Kapanadze Cousin - DALLY FREE ENERGY

[Hoppy]

Don't let's go into the Fraud of Faith Religion and Bolxtics on here, please !
we don't need the mind games, just focus on out target ! many thanks.

I need a much stronger magnifying glass lad. 

[lost_bro]

@lost_bro

That's a nice design and a step in the right direction.
Pls write some more info about the DC-DC connection transformers,  e.g. what is the capacitance between their primary and secondary windings?  Are they regulated?  Power, switching frequency, etc...

I am afraid you design will be lost on most people here, because it looks complex and expensive ...despite that it solves most FET driving problems and even saves money on burned up FETs in the long run and allows to do fast (μs) bidirectional solid state relays ...and the general versatility of the thing!
If it had drain overcurrent protection included, then only drain overvoltage and dv/dt could kill the FETs.


P.S.
Were you ever tempted to do gate charge recovery or HV gate pulse preemphasis?

Good day Verpies

About the Isolated dc-dc SMPS....  I designed the PCB to use the standard footprint ( for the Iso-SMPSs) to accommodate any brand of Iso-SMPS which incorporates the SIP style layout.

I chose (for economic reasons) the ND brand (Chinese made) Iso-dc-dc SMPS.  For the pcb in the photo, which will switch an Infineon FF150R12KS4_B2 IGBT 'Brick' which is a Short Tail, High speed -bridge configuration- IGBT module, I opted for an adjustable voltage range of +15,-5; Using only a -5volt (instead of the standard -8 or -10volt) reduces the overall power dissipation of the AVAGO opto.  (right now it is engineered to the limit of the AVAGO opto).

So, this pcb uses two pieces *12V to dual output 9V isolation 6kv - G1209S-2W unregulated SMPS*. and
two pieces  *6KV dc-dc unregulated single output 5v to 5v 2w SMPS*. One pair for each of the two AVAGO optos.

The dual 9volt output unregulated supplies allowed me to configure the design so the +9 & -9volt on the output are cumulative and float above Ve; Ve=0volt; which gives me a maximum of 18-20 volts above the Ve.  Since the Iso dc-dc supplies are non-regulated, I only have to adjust the Vcc, (input voltage to pcb) within a range of 10.8-13.2 volts to get a variable 14-20volt output @ the Iso dc-dc SMPS for the positive fraction of the Gate drive voltage.

The pcb also has on-board a small adjustable Buck Converter which is used to step down the Vcc (which is variable as stated above) to an adjustable range of 4-7volts.  This is fed into the input of the two pieces  *6KV dc-dc unregulated single output 5v to 5v 2w SMPS*.  Since they are likewise un-regulated, the output is between 4. and 7.volts, but these are connected below Ve accordingly to float the negative voltage (Vee) in relation to Ve.  So the negative voltage is adjustable from -4 to -7 volts.

The switching speed of the ND- iso dc-dc smps is between 70-100Khz depending on the model chosen.
The isolation capacitance is also model dependant, but for the dual output design the spec sheet says:
Isolation Capacitance Input-output, 100KHz/0.1V -- 5 -- pF...............

Unfortunately, the AVAGO  ACPL-K33T-060E do NOT integrate Desaturation protection nor drain overcurrent protection functionality, only UVLO, which is better than nothing.  I opted for the AVAGO  ACPL-K33T-060E optos at the time, because they were one of the first dedicated optos (15-30 volts range) made for use with SiC mosfets, and I originally intended to use SiC Fets with two of these pcbs (parametric resonance, ie; switching capacitances via microcontroller algorithm & zero voltage-crossing detector).  I would now look seriously at the Si. Labs line of RF iso-drivers.......

This Iso-pcb driver board can be used with any type of Clamping protections,  regenerative or dissipative.
This only depends on how you want to configure any additional reset windings, etc. in your XFRMR, paired inductors and so on.

The nice part is that your small signal logic is *isolated* and protected and your FETs should be a little more immune to the R.F. (guess that last part really depends on proper bridge (mosfet) layout on the power side).


"P.S.
Were you ever tempted to do gate charge recovery or HV gate pulse preemphasis? "

Considering that this pcb is just about 1/2 size of a pack of cigarettes, It might be hard to fit more functionality on board. At any frequency the pcb is capable of, dc to +1Mhz, the traces are of sufficiently short length that all transmission lines (traces) would be considered 'short transmission lines' or much less than 1/4 wavelength long of the operating freq.  So the effect on the propagation time delay would not be significant on these relatively slow signals.
I haven't seen much distortion in the signal, but than again I do not have a running kapanadze device to test it completely yet.

I would be interested in your idea of HV gate pulse preemphasis, it you have any literature you can recommend.

Actually after thinking about it for a while now, I am tempted to use a DRSSTC direct-feedback controller and a H-bridge via GDTs to drive the *yoke* primary windings in this device. Damn things are bullet proof and I know from experience that they can deal with the High level of RF. without distorting the waveform. 

take care, peace
lost_bro

[d3x0r]

wish I could speak his language or see the meters slightly better... but this is close to a win?
https://www.youtube.com/watch?v=XhvNOJ-06eg

[NickZ]

  A win?  No, just an inverter, not self running.
  Until he can disconnect the battery and it continues running, there's nothing unconventional or special there.

[d3x0r]

"they're (meter's readings) too good to be true - 12v, 3a DC input, 80v, 1.2a AC output; will tell more after figure out what exactly is on output side....."


36W in 96W out?
Although I suppose that's just measuring the tank curcuit... not what the bulb is actually using from that.