Are mains surge protectors needed in the UK?

[Jonathan Buzzard]

Really?
Please explain what device GD1 in this circuit is, then:
http://jasper.org.uk/w_tom/cct.jpg

Here's a photo of the device itself, to prove it doesn't just exist on
paper. I am using three in this room at the moment. They cost me ukp4
each: http://jasper.org.uk/w_tom/surge.jpg

This protector also protects against common mode surges, something you
claim plug-in protectors cannot do (see the diagram if you don't believe
me.)

This is most amusing really. He stupidly claims something is two expensive
when it is cheap, that it is not in any products when it is required to
meet the British Standard, and when someone has just posted a circuit
diagram clearly showing one. The stupidity of the man beggars belief.

I note you have studiously avoided answering my question about whether
you are employed by, or have an interest in, the manufacturers of the
whole house protectors you constantly advocate. I therefore conclude
your so-called "advice" is biased and partial.

I think that we have to conclude that he does. I notice unlike just about
every other person in the thread he appears to be afraid to show his full
name. I wonder why?

Finally, I have spent some time today on google.groups reading threads
in which you participate. In almost every one you come off worse. Why
do you think this is? Has it ever occurred to you that you may be
wrong?

Does he participate in anything other than discussions on surge protection?

JAB.

 

[Jonathan Buzzard]

With basic knowledge of earthing, then you knew that every
floor on the EMB is, essentially, earth grounded. But then
you knew that from the last time we had this discussion. Did
you forget?

Argh, but according to the doctrine of w_tom, an earth ground is only any
good it is less than three metres in length, so how to you ground even the
tenth floor?

I know exactly what you said last time, and I still maintain that their is
a fundamental logic flaw in the totality of your arguments.

Please tell us how switching power off a British ring main
with so much 'inductance in wires' causes a massive surge? I
am still waiting for that science miracle to be explained. I
am still waiting for your numbers on that wire inductance and
the resulting surge.

I am going to do it with a 100W light bulb, because everything is
much neater, and therefore provides a better example. The reality remains
that a ring circuit has inductance (a bit of straight wire has inductance)
and therefore provides potential for back EMF.

JAB.

 

[Jonathan Buzzard]

On Sun, 11 Jul 2004 17:38:04 -0400, w_tom wrote:


[SNIP]

An MOV does not get hotter only because its let-through
voltage changes; is below 300 volts or above 360 volts. Lets
remember those numbers posted up to. Surges are microsecond
events. If they damage the MOV, then milliseconds of AC
electric cause the thermal link to fail. But a grossly
undersized MOV short circuits. Milliseconds later, AC
electric heats that damaged MOV, pushes too much current to
blow that thermal link, and illuminates that warning light
LP1.

You are forgetting the GDT w_tom I know you don't like them, but they
are cheap, and they are fitted to surge protectors. As a result the MOV
only has to be good enough to deflect the leading edge of the surge till
the GDT takes over to deflect the bulk of the energy in the surge.

JAB.

 

[VWWall]

I think that we have to conclude that he does. I notice unlike just about
every other person in the thread he appears to be afraid to show his full
name. I wonder why?




Does he participate in anything other than discussions on surge protection?

He's an expert on 3 1/2 digit voltmeters. ;-)

Virg Wall

 

[David Maynard]

[snip pointless attempt to reason with half_pint]

David, you're wasting your time arguing with half_wit.

I'm beginning to see your point.

Rather amazing, actually. I mean, even when one points to authoritative
sources he simply declares everyone else ignorant or just plain dumb.

Makes me wonder how people like that arrive at their 'beliefs'.

 

[David Maynard]

That article is so filled with hysterical hyperbole that I can't tell how
much, if any, of it deserves serious consideration.

It was amongst the first hits I got when I googled for "third world
computer recycling" or similar. The original article I was thinking of
was printed in the Guardian, a liberal UK newspaper. These reports are
more even-handed:

http://www.guardian.co.uk/weekend/story/0,3605,849530,00.html

http://www.guardian.co.uk/international/story/0,3604,725756,00.html
[/QUOTE]

Yes. I am aware of the 'recycling' issue itself. It's the hyperbole that I
mistrust.

 

[half_pint]

"Johannes H Andersen" <[email protected]>
wrote in message

"Johannes H Andersen" <[email protected]>
wrote in message

half_pint wrote:

"Johannes H Andersen"

wrote in message


half_pint wrote:

[...]

As said before, voltages also kills semiconductor devices. Why

do
you

think there are such things as electrostatic bags?

To protect semiconductors from static electricity as opposed to
mains electricity.

LOL! How much do you know about electricity?

I have forgotten far more than you will ever know.

Too bad your memory is like a sieve. You might have heard about electric
potential? Measured in Volt in both static and mains cases. A semiconductor
such as a transistor must have the atoms organised in a particular way,
that result in energy bands the for controlled movements of charges through
the lattice. The electrons are not free as in metals or carbon, if

that
was

the case then the electrons would be able to move equally in all directions
and the semiconductor would not do its job.

If you now increase the electric potential over the designed value, then
these delicate structures and the energy bands breaks down. This is not
the same thing as burning a fuse.

They are not delicate structures, no more than a grain of sand is

delicate.

What are you on about warp_saint?

A grain of sand is made of the the same raw material, but it's not a
transistor! Transistor silicon is carefully grown into an almost fault free
lattice. Impurity elements are then carefully inserted to establish the
energy bands. That way P and N materials are created.

Tell me about it.

A zener diode is designed with a Zener knee Voltage for the desired effect.
however the Zeener knee is inside the designed Voltage range. If you exceed
the designed Voltage, you blow the Zener diode for much the same reasons
that you'll blow a transistor. If you ever have experimented with electronics,
you'll have come across many duff transistors etc with no physical sign of
burns whatsoever, just dead because you have used the wrong voltage or used
the wrong polarity.

I think you will find the zener diode is on the output side.
And what do you mean by blow it? It will probably not melt
untill it reaches 1000 degrees, I fancy the wires will melt first.

The only dead transistor I have came across had a big hole melted in the
middle of its casing.

 

[half_pint]

In the case of junction semiconductor devices the failure mode is most
commonly current heating of the junction and the problem with too much
voltage, in those case, is it causes excess current to flow. Same, but
worse, for reverse voltage as junctions that shouldn't conduct do, and vice
versa, which scrambles what the circuit is designed for (bias points are
nonsensical, current flows where it wasn't supposed to, etc.), causing
excessive junction currents and, hence, temperature.

'Electronics', as a generic term, is complex and composed of many device
type and process. And there are many different failure mechanisms,
depending on what kind of devices are being used.

Even the generic term "transistor" doesn't clarify whether it's a bipolar
junction transistor, or a "Field Effect Transistor" (FET), or MOSFET, or
another.

http://www.semiconfareast.com/fmechs_die.html

Even that article doesn't cover everything.

MOSFETs, which are sensitive destructive breakdown from excessive voltage,
are the most common device technology used in PCs.

An amusing story about the first days of FET transistor production
illustrates the problem. Before they realized just how vulnerable the
devices were the failure rates were astronomical, in every sense of the
word. Some PCB manufacturers could get a single board with the things on
them to work. ALL had failures.

The problem was the devices had no static protection in them and things
like conductive foam/tube carriers hadn't been though of yet (what for,
eh?). They typically came 'plugged' into a block of plain old styrofoam
and, as it turned out, the simple act of pulling them back off created
enough static electricity to destroy them.

One manufacturer, before the 'secret' was fleshed out, got so fed up with
the 'out of box' unreliability (100% board failure) that they decided to
pre test the FETs before insertion. So they constructed a device test bed
and sat a fellow down to check them one by one.

The procedure ended up being: pull device from foam block, insert into test
fixture, observe device is bad, throw device away." Not a single one
passed! (for the now obvious reason)

That test ended up being instrumental in finding the problem, however, as
the device manufacturer, in response to these 'claims' of failure (sure
pal), had already implemented a "100% device test" (so it can't be)...
before inserting them into the foam block, of course.

You describe a particticular circumstance where a large voltage is
induces on the device and is earthed by the technician so a large
current will flow.
When the devices are in a circuit board the situation is different and it
is more likey any charge will find a safer way to disapate ( they are
probably
designed in such a manner now any how).
Personaly I cannot recall a cuircuit ever being damaged by static.
Few things will survice several thousand volts going through them
for more than a fraction of a second.

 

[half_pint]

delicate.

Got that bit of information from a voo-doo doll, no doubt.

You think a voo-doo doll would withstand being heated to say 700 degrees?

In a previous post you made a comment about 'a little knowledge incorrectly
applied' and that is a beautiful example.

An example of you and your kind.

You do know sand is mainly silicon?
Try melting some sand some time, let me know how you get on.

MOSFET gate breakdown is not recoverable.

http://www.americanmicrosemi.com/tutorials/mosfets.htm

"breakdown voltage in MOS devices do not depend upon p-n junction stress
but rather upon the thickness and quality of the insulating oxide. When
breakdown does occur, the oxide is punctured and the device is destroyed."

I will get back on that tomorrow. your talking about break down of an
insulator anyhow,
not the silicon device.

 

[half_pint]

Research it and learn, or remain ignorant. Your choice.

Shut up.

Research electrostatic breakdown in semiconductors.

Post graduate research you mean?

true

heat strictly speaking.

I suppose you want to ignore modem lines.

I dont connect my comms line to the mains.

Depends on how high the voltage.


Some are; others aren't.

The fact of the matter is that you know nothing of which you speak.
bollocks.


There cannot be "whatever the input voltage" to the one and only device in
your statement, the zener, if it's "the same."


Well, no. That just happens to be the cheapest and most common means when
they're used as a simple voltage reference. It is not the only way they're
used.



You apparently haven't a clue.
sure

It's like saying silicone is 'essentially sand', and has the properties of
sand. Or that glass is 'essentially sand', and has the properties of sand.

One has to go back to before even the days of alchemy to find such
absurdities promulgated as 'knowledge'.


Simple enough statement.


That's probably how a medieval alchemist would feel about a discussion on
quantum mechanics, and for the same reason.

Are you on drugs?

It is obvious you don't and what's further obvious is that you refuse to
learn a thing about it.


I have already provided you with corroborating authoritative sources for my
explanations.

I challenge you to provide one for your absurd contention that voltage
cannot damage semiconductors.

If I have a billion volts at one end of a 1mA wire fuse and a billion volts
at the
other end will it blow?

Get some education.

Wrong, and I've provided authoritative sources to corroborate it.
Garbage.


If you thought so that certainly doesn't elaborate, but don't bother as
I've had enough of your absurdities anyway.

I explained above, voltage is relative,

You actually stumbled onto something that's halfway correct:

Its all correct anad i didnt stumble I knew where I was going.

you can have
localized heating. It's not, however, very relevant to the point because
the thermal conductivity of semiconductors won't allow so dramatic a
difference, when compared to your claimed '900F' destruction point, as to
matter to the illustration. So we need fools shoving them into 700F ovens,
instead of 900F ovens, before any device would ever fail. It is still an
absurdity.

Btw, the melting point of silicon isn't 900F, it's (much) higher, but I'm
using your numbers to illustrate that even your own 'theory' is

irrational.

I know it is higher but some of the doping agents are lower so I
compromised. :O)

Feel free to quote a text book, or some other authoritative source, that
supports your fantasies.

I have a brain so I need no text books.

Since you made no 'point' there is nothing it could.

probably went over your head.

Not at all. It has nothing to do with ICs being 'immune' from static
voltage damage but it's obviously useless to attempt educating you about
anything.

Its fact though.

I can too. It isn't 'fingers', it's whether your body has built up an
electrostatic charge and then you touch one.

Semiconductors, nowadays, also include protective devices inside them to
reduce the chances of ESD damage. that doesn't mean they're 'immune' from
it and it still proves the point of voltage causing damage.


An excellent description of yourself.



Which would be what?



Didn't find anything, did ya? <chuckle>

I looked elsewhere and there were plenty of other crooks flogging
"untested" items. " A customer probably returned this fully functioning
disk drive worth £20 out of the kindness of his heart."

 

[David Maynard]

You think a voo-doo doll would withstand being heated to say 700 degrees?

I doubt it but then I'm not in the habit of consulting voo-doo like you
seem to be.

An example of you and your kind.

You do know sand is mainly silicon?

Sure do. I also know a semiconductor isn't 'just sand'.

Try melting some sand some time, let me know how you get on.

We aren't talking about sand. We're talking about semiconductors.

I will get back on that tomorrow. your talking about break down of an
insulator anyhow,
not the silicon device.

Yes, the gate insulation, which is REQUIRED for the 'silicon device' to
freaking WORK.

 

[David Maynard]

You describe a particticular circumstance where a large voltage

And how did you determine just how 'large' that voltage was?

It won't get any larger than needed to pop the device because, at that
point, the device conducts.

is
induces on the device and is earthed by the technician so a large
current will flow.

No. It's the voltage which does it (and the current is quite minuscule)

Just as if you take a flat pane of glass, horizontally suspended from the
ends, and place a brick in the middle. Everything is fine. You place
another brick onto the first. It holds. Then another brick, and another
until, eventually, the potential energy from the brick weight is too much,
the glass gives way, and they fall through.

It is the voltage potential, as with the potential energy from the weight
of the bricks, which causes the gate insulation to give way, as did the
glass. And then current flows through the failed gate, just as the bricks
flowed through the glass when it gave way.

When the devices are in a circuit board the situation is different and it
is more likey any charge will find a safer way to disapate

I never said otherwise. That doesn't mean voltage transients can't get in
there and cause damage.

It doesn't take much voltage to pop an integrated MOS device.

( they are
probably
designed in such a manner now any how).

It is nigh onto impossible to make every 'circuit' invincible and much more
practical to clamp surges so they don't get IN to the more sensitive circuits.

Personaly I cannot recall a cuircuit ever being damaged by static.

I dare say there are lots of things in the world you haven't 'personally'
witnessed.

Few things will survice several thousand volts going through them
for more than a fraction of a second.

Well, we're making progress then.

Except that "volts" don't 'go through'. Voltage is a potential between
points. It's current that 'goes through'.

 

[David Maynard]

Shut up.

How elegant.

Post graduate research you mean?

Better to start a little more fundamental.

A simple web search will provide plenty of sources.

heat strictly speaking.

True enough.

I dont connect my comms line to the mains.

Good thinking.

Figuring that makes them, and the computer, safe from surges isn't good
thinking.

bollocks.

I feel the same way.

sure

The oxide gate insulation of MOS devices that we just talked about, for one
obvious example.

Are you on drugs?

Not yet, but you're giving me a headache so an aspirin might be in order soon.

If I have a billion volts at one end of a 1mA wire fuse and a billion volts
at the
other end will it blow?

We aren't talking about devices with 0 voltage across them so your question
is supercilious, at best.

Garbage.

Are you blind too, or is it that you just don't read?

I explained above, voltage is relative,

What you described above was a device with no voltage potential across it
but that has no value whatsoever to the matter of surges, which are large
potentials across the device.

It is, however, how you got your so called 'static shock' from the car
radio without it being damaged.

Its all correct anad i didnt stumble I knew where I was going.

You 'know where you are going' like Columbus 'knew where he was going'. And
the natives have been called 'Indians' ever since.

irrational.

I know it is higher but some of the doping agents are lower so I
compromised. :O)

IC manufacturers don't seem to share your 'compromise', but it doesn't
really matter.

I have a brain so I need no text books.

That would explain it.

probably went over your head.

You first have to 'shoot' for there to be a trajectory.

Its fact though.

I have no doubt you got a 'shock'. It's your 'who needs a textbook?'
analysis of what it means that's cockeyed.

I looked elsewhere and there were plenty of other crooks flogging
"untested" items. " A customer probably returned this fully functioning
disk drive worth £20 out of the kindness of his heart."

wag, wag, babble, babble.

 

[Michael Salem]

Someone wrote:

....

Newsgroups: uk.comp.vendors,uk.comp.homebuilt,alt.comp.hardware,alt.comp.hardware.pc-homebuilt
Subject: Re: Are mains surge protectors needed in the UK?
Date: Sun, 11 Jul 2004

Lines: 856

Never mind surge protectors, let's go for the record!

 

[Mike Tomlinson]

Does he participate in anything other than discussions on surge protection?

Not that I can see. He posted for a while on PC power supplies and how
they were supposed to meet some Intel standard for over voltage
protection, but soon returned to his bête noire.

 

[Mike Tomlinson]

Rather amazing, actually. I mean, even when one points to authoritative
sources he simply declares everyone else ignorant or just plain dumb.

Makes me wonder how people like that arrive at their 'beliefs'.

Voodoo? (grins, ducks & runs>

 

[Mike Tomlinson]

Why would I apologize to someone whose very first post to me
was to call me a twat?

Because you are?

Why would I apologize to
someone so foolish as to claim a "degraded" MOV causes thermal
links to blow?

Your reading comprehension is still astonishingly poor. I did not make
that claim; the manufacturer of the surge protection device I use states
this in their circuit description.

The thermal links are thermal fuses (another cheap, commonly used
component) sandwiched between VDR1 and VDR2. When (if) the MOVs begin
to degrade, they heat up. Eventually, they will heat up to the point
that thermal fuses open, removing the MOVs from the circuit and lighting
the warning lamp LP1. It's simple, basic electronics - beyond you of
course.

You have lightning damage on your Canary Island site
typically because the protection system is defective.

Stop twisting my words. We don't have damage on our CI site.

You are suffering damage in the Canary Islands

No we're not, and I at no point said we were. Stop lying.

However you had a contractor install an earthing
system and still had damage?

We didn't have damage. For heaven's sake, go back and read my original
post. I said only that advice was to abandon site in the event of a
thunderstorm. That's because the building fabric is metal, and it's in
an exposed position on top of a mountain. The advice is for safety of
personnel within, not because of the fear of damage.

Your desperation to prove your point leads to you lying.

[snip hysterical rant]

I can see why you top-post now. It enables you to avoid answering
questions put to you, a profoundly dishonest thing to do. Instead of
answering people's questions in context and thus having a meaningful
discussion, you rant on and on in a top-posted ramble.

I note that his time you carefully snipped my other questions off the
bottom of your quote. Do you work for, or otherwise, have an interest
in, the manufacturer of the whole-house surge protection devices that
you constantly advocate?

 

[David Maynard]

Voodoo? (grins, ducks & runs>

LOL

Novel idea

 

[Mike Tomlinson]

Elsewhere in this topic is a claim that fuses can prevent
surge damage.

That's half_pint's claim, not mine, and is the subject of a separate
subthread. Please stick to the discussion in hand. Diversion is a
classic tactic employed by those losing an argument.

So we ask the MOV professionals to define "degraded". From
a Littelfuse application note AN9310:

I suggest you consult a dictionary or attend that remedial reading
class. You'll see there is a difference between "degraded" and failed".

And what is required to melt that joint? That's right, heat. By the
time this has happened, the thermal fuse has opened and taken the device
out of circuit - before it has had a chance to vaporize.

To meet the human safety requirements imposed by UL1449 2nd
edition, manufacturers install a thermal link as demonstrated
in a schematic from Mike Tomlinson. When does that thermal
link open? When the MOV is degraded? Of course not.

Liar. When MOVs degrade, they heat up. This is clearly demonstrated by
the pictures in the link you posted:
http://www.zerosurge.com/HTML/movs.html

The thermal fuse in the design then takes the MOV out of circuit before
it can fail catastrophically.

He even provided a schematic to show us the thermal link that
blows when MOV degrades!

Yes, I did, didn't I? Instead of waving my hands around and getting
hysterical like you.

A 330 volt MOV degraded so that it operates at 300 volts or
360 volts will blow the thermal fuse? Of course not.

Of course it will, idiot. It'll heat up. Now go and look at the specs
of thermal fuses on the web.

As
demonstrated in a scary picture from Zerosurge, the MOVs can
even be removed and the indicator lamp says protector works
"OK":

Like I said earlier, those are cheap and nasty protectors made for the
American market, posted by a manufacturer with a vested interest in
denigrating the competition so they can sell their own, allegedly
superior, products.

Aa degraded MOV
will not open (blow) the thermal disconnect device LK1. But a
catastrophically failed (short circuited) MOV will.

Wrong.
From http://powerelectronics.com/mag/power_thermally_protected_movs/

"If an MOV is subjected to a sustained abnormal overvoltage, limited
current condition (as required in UL1449), the MOV may go into thermal
runaway, resulting in overheating, smoke, and potentially fire. For end
products to comply with UL1449, the MOV must have some level of
protection to prevent this failure mode. That protection has
traditionally been a thermal fuse or thermal cutoff (TCO) device."

Lets
remember those numbers posted up to.

Your reliance on "numbers", rather than common sense, brings to mind the
old adage "there's lies, damn lies, and statistics." Whoever said that
must have had you in mind.

Surges are microsecond
events. If they damage the MOV, then milliseconds of AC
electric cause the thermal link to fail.

Wrong. MOVs gradually degrade with each surge they pass. You've even
stated this in another post. As they degrade, they heat up. At some
point, they will heat up enough for the thermal fuse to open, taking
them out of the circuit BEFORE they can explode violently.

But a grossly
undersized MOV short circuits.

Prove that the MOVs in the circuit I posted are "grossly undersized" for
their intended application, please.

Milliseconds later, AC
electric heats that damaged MOV,

pushes too much current to
blow that thermal link, and illuminates that warning light
LP1.

Exactly!

You've just totally contradicted yourself.

A surge protector is only as effective as its earth ground.
Facts that don't change no matter what fancy lights report and
no matter how Mike Tomlinson insults others.

<yawn> I only insult liars and idiots. Your posts prove beyond a
shadow of a doubt that you're a member of that august group.

HTH. HAND. FOAD.

 

[Mike Tomlinson]

Why would I apologize to
someone so foolish as to claim a "degraded" MOV causes thermal
links to blow?

You'd better tell these people that they're fools, then:

http://powerelectronics.com/mag/power_thermally_protected_movs/

"If an MOV is subjected to a sustained abnormal overvoltage, limited
current condition (as required in UL1449), the MOV may go into thermal
runaway, resulting in overheating, smoke, and potentially fire. For end
products to comply with UL1449, the MOV must have some level of
protection to prevent this failure mode. That protection has
traditionally been a thermal fuse or thermal cutoff (TCO) device."