| OK Phil, at least; unlike some other posters; your arguments are clearly | written and well argued. So let me pose a question. If a common mode | serge reaches a plug in Transient Voltage Surge Suppressor and that | device does what it is designed to do; which if I understood my previous | training correctly is to keep the voltage as near to the same on all | three wires as possible; isn't that effective protection. No voltage | difference would mean no current flow. No Current flow equals no power | delivered to the protected load. No power equals no damage. Objective | achieved, declare victory, and go home.
No, it is not effective protection against a common mode surge. It can still attenuate the surge in some ways, especially if the ground wire is not part of the arriving surge.
In the case of a common mode surge, there is no voltage between the leads of the MOV, so they would not conduct, assuming the leading edges arrive in sync. But there is voltage propogation, and thus a charging current.
In the case of a mix mode surge (one wire is a higher voltage than the other, but both are high in the same polarity), the MOVs still should conduct. When they do, the best they can accomplish between these two wires is to bring the voltages close together. They could change a mixed mode surge into a (nearly) pure common mode surge.
Now back to common mode. The surge will continue on to the device that was supposed to be protected. It will suddenly raise the voltage of the device as a whole. It is unlikely that the very fast rise time will make it all the way through to digital circuits in a computer, but if it did, that could be destructive to the very sensitive parts just because of the high voltage causing inappropriate breakdown of various barriers of the high density solid state circuitry. Still, in very quick order, the voltage rises, and has no where to go. Translate the "no where to go" to bouncing around from end point to end point of the various apparent subpaths of the device. That can include attached devices like keyboard, mouse, modem, ethernet, video, etc. Those next devices are also charged up by the surge. Some of them provide paths out to other places, such as the modem going to the phone line or cable TV connection. You can then see a high current rise on those paths as well.
When I was in college, home on break, I woke up in the night due to a thunderstorm. I had recently installed a new TV antenna to get a new UHF station, but had not put any protection on it, yet. So I went to the living room and unplugged the antenna lead coax from the TV and tossed the end into the middle of the room. I went back to bed and heard the storm pass. Later on in the morning another storm line came through. During it I heard a very loud lightning hit and my mother who was already up and in the kitchen screamed. I got up to see what had happened. The antenna took what might have been a direct hit. The carpet around the end of that antenna lead was charred about 20 cm radius (not an exactly uniform circle). Nothing else was damaged. The coax lead was fine. The antenna was fine. The TV was fine. I was able to watch the new TV station without the need to replace anything. Insurance covered most of the carpet replacement. I got stuck paying the difference since it was my antenna.
At the end of that coax, the voltage gradient rose very high. At least a lot of the energy from the surge dissipated into the air in some way. I wasn't there so I could not see how wide the flash really was. My mother did say she saw a bright flash, but thought it was just from the lightning as seen through the windows. But it was very very hot for at least a little distance from the end of the coax and apparently not anywhere else.
| I have done remote installs from Alaska to Argentina to Arizona and in | the real world work we did single point grounds and listed TVSS installs | kept the equipment functioning month after month and year after year. | When these sites failed we got phone calls and lots of them. The Alaska | site had a vent control failure that caused the batteries to fail | explosively and a colleague was en route to the site before the acid | stopped fuming. The Arizona site had a physical failure of the | structure years after I left the company that manufactured it. I still | got a phone call. So far not one call for a lightning or surge caused | failure. To what do we owe our clean record of power system | survivability in lightning prone environments. Meir careful adherence | to the IEEE and NIST guidance on protection that some on this newsgroup | keep trying to dismiss as ineffective would be my answer. But I'm sure | some would say that since it violates their belief that low impedance | earthing is the only answer it's just dumb luck.
The guidelines are most certainly not ineffective. To a great degree I believe that a clean and careful installation contributes to how well the various protection methods work. To the extent that a high transient surge behaves like RF, which I believe a significant portion of the energy does, minor non-obvious details like the angles two wires bolted to the same connector have relative to each other makes a big difference in how much of that high transient energy can go from one to another. Other factors can come into play in various installations, some more than others, such as common mode surges being induced from one circuit to another.
One element of protection that I think helps is added inductance following any MOVs. Cheap power strips don't have it. Some, like the Tripp-Lite isobar series, do. It's there more for isolation purposes between different plugged in devices. But it would have the effect of making the path to the protected device appearing to a fast rise transient as high impedance, and letting the (now conducting) MOV be the low impedance path, at least for differential/tranversal surge modes. But even for a common mode surge, it can help attenuate the surge fast rise component. What gets through would be much less like RF in behaviour.
I'm going to be putting up a new TV antenna, soon, to get digital TV from Pittsburgh, since the local cable TV decided not to make HD free in the basic package. This is a different house than the previous antenna. This one will get protection before it gets wired into the house. In addition to the lightning arrestor device located outside at the base of the mast, I will be putting in an additional protection of my own design. This will consist of a 4 foot long PVC pipe, stuffed with steel wool plus a copper drain conductor running through the PVC pipe in contact with the steel wool. The coax will run though the middle of that steel wool. After the pass through the steel wool, it will then wrap a few times through a large ferrite core. All this will be buried in the ground and the copper wire grounded (maybe cadwelded) via an electrode at each end of that PVC pipe (capped on each end). The idea of this is to dissipate the magnetic field of a common mode surge going over the coax. I would not do this kind of thing on power wires (they could prematurely "dissipate" the steel wool). But for TV coax, it should be OK. I don't know how effective it really will be. I won't skip any other protection just because I put that thing on there. I'll still disconnect during storms or times away from home (why I change the TV coax connections to BNC, even if it is an impedance mismatch).
Gun safety courses teach multiple methods of gun handling safety with the intent that every one of them always be practiced. It isn't about being paranoid; it's about the value of what is being protected by these methods. Lightning risk might not be quite the same thing. But in my view, it is just being wise, not paranoid, to use as many different forms of protection as is practical.