I know that the leakage inductance is present due to the imperfect linking of flux between the two transformer windings. I have selected a core for the transformer to be used within a series resonant converter.

Using FEMM, I have simulated the transformer core along with the windings to get some idea about the leakage inductance for the transformer. I have calculated the leakage inductance value from the simulation results to be about 7 μH.

In this case, if I require a leakage inductance value of about 100 μH for the SRC to work for my application, can I keep an external inductor in series with the transformer primary to get the required value? If this is not possible, are there any other ways to achieve the required inductance value?

I am trying to implement a PCB transformer in this case, as inter-winding isolation is important.

  • \$\begingroup\$ Why couldn't you connect one in series? It's normal practice. \$\endgroup\$ Commented Jan 8 at 6:34
  • \$\begingroup\$ @TimWilliams If I connect like that, the magnetizing inductance should be much larger than the selected external inductance right? \$\endgroup\$ Commented Jan 8 at 6:49
  • \$\begingroup\$ Sure. Or put the transformer in front (on the inverter). Why not just use an LLC converter? \$\endgroup\$ Commented Jan 8 at 7:35
  • \$\begingroup\$ @TimWilliams I was given this project to work on. I did not understand what you meant by "put the transformer in front". Could you please explain this in depth? \$\endgroup\$ Commented Jan 8 at 12:26
  • \$\begingroup\$ So that the magnetizing current flows from the inverter rather than affecting the LC circuit. Perhaps you should add a schematic for discussion? \$\endgroup\$ Commented Jan 8 at 12:29

2 Answers 2


From the point of view of your source, adding an inductor in series with the transformer is fine, since it will see the inductance of the transformer (and downstream circuit) summed up with the inductor.

Regarding your FEMM calculations, I would recommend you to make sure your inductance calculation is properly done. For a transformer, there is an important difference between leakage inductance and magnetizing inductance.

enter image description here

In the diagram below, you can see the leakage inductance split in two different inductors (L1 and L2) in the transformer model and the magnetizing inductance (Lm) (note that this model is just one among many transformer models available). The magnetizing inductance is related to the core excitation by the voltage in the primary while the leakage inductance is defined by the stray field between two transformer coils.

  • \$\begingroup\$ From the FEMM, I calculated the total magnetic field energy. I then used energy=0.5*L*I^2, where I is the primary current. 'L' here is the leakage inductance. Is this right? \$\endgroup\$ Commented Jan 9 at 3:03
  • \$\begingroup\$ The risk of mistake is more related to the boundary conditions. To compute the magnetizing inductance you need to input some current in the primary (the flux will concentrate in the core). Now, to compute the leakage inductance you need to input current in both windings with opposite directions (one positive, the other negative), with balanced ampere-turns (the flux will concentrate in the region between the windings). \$\endgroup\$ Commented Jan 9 at 11:26
  • 1
    \$\begingroup\$ I did exactly like that. I hope it is correct. \$\endgroup\$ Commented Jan 9 at 12:14

For a two- or three-element resonant tank having at least one L (e.g. LCC or LLC), it doesn't matter how you obtain the L.

For an LLC series resonant converter, for example, the resonant inductor (namely Lr) is the combination of the transformer's primary leakage and the external inductance (if there's any). If you calculated the required Lr as, say, 20 uH and the xfmr's primary leakage is 5 uH then you can place an external 15 uH in series.


simulate this circuit – Schematic created using CircuitLab

Or, you can design your transformer so that it has a leakage of 20 uH.


simulate this circuit


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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