I only pay $20 per month on electricity bills, and even these unconvetional type of inverters can be also connected to the grid for further savings. I'm already doing this, and have also a few solar panels, connected to these inverters, as well.

Nick,

What type of unconventional inverters are you tying to the grid for further savings?

  Hoppy:
  I'd rather discuss what works.  As you feel that all of these shown devices are nothing but fakes, or bad meter reading, etz... I don't see the point.  I'm not trying to prove anything to anyone, so, I'll leave that for someone else.

  I know about using regular conventional solar systems, for home lighting and such, be it back up power or for full time operation, but that is not what I'm after.
 I'm after a self running device, with a working feed back loop, to light incandescent bulbs.  And it's possible that a regular inverter may work better for this purpose,  or maybe not.   

 
  Hoppy:
   To clarify,  I have connected the Mazilli/yoke (and other circuits) to a 12v, 500mA wall adapter, along with several other adapters, and can produce some useable light, even at this low input. But, that is not what I'm into discussing.
  If you were to see the amount of light that three 100watt bulbs give off, through this 12v yoke circuit, you would not be thinking like you are. However I'm not against using regular inverters, if they work better yet. But, this needs to be proven. Lidmotors video showing a 38watt bulb drawing 4 amps does not convince me that that is the way to go.

  Hoppy:
  I know what you're saying. But, I don't care about the proper measurements, I'll leave that for those that care about that. 
  As mentioned, the non-shocking magnetic output current of this device can't be measured. Meters will not work right. Nor do I believe that a scope can properly read this type of output current, as well. And, If so, maybe someone will report back on their findings. Although it seams that no one is interested in pursuing this, at this time.
  How do we know and can be sure that what the scope reads is for real?
 

   Hoppy:
   The conversation can end,  or it can change to what I'm more into working on, and discussing on this thread.
Which is the feed back path, the overheating issues and cures, loud ringing noise, and the tuning for maximum output brightness from the bulbs, etz...  Not the efficiency issues.
   I don't have lux meters, scopes, signal generators, spectrum analyzers, variable output power supply devices, etz...  But, I'm doing what I can with what I have available to me. And always report my findings, as best as I can. 
  No Z wires, either.

  I would rather show you a working device. No words needed...  just a toast and a big cigar. Ha!

  Gyula:
  Your suggestions are well taken. I'll be more informative, in the future.
   I removed a single turn from the center tap of the flyback coil, (one turn from each side). And connected the flyback up as before. The results was that another (second one, now) single BMR diode, popped after a few seconds running. So, there must still be too much voltage going to the rectifier. Remember that I can't properly read the AC output from the coil on my meter. 
  I also removed the flyback and replaced it with the smaller yoke core that I had on previously. But, I'm out of more diodes, until I look through my junk pile inventory, again. This second yoke core does not put out as much power as the flyback does, at least not with the current windings, so it may work better at keeping the heat down on the rectifier.  I'll adjust the AC output to lower voltage levels. As it looks like the rectifier is one of the bottle-necks in the feed back loop.  The cap that I'm using after the rectifier for the moment was a 200v 400uf electrolytic.
   I did find a set of 4 fairchild DH47Ah DFP 8896, which may be diodes, or something else. I can't find the info on them yet, but they look similar to the diodes.
Edit: Ok, I just found the info on the DFP 8896, they're mosfets instead, so, I'll keep looking for more diodes.
 
                      NickZ
 
 

Hi NickZ,

Say you have 25 turns for your output coil and say the output AC peak voltage is 250V, then a single turn gives as 250/25=10 V voltage. So if you remove say 2 turns then the ouput amplitude reduces from 250 to 230 V only.  This would burn the lower voltage diode equally "well".

Okay on the 200V 400 uF cap, sounds good enough voltage wise that may not get hot easily, especially if you keep a certain load on it.  And if you check the DC voltage across it with a voltmeter, you can continuosly keep an eye on that limit.

Yes the FDP8896 (not DFP) is low voltage high current  power MOSFET. 

Gyula

Hi NickZ,

Thanks for the further detailed info on the transformers.  Now think this over:
When you have a (say) 10 turn primary and (say) a 20 turn secondary coil on a core and you drive the primary, the turn ratio is 10:20 (viewing from primary to secondary direction) and your transformer steps up 2 times the input voltage  i.e.  if you input (say) 10V AC, the secondary output is 20V, right?

Now if you reduce the number of turns at the primary to say 8 turns and the secondary stays 20 turns, the step up ratio increases to 20/8=2.5 times from the earlier ratio of 20/10=2 times.
So when you removed 1+1 turns from the primary, you effectively increased the output voltage across the red wire coil of the flyback core! (i.e. you were giving an even higher "burning bang" to your diodes)

So you may wish to restore the original number of turns for the primary white wire of the flyback core and reduce the number of turns at either the secondary red wire of this core or at the secondary output (also red wire) of the yoke core, both methods will decrease the output voltage amplitude across the red coil of the flyback core.
(Alternatively, you may increase the number of turns for (the center tapped) primary input coil on the flyback core, this will also reduce the output voltage in the red coil of this flyback core.)

Because you may have found so far the turns ratio on the yoke core is okay now,  do not change it, and just reduce the secondary coil (red wire) of the flyback core. If this latter now has 10 turns, then you may wish to make a tap at its 3rd and its 6th turn too.

Regarding your UF306 type diodes, they sound good with their 600V peak reverse voltage and 3 Amper current ratings, they are labeled as high speed switching diodes. The 1N5408 types are for 50/60 Hz, not ok for the several kHz AC power you are dealing with,  so try to avoid using them.  (Of course you can test them to see how they warm up, they have 1000V and 3A ratings.)

It would be good if you had 4 pieces from the UF306 to make a diode bridge. If you have only two such diodes, then you either make a center tap for the red coil on the flyback core and build the rectifier circuit with two diodes SeaMonkey showed (no need for reducing the turns in this case because you have 600V diodes) or you build the voltage doubler I showed with the two diodes and start with a tap on the red wire at say 6th turn. And when you see no heat or only a little, under load, on the diodes and no heat on the capacitor,  then you may use the full 10 turn of the red coil and also attempt to loop back if the heat issue is still ok.

When looping back, it is a good idea to use a few Ohm current limiting power resistor or a light bulb in series with the feedback diode, this can help to save the diode if the feedback current is too high. And this current can be high if the rectified and puffered output DC level from red coil of the flyback core is much higher than your main input DC voltage from the 12V battery. This is where a simple DC voltage meter could also help you, to explore in advance what the DC level is after the rectification and this consider the voltage difference between the input-output you are going to connect. 

Regarding your capacitors as listed, if you start with a tap at (say) the 3rd turn only as the output from the flyback core, then you may try to use first the 200V 470 uF with a load across it of course (I say the 200V cap first because last time you found it did not get heated up but the diodes did.)
The best would be to monitor the DC level across the cap,  do you have a normal DC voltage meter either digital or analog? You mean you abused the meters and this is why you cannot trust them? You can check a DC voltmeter with a 9V or 12V battery in the 20V and then in 200V range, right? IF okay, then you can trust in the DC level across a 470uF puffer cap too, this latter can give enough filtering for the remains of the AC after the full wave rectification.

Gyula