Surge Protector and MOVs

[bostonman]

I've noticed electronics incorporate surge protection into their designs and I've done the same on a recent power supply.

Is it worth using surge protectors since the electronics already has them? Also, does a real difference exist between surge protector models?

[TimNJ]

Not all mains powered products have built-in MOVs. It really depends on what standards the products were designed to meet, and the age of the product. Many products have them, but many others do not.

In general, I think it's worth using a surge protector if it's powering equipment you care about ($$$). I guess that goes without saying.

Yes, there are differences. Cheap surge suppressors might have only a few paralleled MOVs. After a few line surges, their clamping ability will degrade. On the other hand, a top-notch suppressor will have many more MOVs, allowing a longer life. They may also have a thermal protection circuit to keep your house from burning down. (MOVs have a nasty tendency to fail and catch fire.) They may also include line filtering components like common-mode chokes, X-caps, etc.

Tripp-Lite ISOBAR is probably the best surge suppressor on the market.

[rsjsouza]

Tripp-Lite ISOBAR is probably the best surge suppressor on the market.

I haven't heard about this before. Interesting article:
https://www.tomshardware.com/picturestory/684-tripp-lite-isobar-tear-down.html

[helius]

Is it worth using MOV surge suppressors when a device already has them? That depends on whether the clamp voltage is lower than the MOVs in the device. The MOV with the lowest clamp voltage will activate first and absorb the surge, while the others do nothing.

Because MOVs are sacrificial devices, the most effective and reliable surge suppression requires them to be very large, larger than would be installed in a device or even in a power strip. The first line of surge suppression should be at the distribution panel, with the lowest clamp voltage.
As a backup you can have additional, small MOVs in power strips as long as they have higher clamp voltages. As a last resort protection, the MOVs in the device should have the highest clamp voltage.

What is a reasonable clamp voltage for a 120 VAC line? You need to have a reasonable margin between normal power and the point where the MOV activates, or it will leak current, get hot, and burn out. The peak of a pure sine wave at 120 VAC is about 170 VDC, and it makes sense to leave 100% margin for noise that is superimposed on this wave (MOVs cannot filter noise). So 340 VDC clamping would be the minimum, and should be applied at the largest surge suppressor at the distribution panel. In practice, this value can be somewhat higher. The power strip MOVs should clamp higher, at perhaps 380V, and the equipment MOVs should be 400V. If they do not increase in clamp voltage as you get closer to the load, they don't serve a purpose.

[coppercone2]

I use isobars because well, they are not built like shit. Its decently rugged metal construction and I am pretty sure the MOV bank is kapton wrapped.


The best surge protectors though, are the ones that power off with a contactor (meant for remote control with a relay or switch control. They have a seperate control transformer. They are meant for rack mount operation but there is no reason why you can't screw them to a plank of wood.

The best part? super spacious inside. You can put whatever kind of extra goodies for surge protection you want in one of those suckers. GDT, more movs, passives... roomy

Bad? It thunks hard when you turn it on because of the contactor. Sounds like someone hitting something with a hammer. Also, sometimes they power down when there is a brown out because the contactor slips. So it will thunk on you (unlike a mechanical switch). If you want the best of both worlds, put a power switch across the contactor, so you make the circuit with the contactor then short out the contactor with the mains switch to save the switch life. If you put it under a table it can sound a bit scary when it hits. You can also fit a arc supressor in there across the contactor to trick it out further.


Proper term: mains distribution unit (MDU)


How to make it work in the home under a table : use a saw (recommend band saw) to cut off the handles. Use a file to debur it and take it apart and drill holes in it to attach it to a plank of wood slightly bigger then the MDU to make a flange. Or make a real flange out of metal (be sure to ground it).

[bostonman]

They may also have a thermal protection circuit to keep your house from burning down. (MOVs have a nasty tendency to fail and catch fire.)

Really? This is concerning. Why are they a common component for protection in new electronics such as TVs? Anything to safeguard this from happening?

The Isobar is something I've never heard of, however, it seems a bit bulky for things like wall mounted TVs. Currently my TV is plugged directly into the wall and I'm trying to figure out the best method for protection. I always got sucked into those nice APC () large plastic surge suppressor/outlet strip where it's guaranteed protection from X Joules.

As for MOVs and placing them in parallel. When I designed a simple power supply, I was unfamiliar with such protection and decided to include something, so I read to place three MOVs: hot to neutral, hot to Earth ground, and neutral to Earth. From reading, this was overkill, but, I got them as a sample, and figured I'd use them. Now I'm discovering that I should use more in parallel? If I'm understanding correctly, then I could ideally use six, nine, twelve, etc... (three in parallel connected to the mains as listed above)?

[TimNJ]

I think for the most part, they are okay as long as there is a suitable fuse before the MOV. If the MOV fails catastrophically and starts drawing a crap ton of current, the fuse should blow relatively quickly, or at least fast enough to prevent the MOV from lighting on fire.

In cheap surge suppressor strips, the problem is that the MOVs are sitting right across the line with the only downstream protection being a circuit breaker at the panel, or a circuit breaker in the strip (if there is one). Depends on the installation, but if thermal/bi-metallic breakers are used, those can take many second before they respond. That might be too long. The MOV might already be cooked.

[bostonman]

This is good information. I always thought the Belkin Surge Suppressors were great based on how well built they appear. Maybe this isn't the case though since mine (an older one) that's used on my PC doesn't have fuse protection, and displays a pretty green LED stating the protection is still working (although I doubt the MOVs are actually measured for having failed).

I think many surge suppressors offer noise suppression or whatever. My friend would spend much more money on these type for his stereo system to clean the 'dirty' AC noise. I would tell him it was a waste of money and the filtering is done inside the electronics (stereo, amplifier, etc...).

This being said, is using one with a common-mode choke worth it? My opinion: each additional component adds a greater risk of failure/fire. So personally I'd rather keep (in this case) a surge suppressor simple.

Two design questions though: should multiple MOVs be used in parallel across each line (hot-neutral, hot-Earth, and neutral-Earth)? If so, should they all be the same size?

I never heard of GDT's until reading these posts and did some research on them. Seems the risk of fire is minimum, and assume these are better than MOVs, but guess they are used less because of size and cost, is this true?

[TimNJ]

If your surge protector has a "protection active" indicator, then it's probably pretty well protected against catastrophic failure. It probably has a thermal fuse near/touching the MOV(s). When it heats up enough, it breaks the circuit before it releases the magic smoke.

The attached image shows a typical implementation, taken from the Littelfuse TMOV datasheet. TMOV, by the way, is a good way to simplify your design; It incorporates the thermal fuse directly into the MOV.

Belkin and most name brand power strips probably won't burn down your house. But, a shoddy strip without adequate protections can be quite dangerous.

I've never used MOVs from L-to-Earth or N-to-Earth. I'm not sure how much it would really help. I think Line-to-Neutral is probably most important. (By the way, for MOVs connected to earth, you may need a GDT in series for reasons I can't quite remember.)

You can generally use as many MOVs in parallel as you want; Maybe just be cognizant of how you are going to protect them all.

Done correctly, I don't see how adding a common-mode choke and some properly rated capacitors would increase chance of failure. Those are pretty safe components. Just make sure to use X-rated and/or Y-rated capacitors where applicable and make sure you have adequate clearance/creepage between windings on your common-mode choke, if you are hand winding.

[rsjsouza]

This is good information. I always thought the Belkin Surge Suppressors were great based on how well built they appear. Maybe this isn't the case though since mine (an older one) that's used on my PC doesn't have fuse protection, and displays a pretty green LED stating the protection is still working (although I doubt the MOVs are actually measured for having failed).

Complete anecdotal evidence but, in my experience, the uglier the power strip the safer it is. I have seen really fancy power strips with absurd "protection" circuits and the usual ugly ones have more decent stuff built in.

I think many surge suppressors offer noise suppression or whatever. My friend would spend much more money on these type for his stereo system to clean the 'dirty' AC noise. I would tell him it was a waste of money and the filtering is done inside the electronics (stereo, amplifier, etc...).

Anyone can spend their money in whatever they want, but you are right that some filtering is naturally done due to the physics of transformers in power supplies - that is entirely dependent on the nature of the noise in the power lines. If the noise is of higher frequency due to other appliances, high power motors, etc., common-mode chokes have a more critical role in dampening it. However, if the voltage has a very large and somewhat slow swing (several 50/60Hz cycles), then a surge suppressor would not help much - too little energy is stored in its inductive and capacitive components.

In ancient days there were voltage stabilizers used with sensitive electronics (early TV and radio sets) that were true autotransformers with sense circuits that switched the taps depending on the input voltage. Nowadays these are called "power conditioners" and are still required in areas where voltage wildly varies - usually on installations at the end of the line, rural, ancient, or supplied via a generator.

This being said, is using one with a common-mode choke worth it? My opinion: each additional component adds a greater risk of failure/fire. So personally I'd rather keep (in this case) a surge suppressor simple.

See above.

Two design questions though: should multiple MOVs be used in parallel across each line (hot-neutral, hot-Earth, and neutral-Earth)? If so, should they all be the same size?

That depends on the distribution system, but the Neutral-Earth could be, in theory, smaller than the other two. However, given that other factors can cause the neutral line to float and bring it too close to the live wire, I would simply use everything the same.

I never heard of GDT's until reading these posts and did some research on them. Seems the risk of fire is minimum, and assume these are better than MOVs, but guess they are used less because of size and cost, is this true?

GDTs are slower to react when compared to MOVs, which are slower when compared to fast diode/transistor-based clamping systems (such as the ones inside a DMM, for example).

All in all, modeling a surge is complicated and therefore the solution are tailored to cover the widest range of possibilities, thus you see a combination of different safety mechanisms.

[bostonman]

I've used MOVs once, and watched a tutorial (through the company I got them from) on 'calculating' a proper MOV. If I remember correctly, it suggested three (hot to neutral, hot to Earth, and neutral to Earth), so that's what I used.

It didn't make sense to me at the time, and didn't really care if it added an extra $0.50 to the cost of my design because it was a one off. Ideally though I would like to learn the proper way to include such protection, however, as it was mentioned, modeling a surge is complicated. The power supply I designed was very simple, but I was practicing with proper design techniques without making it too complicated.

From what I just learned, I had good intensions, but it had lots of flaws. I was watching a YouTube video yesterday on spark gaps and didn't know anything about them. It went on to explain those blue capacitors, etc... Also, I need to research what X-rated and Y-rated capacitors are. hahaha  I guess lots more knowledge is needed before I design another power supply.

[Kjelt]

That is pretty much how i did it,
However i put a fuse just before the mov from fase to neutral, in case the mov breaks down damage is restricted.
Then on the earth line from the two movs i put a sparkgap in series to protect for lightning.

[T3sl4co1l]

This being said, is using one with a common-mode choke worth it? My opinion: each additional component adds a greater risk of failure/fire. So personally I'd rather keep (in this case) a surge suppressor simple.

Right, it probably doesn't matter, at least as long as the attached equipment is FCC/CE compliant in the first place.

It could even make things worse, i.e., power line communication anything.

Inductors are a not-uncommon sight in surge suppressors because of the added impedance, particularly air-core or rod-core types (which retain reasonable inductance at high peak currents).  More of the surge is dropped harmlessly across the inductor, reducing peak current and energy dissipated by the MOV(s).  Common mode chokes may or may not be useful in this regard; I would expect normal mode (i.e., independent inductors on each line) to be most useful.  Therefore the CMC could probably be removed if filtering isn't so important.

Tim