I've been researching on whole house surge protection devices and came across an Illustration of lightning current flow from a direct strike near the (protected) structure.

From my understanding Type 2 residential SPDs divert surges from L-N, L-PE and N-PE. But the above illustration shows a Type 2 surge I never considered before, where inductance from a nearby lightning strike to the ground causes surge in the Ground (PE) wire. Can a normal type 2 SPD protect from this type of surge event?

P.S. I have a type 3 SPD (to protect terminal equipment) which I currently use and a whole house type 2 SPD which I haven't installed. It is my understanding that both essentially divert surges to Ground (PE) but I honestly have no idea what they'd do when it is the ground wire that causes the surge.

  • \$\begingroup\$ No, but the foundation ground would distribute the house to the equal potential. Also the PE wires are connected to a single point, so there should be no such potential difference. \$\endgroup\$ Jun 14, 2019 at 11:05
  • \$\begingroup\$ @MarkoBuršič Which path would the induced ground (PE) surge current take? I'm afraid of a scenario where it might cause a surge current to form in the neutral wire (as the neutral is bonded to ground at distribution) and in the PE wire connected to appliance's metal enclosures. Would the latter cause the infamous arc/plasma flashover (high V, hopefully low I) that some people experience during thunderstorms? \$\endgroup\$ Jun 14, 2019 at 11:22

2 Answers 2


The diagram you link is ineffective - an indirect lightning strike bounces the local earth and, the whole earth system of the building follows that bounce (or surge). Usually (and provided there is only one earth point), there is no damage to building electrical infrastructure or appliances. This is because they all bounce together i.e. they remain at equipotential even if during that surge they might have been bounced to a kV or so.

The problems come when the building electrical infrastructure extends away from the building such as AC wiring and telephone wires because, if the far-ends are "earthy" those far-ends are not necessarily "bounced" by the same surge. The result can be tens or hundreds of amps flowing through that extended electrical infrastructure. That current now flows through the building electrical infrastructure and can lead to high levels of non-equipotentials and that can damage local equipment.

Surge protection devices can divert surges currents to earth and prevent common-mode AC surges damaging (say) computer power supplies. These SPDs are needed when the equipment cannot handle several thousands of volts to earth. An alternative is to design the equipment to handle that common-mode surge - more modern ethernet is a good example - generally the isolating magnetics are good for 6 kV surges.

Surge protection devices are also used to prevent differential effects damaging equipment. This is usually because a common-mode surge (as it normally begins) can produce differential artefacts that may rise to several hundred volts. These differential artefacts arise because two wires in a cable may not propagate the common-mode surge identically and therefore a differential surge is created.

  • \$\begingroup\$ I’m unclear why you say that diagram is ineffective. I would estimate most lightning damage is caused by indirect strikes and indeed may come in via the ground wiring. The problem is that the ‘whole earth’ doesn’t bounce, rather there is a voltage gradient caused by earth resistance. \$\endgroup\$ Jun 14, 2019 at 12:02
  • \$\begingroup\$ @ErikFriesen I said "the whole earth system of the building follows that bounce". Yes, there is a voltage gradient in the earth due to earth resistance but that isn't the "whole earth system of the building". Maybe I'm misinterpreting what your comment is about? \$\endgroup\$
    – Andy aka
    Jun 14, 2019 at 12:05
  • \$\begingroup\$ @Andyaka Thanks for the answer. I would be grateful if you could answer the below doubts: 1. Would a small house be unaffected since there might be no voltage gradient? 2. Could an SPD in common mode deal with such surge currents (ground originated) effectively as its normal current diverting path is now responsible for the surge. 3. How would the SPD and an appliance react to a line voltage of 230V and earth/neutral potential of 1 kV (assuming there's a ground surge)? Might be related to [2]. \$\endgroup\$ Jun 14, 2019 at 13:01
  • 1
    \$\begingroup\$ 1) there would be a small gradient but it depends on whether there is a large surge current due to external earths on cables coming into the house. There needs to be diagrams about the various other scenarios and those diagrams may raise queries and the diagrams might need to evolve before a decent answer can be given to your other points. I would suggest looking at EN 61000-4-5 and I’ve seen unofficial versions on the web in the last month in order to understand surge testing because tests emulate reality. \$\endgroup\$
    – Andy aka
    Jun 14, 2019 at 13:11
  • \$\begingroup\$ @Andyaka The diagram is from this document, there are several others in there as well: www-public.tnb.com/eel/docs/furse/ESP_-_Introduction.pdf \$\endgroup\$ Jun 15, 2019 at 18:53

Metal oxide varistor (MOV) based surge protection devices (SPD) which include N-PE protection would also protect from PE-N (reverse direction, induced current from ground lightning strike) surges since varistors function similarly in both directions.

A varistor is an electronic component with an electrical resistance that varies with the applied voltage. Also known as a voltage-dependent resistor (VDR), it has a nonlinear, non-ohmic current–voltage characteristic that is similar to that of a diode. In contrast to a diode however, it has the same characteristic for both directions of traversing current.



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.