Air gap

7 posts / 0 new
I calculated with PI Expert a fly back transformer with an air gap of LG = 0.35mm. Is that the total gap distance ? do I have to use a core half with a gap of 0mm and another core half with a gap of 0.35mm ? Thanks

As long as total gap is 0.35 mm you can do any combination of gaps on each core half (ie 0 + 0.35 mm, 0.175 mm + 0.175 mm etc).



i m also doing experiment on my transformer by altering gap . i know one reason for introducing gap is that it helps in preventing core saturation but if ne other thing is there then kindly let me know.

Hello Vaib, "core saturation" is just one aspect related to "gaps" in the magnetic flux path. It also increases the "reluctance" of the magnetic circuit - which in turn helps to store "energy" when the current is flowing through the primary winding when the switch is closed. when the switch opens, flow of current stops, reversal of polarity happens and this "stored energy" in the gap is transferred to the secondary. long and short - gaps are sensitive to the performance of the transformer and the whole powersupply - just follow the "suggested" "calculated" value of gap in the PI "expert design software" - experiment, play and be safe... good luck - GPM

Hi Jom,

You are correct the air gap prevents the core from saturating. Experimenting can be interesting but for your final design I'd strongly recommend you follow PI Expert or PI Xls for guidance on the gap size. The gap will determine the value of the flux density (and therefore at what primary current core saturation will occur) as well as primary inductance (and therefore power delivery).

See AN16 page 20 for the mathematical relationships (this is for TOPSwitch but applies to any flyback):



Is it possible if the calculated air gap is 0.35mm and on the actual transformers the air gap is 0.70mm that the the clamp circuit has to dissipate too much (some transorbs goes defect after a while) The disign is made with a TOP switch TOP249Y

Hi jom

This does make some sense. If the gap is larger than specified then the value of the primary inductance will be much lower than designed for. As output power is proportional to 0.5 x Lp x Ip^2 x f the TOPSwitch will still try to regulate the output by increasing the primary peak current. As the energy dissipated in the clamp (transorb in your case) is proportional to 0.5 x Ll x Ip^2 x f (where Ll is the leakage inductance) then the energy dissipation in the clamp can increase significantly.

Other possibilities are: leakage inductance too high, clamp voltage too low (need to be ~1.5 times the value of VOR). The TOP249Y is a large device so I'd expect a leakage inductance value <1% of the value of the primary inductance.

To help understand more specifically can you provide the schematic and transformer design and measured primary inductance and leakage inductance.

If you don't have an LCR meter then this AppsTV segment might help you: