Lightning takes out LEDs?

[RobbieW]

Following on from another thread on Lopolights, I'm curious about the effects of large transient voltages on yacht equipment. We are mostly aware that a lightning strike will take out electronic equipment on board, but what is the effect of similar voltages at a lower level? I'm thinking about nearby electrical storms or even HF transmissions.

There is evidence that an electrical storm within a few miles can take out network cards on PCs. A large enough transient is induced into the network cable that it takes out sensitive equipment attached to that cable. When we install LED navigation lights, they are usually at the end of the longest cables on the boat - whether at the mast head or on the bow. These wouldnt need to be in operation to be affected by a transient, could this explain some of the failure rate for LED navigation lighting.

So my question for any expert out there is does this theory hold water ? If it does is there anything we can do to try and minimise the effects?

 

[Davegriff]

Not an expert, but I'd say yes.

http://www.edn.com/design/component...6/Circuit-protection-for-outdoor-LED-lighting

 

[nimbusgb]

Not an expert, but I'd say yes.

http://www.edn.com/design/component...6/Circuit-protection-for-outdoor-LED-lighting

That article looks at mains driven led lighting that may have many miles of cable attached to it. That length of cable could well get large spikes induced into it by nearby ( to a cable miles from the actual point of consumption ) strikes.

On a yacht the circuitry is isolated and the total run of wire to a LED is probably less than 100m or so. Whats more on a sailing vessel the longest 'receiver', the run up the mast is enclosed in an aluminium tube. Inducing any meaningful voltage in the boats circuitry from a lightning 'near miss' is going to be nigh on impossible! In fact I'd hazard that it would have to be so close that a strike would rather hit the boat than something else!

That having been said a couple of big capacity 16volt transorbs or Zener diodes around the boat would be a good insurance policy.

 

[Chalker]

Leds like all other diodes have a maximum forward current and a maximum reverse voltage capability. The cable feeding power can act as an antenna or a single turn of a transformer with a lightning flash or SSB transmission. So yes, either could be the cause of LED failure. It could also damage the other electronic components in the regulator feeding the LEDs.
The regulator is there to control the LED current so that the light is constant for 11-16 volts or 11-30volts depending on the spec.
A diode (in the non-conducting direction) and capacitors across the supply as it enters the light will reduce but not prevent the possibility,

 

[rogerthebodger]

Living in an area of high level of lighting with some experience of lighting protection on my networks and on my boat. lighting was not the cause of the faults on my NASA led mast head light.

Reading the previous thread and having personally repaired my tricolour LED nav light It would be very interesting to find out from the manufactures of the varyous LED nav lights what is infact the most common mode of failure.

In my case a number of the red and green LED's failed open circuit and in replacing the faulty LED's I got the unit working again.

The NASA light used a micro controller to up the supply voltage to around 50 volts to drive 3 sections of series led's, hence if 1 led does open circuit in a section the whole section will not operate.

IMHO if the unit was damaged by lighting it would be the controlling circuit that would be damaged not just some of the LED's

Have no direct experience of the lopolighs so cannot comment directly.

 

[Chalker]

You must have posted whilst I was writing!
I think that the unshielded wiring or even the slight imbalance of the cable in the mast could be sufficient, but we agree on the possibility and the basic reduction method in the effect.
I also agree that a zener diode or transorbs are other helpful methods.

 

[ostell]

In the days when I actively worked fixing computers and ADSL, with its modem/router, was not quite so popular I would buy a few modem cards after a storm had passed over as I knew I would be getting calls for dead modems for a few days.

 

[Davegriff]

That article looks at mains driven led lighting that may have many miles of cable attached to it. That length of cable could well get large spikes induced into it by nearby ( to a cable miles from the actual point of consumption ) strikes.

On a yacht the circuitry is isolated and the total run of wire to a LED is probably less than 100m or so. Whats more on a sailing vessel the longest 'receiver', the run up the mast is enclosed in an aluminium tube. Inducing any meaningful voltage in the boats circuitry from a lightning 'near miss' is going to be nigh on impossible! In fact I'd hazard that it would have to be so close that a strike would rather hit the boat than something else!

That having been said a couple of big capacity 16volt transorbs or Zener diodes around the boat would be a good insurance policy.

Well I said I was no expert, but always willing to learn. So do I understand correctly that you're saying that we needn't worry about about any electronic equipment on board being affected by lightning other than a direct hit? Why the caveat? (Internal spike?)

 

[nimbusgb]

imo A strike would have to be so close as to be considered a direct hit in order to induce enough current to cause a problem in boat wiring.

I have flown a sailplane under cumulonimbus and had several lightning discharges, cloud to ground, very close by, perhaps 300m or so. The panel of was full of sensitive micro electronics ( flight computer, 3 gps's, variometers, moving map display and engine control panel ) and they survived without the slightest glitch. ( I hasten to say that I had pushed my luck and I hightailed it pretty sharpish! The charge gradients under that cloud were so big that I was getting static shocks off the panel and controls. Scratch one of the nine lives!)

Yes the caveat would cover on board generated transients. I have seen spikes of 100v ( captured with a 60 mHz storage scope ) fed back into low voltage circuitry by solenoids and relays dropping out. Modern Regulators on alternators are usually quite clean but a dicey ignition switch can still cause significant alternator dump transients.

 

[ostell]

High current is not directly the problem for the electronics, it's the induced high voltage which can break down the semiconductor elements. There's no need for a direct hit, or even a close hit, just a thunderstorm passing overhead that induces the high voltage in the wires.

I've seen a thunderstorm approaching while on the beach, sufficient voltage gradient to make my hair stand on end. Kids thought it extremely funny as I had to rush them off the beach. No strikes followed, only a great hailstorm which played havoc with my plastic conservatory roof.

 

[nimbusgb]

High current is not directly the problem for the electronics, it's the induced high voltage which can break down the semiconductor elements. There's no need for a direct hit, just a thunderstorm passing overhead that induces the high voltage in the wires.

Yes and Faraday, Maxwell and Lenz pretty much tell you that even an n million volt, n million amp current in a lightning discharge needs to be extremely close to an isolated piece of wire ( a boat not a mains connected system ) to induce anything approaching damaging levels

 

[RobbieW]

We seem to be having a bit of a theory vs empirical discussion here. Part of my experience that led to the question was having a Bebi LED anchor light fail. That light hangs in the foretriangle on the end of a 12m cable running back to a 12v plug in the cockpit. The night it failed we'd had several hours of electrical storm rolling around the anchorage in Santa Ponca, Mallorca, though I saw no ground strikes - more sheet lightning. When I reported the failure back to Bebi, and added the electrical storm bit and some resistance measurements I made on the failed unit, the response I got suggested 'near field' events as a cause - Bebi replaced the light without question so full marks there.

I also have friends who see thier LED cabin lights switch when they operate thier HF transmitters - granted there could be a number of causes for that behaviour.

nimbusgb, could it be that in your sailplane experience the cockpit equipment was designed with the thought that lightning events were a distinct possibility and thus protected accordingly?

In general yachties dont see electronic equipment failure from lightning except from a direct strike. However, anecdotally at least, yachties dont seem to be getting the expected mtbf from thier LED nav lights.

 

[nimbusgb]

nimbusgb, could it be that in your sailplane experience the cockpit equipment was designed with the thought that lightning events were a distinct possibility and thus protected accordingly?

Doubt it given that the moving map and GPS unit were COTs units not originally intended for aircraft installation and the entire panel was built by me anyway with no special considerations.

 

[Iliade]

Slightly O/T, but...
We live less then ten miles from Gatwick airport.
If we put a low energy (fluorescent) type of light bulb in the top hall light it occults nicely, albeit at a very low level, when the lamp is switched off.

All we can ascribe it to is the house wiring picking up the radar emissions...

 

[nimbusgb]

Slightly O/T, but...
We live less then ten miles from Gatwick airport.
If we put a low energy (fluorescent) type of light bulb in the top hall light it occults nicely, albeit at a very low level, when the lamp is switched off.

All we can ascribe it to is the house wiring picking up the radar emissions...

Perhaps, but very different to a boat. A house is connected to the grid, miles and miles of cables. A boat isn't.

If my lights were occulting i'd be very worried. It wont be gatwick but it could be a localizer or some other radio source. If it is radiated rf energy I'd be after buying a lot of tinfoil!

 

[grumpy_o_g]

Slightly O/T, but...
We live less then ten miles from Gatwick airport.
If we put a low energy (fluorescent) type of light bulb in the top hall light it occults nicely, albeit at a very low level, when the lamp is switched off.

All we can ascribe it to is the house wiring picking up the radar emissions...

The flourescent is being hit with radio waves so the gas will be excited slightly and you could see some variations in the light produced but I would have thought you'd need to be a lot closer than ten miles. I'd be amazed if induced voltage in the wires was the culprit as well for the same reason. If it's the radar then you should actually see an increase in brightness every few seconds rather than occulting.

 

[grumpy_o_g]

I think it would need to be a lot closer than a few miles. A few hundred yards perhaps but not a few miles away. For an electrical storm to take out a NIC froma few miles away it would need to be exceptional circumstances and even then remember that NIC's are far more sensitive than LED's. I don't know but I wouldn't be surprised if a lot of the failure rate is down to poor quality control on the LED's themselves.

 

[Marsupial]

I think it would need to be a lot closer than a few miles. A few hundred yards perhaps but not a few miles away. For an electrical storm to take out a NIC froma few miles away it would need to be exceptional circumstances and even then remember that NIC's are far more sensitive than LED's. I don't know but I wouldn't be surprised if a lot of the failure rate is down to poor quality control on the LED's themselves.

we seem to be drifting a bit, but I have seen lots of cases where lightning has taken out boat electrics without actually striking the boat, but the strikes were yards not miles from the boat, all the owners knew was there was a storm and now nothing electronic works, so LEDs would fall into that category I think.

I used to live in Rayleigh Essex a fair few miles from Holehaven; the west end of Canvey island and yet we always knew when the methane boats were berthing as the Holehaven radar would cause our TV to occult.