Kapanadze Cousin - DALLY FREE ENERGY

[Hoppy]

   OK, Void. Thanks for the info.
  Problem lies in not being able to read the numbers or able to reach the knobs when the scope is 3 feet away.
I have it about two feet away, where I can see it and reach it. I hope that it's far enough away.
  BTW: My Tectronix scope just bit the dust. So, I'm concerned about the Siglent.

Nick,

Do not make any direct connections with your new scope probes unless you have at least x100 probes.

Looking at the scope shots you posted earlier, it appears that you have no dead time and possibly suffering from current 'shoot-through'. The voltage spikes on the yellow scope shot appear to be clipping due to the mosfet body diode going into reverse breakdown.

[NickZ]

   Itsu:
   Remember that I also have the MUR1530 between the source and drains.
   If those scope shots are not the avalanche that I can see, I don't know what is happening, then.

   You are right about the HV not affecting the TL494 or driver voltages, it doesn't affect them. But the snubbers resistors DO heat up when the Kacher is also on, and a heavy load is used, like 300w or higher. Using a smaller bulb, such as a 100w bulb, lights to blinding output. It takes me a minute or two to be able to see anything again, afterwards.

   The snubber resistors will start to smoke after one minute running time, or less. So, I doubt that the new carbon resistors that I'm waiting on will make much difference. But, we'll see.
   In any case, I'll do whatever it takes to get this working right.
   
   My older scope is dying, of old age I think. But, it still shows a dim signal, for now, and so is still somewhat use able, I hope.
   I'll use my single 100x probe on the new scope, for now.
   
    Thanks for your advice, guys.

[itsu]

Hoppy,

this "it appears that you have no dead time" was extensively researched in this thread some time ago, and it turned out that you cannot
judge the duty cycle / dead time from looking at the drain source voltage, it is needed to look at the current there to make a real assesment.




Nick,

the parallel MURs will not protect the MOSFET from avalanching i think, the weakest diode (either the external MUR or the internal body diode)
will start conducting and then the current will flow and starts flowing (despite if the gate voltage is on or off) until a critical temperature threshold is reached.
At least this is how i understand it.

There are many documents to be found on the net when searching on MOSFET avalanching, like this one on how to protect against it:
http://www.semtech.com/images/datasheet/transient_voltage_protection_of_mosfets.pdf


I cannot explain why your snubber resistors are heating up when the kacher is on, perhaps they are a too low value and we need to recalculate the RC values
with using your new scope (you know the drill by now, following the 5 steps).


Itsu

[AlienGrey]

Have a look at the two pic's they are of a C-Mos Logic emulator.

It works on a Katcher VFO clock driver (tuned coil required) and a divide by 50 + 2 +2 to give an EW push-pull drive with variable dead time using 5 logic chips + 2 added drivers not shown.

So if Katcher was for the sake of argument was 2 Mhz the push-pull would be 20kz (10 +10 kHz).

Notice the dead time.

Allen

[NickZ]

Hoppy,

this "it appears that you have no dead time" was extensively researched in this thread some time ago, and it turned out that you cannot
judge the duty cycle / dead time from looking at the drain source voltage, it is needed to look at the current there to make a real assesment.


 




Nick,

the parallel MURs will not protect the MOSFET from avalanching i think, the weakest diode (either the external MUR or the internal body diode)
will start conducting and then the current will flow and starts flowing (despite if the gate voltage is on or off) until a critical temperature threshold is reached.
At least this is how i understand it.

There are many documents to be found on the net when searching on MOSFET avalanching, like this one on how to protect against it:
http://www.semtech.com/images/datasheet/transient_voltage_protection_of_mosfets.pdf


I cannot explain why your snubber resistors are heating up when the kacher is on, perhaps they are a too low value and we need to recalculate the RC values
with using your new scope (you know the drill by now, following the 5 steps).


Itsu

   Hoppy: There is no problem with the dead time controls. I can raise or lower the duty cycle as needed. And there is always the minimum dead time between cycles.
   Adding another load by connecting up the feed back path is not going to lower the heating, but raises it instead. Just like it does when I place any more bulbs on the output. In any case we'll see what happens, soon enough.
   I'll show better scope shots when I get the chance.

   Itsu:  I get my carbon resistors for the snubbers tonight. So we'll see if that changes anything, but somehow I doubt it.
If there is no change, we'll have to rework the snubber values, using the new scope. I'm hoping that it's just a matter of getting these snubbers values right. However, the mosfet are running cooler even as it is now. And I'd rather burn up the resistors, than the fets.
   I have tried on different snubber resistors, and capacitors, but the heating is still present. Everything else seams to be ok.
   I'm still aiming for the ringing radio signal which I've lost. But, perhaps it's not an important consideration either, but I still need to see that for myself.
   The 15KHz or 27KHz are where the sweet spots are at on my device. Yet, I don't know if the increase in output there is enough to self run on. We'll see after I install the carbon resistors. As I still can't run the device for more than a few seconds, so it's been hard to have enough run time to fully test this rig up to now.