to ground.

However the inverter after LM393 or TLC272 output is required.

The current sensor i am using a blue toroid mounted on the thick wire from 3 turns from toroid/yoke after 0.47ufx2 WIMA capacitor.
The 10ohms resistor and 2 x 1n5818 diodes connected as bi-directional is connected in parallel from output of blue toroid using 20 turns.

The clock capacitor for 4046 needs to changed to 1nf and resistor at pin 11 and pin 12 needs to be changed to a 100k  pot or with suitable resistors.

---------------------------
My experiment with 4017 used as frequency divider was a failure with the multilayer coil it was just stuck at 36.4khz.

But final testing merely using breadboard late last night by changing clk capacitor from 1nf to 471pf for the pll solve the problem.

I would be available on Thursday onwards to commence experiment.


--------------------------------------------
Side track which is indirectly related to my project-

I have attached a image of the world smallest modified sine-wave 220volts 150watt inverter with 300watt surge capability.

I wonder how they managed to fit a 50hz transformer in there and tested to be electrically isolated from the 12volts supply and comes with 4 types of circuitry protection.

[NickZ]

Good day NickZ

will the current draw be reduced by 1/2.  Or not?    Yes, for the *same* size load............Total power consumption remains constant unless you have achieved O.U.

fet overheating issue  There are many causes related to MOSFET overheating issues:

Overstaying it's welcome at the *Miller Plateau* ie;  slow Tr on gate signal.........  increase decoupling capacitance on driver, use a more powerful driver. Speed-up/clean-up  Tr (rise time) on gate signal.

Not reaching the *full-on* voltage level necessary, ie; from spec. sheet.

Not turning Off completely; Check with scope

Voltage going *too far* below ground while switching off; check with scope.

Spurious oscillations on GATE due to feed back circuit; Check with scope.

To low of an inductance on primary coil/ to fast of a slew rate.

Too High a Frequency; check specs...... spec RF FETs available.......

Vds overvoltage, ie; usually burns up fast............

Over current conditions, ie; usually burns the FET.

Faulty Source to ground connection, not common.

Insufficient heat sinking.............

This list in not comprehensive,,,,,,,,,,,,,, just wanted to throw out some ideas for ya....


take care, peace
lost_bro


EDIT..... not being sarcastic but,  you cannot do this kind of work without an 0-scope................. believe me.

   Yet, the guys with ALL the proper test gear have not shown a higher bulb brightness (lumin) levels, nor have no fet overheating when drawing over 500 watts to 1000 watts from their device. 
   I've checked most of the things that you've mentioned already, and I've also exchanged most of the components in the circuit several times, but I'm still looking for a solution, if there is one, other than making a new push-pull driver.
The fets that I use, IRFP260N should handle 50A, and 200v. 

[lost_bro]

   Yet, the guys with ALL the proper test gear have not shown a higher bulb brightness (lumin) levels, nor have no fet overheating when drawing over 500 watts to 1000 watts from their device. 
   I've checked most of the things that you've mentioned already, and I've also exchanged most of the components in the circuit several times, but I'm still looking for a solution, if there is one, other than making a new push-pull driver.
The fets that I use, IRFP260N should handle 50A, and 200v.

Good evening NickZ

OK, the first thing I would do the scope-out the Gate-Source signal on both FETs.

Then:  scope-out the Drain-Source on each FET.

This information will give you 99% of the picture as far as the MOSFETs are concerned..........

You will find your answers there.

That much being said, from what I have seen it is normal for a Mazzilli circuit to heat up the FETs a bit.  Check out this on ebay:
http://www.ebay.com/itm/New-ZVS-Tesla-coil-flyback-driver-SGTC-Marx-generator-Jacobs-ladde-12-30V-DC-/121541256158?pt=LH_DefaultDomain_0&hash=item1c4c6c63de

It has a built in fan on the heatsink.....

take care, peace
lost_bro