Re: Ferrite Cores - Where's the Gap? (fwd)

[Steven Roys <sroys@xxxxxxxx>]

---------- Forwarded message ----------
Date: Mon, 5 Apr 2004 18:20:34 +1200
From: Malcolm Watts <m.j.watts@xxxxxxxxxxxx>
To: High Voltage list <hvlist@xxxxxxxxxx>
Subject: Re: Ferrite Cores - Where's the Gap? (fwd)

On 4 Apr 2004, at 22:58, High Voltage list wrote:

> Original poster: Steven Roys <sroys@xxxxxxxx>
>
>
>
> ---------- Forwarded message ----------
> Date: Mon, 05 Apr 2004 11:53:41 +0930
> From: Matthew Smith <matt@xxxxxxxxxx>
> To: High Voltage list <hvlist@xxxxxxxxxx>
> Subject: Ferrite Cores - Where's the Gap?
>
> Hi All
>
> Flyback transformers have gapped cores, yes?  Why is it that not one
> core that I have lying around has anything between the two halves of
> the core?  Does the fact that they are not one single piece count as
> an air gap?
>
> Cheers
>
> M

It is important to understand that the shape or winding type of a
transformer does not define it as a "flyback" transformer. What is
defined as "flyback" is a particular circuit topology which has the
following characteristics:

- the duty cycle is less than 50%
- the supply is single-ended (i.e. not pushpull or bridge)
- energy transfer is via energy storage in the core (switch on) and
  energy release from the core (switch off)
- a steering diode (rectifier if you like) at the output so that the
  secondary does not pass current while the switching transistor is
  on.

Since only one winding of a typical two-winding transformer passes
current at any instant *when used in a flyback supply*, it is
essentially being used as two chokes on a common core. The usual
turns-ratio considerations apply when determining how many volts will
appear across one winding when the other is passing current (i.e. has
a voltage impressed across it). It is all very predictable and you can
use this knowledge to calculate a safe rating for both transistor and
diode in this type of supply. More often than not, the winding
capacitances will dictate the efficient operating frequency and I have
actually seen supplies self-oscillate with no external feedback
applied to the transistor base (the internal transistor capacitances
do the job rather too well), the frequency being determined by the
winding inductances and self-capacitances.
     Snubbing may or may not be necessary depending on the amount of
leakage inductance present (normally a function of the proximity and
interleaving if any of the windings). In general, the smaller the
volume a winding occupies, the smaller will be its leakage inductance.
     A flyback supply will always have a gapped core since energy
transfer is accomplished by storing energy in the core (an undesirable
property in pushpull and bridge topologies) and the core material
itself will store very little energy without going into saturation.
Despite the undesirable side-effects of gapping in other types of
supply, transformers used in forward converters, bridge converters
etc. will often also have some core gapping to obviate flux
staircasing.

Malcolm