I am considering building an AC protection circuit using the following schematics. I am a physicist and have little practical experience with implementation of electric circuits. Looking forward to your input on it.

  1. The 5A fuse is obviously for overload protection. (Even 1A may be acceptable. I put 5A for transients.)
  2. The ten 33V Zener diodes will not trigger till 330V (plus some forward voltage) which is higher than the AC peak. However, if due to some reason the AC voltage spikes the Zeners will trigger and blow up the fuse and ensure voltage remains the within the safe limit at output.
  3. The capacitors are there to filter DC from AC. (A power transformer and some other things up in the line.) I added some high resistance bleeder resistors. However their necessity seems to be debated on this forum.

My reason for using Zeners instead of varistors is that I will only need to reset the fuse in case a fault occurs. I am concerned if the Zeners will be able to handle the surge, and I haven't seen such implementation anywhere giving rise to lack of confidence. Maybe I am missing something? This is a hobby DIY project, so keeping cost down for mass production is not a priority. Please give your input on the circuit before I put it to test.

Circuit diagram

  • 2
    \$\begingroup\$ There is a good possibility that the Zeners will fail before the fuse blows. \$\endgroup\$ Commented Oct 21, 2021 at 21:35
  • \$\begingroup\$ That's my main concern too. The surge current is rated at .15A. The lower voltage zener diodes can handle more, but I will need a LOT more diodes then. \$\endgroup\$
    – Likhon
    Commented Oct 21, 2021 at 21:44
  • 1
    \$\begingroup\$ Which zener diodes would those be? They only have 33V over them when rated current of some specified amount flows through them. Why there are electrolytic capacitors in series? What is the purpose of this circuit? If there happens a short circuit on the output, the first thing to blow up would be the cap with 150V over it in reverse. \$\endgroup\$
    – Justme
    Commented Oct 21, 2021 at 21:49
  • \$\begingroup\$ There are no part numbers given, but Zeners are usually limited to 10s of mA. This circuit has no current limiting for the Zeners, so they'll burn up real fast, way before they conduct 5A. \$\endgroup\$
    – Aaron
    Commented Oct 21, 2021 at 21:50
  • 1
    \$\begingroup\$ That crowbar is designed for DC. I'm sure it can be made to work on AC. \$\endgroup\$
    – Aaron
    Commented Oct 21, 2021 at 21:55

2 Answers 2


Generally zeners are not for high current suppression (unless you find one with a datasheet that says otherwise). Typically TVS (transient voltage suppression) diodes are used for this.

Why don't you use a MOV (metal oxide varistor) like just about every other surge protector uses? You have to mount a thermal fuse next to and touching the MOV to open up the circuit to prevent fires caused by the MOV. I don't think you understand varistors -- they don't destroy themselves when suppressing a small surge, at least not right away. Zeners would be even worse.

If you do go the route of using diodes, a bridge rectifier with the output going to the voltage clamping diodes may be an easier method. If you want to replace the fuse each time a fault occurs, I suggest using a zener diode to trigger a thyristor (SCR or triac). A diac and some capacitors might also be needed to trigger the thyristor properly. There are other things that can be done, such as using an SCR to trigger a relay to disconnect the circuit.

edit: The thyristor method may be necessary because if there isn't enough Voltage to exceed the 5 Amps and blow the fuse, a long lasting mild surge will blow out just about any surge suppression device. Surge suppression devices are meant for small fractions of a second use. 5 Amps at 240VAC is over 1000 Watts. You would need over 1000 1 Watt zener diodes!


Please note that 220 V was used 20 years ago.

In Europe the standardized mains voltage is 230 V.

Further: 230 V is the RMS value, therefore the actual voltage that feeds your circuit is:

v(t) = 1.414 * 230 * sin(6.28 * f * t)

You may use an inductor L of 10 to 200 micro Henry in series to the fuse.

L will act as a resistor for transient events and will stand any voltage because it's a wire.

Forget Zener diodes for mains protection. All the world uses varistors.

Take a look at NXP's AC/DC converters reference designs.

They always test the circuit for EMI performance and you may see what kind of components they use for EFT transient tests as well as for Surge tests.


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.