I have a friend who wants to put surge protection at his panel. His concerns are everything in the house with electronics including furnace, microwave, stereo, TVs, computers and etc. My questions are whether surge protection actually works and which product works best.

Most damage, that is brought to my attention, is from lost neutrals. I don't remember dealing with anyone who lost equipment that might have been saved with surge protection.

My own advice would be to pick the favorite equipment and use a USP, but they are costly and annoying because they beep when the power goes out.

What do you think?

Start out looking at your Ground electrode. Be sure the protectors for phone cable and satellite connect there. Use a panel protector and point of use protection that picks up all inputs on interconnected equipment (TV, phone, computer).
UPS is really just of you don't want something to go off.

As per my training,
I'm told that there are two types of surge protection.
The first known as coarse protection is fitted into a panel between the incoming Hot wire and Ground.
All these things do is "clamp" over-voltage events as they happen on the incoming mains.
However, they won't clamp everything that will damage sensitive electronic equipment.
For this you need fine protection for your individual appliances.

One caveat though, most of the surge diverters I've installed here, have a consumable plug-in diverter module.
Now if that operates, it needs to be replaced otherwise you have no protection.

What I have done in the past, is installed a small alarm to the side of the panel, run off an auxillary contact on the side of the diverter module, if the diverter module trips, the alarm goes off and will beep until the module is replaced.

Some may say that that is rather heavy-handed, but at the end of the day, if people want protection, it needs to be
kept up to date, because if you have something like this installed and it trips, Murphy's Law says that you are more than likely to have a surge, between when it trips and when you replace the module.

The two basic types of protection are "whole house" and "point of use". Types of surges include "ground seeking" (e.g. a lightning strike, or a MV phase contacting an LV line, etc), "differential", and "inductive".

Point of use protectors protect against the differential surge, which is the most common at the point of use. As all voltage is a difference between the potential of 2 points (conductors in this case), the protection is to minimize and limit this voltage difference. That difference can exist not only between line and neutral, or between lines, but also between any of these and EGC. Additionally, the voltage difference can be between anything else metallic (cable coax, antenna feed, CAT5 to network, twisted pair to the telco, metal water pipe, metal gas pipe, A/C freon tubes, heating ducts, metal building frame, etc). A scary long list!

Correct protection at point of use is to interconnect every metal together so that the voltage potential is minimized. The closer connected, the better. Since direct connection isn't practical (fault currents on power, lost of signal on communication wires), the metal-oxide varistor (MOV) device is substituted as the best alternative, since it will have a high resistance at low voltage, and a low resistance at high voltage.

Everything must be connected. Leaving some metal conductor out of the protection does not leave you partially protected ... it leaves you unprotected. This is because electricity does not just flow over the least resistive path. It flows over all paths it can access. So even if you have proper protection for your TV on the power conductors with a POU surge protector (power strip), damaging currents can still flow from the rising voltage on the power (if that's where it comes from) to the cable TV coax or the satellite dish coax (through the receiver/converter box, and out through that coax).

The coax and phone connectors on the power strip protectors are not there as a selling gimmick to make people think that gives them other options. They must be used if those kinds of cables are present and connected.

Whole house protection works similar to point of use protection, but on a larger scale. However, because some surges can occur inside the house for various reasons (such as an inductive surge from a lightning strike on a tree at the opposite end of the house from the service entrance), the whole house protector is not a substitute for point of use protection. Even with whole house protection, you get more protection by also having point of use protection.

But even whole house protection needs to be done right. That means everything metallic must come in together at a very close proximity. And this can be hard to do with some code requirements for minimum spacing of things like an entrance meter and a gas meter. The whole house protection can be ineffective if there is an alternate path for the surge to exit. Cable TV must come in at the service entrance. The phone line(s) must come in at the service entrance. The antenna feed must come in at the service entrance (and this gets tough for amateur radio operators who have their "shack" inside the house ... so I recommend hams have a real separate shack where possible to simplify the antenna wiring). Water and gas also need to come in there as close as you can get them, and be well bonded with large conductors to the same grounds as the entrance panel and protector are connected to, closer to the protector.

Both whole house and point of use protection work on the principle of equalizing the voltage potential so that the difference between any two points is reduced. When a surge event happens, the entire house can rise in voltage relative to earth perhaps many thousands of volts. As you know, it is not the voltage that kills, but the current (and it doesn't take much). It's the same for electronics, with some devices being damaged by even less current than would kill a human (though many other things would need a lot more current).

Whole house protection can also protect against common mode surges with fast rise times. But these surges will weaken quickly as they travel over the conductors due to the induction that would be present for current flowing in a single direction without a return current. A well grounded whole house protector can generally protect against a direct lightning strike on an in-air service drop. Beyond that, the inductive effects of the cables will reduce the first voltage rise (high frequencies are impeded much more by the inductance).

Many UPSes do have integrated surge protection. But have yet to see one that also includes auxiliary connectors like 8P8C (RJ-45) for ethernet, or for phone or coax. Instead, I recommend having separate point of use protection ahead of the UPS, and minimizing the cabling between these components to minimize inductive surges.

I also recommend the use of wireless networking at home, and either/both wireless and fiber optic in an office. But if you do need to run ethernet cables (like in an office where conversion to other means is difficult), then be sure every individual end use is protected with a power strip protector that includes an ethernet port pair. All interconnected equipment must be on the same protector. Two separate protectors, especially on separate power circuits, where devices on different protectors are cross connected by metal conductors, is just NOT protected. If a computer station needs more outlets, then cascade one protector into another (but limit other connector types like ethernet to the first).

I've only found ONE power strip protector that includes ethernet. I found it here (and bought a couple dozen to protect the office where I work, which had everything wired before I came to work there, which made quite a nightmare).

http://www.cyberguys.com/product-details/?productid=30778

These protectors also have a nice sideways plug. But don't let that tempt you into using two such protectors on the same outlet in parallel, unless nothing is interconnected between each protector's connected equipment.

Electrical contractors are, I presume, less likely to be called on to address a surge incident. In most cases, one or more pieces of equipment will just be invisibly damaged and not function properly (dead ethernet port for example), or at all (dead power supply or dead CPU). People usually won't see it as an electrical wiring issue. As long as the power works and there isn't that burning smell going on, they won't think to call an electrician. An open/loose neutral, however, tends to cause widespread power effects (even if the 240 volts won't cause damage to autoranging power supplies) that people do notice. And that can also cause that burning smell with 120 volt devices are being burned by say 180 volts from a neutral-less circuit out of load balance (which gets worse as people try to shut more and more things off individually, instead of shutting the main off). In this case, they will know they need an electrician. So I suggest that electricians will see more loose neutral cases than surge damage cases, even if the latter happens more (which I'm not sure if it does).

Usually the ethernet cable is not a problem as long as it stays in the building. Where we used to blow stuff up was when the ethernet ran between buildings. I am not even sure those strips would fix that problem. We made the problem go away with a fat bonding wire that was significantly shorter than the ethernet cable, bonding the machine frames together.
We also clamped ferrite beads on the ethernet cable at each end.
There was some engineering dispute about how that worked, but it worked. The basic idea came from the State Farm data guys in Winter Haven on parallel printer cables but we adapted it to ethernet.

I have twice encountered a case of a surge causing damage to equipment that was not connected to anything at all. In one case when I was around 12 years of age, a lightning strike in our back yard killed the front end transistor on my portable battery operated radio. The antenna was vertical for FM reception, and so it was exactly aligned with the lightning path for optimal RF energy transfer.

The other incident involved my stereo system. I had it in an upstairs spare bedroom. Because the speaker cables were expensive highly stranded AWG 6 conductor pairs of 50 feet originally intended for downstairs use where I needed that length, I decided not to cut them because I might want to take things back there in the future. I should have purchased some new shorter cables and put the long ones in storage. Instead, I just coiled them in an empty closet midway along the path. Having no surge protectors, I simply disconnected the stereo system from power when I was not using it and during any lightning storm.

Upon reconnecting it after one storm, I found it to be dead. Electronic forensics found that all 8 amplifier final transistors were dead. Also, the tweeter coils on the speakers were burned up. Tearing down the tweeters revealed the coils severely melted. It was then I realized that the coiled cables enhanced the effects of any inductive energy transfer from nearby lightning strikes. Even though the current would have nowhere else to go, it can still bring up a substantial current given the length of wire.

Surge protection is not a perfect thing. Even with the best of protection, substantial surges, especially from very close lightning, can still get through. It just needs to develop a voltage between points somewhere for long enough to cause a big current. And that can happen even if you disconnect everything in a storm (although if that is easy and safe enough to do, it does help some).

Ethernet between buildings is just asking for trouble. Fiber is the best choice there. But you did take the correct steps to minimize the effects. The ferrite beads would not have stopped a differential mode on the ethernet cable, but if it was getting a surge from the voltage difference between buildings (step potential in earth when lightning makes a nearby hit, especially if it hots a building), this would be common mode on the cable, not differential. So the beads would help by having even more of an inductive effect than the cable by itself. Wrapping the cable through a large toroid a few times would help even more on common mode surges (without effecting the differential voltage since both sides are parallel in the cable).

I'd prefer to run fiber optic and use an ethernet converter on each end. They are cheap enough. Even gigabit ones are showing up.

This was before fiber was commonly available for ethernet. It was a Holiday Inn that started out as 3 separate buildings that eventually effectively became one but there were still 3 services. There was also a pool bar with a POS terminal.
We started by bonding all of the electrode systems together and them moved to the actual machines that were blowing up (Pool bar being the worst)

When we fixed the problem I had to explain to Holiday Inn what we did because they were having the problem in other hotels.

The thinking was always along the line that the bond wire shunted out the common mode transient and the ferrite/and the looped data cable delayed the shot before it got to the card, long enough so it was dissipated.

There was an IBM engineer who said simply tying a knot in the power cord would stop some common mode transients. We didn't really know if it worked but you usually saw a few knots in the power cord of IPR's machines after that.

A knot is a mix of the cable going in both directions within the knot (depending on the topology of whatever know was used). The magnetic field will effectively cancel out at the inner parts of the knot, and just wrap around it on the outer parts. So the current would just flow through the knot easily. A coil would be better. At best the knot would reduce some UHF and higher RF.

Trumpy makes a good point - that when the surge protection works, it could need to be replaced. I suppose it must also be robust enough to survive the duration of the surge. A lightning strike can't be safely dissipated inside a house, so it doesn't need to survive that.

With a power bar, the return path to ground (or neutral if conducting to ground isn't legal) can be quite long. Might it be too long to be effective?

The theory of the point of use protector is that it shunts transients within the protector so you need a lot of joules in the rating. Basically you are just heating up an MOV or two. The ground connection is just there to provide a path for that delta. You still should be just burning off the transients that were generated within the building there though. The bulk of the incoming transients should have been stopped at the Dmark or the service disconnect panel where you have a very short grounding path. I would drive a rod there, even if I had a Ufer, just for a local ground reference (bonded to the other electrodes).

I have had 2 direct hits to the mast where my weather station is mounted and all I ever lost was the serial port in the PC that runs it. That was the first strike. After working on that line (ferrites and a "phone line" style protector in a strip), the second strike left me unscathed.

You can make some fairly good protectors yourself if you carefully match the MOV to the expected line voltage.
Hosfelt will sell you them in all sorts of different voltages and joule ratings.

And everything else. The MOV has a low resistance when the voltage is high, so the dissipated energy is low. Unless a destructive fault current flows (like a direct strike with no alternate path to ground), the MOV is going to hold the voltage drop at the clamp voltage where it goes back down in resistance. The conductors will dissipate a lot of it, too, because it is current flowing. A fuse or breaker in the circuit to the POU protector could blow/trip, as well.


I agree.


Yup! But you won't have the advantage of it being a listed protective device and the insurance that comes with some protectors. Your own insurance might not pay out in the event of a surge too large to be protected, claiming that it wasn't so large and your unlisted protector was faulty.

I think I may need to do just that, though, because I have yet found a POU protector for a L-L 240 volt circuit (NEMA 6-15/20P and a few 6-15/20Rs although I'll take C13/C19 as an alternative). 330V MOVs on L1-G and L2-G, and 660V MOVs on L1-L2. I do have some power strip boxes with the traditional looking receptacle in the box in 5-15R form. So I just need to find 6-15R in that form, and rewire the guts for 240V and L+L+G circuit type.

I like the idea of a whole house protector that includes cable and phone surge protection. That and a point of use type protector at each piece of equipment.
We had a nearby lightning strike sneak into our house via the outside overhead phone and cable lines a few of years back and it really raised hell. It damaged the pairs in one of the phone wires in the basement and burned up a cordless phone base in the upstairs bedroom as well as our kitchen wall phone/answering machine, a 56K modem PC card, a small APC UPS unit and a couple of older 19" CRT TV’s. Our HO insurance paid the claim promptly, so at least that part went smoothly with no hassles.
I notice that most if not all surge protector manufacturers claim that in addition to the regular warranty, their product usually include some rather substantial amount of insurance protection against damage or loss, but has anyone ever actually collected from these or are they just a marketing tactic?

That insurance is buried in the price somewhere. I have never heard of anyone actually having to get it honored so I am not sure how well it works. We had a deal with Panamax (one of the first multi input protectors) and sold them. We did get a couple of Panamax units replaced but the equipment survived.