I am designing a SEPIC power supply using the LT3758. I have simulated the circuit in LTSpice and all the waveforms are as expected except the L1 current. Here I see massive overshoots when the MOSFET switches on (eg normal current is about 3A pk-pk and overshoot goes up to over 20A). The spike is extremely narrow (< 1nS).

I am new to LTSpice, so might be missing something obvious.

Is this spike real or a simulator aboration? What could be causing it and what can I do about it?


SEPIC schematic

Input +28V, Output +40V 1A.

[Edit] I have posted a follow on question here LTSpice SEPIC design low freq ringing

  • 1
    \$\begingroup\$ Are you using realistic values for parasitics? \$\endgroup\$ – PlasmaHH Jan 1 '16 at 10:59
  • 2
    \$\begingroup\$ If you put 0.01 ohms in series with the inductor what happens. \$\endgroup\$ – Andy aka Jan 1 '16 at 11:21
  • \$\begingroup\$ I see you are using coupled inductors; if you do not use a realistic model (preferably from the manufacturer), your results can contain 'interesting' artefacts. I have had this issue and it mostly disappeared when I used the manufacturers model. \$\endgroup\$ – Peter Smith Jan 1 '16 at 16:53
  • \$\begingroup\$ The V1 source also has 0 resistance. Add the expected ESR of your power source to that. \$\endgroup\$ – rdtsc Jan 1 '16 at 17:35
  • \$\begingroup\$ And why V=50 on C5? \$\endgroup\$ – rdtsc Jan 1 '16 at 17:37

The only place this can come from is stored charge in D1 (or gate drive on Q1, but that's not the case).

Your currents look non-linear with time -- likely your circuit is not very efficient.

  • \$\begingroup\$ LTSpice claims the circuit is 88% efficient. I had some suspicions about the diode specs but changing the inductor coupling seems to have sorted that problem - see comment above. \$\endgroup\$ – David G Jan 2 '16 at 0:14

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.