Kapanadze Cousin - DALLY FREE ENERGY

[verpies]

Phase shifting degrees, depend on the frequency that you work with.

Only if you use a fixed delay technique. 
With this technique, the phase shift is created by time delaying a set number of microseconds, not by a set number of degrees.
Because of this little difference, the same time delay creates different degrees of phase shifts at different frequencies.

Hey guys. Can anyone direct me to a schematic for a circuit that has been tried and tested which
will trigger off a pulse waveform, such as an output from the TL494, and which will create another pulse train
of the same frequency with a continuously adjustable phase delay, allowing setting the phase delay
angle from 0 to 90 degrees?

To do that, I'd need to know whether the e.g. TL494 outputs always stays at the same frequency ...or is that frequency to be variable (manually or automatically) ?

In case of the latter - do you want the phase shift to stay fixed as the output frequency of the TL494 is varied ?

[itsu]

I don't know what you mean by "bucking" since the these two halves of the primary winding are not supposed to be driven both ON at the same time, thus the fluxes generated by these windings cannot buck.

Anyway, since the goal is to produce bipolar alternating flux in the core, then these primary winding halves are supposed to be connected in such manner that when the 1^st MOSFET is ON, then the flux in the core is generated in one direction and when the 2^nd MOSFET is ON, then the flux in the core is generated in the other direction. 
An alternating flux, generates an AC across a resistive load connected to the secondary winding.

The black dots A & B, on the diagram below, illustrate the proper split-primary winding connections according to the dot convention.
The orange parts of winding W1 (dots C & D) as well as C3/D3 and C4/D4 are components of the optional lossless clamps.
...and that is the only reason not to use the lossless clamps which are more effective and more efficient than RCD snubbers.

Verpies,   


thanks for reminding me to that lossless clamp design, i did thought about using it, but my main problem with it is that the present
2x 12 turn primaries together with the 28 and 3 turn secondaries setup creates a certain voltage relationship.

I would have to use trail on error and much more secondary turns to get to a similar relation with your design i think.

But if all fails, i might give it a try anyway.

Regards Itsu

[verpies]

...my main problem with it is that the present 2x 12 turn primaries together with the 28 and 3 turn secondaries setup creates a certain voltage relationship.

Actually the lossless clamp technique does not change the turn ratio, nor the voltage ratios, because the additional primary windings (4x 12 turn in total) do not change the number of effective primary turns.  That quadriple primary winding still behaves as a 2x 12 turn winding, as far as the transformer is concerned. 
In other words: the 2^nd primary winding (orange) does not participate in transformer action - it just catches the spike energy from the 1^st primary winding (black).

In any case, pay attention to your dot convention so your primary windings generate a bipolar flux ...not a pulsating unipolar flux.

[itsu]

Ok, thats clear, i thought that one requirement of the lossless clamp design was that the primaries consists of severall layers around the circumference of the toroid
making much more then 2x 12 (or 4x 12) turns primary which distorts the primary/secondary relationship.

Good to know it can be done with 2x 12 (4x 12) primaries too.


Itsu

[Void]

To do that, I'd need to know whether the e.g. TL494 outputs always stays at the same frequency ...or is that frequency to be variable (manually or automatically) ?
In case of the latter - do you want the phase shift to stay fixed as the output frequency of the TL494 is varied ?

Hi Verpies. I would be adjusting the TL494 driver circuit frequency manually through a small range.
The main aim there would be to manually tune for series LC peak resonance in a secondary loop.
It would be nice to have a circuit that maintains the same phase shift if the TL494 driver's
frequency is adjusted up or down a bit, but I don't think it is necessary for basic testing. The Allega phase shift
circuit appears to be time constant based, but it should get the job done. Will just need to readjust
the phase delay each time after adjusting the PWM frequency.

Verpies don't go to any trouble designing a new circuit. I should be fine for now with the Allega phase shift
circuit for the testing I have planned. All I really need to do is for a given PWM driver frequency in the range
of say 10 kHz to 30 kHz or so, to be able to set the phase delay of a second pulse train of the same frequency
anywhere within 0 to 90 degrees.  It sounds like Allega's circuit will do the trick here.