TVS Diodes for Lightning Surge Supression?

[colemj]

Originally Posted by EwenT

I was not involved in their installation and see a couple of fundamental errors in this respect.

For those of us contemplating adding TVS diode protection, please explain what you consider proper installation.

Mark

[EwenT]

Originally Posted by colemj

The main strike exited the boat through our direct mast grounding system.

Mark

Mark,
What did you observe on the mast grounding system? Do you have any photos?

[colemj]

Originally Posted by EwenT

Mark,
What did you observe on the mast grounding system? Do you have any photos?

Wait, you have to answer my question first.

OK, I'll go first. Our mast grounding system is a commercial system called a "Strike Shield" (no longer in business). It consists of a mounting plate bolted to the mast above deck with a thick 4/0 cable connected to that plate and dropped overboard through the trampoline (we are a catamaran). The water end of this cable consists of a machined copper electrode with 15-18 feet of linear edge surfaces machined into it (it is ~12" long with 18 machined edges in it).

When we were struck, we noticed that the thick heat shrink plastic used to cover the connection of the wire to the mast connector was split opened and the electrode was missing the few barnacles and all the sponge/weed that often grows on the machined edges (we keep this in the water all the time). I usually have to pull this up and scrap it hard every so often, so finding it pretty clean was a surprise.

I don't have any photos, but could take one of the split heat shrink plastic since I haven't figured out how to correct that yet. It isn't very interesting, though.

Mark

[Nicholson58]

Originally Posted by Andina Marie

You can use bi directional protection diodes on DC, you are just using half of it.

And during a lightning strike the voltage across the device could be reversed momentarily so the bi directional diode "could" help.

HOWEVER, the amount of energy they can absorb is microscopic compared to the energy in a lightning strike, or even the induced energy from a near-by strike so the chance they will do any good is probably not worth their cost. The money would be spent much more effectively on insurance .

At the buss-bar or power switch level of the circuitry they would be totally useless. If you are going to stop a freight train you don't string a chain across a main line. At the device to be protected with all the branch line wiring between to add some "cushion" the train will have lost a lot of energy so the chain across will be more effective but still nearly useless.

We have been struck several times in the last 5 years. I have come to the conclusion (observations) that my devices wired in by means of air gap double pole rocker switches have, so far, spared every device in which I have installed them. Nearly everything else destroyed. I have even added two-pole air gaps to my VHS antennas. I'm sure this is not perfect but handles the issue that most electronics are not really 'off'. The power switch is software. Locate the switches as close to the device as possible.

[hellosailor]

Mark, are those machined edges what we call in the US "fluting" of the metal? Like fins on a classic simple heat sink, made by chewing slots on the metal mass?


Polyphaser has a large industry following, and their lightning protectors aren't TVS diodes, but rather, "air gap" to ground. Except they do vacuum gaps instead of using unreliable air, as I recall it.


That also used to be a common DIY way to set up a lightning ground path. Install some old spark plugs in the lines you want to protect, with the top of the plug connected to your wiring, and the base connected to your lightning ground path. Adjust the spark plug gap to a minimum, usually whatever it was is enough. Forty thousand volts comes down the wire, and that spark plug gap might as well be solid metal, so there's a good chance the strike will go to ground that way.

[colemj]

No - here is a picture of one. There are 6 lobes with 3 edges machined into each one.

Mark

[hellosailor]

Yeah, that's basically the same as "just" six very wide flutes, machined into a rifle barrel or other piece of rod stock. Except here someone has made the ridges a bit beefier, and made way lot fewer of them. Looks like the same concept for the same reason (increased surface area) but made special and proprietary.
let's see if I can attach a quick googleimage.

[transmitterdan]

Originally Posted by colemj

Lightning surge will almost assuredly come in through the ground side of the circuit. After our strike, we could trace the path completely through the ground side - enter the boat through the VHF coax shield, jump from the VHF into the boat ground (most likely through the case), blow the fuses on the negative wires while leaving the positive wire fuses untouched, enter electrics through the case ground and exit through the engine ground taking out the alternator's internal regulator.

And that was only the surge component of the strike. The main strike exited the boat through our direct mast grounding system.

In autopsying a couple of other boats (including Domino), they too had the surge come through the negative side of the circuits.

Mark

This is an important observation and is usually what happens. Lightning doesn't strike the +12V bus. It strikes the mast which in most boats is connected either directly or indirectly to the negative bus of the boat's electrical system. When high current flows it causes a large differential voltage across whatever it is flowing in. Since most of the current is flowing in metal bits that eventually connect to the water that means the current will be mostly in the negative bus of the electrical system. So the negative bus will have hundreds to thousands of volts differential from one part of the ground bus to another part. And some electrical gear has multiple grounds (e.g. radios, sailing instruments, data networks) which makes the problem harder to analyze.

Even though the common thread is the negative side the damage is caused by too much voltage between the minus and plus bus. So a transient protector across the 12V pair of wires feeding the device can help.

Radios are an especially hard case because the radio is connected to minus via the coax and the 12V power leads. Polyphaser RF protectors like this one (PolyPhaser Coaxial Lightning Protectors IS-50UX-C0 - Free Shipping on Orders Over $99 at DX Engineering) contain a gas filled arc tube across the two conductors of the coax. The gas does not waste a lot of power at RF frequencies and it can help set the arcing voltage more precisely than a plain air or vacuum gap. There is no diode in these devices. The combination of a coax gas tube and a 12V transient protector can help protect VHF and SSB radios.

A DC 12V power transient protector may have a gas arc tube then a fuse following by a transient protection diode (e.g. http://www.l-com.com/multimedia/data...HGLN-DC-12.PDF).

Doing this for every sensitive electronic device in the boat can run into significant money. And no reputable manufacturer will give a guarantee of success against damage.

Some light reading: http://www.onsemi.com/pub_link/Collateral/TND335-D.PDF

[Nicholson58]

Originally Posted by colemj

No - here is a picture of one. There are 6 lobes with 3 edges machined into each one.

Mark

Aluminum I suppose? If so, I suspect it is an extrusion. Machining is relatively very expensive.

[colemj]

Originally Posted by Nicholson58

Aluminum I suppose? If so, I suspect it is an extrusion. Machining is relatively very expensive.

No, it is tinned copper and formed by machining.

Another picture from the company's website:

Mark

[Nicholson58]

Originally Posted by transmitterdan

Even though the common thread is the negative side the damage is caused by too much voltage between the minus and plus bus. So a transient protector across the 12V pair of wires feeding the device can help.

Radios are an especially hard case because the radio is connected to minus via the coax and the 12V power leads. Polyphaser RF protectors like this one (PolyPhaser Coaxial Lightning Protectors IS-50UX-C0 - Free Shipping on Orders Over $99 at DX Engineering) contain a gas filled arc tube across the two conductors of the coax.

Some light reading: http://www.onsemi.com/pub_link/Collateral/TND335-D.PDF

Does this impact the range of the VHF or the AIS class B?

Do you have any experience that indicates that it really works? It is not clear from the manufacturer's text that it handles really high DC. (lightning)

As you noted, the problem is the negative central buss. This is why I have added double pole air gap switches to isolate the devices from both + & - boat wiring when we are not on the boat or during storms. I have these also in the antenna wiring for both core & shield.

[colemj]

Originally Posted by hellosailor

Yeah, that's basically the same as "just" six very wide flutes, machined into a rifle barrel or other piece of rod stock. Except here someone has made the ridges a bit beefier, and made way lot fewer of them. Looks like the same concept for the same reason (increased surface area) but made special and proprietary.

Similar - the difference is in the amount of linear edge surface. Edge surface or sharp points are what is supposedly necessary for efficient dissipation - the more the better.

Mark

[colemj]

Originally Posted by Nicholson58

Does this impact the range of the VHF or the AIS class B?

Do you have any experience that indicates that it really works? It is not clear from the manufacturer's text that it handles really high DC. (lightning)

As you noted, the problem is the negative central buss. This is why I have added double pole air gap switches to isolate the devices from both + & - boat wiring when we are not on the boat or during storms. I have these also in the antenna wiring for both core & shield.

There will be no effect on the range or performance of equipment. They work by inserting a ionizable gas cartridge between the conductor and the shield. When struck, the gas ionizes and shorts the conductor to the shield. They are externally grounded, so presumably the shorted shield/conductor then goes to that bond point. A proper installation has one device on each end of the coax.

Your double pole air gap switch on the coax is the equivalent of simply disconnecting the coax from the radio. I have always wondered what would happen during a direct antenna strike at the disconnected end of the coax. Something tells me that that energy has to jump somewhere…

My only example has been our strike, which hit the antenna and mast - most of which passed through the lightning bond to water. The part that did come into the boat through the coax blew the face right off the radio. I suspect that it would have easily jumped an air gap switch.

Maybe a better way would be to use a switch to short the coax to the conductor?

Mark

[Cottontop]

For the OP, if you haven't found it yet, Ewen Thomson has a site, marinelightning.com, that has a lot of information on it.

[Nicholson58]

Originally Posted by colemj

A proper installation has one device on each end of the coax.

Your double pole air gap switch on the coax is the equivalent of simply disconnecting the coax from the radio.

Maybe a better way would be to use a switch to short the coax to the conductor?

Mark

If one of these is required on each end of the antenna does that mean top of the mast too? One could be easily placed at the foot of the mast in the cabin floor.

Our last hit:
stereo gone
VHF gone
60 amp 2-pole shore breaker - one leg gone
18 Buss fuses
All LED nav & deck lighting
All wiring in the bow cabin - fused
3 interior LED cabin lights
Raymarine ST 60+ suit of wind, speed, depth
Watchmate 850 AIS
Awlgrip on the transom crazed
Windex on the main mast - vaporized
VHS antenna on the main mast - vaporized (bits of fused SS on the deck)
1.5 KW aux alternator on the Gen set - diodes gone
2.5 KW Xantrex - brains scrambled - OK after total re-boot
Autopilot unreliable

What is particularly interesting is that our Simrad NSE8 with its broadband radar & depth scanner were all wired with air gap rocker switches so that they could be individually controlled. NO damage. In the interim while we were adding more switches, I removed the power & antenna from the VHS and popped the face from the stereo.

I did examine the guts of the VHF & the LED lights. LED lights had the diodes of the rectifying circuits blown. The VHS had burn evidence near the power-on part of that one board. I suspect a replacement board would fix it - good luck getting that.

I rely a lot on the Dakota hand-held GPS plotter. It stays clipped to my lifejacket. If it gets hit I probably won't know or care.