SURGE SUPPRESSOR

In a MOV's usual wear-out mechanism they typically fail shorted. When the MOV fails it blows a fuse and with the fuse blown the LED goes out.

It's possible (but rare) for the MOV to fail open (eg. one of the leads breaks due to external vibrations), If that happens the fuse will not blow and the LED will stay lit.

Thanks for that info. So when purchasing, how does one tell a good Suppressor from a not good ?

-or- all they all about the same ? (as long as the Joules are the same) Or can you just open it up and add as many MOVs as you feel needed ?

Sid.

The ones that offer a warranty/insurance policy on the equipment connected to them up to X-thousand dollars might be good place to start

The better-quality ones have two LEDs, one that says "Grounded" and one that says "Protected" and I would imagine some internal circuitry to determine whether this is actually so.

Why bother?

Read carefully what that insurance policy actually require you to do, since you might need to upgrade your electric system, such as improving grounding or even add spark gaps at the mains entry :-). After this upgrade this add-on suppressor might not be needed at all.

Those surge suppressor that are plugged in the socket protects only equipment (or extension cord) connected to it suppressor , but may fail miserably, if there are other connections e.g. to telephone, CATV or antenna systems or other devices connected to a different mains socket (even it has a same kind of suppressor).

One doesn't. Brand loyalty is good, but you cannot usefully test components for a once-in-a-lifetime lightning stroke (and even if you could, the surge can jump from adjacent wiring and completely bypass a single 'suppressor'.

You can be sure that a reputable manufacturer won't give his products a lightning-test before selling them to you...

In the US, MOVs are connected H-N, H-G N-G.

Excellent information on surges and surge protection is at

"How to protect your house and its contents from lightning: IEEE guide for surge protection of equipment connected to AC power and communication circuits" published by the IEEE Some of the information is somewhat specific to the US

From pdf page 35, the is no standard method (for instance by UL) to measure Joules (at least in the US). Some manufacturers use deceptive methods. As a result some other manufacturers don't list Joule ratings. Clamping voltage and surge current ratings can be used to compare. A lower clamp voltage is not necessarily better. It may degrade MOVs when the voltage is not harmful to the connected device.

I buy devices from a major brand.

As they fail, MOVs start to conduct at a lower voltage. They finally start conducting at 'normal' voltage, resulting in heat and run-away to a short (as above). The disconnect is likely partly triggered by this increased heat.

Adding MOVs to an existing product would re-engineer the protection mechanism. Protectors with high ratings are readily and cheaply available

Damage to equipment may result from high voltage between power and signal wires. This is shown in the IEEE guide on pdf page 40+. If a plug-in suppressor is used, all wires to a set of protected devices need to go _through_ the suppressor. If not, an insurance claim would certainly be denied if. I suspect a claim would be denied if there was not damage to some of the protection in the suppressor.

Francois Martzloff was the surge expert at the US-NIST. He investigated how much energy might be absorbed in a MOV in a plug in suppressor. Branch circuits were 10M and longer, and the surge on incoming power wires was up to 10,000A. (That is the maximum that has any reasonable probability of occurring and is based on a 100,000A strike to a utility pole adjacent to the house in typical urban overhead distribution. Only

5% of strikes are stronger.) The maximum energy at the MOV was a surprisingly small 35 joules. In 13 of 15 cases it was 1 joule or less.

There are 2 reasons the energy is so small. One is that at about 6,000V there is arc-over from the service panel bussbars to the enclosure (this appears to be a 'feature' of US panels). After the arc is established the voltage is hundreds of volts. Since the enclosure/ground/neutral are connected to the earthing system (US) that dumps most of the incoming surge energy to earth.

The second reason is the impedance of the branch circuit wiring. A surge is a very short event so current components are relatively high frequency. The wire inductance is more important than the resistance and the branch circuit impedance greatly limits the current to the MOVs, which greatly limits the energy.

The maximum was not even for the largest surges. The largest surges forced the voltage at the service above 6kV and arc-over. For some smaller surges (with the shortest branch circuits) the MOV at the protector held the voltage at the panel below 6kV and there was no arc-over (but the voltage was higher than an arc voltage.). One of them resulted in the maximum energy of 35 joules.

bitrex bullshitted:

==============

** Nothing wromg to see there.

** It make no actual claims other than ones about itself.

1. The thing is "splash proof" so will not short out or shock just because of light rain.
2. Only 3 pin plugs can be used with it and it has a current overload breaker like all such panels do.
3. The A and N pins are shuttered.
4. It has a MOV inside for good luck.

..... Phil

The professional method is to have proper spark gaps in the main distribution panel with known turn-on voltages. Better yet, have the spark gap outside the house.

In countries with long mains lines, the customers at the end of the line is in the worst situation, since over voltages are largest. Think about an open transmission line, in which the forward wave is reflected back from the "open" end of the transmission line, increasing the voltage. Getting spark gaps at into the last pole of the long line would be the best solution. You could try to get the utility company to install such devices and even offer to pay for the installation.

With a "standard" 8/20 us lightning pulse the slew rate of 1 kA/us and

10 uH line inductance, the voltage drop is 1 kV, greatly simplifying the job for the MOV. In fact in large sites, a stepwise protection system is used, with spark gaps at mains entry, some lower voltage protection at intermediate panels and finally MOVs close to or in an equipment. These all rely on wiring inductance+resistance between each protection step There is no point of connecting MOVs directly in parallel with a spark gap, the MOV will be fried quite quickly.

The claims on this other one by the same manufacturer go a bit further:

"IPX6 Waterproof and Shockproof"

IP standards don't have anything to do with protecting the user AFAIK

"Worry-free power for anywhere near water."

Sort of like worry-free skydiving?

IPX6 is more than a "water splash" it's high-pressure jets. I don't think they do this test with anything connected to it or powered up they're just looking for water intrusion in its "natural state" with all the shutters closed.

Even strips with integrated GFCIs don't claim to be "shock-proof"

Beats me. Amazon don't care.

This other one advertises "shock proof":

A surge protector that also offered integrated GFCI would be nice but they seem hard to come by as these are somewhat contradictory goals. Maybe putting the surge protecting device upstream of the GFCI would work in some cases.

Yes, over distances like a hundred miles the electrical length of the line at 50-60Hz is enough to cause reflection-effects like with RF in an unterminated cable and give a nice boost at the end, there's an analysis in e.g.:

Maybe it's of unknown provenance. I came across some cheap Chinese multiple adaptors for 240V use (as they claimed anyway) and none of the earths were connected to anything, despite having 3 pin sockets for live neutral and earth. And they came from a well known DIY chain in the UK too. Natch I reported it and they got withdrawn/recalled.

Lightnings are RF. Just listen with a radio at LF/MF bands and even up to lower part of the HF band and you can hear crackle all over, especially in the summer.

While the "standard" lightning current pulse is 8/20 us, some sources claim a 3 ,s rise time, i.e. 10 us cycle time or 100 kHz frequency. This translates to a 3 km wavelength.

In the 230/400 V world low voltage distribution lines can be longer than a kilometer, which is a significant part of the wavelength and transmission line issues must be observed. Since the power line RF impedance is a few hundred ohms and one or both ends of the transmission line has an open circuit., the RF pulse will reflect a few times back and forth, until attenuated by resistive losses or flashovers.

I have never heard of or seen spark gaps in, or added to, panelboards here (US) other than what appears to be an intentional design to arc-over. A short duration 10kA max surge is well within the capabilities of MOV based protection.

Also don't know of spark gaps in bigger commercial/industrial service switchgear.I have seen MOV based protection. The max surge should still be something like 10kA (less because of 4 wire 3-phase, more because transformers only supply one building).

It is common to have a lightning arrester across the primary (distribution side) of pole mounted transformers. They needed to be there to protect the transformer. Rural overhead distribution lines have lightning arresters. I saw a stretch that had an arrester on every other pole. A line-to-earth arrester will drastically lift the ground potential at that pole (crappy earth electrode).

I think the analysis by Martzloff that found a maximum probable 10kA surge per service wire had a strike to the primary wire and an arrester across the primary of the supply transformer. The surge was (H-N-H) to earth at the transformer (surge current lifted 'ground' at the transformer, as above). Earthing the N at the service results in H-N surge.

Arresters may not be a simple gap - that can result in "follow-on" arc from normal power immediately following an event.

Don't know if "mains lines" refers to primary side, secondary side or both.

For instance Phoenix Contact used to make spark gaps for 230/400 V and

400/690 V. They had blow holes to exit the hot plasma, so mot nice in an EXplosive environment :-).

While we visually see a single lightning, it usually consists of multiple bolts within half a second. The surge protector of any type must be able to handle these multiple bolts without overheating.

Ya, the same math of course applies to "shorter" transmission lines with higher equivalent frequencies

I thinking about what happens if a surge is generated internal to the utility from e.g. a switch-over or break in a long-distance high voltage transmission line; if those are always arrested at a substation prior to being coupled into low voltage distribution or if they can propagate down to the user. Those lines must be very dispersive at high frequency over tens of miles so the RF from a lightning strike or transient of a cut-out would disperse over that distance I would think but there's still the possibility of transmission line effects from the low frequency/dispersed transient over that distance