3
\$\begingroup\$

An article I have read regarding switching power supplies (https://sound-au.com/articles/smps-primer.html) says that push-pull, half-bridge, and full-bridge switching supplies don't use an air gap in the transformer.

Now, flyback converters need an air gap since their unidirectional operation puts a DC component in the transformer's primary current.

But would it be advantageous to use an air gap in a high-power (say 24VDC out at 50 Amps) or very high power (340VDC out at 800 Amps for a high-power air ionizer) full-bridge converter's output transformer? It seems that it would be great to do so, since the transformer would be much more resistant towards saturation if the microcontroller ever fires the switching transistors on one side longer than the other, but what are the disadvantages?

\$\endgroup\$
8
  • 2
    \$\begingroup\$ No such thing as prevent any DC component. You can compute the reluctance but I usually just see it as lower Al. So, for your full bridge, reduce the Al, introduce DC and start integrating your voltage-time area to see where your flux level ends up. \$\endgroup\$
    – winny
    Nov 27, 2017 at 21:52
  • \$\begingroup\$ I said that it prevented the DC component from saturating the core, not that there was no DC component... \$\endgroup\$
    – El Ectric
    Feb 22, 2018 at 21:44
  • \$\begingroup\$ Same thing. You can build a flyback converter (or any topology with unidirectional core excitation) with zero air gap if you want. Unfavorable operating point yes, but no saturation despite DC component if you design accordingly. \$\endgroup\$
    – winny
    Feb 22, 2018 at 22:21
  • \$\begingroup\$ Thanks! I'm currently looking at a gapped core in a transformer for a high-power full bridge converter. \$\endgroup\$
    – El Ectric
    Feb 22, 2018 at 23:16
  • \$\begingroup\$ Strange operating point. Does it have super low input voltage and very high frequency? \$\endgroup\$
    – winny
    Feb 22, 2018 at 23:19

2 Answers 2

3
\$\begingroup\$

Adding an air gap to the the flyback transformer increases the reluctance of the magnetic circuit. Doing this does a few things:

  • Decreases the effective permeability of the core
  • Decreasing permeability (also decreasing inductance) increases the amount of current that can be supplied to the transformer before saturating.

enter image description here

enter image description here

These equations show how introducing a gap changes the effective permeability (z is the 'gap factor'). Lg is the gap length introduced and Le is the length of the path the magnetic field loops around (not including Lg). Here is a picture of how the gap affects the BH curve. You can see that you are able to apply a larger H field before the B field rolls off and the core is saturated.

enter image description here

Since a larger primary side current can be used before saturation occurs, effectively more energy can be stored in the primary side magnetizing inductance.

Other topologies don't need an air gap like the flyback since they are transferring energy during both the on and off time of the switching. Energy is not stored in the magnetizing inductance in the same manner. The flyback is different since it stores energy when the primary side switch is on and transfers energy to the secondary size during the off time.

I would recommend you build a flyback (with safety precautions). You can wind your own transformer and file the core yourself if you're so inclined.

\$\endgroup\$
2
  • \$\begingroup\$ Is it possible to build a push pull/half bridge/full bridge converter where energy is stored in the transformer when the transistor(s) is (are) on, and then when the transistor is off the energy is released to the secondary side? A kind of converter that does what the flyback does in one cycle, and then the same thing in the opposite direction. \$\endgroup\$
    – El Ectric
    Feb 23, 2018 at 5:21
  • \$\begingroup\$ If we take the forward converter as an example, we don't really need the transformer for energy storage since this converter has an output inductor. The energy is stored in this inductor in the same way energy is stored in the flyback transformer. Even if we did introduce an air gap into the transformer for storing energy, we aren't able to retrieve it with this topology during the off time like we are with the flyback. This has to do with the configuration of the secondary side diodes. For this type of topology, magnetizing inductance is made large to reduce wasted energy storage. \$\endgroup\$ Feb 23, 2018 at 13:00
4
\$\begingroup\$

Air gap just adds more reluctance - increase of magnetic resistance. Like adding a resistor in the circuit to limit the current (flux) with the scope to avoid core saturation.

Adding the gap does not filter out the DC component, it behaves the same with AC, too.

Take some examples how o calculate the HF core step by step, you will then come to the point when to use or not the air gap. There are no cons/pros, once you have to add the gap - it just can't work without it. And it has to be quite exact, too.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.