Clamping voltage on surge protector?

[Uncle Peter]

"The theoretical lowest possible let-through voltage for 120 V power lines was 180 V. New technology, high quality surge suppressors can now clamp voltage at 130 V."

[Andrew Gabriel]

Not a short spike. It won't get past the SMPSU storage capacitor. (Power factor correction circuit might be slightly more vulnerable.)

A MOV won't handle anything longer than a short spike, without risk of failing in an unsafe mode.

[Uncle Peter]

Ok, I'll assume I've protected the computers well enough then. I've got a spike arrestor on the phoneline too.

Oh. I assumed if it shorted it'd blow a fuse or breaker before it melted. I mean if you short a wire in a mains plug, the fuse blows before any hint of a fire.

[Capitol]

Yes, but it may well be too slow to protect the circuitry. Protection is a risk reduction process, not a risk removal process.

[Uncle Peter]

I guess it would need to be use in conjunction with other devices. What about a crowbar along with a MOV set to 350V?

[Capitol]

If the source impedance is low enough. Also, for many units 130V is the level at which clamping starts, the let through voltage can be much higher if it is a varistor system. Large Zener diodes have a lower series resistance, but the clamping current can destroy the device if large enough. Also, Zener diodes do not like RF energy, so lightning strikes can destroy them at quite low levels of energy (punch through occurs).

[Capitol]

Correction If the source impedance id high enough. Crowbars only protect to an extent, if the fuses arc through, the crowbar will self destruct. That's why fuses have a breaking capacity rating.

[Andrew Gabriel]

The first minicomputers I worked on were switched off by crowbarring the DC supplies. The original systems supported core store which is non-volatile, but you don't want it to get scribbled on by the computer collapsing in an uncontrolled way as the various voltage rails decay. So, once the supply controller had decided the computer was powering off (which lots of events could trigger), it told the CPU so it could save the current registers to memory, and then it crowbarred the supplies by shorting them out with large SCRs in a controlled sequence, designed to protect the core store contents.

When you powered it up, if it was fitted with core store, it would carry on running from where it stopped.

[westom]

was 180 V. New technology, high quality surge suppressors can now clamp voltage at 130 V."

[Mike Tomlinson]

In article , Harry Bloomfield writes

Time to trot this out again :-)

This is a photo of two failed 13A power strips. Each has two neons, one to show 'power on' and another to show 'surge protection working'.

The 'SP working' neons were still lit despite the MOVs having vaporised :-)

[Andrew Gabriel]

They've been responsible for some fires in the US, and I believe UL are cracking down hard on MOV devices.

I know we had to stop using them in computer systems we manufactured sometime around 1980, due to regulatory changes due to fire risk.

In power strips, they are a gimmick. Power strips are produced for under £1. The MOV devices cost in the order of 10p in bulk, but the addition of this 10p device enables you to market the power strip for at least £5 more. Hence you can ship a higher margin product, even though the added value pretty much useless.

I generally make a point of making sure power strips that I buy do

*not* have any MOV devices in them.

[Graham.]

Its reassuring to know that my view on these things is shared by someone as knowledgeable as you.

Here are my pictures

What I was trying to illustrate here is how the L & N use the MOV PCB as a junction block, relying on two blobs of best lead free solder.

If you were designing such a thing wouldn't you ferrule and spot weld the incoming mains *directly* to the bus bars, and hang the PCB on the bus bars separately?

[Uncle Peter]

Learn what those numbers are actually report. Normal minimum let-through voltage for 120 volt protectors is 330 volts. Normal let-through voltage for 230 volts would be almost double that. 180 volts is reporting something completely different.

Also appreciate what a protector does and does not do. For example, excessive voltage by an electrician miswire would only destroy a protector

- do no protection. And if the protector's thermal fuse (not to be confused with a 13 amp line fuse) does not trip fast enough, then a fire may occur.

That protector only claims to protect from one type of anomaly that typically causes no electronics damage. Due to robust protection already inside appliances. Robust protection that even says why 650 volts during a surge does not damage electronics.

The many who did not know 650 volts is a standard should cause you pause. Since so many made recommendations without even knowing basics.

Finally, typically destructive surges can be hundreds of thousands of joules. A protector adjacent to an appliance can only stop that current or absorb that energy. How many joules does that protector claim to absorb? Compare that number to hundreds of thousands of joules. Then ask yourself, "What does it claim to do?" That should also give you pause.

You may want to learn of a completely different device (also called a protector) that would address your concerns.

Surely if the surge protector could just short the line to anything that was over the 240V RMS, it would blow a fuse or breaker, and not cause a fire?

[meow2222]

The designer & manufacturer of the PC PSU has done that.

Overhead phone services already have those built in.

Youre buying into marketing bs and wasting your time.

NT

[Uncle Peter]

Then I guess the 4 quid surge protector is protecting several 40 quid PSUs.

Then why did my dad's network card burn out when he got lightning through his phone line, through the router, and into the ethernet port?

[meow2222]

they dont protect anything, other than someone's profits

Presumably because it was hit by lightning. No tiny surge arrestor can do anything about that.

NT

[Uncle Peter]

At 4 quid I very much doubt there is much profit. That includes postage and packing and packaging and Ebay fees.

It might do enough to reduce the effect.

So, why did the plug in surge protector melt when at my 2nd last place of work an electrician put 2 phases onto a circuit instead of one? If they do nothing below 650V, it should have just sat there.

[jrwalliker]

Computer equipment sold in Europe needs to comply with EN55024 (Information technology equipment - Immunity characteristics - Limits and methods of me asurement) to obtain CE marking.

This requires the following immunity on the mains input: Surges: 1kV peak L to N, 2kV peak L or N to E Fast transients: 1kV peak at 5kHz repetition rate.

Surges and transients are applied with both polarities and random phase rel ative to the mains waveform, so some of the surges are on top of the peak m ains voltage.

Immunity in this case means that the equipment continues operating correctl y after the test and that stored data has not been corrupted during the tes t.

Therefore, there is no point in clamping just above the mains voltage. A 6

50V clamping device is good enough. However, I agree with the other poster s who make a point of avoiding TVS "protected" mains adapters which don't d o anything useful but which can explode under extreme conditions. There ar e some very substantial DIN-rail protectors which protect the whole propert y and might be useful, but they are very expensive.

John

[meow2222]

and packing and packaging and Ebay fees.

Of course there is, or they wouldnt be sold. Its basic capitalism principle s. You can get them in retail quantities for 50p last time I looked.

ugh his phone line, through the router, and into the ethernet port?

do anything about that.

You've got sometimes a billion plus volts crossing over a mile of air. You' ve got maybe 300,000 amps. You've got a discharge path exceeding 50,000 deg rees C. Do you seriously think a 1cm disc of any material on earth is going to make the slightest difference?

work an electrician put 2 phases onto a circuit instead of one? If they d o nothing below 650V, it should have just sat there.

who cares? All you get for your money is a fire risk. Even the insurance th at often comes with them is too often worthless, and at best worth very lit tle. Its just a game, a profitable game.

NT

[Uncle Peter]

It's not very much profit though is it?

So lightning protectors are a con? And they are legally sold? I don't think so. Otherwise I could make up any old shit and stick it in a plastic box and sell it.

It disproves the 650V shit.

It probably helped trip the breaker by causing a short.