Lightning Damage Anyone?

Hang out, sip some ice tea, and shoot the breeze with TR regulars.

Moderators: emkubed, Captain Ned

Lightning Damage Anyone?

Postposted on Sat Aug 04, 2012 9:18 pm

I'm sure everyone's had something get zapped before, any one want to share?

We just had a strike 50 feet from the house into a tree. There doesn't seem to be any major damage to the tree trunk, but there's an 8 inch deep trench running about 10 feet along a major root. I had just unplugged all our computers less than a minute before it hit, but I'll have to rethink this strategy as it didn't stop the magic smoke...

We've got a wireless router upstairs for the kitchen computer but connect the two computers upstairs to the wired gigabit ports. I followed normal procedure and unplugged all the devices and the DSL line but left the computers connected to the router as usual. Thirty seconds later there was an orange fireball in the front yard as the house shook from the thunder. Fast forward an hour after dinner and the router won't power up. DVM shows voltage leaving the power supply while the bottom of the router gets very hot where a chip inside has become shorted I suppose. Hook up the backup router and get my computer up with no trouble. Check my brother's and no network adapter is recognized. Smoked. :x

The only thing I can figure is that the 100 feet of ethernet cable picked up some induced current from the very close lightning bolt and fried the stuff. I guess I'll be unplugging those cables as well from now on. In addition to the computer equipment we lost a wireless phone and the breaker for our well pump tripped. The trench points directly to our well which is about 20 feet from the end. We turned it back on and it appears to be working normally, but we'll be watching it closely.

The main trench is about 8 feet long. There are chunks of clay 20 feet from here. Don't mess with Mother Nature.
Image
bobboobles
Gerbil Team Leader
 
Posts: 298
Joined: Tue Oct 27, 2009 4:46 pm

Re: Lightning Damage Anyone?

Postposted on Sat Aug 04, 2012 9:37 pm

That's pretty cool!
Intel Core i7-875K, Asus P7P55D-E Pro, Win 7 Home Premium
MSI GTX 560 Ti OC, Mushkin 2x2GB DDR3-1333, Corsair TX650
Cooler Master Hyper 212+, Logitech Z-2300, ASUS Xonar DX
Samsung Spinpoint F3 1TB, Dell Ultrasharp U2410, Antec P183
DeadOfKnight
Gerbil Elite
Gold subscriber
 
 
Posts: 634
Joined: Tue May 18, 2010 12:20 pm

Re: Lightning Damage Anyone?

Postposted on Sat Aug 04, 2012 9:46 pm

You're completely right about the induced current. There simply isn't much you can do about a strike that close.
"We demand rigidly defined values of doubt and uncertainty!"
CasbahBoy
Gerbil
 
Posts: 28
Joined: Fri Jul 18, 2003 4:30 pm
Location: Cleveland, OH, USA

Re: Lightning Damage Anyone?

Postposted on Sat Aug 04, 2012 10:13 pm

My mailbox saved the house from a lightning strike a few years ago. Thing was masonry with a cinder-block core and chicken wire + mortar holding the two together; it must have been that chicken wire.

At any rate, one day I was reading in the bedroom during a thunderstorm and KRAPOW. One breaker in the house was thrown vs. a huge divot in the mailbox and broken masonry all over the front yard.
Think for yourself, schmuck!
i5-2500K@4.3|Asus P8P67-LE|8GB DDR3-1600|Powercolor R7850 2G|1.5TB 7200.11|1988 Model M|Saitek X-45 & P880|Logitech MX 518|Dell 2209WA|Sennheiser PC151|Asus Xonar DX
bthylafh
Grand Gerbil Poohbah
 
Posts: 3068
Joined: Mon Dec 29, 2003 10:55 pm
Location: Southwest Missouri, USA

Re: Lightning Damage Anyone?

Postposted on Sat Aug 04, 2012 11:05 pm

bobboobles wrote: Don't mess with Mother Nature.

Yeap. I've seen some damage done from a near-direct lightning strikes, including damaged home appliances and other electric equipment. No photos, though :( Also, I (barely) remember seeing a small "ball lightning" when I was a child - don't remember much except that for some reason I felt very scared (that **** got inside the house) and that it left some burn marks on some furniture when it exploded :-? My grandmother was with me at that time, and she also remember seeing that.
My subscription allows you people to exist on this site and makes me a better human being than you'll ever be
JohnC
Gerbil Jedi
Gold subscriber
 
 
Posts: 1741
Joined: Fri Jan 28, 2011 1:08 pm
Location: NY/NJ/FL

Re: Lightning Damage Anyone?

Postposted on Sat Aug 04, 2012 11:20 pm

Sounds like you had a (comparatively rare) positive strike, to dig up the root like that. But yes, induced potential in building wiring is quite common for nearby strikes. Could have been worse, though: about fifteen years ago, there was a strong positive strike that took out a tree near Ft. Collins, Colorado and dug up the roots in a similar fashion to what you saw, but it induced a high enough potential in the adjacent house that it blew up all of the electrical wiring. There were pictures in the paper showing ray gun blast patterns around every single light figure and coverplate in the structure.

bthylafh wrote:My mailbox saved the house from a lightning strike a few years ago. Thing was masonry with a cinder-block core and chicken wire + mortar holding the two together; it must have been that chicken wire.

Nah, the mailbox was just unlucky: it was the first thing that the leader got close to on the ground, and then whammo, the return stroke took it out. A lightning bolt that can ionize an air channel for a mile or more, does not care either way about a little bit of chicken wire. Also, inquiring minds want to know just what kind of mail you actually receive in that mailbox :P

For electrical substation design, we estimate that every lightning mast or static wire placed in the air will produce approximately a 30-degree "cone of protection" extending down from its highest point. This is a bit conservative but also a bit of a guess, because lightning is highly unpredictable, and there is no exact way to study it. </dayjob>
He who laughs last, laughs first next time.
ludi
Gerbil Elder
 
Posts: 5320
Joined: Fri Jun 21, 2002 9:47 pm
Location: Sunny Colorado front range

Re: Lightning Damage Anyone?

Postposted on Sun Aug 05, 2012 1:28 am

CasbahBoy wrote:You're completely right about the induced current. There simply isn't much you can do about a strike that close.

Unplugging any long cables will help though. I always unplug both power and network, but then again I'm too lazy to unplug the cable from my TV..
Firestarter
Gerbil XP
 
Posts: 468
Joined: Sun Apr 25, 2004 10:12 am

Re: Lightning Damage Anyone?

Postposted on Sun Aug 05, 2012 7:59 am

Ethernet is probable the most vulnerable for induced voltages since it typically unshielded twisted pairs. It depends on common mode rejection for transmission of signals, but can only tolerate 2kV before a breakdown. A strike close like that may be 20kV or more induction in anything over 2' long that does not have a means to dissipate the energy safely. Those twisted pairs? Typically terminated through the transformer of the NIC on a center point into an R/C network, DC isolated in cases with a poor earth ground in the system.

An FYI, my lightning detector consists of a 2' long antenna with a 5M ohm termination with resonant tank LC. This will "see" about 3V pk/pk signals from strikes half way across the country with NO amplification. 50' away? That is why I have a transorb on it. I saw one while I had a scope hooked up to it that was over 100V. It was 50 miles away or so.
liquidsquid
Minister of Gerbil Affairs
 
Posts: 2437
Joined: Wed May 29, 2002 9:49 am
Location: New York

Re: Lightning Damage Anyone?

Postposted on Sun Aug 05, 2012 9:30 am

liquidsquid wrote:An FYI, my lightning detector consists of a 2' long antenna with a 5M ohm termination with resonant tank LC. This will "see" about 3V pk/pk signals from strikes half way across the country with NO amplification. 50' away? That is why I have a transorb on it. I saw one while I had a scope hooked up to it that was over 100V. It was 50 miles away or so.

You might want to repeat that in English, if your goal is to enlighten us poor saps with no education in these matters :D
Firestarter
Gerbil XP
 
Posts: 468
Joined: Sun Apr 25, 2004 10:12 am

Re: Lightning Damage Anyone?

Postposted on Sun Aug 05, 2012 10:06 am

OT: I saw the topic and thought "Huh... Damage must have become a superhero somehow. This'll be a good read."

Slightly disappointed in one way, but this was cool in another.
Sony a7
Sony Zeiss 55/1.8 SSM, 24-70/4 SSM
Minolta 17-35/2.8-4 D, 35-105/3.5-4.5, 100-300 APO
TheEmrys
Minister of Gerbil Affairs
 
Posts: 2067
Joined: Wed May 29, 2002 7:22 pm
Location: Northern Colorado

Re: Lightning Damage Anyone?

Postposted on Sun Aug 05, 2012 5:03 pm

Firestarter wrote:
liquidsquid wrote:An FYI, my lightning detector consists of a 2' long antenna with a 5M ohm termination with resonant tank LC. This will "see" about 3V pk/pk signals from strikes half way across the country with NO amplification. 50' away? That is why I have a transorb on it. I saw one while I had a scope hooked up to it that was over 100V. It was 50 miles away or so.

You might want to repeat that in English, if your goal is to enlighten us poor saps with no education in these matters :D

On his 2 foot long antenna setup, lightning 50 miles away gives about the same voltage as plugging it directly in to the wall socket!!! I'm keeping well clear of that.
notfred
Grand Gerbil Poohbah
 
Posts: 3647
Joined: Tue Aug 10, 2004 9:10 am
Location: Ottawa, Canada

Re: Lightning Damage Anyone?

Postposted on Sun Aug 05, 2012 5:08 pm

notfred wrote:On his 2 foot long antenna setup, lightning 50 miles away gives about the same voltage as plugging it directly in to the wall socket!!! I'm keeping well clear of that.

On top of a 5 megaohm resistor to boot.
There are people that embrace the Oxford comma and people who don't. Never get between these people when drink has been taken. I use the Oxford comma and always will. The rest can sod off.
Captain Ned
Global Moderator
Gold subscriber
 
 
Posts: 19736
Joined: Wed Jan 16, 2002 6:00 pm
Location: Vermont, USA

Re: Lightning Damage Anyone?

Postposted on Sun Aug 05, 2012 8:58 pm

Yeah I know how it feels. The transformer across from my place got hit....we thought a bomb had gone off!! Luckily no major damage. :)

James
salazarj
Gerbil In Training
 
Posts: 1
Joined: Sun Aug 05, 2012 8:49 pm

Re: Lightning Damage Anyone?

Postposted on Sun Aug 05, 2012 9:35 pm

Image
metalhead
Gerbil Team Leader
 
Posts: 248
Joined: Mon Dec 29, 2003 4:57 pm
Location: St.Louis

Re: Lightning Damage Anyone?

Postposted on Sun Aug 05, 2012 10:49 pm

Unless you plan to go into full-on-OCD-paranoia mode, unplugging all power and I/O cables into every device every time there is a thunderstorm, I doubt there's much you can do about induced voltages from a near-direct hit like that.

Firestarter wrote:
liquidsquid wrote:An FYI, my lightning detector consists of a 2' long antenna with a 5M ohm termination with resonant tank LC. This will "see" about 3V pk/pk signals from strikes half way across the country with NO amplification. 50' away? That is why I have a transorb on it. I saw one while I had a scope hooked up to it that was over 100V. It was 50 miles away or so.

You might want to repeat that in English, if your goal is to enlighten us poor saps with no education in these matters :D

My translation: Even a short length of wire can develop significant induced voltages from lightning strikes which occur miles away. Trying to protect your sensitive electronics from the induced voltages which would result from a direct hit on your property is probably an exercise in futility.
(this space intentionally left blank)
just brew it!
Administrator
Gold subscriber
 
 
Posts: 36886
Joined: Tue Aug 20, 2002 9:51 pm
Location: Somewhere, having a beer

Re: Lightning Damage Anyone?

Postposted on Mon Aug 06, 2012 6:44 am

Ah yes, typing up a post with only a few hours of sleep leads to that.

While the voltage may be high, typically the current is very low on my antenna. Put your finger on it and the voltage would never get above a few milivolts.

Most antenna systems have ~50 ohm terminations, partly for easy realization of circuit geometries, but also to dissipate atmospheric energy before it damages something. On top of that, a transorb (a reverse-diode which has an abrupt breakdown voltage, like a zener diode only specialized) is placed across the center antenna conductor. Its intention is to absorb and dissipate the excess energy instead of blowing sensitive circuits. The transorb is the sacrificial component. I have no idea why there are not typically used on Ethernet cards, but they are not. Probably to sell more NICs.

This is why I have a wireless card in my PC now, I've lost 2 PCs due to nearby strikes due to the ethernet weak link. I will let the wireless router take the hit, it is much cheaper.

You definitely were graced by a high-power positive stroke. They will typically originate on the top portion of a storm cloud and exit the main thunderhead all the way to the ground. What makes them uniquely strong is they will discharge the entire storm's charge in one shot, where negative strokes are almost always more localized in a cloud. They are triggered by a normal discharge negative stroke in the another part of the storm that abruptly collapses the storm's field in a small area, and the positive stroke occurs when this collapsed field tries to fill back in from all of the surrounding charge at once due to the imbalance, and then there is a cascade of discharge events like a house of cards. There is some debate of added energy due to sprites above the storm dumping in additional charge from the upper atmosphere. It takes a large tall storm cloud to pull these off and provide the necessary quantity of charge to cut a channel that long and sustain it.

So yeah, dealing with large cosmic powers like this and still protecting circuits only 45nm across don't go well together. Something has to give.

As a side-note, out in the ocean there is an additional scale of lighting stroke called a "superbolt" a few magnitudes greater discharge energy than even a positive stoke, and have been known to blow large holes in the hulls of ships. I suppose when you drive something between the plates of a super-huge high voltage capacitor, you can expect that to happen.
liquidsquid
Minister of Gerbil Affairs
 
Posts: 2437
Joined: Wed May 29, 2002 9:49 am
Location: New York

Re: Lightning Damage Anyone?

Postposted on Mon Aug 06, 2012 7:00 am

Wonder if your clay is the right consistency for there to be Type II fulgurites in that trench? Dig very carefully.
There are people that embrace the Oxford comma and people who don't. Never get between these people when drink has been taken. I use the Oxford comma and always will. The rest can sod off.
Captain Ned
Global Moderator
Gold subscriber
 
 
Posts: 19736
Joined: Wed Jan 16, 2002 6:00 pm
Location: Vermont, USA

Re: Lightning Damage Anyone?

Postposted on Mon Aug 06, 2012 7:05 am

I lived through a direct lightning strike on my house. Back in the CB radio's heyday my father had a base station set up with an antenna probably 15' above the highest point of the house(he used it in his business). I must have been 12 when it happened. My bedroom(where I was) was directly adjacent to the living room. The cb was in the corner opposite my room's door and a tv was diagonal across the room from the cb. BOOM!! Fried the tv and sent the front cover of the cb flying across the room!! Sorry but no pix - this was back in the 70s. It DID leave a crater around the base of the antenna outside(it was grounded too but NOthing helps a direct hit) close to 6 inches deep!!! Thankfully we were all okay other than having to avoid the steaming leftovers.... :wink:
piccolopete
Gerbil XP
 
Posts: 397
Joined: Mon Apr 21, 2003 9:27 pm
Location: Pittsburgh, Pa

Re: Lightning Damage Anyone?

Postposted on Mon Aug 06, 2012 8:20 am

I currently live in the country in the world that receives the most lightning strikes per year (Brazil). I also happen to live in Vale do Paraíba, a small region in this country that is the national champion in lightning strikes. It even has its very own lightning research institute. On top of this my house lies atop a 100 m high hill, and is the last installation (the termination, sort of) on both the power line and the phone line. There are a LOT of lightning strikes and current spikes happening around here in summer. I have never had a direct hit on my house I think, but several within 50 yards, mostly into the power lines (which are all above ground around here). It makes for a spectacular show but wreaks havoc on electronics.

I guess I have fried something like two mobos, three monitors, two routers, two wireless phones and four ADSL modems due to lightning strikes over the years. Nowadays I have installed physical switches for both electricity and the phone line going to any sensitive equipment, and as soon as I leave home or hear thunder in the distance I disconnect everything and run on 3G and my UPS (powered by three car batteries).

I hadn't thought about the possibility of a close lightning strike inducing enough current in the 2 m Ethernet cable going between my computer and the router to fry them, though. Maybe I should go wireless too? Is there any easy way of shielding this cable, like wrapping it in aluminum foil or whatever? Can you buy shielded Ethernet cables? What about USB cables, are they less susceptible to induced current for some reason? Lots of questions for the knowledgeable people in this thread! :)

And another one: The grounding in my house is through a copper rod beat into the ground just outside the house. The switches that separate the sensitive equipment from the grid switches off both current phases (I have two phases with 110V each for a 220V total) and also the ground wire, so that the equipment is totally disconnected from any external electrical wiring. My (maybe faulty) reasoning for this was that a close lightning strike could induce current or create a spike even in the ground wiring, which might fry the electronics even if the electrical phases were disconnected. Is this so? Or would I be better off leaving the ground wire connected, for some reason? I'm definitely not an expert in electrical matters, so any insight is appreciated.
KilgoreTrout
Gerbil In Training
Silver subscriber
 
 
Posts: 3
Joined: Sun Jun 21, 2009 1:37 pm

Re: Lightning Damage Anyone?

Postposted on Mon Aug 06, 2012 8:57 am

USB is much less susceptible than Ethernet since it usually has a shielded ground shell around it, and other ends are not connected to equipment from a remote power source. You can purchase wiring for Ethernet that has shielding on it, but you have to be very careful of having both ends terminated to ground, which means coming up with a means yourself. There is no specification for attaching a shielded ground to an Ethernet cable connector or even the panel of a NIC card, it is all supposed to be isolated, which is fine as long as you do not have an arc-over creating a connection. This isolation is to protect you from poor grounding issues in your house or building causing a fire or electrocution when trying to wire two things up which may not necessarily be grounded together on the same power domain. The connection scheme I can only surmise comes from telephone wiring for the same reasons.

Power over Ethernet systems are probably much less prone to this since they do have ground return paths.

To properly shield your Ethernet you would need the shielded cable, then at each end you would connect the shield to the earth ground of the case of the equipment. This is fine as long as both halves of the connection are attached to the same ground which is typical in a home. If however one ground is connected to the case of the PC which is connected to the surge isolator you speak about, and the other is connected to the grounded case of say a DSL modem which is not isolated, then you have much worse problems when one half of the ground opens. Now the entire house's ground is connected through a puny wire and tiny equipment when the isolator opens.

A good way to resolve anything about this is to go wireless or optical. Wireless results in latency issues if you are gaming (doesn't effect me much, my DSL already has latency issues), and optical is a PIA and can be expensive. However both of them couldn't get any better in protecting your PC from inadvertent destruction from lightning. Ethernet wiring is the weak link in these cases from what I have experienced.
liquidsquid
Minister of Gerbil Affairs
 
Posts: 2437
Joined: Wed May 29, 2002 9:49 am
Location: New York

Re: Lightning Damage Anyone?

Postposted on Mon Aug 06, 2012 9:53 am

liquidsquid wrote:On top of that, a transorb (a reverse-diode which has an abrupt breakdown voltage, like a zener diode only specialized) is placed across the center antenna conductor. Its intention is to absorb and dissipate the excess energy instead of blowing sensitive circuits. The transorb is the sacrificial component. I have no idea why there are not typically used on Ethernet cards, but they are not. Probably to sell more NICs.

The tank circuit on your lightning detector is tuned to what... a few hundred KHz? I'd be willing to bet that a transorb capable of handling large induced surges would degrade the signal on a gigabit Ethernet link (where the operating frequencies are orders of magnitude higher) way too much to be practical. Ability to shunt away larger currents = significantly larger junction size = buttloads of parasitic capacitance. That's probably why NICs tend to rely solely on the transformer to provide ESD and surge protection.

More English translation for the non-EEs:

Lightning emits electromagnetic energy across a very broad range of frequencies. However, lightning detectors tend to operate towards the lower frequency end of the spectrum (in the KHz), since lower frequencies are detectable over greater distances (they tend to follow the curve of the Earth instead of dissipating or radiating off into space) and there is less interference from man-made sources.

Gigabit Ethernet uses signals in the hundreds of MHz.

A transorb is similar to a fuse, except that it trips on over-voltage instead of over-current. Instead of being wired into the circuit in series with the device to be protected, it is wired in parallel. If the voltage exceeds the transorb's trip point, it shunts electrical energy around the protected device, (hopefully) preventing it from being damaged.

To protect against large surges a transorb must be large (where "large" is relative, we're still talking about a physically small electronic component), since it needs to be able to shunt the entire current of the surge around the device being protected without completely disintegrating. But because of how transorbs are constructed, they also tend to act a little like capacitors; and the larger the transorb, the more it acts like a capacitor. The upshot of this is that the transorb will tend to shunt high-frequency energy around the protected device, even when the voltage is below the transorb's trip point. If that high frequency energy happens to be the data signal from a NIC or network switch at the other end of a network cable, you've got a problem - you've basically just thrown your data signal away.
(this space intentionally left blank)
just brew it!
Administrator
Gold subscriber
 
 
Posts: 36886
Joined: Tue Aug 20, 2002 9:51 pm
Location: Somewhere, having a beer

Re: Lightning Damage Anyone?

Postposted on Mon Aug 06, 2012 10:27 am

I thought about it some more, and the reason a transorb is not used on Ethernet is it is supposed to have at least 2kV of isolation potential to earth ground. You can easily get transorbs that only present tiny amounts of capacitive loading which would not even slightly effect signal integrity of a strong Ethernet signal (they are VERY stout), much less than you would see in a 100' twisted pair. You however cannot combine it with 2kV of isolation. You would have to go against the Ethernet spec and refer everything to earth ground on both sides of the connection, then put in the transorbs on each line on each end for 8 minimum. It would be possible and relatively easy to design them this way, but again when you run into grounding issues, you will have some serious problems.

Grounding issues happen a LOT. Imagine two buildings right next to each other. One has a great 6 foot long copper rod pounded into wet conductive clay for building ground. The other building's ground rod was knocked off by a lawn tractor. Now Mr. IT tries to run a shielded Ethernet between the two buildings, and then someone fires up a welder in the ungrounded building. Ooops. With the isolation it wont be an issue, only maybe 240V of 60Hz present. The 2kV isolation deals with it.
liquidsquid
Minister of Gerbil Affairs
 
Posts: 2437
Joined: Wed May 29, 2002 9:49 am
Location: New York

Re: Lightning Damage Anyone?

Postposted on Mon Aug 06, 2012 11:24 am

liquidsquid wrote:You can easily get transorbs that only present tiny amounts of capacitive loading which would not even slightly effect signal integrity of a strong Ethernet signal (they are VERY stout), much less than you would see in a 100' twisted pair.

It isn't just the amount of loading that matters; where it is placed along the line is important too, otherwise you will get signal reflections. The capacitance does not need to be comparable to that of 100' of twisted pair to cause a problem if it is all in one spot.
(this space intentionally left blank)
just brew it!
Administrator
Gold subscriber
 
 
Posts: 36886
Joined: Tue Aug 20, 2002 9:51 pm
Location: Somewhere, having a beer

Re: Lightning Damage Anyone?

Postposted on Mon Aug 06, 2012 1:38 pm

When I was a boy, I had a lightning strike similar in magnitude to the one you describe cause a tree in the back yard to literally explode. Chunks of the tree two to three feet in length were thrown over the house into the front yard. We lost a window. This was pre-computer.

More recently I had a lightning strike to a tree in my neighbor's yard induce enough voltage in our power drop (which passes next to the tree) to produce an arc in a light switch a couple feet from where I was standing. A little disconcerting to be in the kitchen and hear BOOM followed by crackling sounds come from electrical appliances. UPSes on both my workstation and entertainment center took the hit, saving all my gear.
mnecaise
Gerbil First Class
 
Posts: 102
Joined: Thu Oct 30, 2008 12:17 pm
Location: Hampton Roads, Virginia

Re: Lightning Damage Anyone?

Postposted on Mon Aug 06, 2012 8:42 pm

Neat stuff folks! While nothing so cool as a fulgurite was found I did notice an odd gray dusting on some chunks of clay. Through my macro lens it certainly appears to be melted something... I don't think there's very much, but there are bits and pieces of quartz/sand which I suspect is where the gray dusting came from. The ground was fairly moist for once and pretty much exploded from the heat so if there was something in there it got blown to bits along with the rest of the stuff.

Image
Image
bobboobles
Gerbil Team Leader
 
Posts: 298
Joined: Tue Oct 27, 2009 4:46 pm

Re: Lightning Damage Anyone?

Postposted on Tue Aug 07, 2012 11:22 am

just brew it! wrote:
liquidsquid wrote:You can easily get transorbs that only present tiny amounts of capacitive loading which would not even slightly effect signal integrity of a strong Ethernet signal (they are VERY stout), much less than you would see in a 100' twisted pair.

It isn't just the amount of loading that matters; where it is placed along the line is important too, otherwise you will get signal reflections. The capacitance does not need to be comparable to that of 100' of twisted pair to cause a problem if it is all in one spot.


To some degree you are correct, but Ethernet signalling is a tad different from say a high-speed digital single-ended bus in the sense it sends data in a more sinusoidal fashion, not relying on ultra-fast rise times and transitional crossings so reflections are not as significant. It is the positive level in respect to the negative level that changes the bit from one state to another. If a load is presented equally on both halves of the diff pair, differentially it will be transparent to the receiver and the data will be received, though the driver may be presented with a difficult load to drive. You could get reflections, etc on individual lines with unequal loading (say a corroded connection) causing mayhem in the transmission of data, but if they are equal and opposite on both halves, they cancel out at the receiver. The only real problem is if losses become great enough that the differential voltage cannot be achieved in order to differentiate a '1' from a '0', or if you get the pair to resonate/ring from bad impedance matching.
liquidsquid
Minister of Gerbil Affairs
 
Posts: 2437
Joined: Wed May 29, 2002 9:49 am
Location: New York

Re: Lightning Damage Anyone?

Postposted on Tue Aug 07, 2012 3:37 pm

Back when I first got my pilot's license the Canadian MoT used to send out accident reports on a quarterly basis (on paper, this being before the web), and they made for morbidly fascinating reading. Thunderstorms and lightning strikes were particularly interesting to me, especially since much of my recreational flying was in a glider in convective conditions. Lightning strikes on aircraft used to be a lot more common when big commercial aircraft weren't able to fly above the weather and strike mitigation was less well-understood (fatal strikes on jet aircraft in the 60s and 70s caused several design changes). But it can still be a problem for aircraft that spend their lives down in the bumpy parts of the atmosphere -- and of course even the big jets pass through that regime on t/o and landing. And meanwhile we're always developing (and increasingly relying on) electronic control systems that may be more sensitive and building more of the aircraft structure out of composites which may be less or even non-conductive (the conductivity of the aluminum skin of most aircraft is actually an advantage in most cases, as it allows charge to spread out rather than dangerously concentrate, and it can act as a Faraday cage for things inside it; for this reason, composite elements on recent commercial aircraft include conductive components expressly for lightning mitigation).

Some of the more interesting incidents and accidents are summarized in this report, starting at page 74 (though I warn against reading if you are a nervous flier). My favorite (in as much as it wasn't fatal and it demonstrates the mysteriously freaky power of positive lightning strikes) is probably this one:
Uman et al wrote:The event involved a Convair aircraft, Flight 517, taking off from the Salt Lake City Airport on October 15, 1965. At the time of the event, there was some light rain in the area but apparently no lightning other than the event to be described. During takeoff, an extremely loud noise occurred. The first officer stated to the pilot that he believed they had sustained a lightning strike, subsequently confirmed by observers in the control tower, based on his observation of a blue-white glow around the nose of the aircraft at the time of the explosion. The aircraft returned to the airport. Three large holes were found in the runway which exactly matched the dimensions of the two main landing gear and the nose wheel. The largest hole, under the right main gear, was nearly 2 m in diameter and 15–20 cm deep. Pieces of asphalt as large as 0.3 m had been hurled 30–50 m down the runway. The aircraft suffered numerous burns to the wheel rims and fuselage just aft of the nose wheel-well. The rotating beacon, the grounding wire on the right main gear, and the fixed vertical stabilizer cap were burned off
Another incident mentioned in that report but fully documented here is the freakiest for me, and not just because it involves a glider. Like most modern sailplanes, this glider's structure was entirely fiberglass, which is insulating, but contained numerous metal parts. In particular, the control linkages are mostly aluminum or steel, and in this case a positive strike apparently entered the aircraft structure through the aileron fittings and passed through the wing via the control tubes. In the process it flash-heated various components and the air inside the wing, which created enough overpressure to both blow out the canopy and delaminate the fiberglass (basically partially blowing the wing apart from the inside). The freakiest thing, though, was the damage to the control tube itself: the electromagnetic flux travelling along it was so high that it caused the tube (16mm aluminum with a 1 mm wall thickness) to collapse on itself (bottom three pictures here), becoming essentially a twisted solid rod thanks to Lorentz forces and Ampère's Law. That may have been aided by heating, but the aluminum is quite conductive and metallurgical analysis suggested the metal didn't get hotter than 200°C. Yet he investigators were completely unable to duplicate this effect with the highest (312 kA) electrical power available to them. Some things about lightning remain mysterious.
ludi wrote:Sounds like you had a (comparatively rare) positive strike, to dig up the root like that. But yes, induced potential in building wiring is quite common for nearby strikes. Could have been worse, though: about fifteen years ago, there was a strong positive strike that took out a tree near Ft. Collins, Colorado and dug up the roots in a similar fashion to what you saw, but it induced a high enough potential in the adjacent house that it blew up all of the electrical wiring. There were pictures in the paper showing ray gun blast patterns around every single light figure and coverplate in the structure.
I remember reading about that one (or one very much like it). As I recall they had to strip all the wiring out of the house and all the appliances had to be replaced. (If it's the one I'm thinking of I recall that the strike entered through the dryer vent and the dryer, completely vaporizing that accordion-style dryer hose).
UberGerbil
Gerbil Khan
 
Posts: 9924
Joined: Thu Jun 19, 2003 2:11 pm

Re: Lightning Damage Anyone?

Postposted on Tue Aug 07, 2012 3:48 pm

The bit about the compressed aluminum tube reminded me of this site.
(this space intentionally left blank)
just brew it!
Administrator
Gold subscriber
 
 
Posts: 36886
Joined: Tue Aug 20, 2002 9:51 pm
Location: Somewhere, having a beer

Re: Lightning Damage Anyone?

Postposted on Tue Aug 07, 2012 4:05 pm

just brew it! wrote:The bit about the compressed aluminum tube reminded me of this site.
Wow, that's really cool. Now I want to make some of those and do a little casual "magic" with coins I borrow from others. I'm surprised they don't have any Canadian coins (the "Toonie" would be particularly interesting)
UberGerbil
Gerbil Khan
 
Posts: 9924
Joined: Thu Jun 19, 2003 2:11 pm

Re: Lightning Damage Anyone?

Postposted on Tue Aug 07, 2012 4:23 pm

They do have the toonie on that page:

A Canadian "Toonie" has the center loosened at 6,500 joules, but the center is still held captive. At 14,000 joules, Rob Stephens, a friend and fellow coin shrinker from Ontario, Canada was able to separate the coin into two independent pieces. However, his blast shield failed during the shot, and he sustained considerable damage to his lab from shrapnel from the exploding work coil
ChronoReverse
Gerbil Elite
 
Posts: 732
Joined: Wed Dec 12, 2007 3:20 pm

Next

Return to The Back Porch

Who is online

Users browsing this forum: No registered users and 9 guests