0
\$\begingroup\$

Our condos' rooftops don't have lightning rods per se, but they do have satellite dishes and AC units mounted there as well. Would these, if properly grounded (and these should be) act as lightning rod "hacks?" This question just hit me like, well, a bolt of lightning !!

\$\endgroup\$
  • \$\begingroup\$ Don't use parabolic dishes to disperse the energy from a lightning strike. The giant spark would fry the receiver module, by striking nearby. Place a real lightning rod on the roof. \$\endgroup\$ – user2497 Sep 14 '17 at 12:43
  • \$\begingroup\$ True, but those receivers are no more expensive and hassle than replacing the lightning rod thingy that needs to be replaced after such a strike. Or a penetrated roof for that matter. Am I right? \$\endgroup\$ – Tim Spriggs Sep 14 '17 at 12:51
  • \$\begingroup\$ Sure, it will handle the strike fine. Just make sure the ground wires won't melt. Ask an electrician for the types you need. You need either thick rods, or thick filament wire with low resistance. \$\endgroup\$ – user2497 Sep 14 '17 at 12:54
  • 1
    \$\begingroup\$ Of course, it is cheap to replace receivers, but having lightnings strike dishes and getting conducted into the receivers means that they are not doing the job as lightning rods: keeping the lightning out of the house and away from any person. A receiver can be touched, or if a person is standing close by the lightning can jump over to that person and potentially kill her. Never ever use roof top equipment without proper lightning arrestors and safety procedures. Unlike equipment, human life cannot be replaced. \$\endgroup\$ – Attila Kinali Sep 14 '17 at 14:25
  • 1
    \$\begingroup\$ The satellite receivers might be not the only electronic devices destroyed in your house. The lightning strike may pass from the receivers to the power grid and destroy other electronic devices in your house and also in your neighbors houses too. A fire in the house caused by the strike is not impossible. \$\endgroup\$ – Uwe Sep 14 '17 at 15:29
1
\$\begingroup\$

Lighting is a lot of energy dumped in a small amount of time. The grounding wire used (most likely from the existing power network) is most likely not rated to handle that kind of surge.

Proper lighting rods and their grounding wires are very generous in their current carrying capacity to minimize the energy absorbed due to internal resistance and heavy enough to make sure there is enough thermal mass to stay cool enough to not melt the wire and/or set fire to your house.

\$\endgroup\$
  • \$\begingroup\$ What about the AC units though? They use thick wires for the grounds and legs. Won't those handle the strike? \$\endgroup\$ – Tim Spriggs Sep 14 '17 at 13:03
  • \$\begingroup\$ Here is some very pertinent information on just this topic...ecmag.com/section/your-business/… \$\endgroup\$ – Tim Spriggs Sep 14 '17 at 13:54
  • 1
    \$\begingroup\$ @TimSpriggs no, AC units use tiny little 30amp cables, wimpy things which melt or perhaps vaporize. Instead you need a copper wire about 1cm thick. Also, if the ground-wire isn't straight down vertical, it can launch a lightning-leader sideways from any of its curved parts. Also, if lightning gets into your AC outlets, a major stroke could destroy every piece of electronics in the building (since even the green ground suddenly develops many kilovolts with microsecond rise time.) Large lightning strikes can turn a large breaker-box into flying shrapnel. \$\endgroup\$ – wbeaty Sep 14 '17 at 17:21
  • \$\begingroup\$ I just googled some examples of homes getting struck and setting them afire. I think most of us, at least me, underestimated what it could do. Thanks for yours, and others, input. \$\endgroup\$ – Tim Spriggs Sep 15 '17 at 20:36
1
\$\begingroup\$

The magnetic field of a lighting strike may trash any "loops" nearby, with hundreds of volts induced into that loop.

For example, assuming the strike has 50,000 amps with risetime of 1 microsecond. Run your grounded rod 1 meter away from the A/C control board, with no intervening magnetic shielding (no steel plates or cases). Let the A/C control board have 4" by 4" (0.1 meter) loops on the PCB foil. How much voltage gets induced into that loop, given a worst-case 3_D topology? Use the formula

Vinduce = [MU0 * MUr * Area / (2 * pi * Distance)] * dI/dT

which simplifies (with MU0 being 4 * pi * 10^-7 Henries/meter, MUr = 1) to

Vinduce = 2e-7 * Area/Distance * dI/dT

Substituting, we find

Vinduce = 2e-7H/m * 0.1meter*0.1meter/1meter * 50,000amp/uSecond

Vinduce = 2e-7 * 0.01 * 50 GegaVolt/second = 2e-9 * 50 e+9 = 100 volts induced into the ground-foil loop of your A/C control PCB.

Thus huge currents flow, and the PCB may get vaporized.

Summary: keep the GND rods away from electronics.

\$\endgroup\$

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.