I would like to simulate a circuit with a resistor and zener diode for various voltages in an excel spread sheet and need a way to model it.

  • \$\begingroup\$ Watch out for tempco, the vz rise due to a rise in temp due to increase in current is typ worse than esr \$\endgroup\$
    – sstobbe
    May 20, 2017 at 15:23

1 Answer 1


Vf= Vz + Zzt*If

Background of my research

  • where Zzt ~= 1/Pd rating of part +/- 25% and depends on heatsink design at 85'C since Schockley effects reduces Vz with rising temp. Vz depends on assumptions for test current, and I typically use 5~10% rated current for Vz and Iz at the onset on saturation. Since ESR is the incremental R from bulk resistance, this is related to the part Rja , Pd rating and heatsink Rca ['C/W]. Thus the junction temp rise is a product Rja * W =['C/W /[Ω] ] at rated power for Pd = (ESR * If+Vz) * If.

  • you can use this model for all diodes including LEDs , where the best thermally designed parts have ESR or Zzt=k/(Pd) with k=0.5 while most are k=1.

    • e.g. 3W LED looks like a zener with Vz=2.8V and ESR =1/3 ohms, then if in a string, ESR' = xS/yP for an array of xSyP White LEDs each with k/Pd = ESR per cell.
  • My model assumes Tj=80'C max. which I have computed on many spreadsheets for many different diodes and zeners, where ESR drops slowly with rising current /near&above\ rated current and temperature for pulse mode.

I have jokingly in the past here called this Stewart's Theorem or Stewart's law for ESR on semiconductors but it is a good model for even BJT power switches in saturation and CMOS drivers and ESD diode protection inside CMOS.


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.