OT - lightning strike and weirdness

[Tony Ennis]

We had some rough weather the other day. During a downpour, lightning hit close enough that the flash and the sound had no separation. I heard what I thought were bits of wood hitting the house. When the rain stopped, I went outside. No obvious strike. But I found that an LED bulb in the barn was dead, the security light on the barn is not working, and a phone charger in the house exploded. The overload fuse (?) was vaporized, leaving a black smear where it used to be, and blowing the cover off the charger; it caromed around the kitchen; this is what I thought was wood hitting the siding. The wife's phone, which was plugged in, was inexplicably unharmed.

The lights never flickered and nothing else was damaged as far as I can tell.

How can this be? How can I get hit hard enough to blow up the charger and kill an LED bulb 100' away, and not lose power?

[lakeside53]

maybe hit the only one live wire; your lights and stuff was all on the other

[J Tiers]

More likely the items that failed were poorly protected, and the very high speed pulse from a nearby strike was able to destroy them.

Most motion lights have a capacitor type voltage dropping system. That is extremely sensitive to line spikes.

LED lights usually are not that sensitive, but are cheap, and not generally made to withstand much overvoltage.

[johansen]

likely that particular circuit some how got enough volts from line to line to blow the charger and the led bulb.. the amps rise at the rate of 0-50,000 amps in the first 10uS and that wavelength is on the order of 3kilometers. this is only 50Khz so the current should not inductively/magnetically induce voltage at 50khz in your household wiring, and the fact that the conductors in indoor wiring is usually less than half an inch apart (including all the way to the pole transformer) mean this component isn't significant.

however there is also capacitive coupling between the lightning strike and the high voltage utility line and any other conductors in the vicinity, which puts a voltage spike through the pole transformer (this is not hindered by the transformer because of the high frequency content cannot saturate the core)


there is also the voltage created in/at the ground due to the 50,000 amps which puts voltage on your neutral/ground (bonded at the house) this current then flows through the utility drop up to the transformer (because the transformer is also grounded) and puts voltage across your neutral wire. (which is of a 50Khz frequency or so, and so that voltage is added on top of the 120vac 60hz) so you then get whatever voltage across your neutral added to your 120vac directly) but this voltage is entirely the difference in voltage between the ground rod at your house and the ground rod at the pole transformer.

if you figure your ground rod was 10 ohms then you can have the ground potential rise to 500 thousand volts. (not sure if it ever gets that high but 10 ohms for a ground rod is within the realm of possibility. *

the fact that the phone was not damaged only means there was not enough line to ground voltage at that circuit to arc through the charger and from the phone to something else.. the charger itself is nearly 100% effective at preventing a line to line voltage spike at the 120v side from pushing more than 5 volts through to the dc side, because the arc flash inside the charger is equally effective at connecting the 120vac side to the 5v side as it is shorting out the 5v side to itself.

*this is why lighting can kill a few hundred animals in a field. there is literally thousands of volts per meter on the surface of the ground, that current flows through the animals, killing them.

[wdtom44]

Lightning is weird stuff, you can never tell how it is going to work.

[J Tiers]

Ground rods are accepted if they are less than 25 ohms resistance to the general earth, as measured.

Most any 60 Hz transformer will not be very efficient at the 50 or 100kHz area. But there is coupling through the capacitance of the two windings, which is very efficient at high frequencies, and is not particularly dependent on the step down ratio of the transformer. The spike voltage goes right through that.

The wiring is generally considered to be a lossy transmission line. There is attenuation dependent on frequency.

You likely had a line to line voltage that killed the charger, and that may have been attenuated by the act of killing the charger. The LED bulb may have performed the same service out in the shed.

The weakest component protected the reكt.

[ed_h]

In a previous life, I was a design engineer for an outfit that made remote meter readers. Our system worked over the phone lines, and we had a device that interfaced to the phone line pair. Part of my job was to analyze field failures. I often saw a failed device with obvious lightning damage. More than once I saw burnt tracks on the PCB that led to a vaporized transformer. Sometimes, a couple of small diodes on the way were undamaged.

Lighting pretty much does what it wants. We can make general statements about it's behaviour, but trying to predict or explain specific instances is usually not easy.

Ed

[AD5MB]

The overload fuse (?) was vaporized

possibly an MOV, Metal Oxide Varistor, designed to be a surge protector. The lightning may have hit a power line away from your house

I once lost a $12 video camera, a $500 monitor, a printer and one component in a computer power supply due to a lightning strike. 3 other computers, 23 other video cameras and a dozen smaller less expensive monitors still work fine.

[J Tiers]

We do tend to say that "lightning does what it wants".

It does seem that way, but the effect of lightning is due to the induced voltages and currents, or current directly inserted by a strike. Those normally involve very "steep" wavefronts , involving high frequency and high voltages/currents. The apparent random damage is due to the way those high frequency voltages and currents behave according to natural laws. The analysis of the circuit, if that were possible, and if the voltages etc were known, would show the same results that are seen, but in general, there are far too many unknowns to do any sort of meaningful analysis.

As a rule, you can assume that there is almost no practical means of protecting a small product from the effects of a direct or nearby strike. A discharge that has already traveled a mile or more through the air is simply not going to be affected by any small device, and anything small cannot withstand the power contained in the strike. There are ways to protect buildings and the power lines etc that enter the buildings from even direct strikes, but they are not cheap.

After traveling a way on power lines, the lightning "signal" ends up being modified. There are several accepted test waveforms that are used to assess the vulnerability of a device to lightning surges on the power line or on data lines. It is entirely possible to protect even relatively small devices from these waveforms. It may not be done, in the case of a low cost LED bulb, but it is possible. And, to the extent that the test waveforms emulate actual lightning effects, it is then possible to show protection from more distant strikes, as well as from other transients that may occur on a power line from the operation of contactors, and other switchgear.

The typical power outlet will arc over at several thousand volts, which puts an upper bound on the voltage that is likely to be applied to anything plugged-in.

However, not all damage may be due to the actual voltage and current from the strike. In some cases, the strike voltage (or the resulting voltage on the power lines) starts an arc, but it may be that the damage is really the result of "follow current", current from the normal power on the wires, which continues the arc that was started by the high voltage from the strike. That depends on the high voltage ionizing the air, and/or possibly creating a "carbon track" on the surface of a PC board or other surface. Then the regular power voltage may continue to put current through that path, and add to the damage. The power voltage must be high enough to be above the "arc extinguishing voltage" for the distance the arc is through air, but that can be a fair distance, as anyone who has done welding will understand.

The effects of "follow current" probably account for much of the inexplicable damage. For instance, the initial "spike" voltage and current may degrade an MOV so that its withstand voltage is less than the power voltage, even though the device is not otherwise visibly damaged. Then the power line continues to cause a current through the MOV, and may burn it up, or cause it to vaporize, depending on the rating of the MOV, and the available current. Circuit breakers may not be tripped by the current, since it does not really take much to heat up and even vaporize a fairly small part.

[Jon Heron]

Lighting pretty much does what it wants. We can make general statements about it's behaviour, but trying to predict or explain specific instances is usually not easy.

That right there covers it. I have been involved in many lightning strikes and you never know what it is going to do, it does not just take the path of least resistance, it takes all paths available to it.
I have seen where a microwave exploded but the brand new big screen TV 15' away was untouched. This bizarre stuff happens all the time with lightning.
Cheers,
Jon

[J Tiers]

Oh, it's following the rules...... The problem is that you do not KNOW the rules for the actual event, mostly because you don't know what voltage appeared where, in what order. And you never will.

That's why it seems so counter-intuitive.

It's just about impossible to figure out what will happen because there are too many unknowns. After it happens, there is little point in trying to figure out why it happened.

Lightning always DOES find the best path for it. That's why it is jagged in its path through the air.... small differences in the air may send it one way, the other way, or both ways. The same with its path through wiring.

[Arcane]

Years ago when I was working on an electrical utility crew (Journeyman Lineman) and we were crimping on permanent 4/0 copper risers from an underground three phase line to an overhead three phase line about half a mile long. Two of us were doing the work "rubber glove method" while we wore 40Kv rated rubber gloves while standing in individual 50Kv rated insulated buckets at the end of a fiberglass boom that was rated for IIRC 100 KV. There was a storm on the horizon and as we were crimping up the last connection a lightening strike happened less than half a mile away. It made the both of us tense up completely. It felt as if every almost muscle in our bodies contracted and hard! The DC pulse must have been felt by the nerves in our bodies. Thankfully it never stopped our hearts.

[Jon Heron]

Arcane, that is some serious pucker factor, without the lightning!
Glad your still with us!
Cheers,
Jon

[darryl]

I think JT is on to something with the follow current idea. If the actual current in the lightning strike were to pass through a device, or a house electrical system for that matter, there would most likely be lots of damage. But a high voltage, high speed pulse can induce a high voltage across something without needing a direct connection to it. It's entirely possible, and likely in my opinion, that much damage is caused by line current that is flowing because the insulation has been broken down by the high voltage pulse. The induced current doesn't have to be very high at all- and in fact doesn't flow until a conductive path has been established. A voltage spike causing a conductive path to be established could then have a current flow component, or it could just be the catalyst that breaks down the insulative ability in wiring or components, allowing line current to flow. This would most likely create 'pockets' of damage instead of having everything blown to bits by the strike itself.

When you consider the many instances where lightning appears to have 'bypassed' some electronic things, yet blows apart random things or things that aren't conductors, you have to wonder what the 'rules' are. You probably have to consider yourself lucky if lightning hits a lightning rod instead of snaking down between building and blowing up a laundry basket instead. At the high speed of electricity and the high rate of voltage development (rise time), being a conductor doesn't mean much. Lightning doesn't have to follow a conductor- it is following the rules but we aren't well enough versed in the rules to fully understand what's going on.

[Jim Stewart]

I think JT is on to something with the follow current idea. If the actual current in the lightning strike were to pass through a device, or a house electrical system for that matter, there would most likely be lots of damage. But a high voltage, high speed pulse can induce a high voltage across something without needing a direct connection to it.
(snip)
At the high speed of electricity and the high rate of voltage development (rise time), being a conductor doesn't mean much. Lightning doesn't have to follow a conductor- it is following the rules but we aren't well enough versed in the rules to fully understand what's going on.

Yes.

And earlier posts about risetime of the pulse are much underestimated. The peak risetime (second derivitave of V/t IIRC) is more like 0.3 microseconds, roughly meaning that the equivalent frequency is more like 30 Mhz, with (very high energy) harmonics likely to be in play at something like 100 Mhz. Here's what IEEE has to say about it:

http://ieeexplore.ieee.org/abstract/document/6344272/?reload=true

A long time ago, when we still used stone axes to solve most problems, I worked at Tektronix. One of the very exotic oscilloscopes in my group was the 519, with a vertical bandwidth of 1.9 GHz. In 1960! It had no vertical amplification at all, just a long and beefy delay line, but a very fast timebase (for 1960). Its purpose was tracking transients on HV power lines, particularly from lightning strikes.

-js