Please feel free to show us an MOV datasheet that says
... good surge protectors are destroyed by big surges just
as they are designed to do. The good one FAIL closed
circuit where protection is even better!

That claim is classic urban myth. MOV data sheets define
normal operation. MOV is at end of life typically when it
degrades by about 5%. How can it degrade 5% and yet
vaporize? It cannot. Bottom line remains - a properly sized
protector shunts the transient and remains fully operational.
Eventually MOV degrades; does not vaporize. Vaporizing is
when the MOV grossly exceeds manufacturer specification - is
grossly undersized for the task. But purveyors of undersized
and ineffective protectors want consumers to believe their
overpriced protector should vaporize on every surge. Scam is
the better word.

In the meantime, there is no topology in electronic
circuits. Electrically, a shunt mode protector is not
"between" the appliance and a surge no matter how much junk
science topology is rationalized. But then this thread is
full of myth purveyors promoting such junk science reasoning -
such as MOVs are designed to protect by vaporizing. Which
plug-in manufacturer do you work for, Ron Reaugh?

Ron Reaugh wrote:
"w_tom" wrote in message
...
Was UPS between AC mains and computer? No.

Wrong. Topologically the UPS is "between".

UPS and computer both connect to AC mains just like
light bulbs. In fact it would be same protection if both
computer and UPS shared same wall receptacle.

WRONG! About a critical 10 nanoseconds WRONG nevermind the
impedances and common mode condiderations!

Any transient from the receptacle confronts UPS and
computer equally.

Not in the critical time domain.

However protection inside a UPS is often so grossly
undersized that a surge too small to damage a computer
might still damage the UPS. Furthermore, some computers
can even act as surge protectors - shunt a destructive
surge so that it does not seek earth ground via other
computers.

Until you define specific circuits - including how
every wall receptacle is wired, then I cannot provide
more information.

I cannot say exactly why that particular event happened.
But above is one reason why a UPS may be damaged and
computer is not.

NO, the first component with surge suppression topologically
is usually the one that takes the HIT. Do you suppose that's
by design?

Computer power supplies have internal protection.
Protection so sufficient that there is little adjacent to a
power supply that can enhance protection. But computer
internal protection can be overwhelmed if destructive
transients are not earthed before entering the building.

OH, you mean unless the building is a heavily constructed
Faraday cage and all wiring has feedthru bypass and surge
suppression, then a destructive transient could get through
and that has NOTHING to do with you high transient impedance
ground wire.. Why is it that we all knew that?

Bottom line is this. You had UPS failure. Therefore you
have no effective surge protection. Even surge protectors
must not be damaged due to a surge.

HUH, frequently good surge protectors are destroyed by big
surges just as they are designed to do. The good one FAIL
closed circuit where protection is even better!

To provide a better answer, do as I do - autopsy the dead
body. Replace the defective part to learn what has actually
been damaged. Autopsy only complete when the failed unit is
fully functional.
...
From Sun Microsystems planning guide:
http://www.sun.com/servers/white-pap...ning-guide.pdf
Lightning surges cannot be stopped, but they can be diverted.
The plans for the data center should be thoroughly reviewed
to identify any paths for surge entry into the data center.
Surge arrestors can be designed into the system to help
mitigate the potential for lightning damage within the data
center. These should divert the power of the surge by
providing a path to ground for the surge energy. Protection
should be placed on both the primary and secondary side of
the service transformer. It is also necessary to protect
against surges through the communications lines. The
specific design of the lightning protection system for the
data center will be dependent on the design of the building
and utilities and existing protection measures.

Well lets see. The Boeing 707 was an entire 'faraday cage'
of aluminum. And inside that aluminum 'faraday cage' was a
completely enclosed fuel tank - another faraday cage. How did
lightning get through two faraday cages to explode the fuel
tank on that Boeing 707? And why was the correction to
install more grounds inside that 'faraday cage'? Could it be
that no sufficient 'faraday cage' exists? Yep.

Ron, when you get some real world experience with 'faraday
cages', then come back and share your experiences. In the
meantime, grounding inside that 'faraday cage' is essential
for safe airline operations - so that lightning will pass
through the inside of that 'faraday cage' without doing
damage. Lightning caused damage inside that 'faraday cage'
over Elkton MD.

Ron Reaugh wrote:
"w_tom" wrote in message
...
Transmitted an incomplete post. Sorry for the mistake.
Now for that airplane. Notice earth ground is the tail
section:
http://bm6aak.myweb.hinet.net/file/456.gif

OH, cool but of course it simply proves my points.

Of course this airplane ground is completely beyond the
scope of the current discussion.

You mean beyond you.

Airplanes are more difficult to ground.

No, airplanes are impossible to proactively ground while in
flight(save a high energy beam) and much more importantly the
is no need to ground an airplane in flight. There is just a
need to have a good continuous Faraday Cage. Damn, how did
anyone ever survive when planes were made of wood or paper(or
are they non-conductors)?

A Pan Am 707 was destroyed by lightning over
Elkland MD because internal grounding was not sufficient.

Oh, you mean the Faraday Cage was discontinuous or flawed
or maybe it was a super bolt of the kind that has punched
holes in heavy gauge steel petroleum tanks. In any case I'll
bet that the cockpit radio was undamaged at least until
impact.

An airplane must be grounded so that any part can become
an earth ground; making airplane design more challenging.

HUH?

We, on the other hand, are having enough trouble discussing
simple structural earthing - a well proven 1930 technology.
Why then complicate it with airplanes and other irerelevant
questions?

Because device/PC protection design has little to do with
earth grounding.

Ron. Did you read your citation before posting it? Where
is the reference to common mode protection? Where are any
numbers that apply to common mode protection? Once APC used
to provide far more numbers when they claimed to protect from
Normal mode transients. Now they don't even make those
claims. Numbers provided by Ron's citation:
Surge energy rating (one time, 10/1000 µs
waveform): 160 J.
Surge current capability (one time, 8/20 µs
waveform): 6500 Amp peak maximum.

Hell. They don't even list dBs for the noise filter. What
kind of spec is that? Noise filter for what? Incoming AC
line? Output power line? Clearly these are specs for the
technically naive.

After one surge, the entire UPS is toast? Look at those
pathetic numbers. Only 160 joules? Only 6500 amps?
Effective protection starts at about 1000 joules and 50,000
amps. Thank you Ron for demonstrating pathetic protection
from that plug-in UPS.

Oh - where do they mention anything about 'faraday cage'
protection?

In the meantime, Ron describes normal mode protection:
In order to protect a device from an undesirable voltage
arriving over the power cable one simply shunts that unwanted
voltage such that it appears equally on all the wires on that
power cable ...

Where is the common mode protection? Not in that citation.
Not in what Ron describes. Just another reason why that
plug-in UPS does not provide effective protection.

Ron Reaugh wrote:
"w_tom" wrote in message
...
Ok sir. Explain to us how a plug-in UPS provides common mode
protection. Also cite the manufactuer's spec that claims that
common mode protection (and good luck).

No problem. He
http://sturgeon.apcc.com/techref.nsf/partnum/990-7015/$FILE/7015-1.pdf
Section 9.4

Do you understand what common mode means with respect to the
above spec and do you understand how that differs/same as
the general concept of common mode and how that relates to
these issues?

In order to protect a device from an undesirable voltage
arriving over the power cable one simply shunts that unwanted
voltage such that it appears equally on all the wires on that
power cable i.e. AC-hot, AC-neutral and the ground wire
which connects to the chassis of the device. That shunt is
done with capacitors and surge diodes or MOV devices etc. or
in the old days on your phone line with a spark gap. That's
basically what a surge suppressor does. The device's input
components therefore see no intolerable VOLTAGES and it
survives. It makes no difference to the device if the whole
device(chassis and all) jumps to a million volts during the
episode. Ever heard of a Faraday Cage?
...

View specs for a plug-in protector. It claims to protect
from a type of surge that does not typically exist. That way,
the myth purveyors will assume it protects from all types of
surges. And so we have a common quote - "a surge got past my
surge protector". In reality, the surge took a left turn to
get the surge protector and a right turn to attack the
computer - simultaneously.

Get a surge protector that protects from a type of surge
that actually does damage. So which one do you install? The
one that costs tens of times more money per protected
appliance, OR the one that costs so much less and even
protects from the destructive type of surge? The latter is
the single, properly sized, and properly earthed 'whole house'
protector. Now that the best protector is also the most cost
effective, we can compare that price to what we might lose
without installing it. IOW the 'whole house' protector
provides a basis to decide protection for everything inside
the house.

Is a PC surge protector needed in the UK. No: if protectors
is the ineffective plug-in type that may even contribute to
damage of an adjacent computer. Maybe: if it is the less
expensive and more effective 'whole house' type.

Pyriform wrote:
w_tom wrote:
Furthermore, the little
surge protector does not absorb even modest transients.
Absorbing is not what they do.

I say absorb; you say shunt. We mean the same thing. Energy
that would have entered the 'protected' load instead goes
somewhere else. So unless you are arguing purely on the basis
of semantics, your claim that even "modest transients" are
not absorbed by plug-in surge suppressors is clearly false.
What I want to know is whether such transients are actually
found, and whether they pose a threat to 'unprotected'
equipment.

Your "unless it protects against everything, it protects
against nothing" argument is not entirely convincing.

Why is 'absorb' verse 'shunt' an important difference? If a
protector absorbs surges, then it sits between surge and the
protected device - in series. It must stop or block surges.
But then it must be a series mode protector. It is not. It
is shunt mode. Shunt mode protectors don't function if they
absorb surge energy. Difference between 'shunt' and 'absorb'
is important to keep the consumer confused.

A shunt mode protector connects off to the side; is not
located electrically between protected device and incoming
transient. A shunt mode protector connects as if it was
another appliance - albeit much closer to the single point
ground.

Absorbing is what the plug-in protector manufacturer hopes
you assume. That way they need not discuss earthing and hope
you assume it is a series mode protector. But they are,
instead, shunt mode protectors. Effective shunt mode
protectors must be connected short to earth ground. If you
*assume* it absorbs surges then they can avoid an earthing
discussion; let myths purveyors promote their ineffective
product.

Series mode protectors absorb. Shunt mode protectors are
similar to electric switches or electric wires - they shunt.
If others believe that it absorbs, then critical earth ground
may be overlooked. Essential to selling that ineffective
protector is to avoid all mention of earthing. And so they
hope other will *assume* it absorbs. If it shunts, then one
may ask what it shunts to? Those would be embarrassing
questions.


Pyriform wrote:
w_tom wrote:
Furthermore, the little surge protector does not absorb
even modest transients. Absorbing is not what they do.

I say absorb; you say shunt. We mean the same thing. Energy
that would have entered the 'protected' load instead goes
somewhere else. So unless you are arguing purely on the
basis of semantics, your claim that even "modest transients"
are not absorbed by plug-in surge suppressors is clearly
false. What I want to know is whether such transients are
actually found, and whether they pose a threat to
'unprotected' equipment.

Your "unless it protects against everything, it protects against
nothing" argument is not entirely convincing.

"w_tom" wrote in message
...
Please feel free to show us an MOV datasheet that says
... good surge protectors are destroyed by big surges just
as they are designed to do. The good one FAIL closed
circuit where protection is even better!

That claim is classic urban myth. MOV data sheets define
normal operation. MOV is at end of life typically when it
degrades by about 5%. How can it degrade 5% and yet
vaporize? It cannot. Bottom line remains - a properly sized
protector shunts the transient and remains fully operational.
Eventually MOV degrades; does not vaporize. Vaporizing is
when the MOV grossly exceeds manufacturer specification - is
grossly undersized for the task. But purveyors of undersized
and ineffective protectors want consumers to believe their
overpriced protector should vaporize on every surge. Scam is
the better word.

Gibber ignored.

In the meantime, there is no topology in electronic
circuits.

And this wacko nonsense is from the guy who brought up DA design! At high
frequency it is most ALL about topology!

Electrically, a shunt mode protector is not
"between" the appliance and a surge no matter how much junk
science topology is rationalized.

Did you ever hear about the speed of light or about 1 foot per nanosecond?

But then this thread is
full of myth purveyors promoting such junk science reasoning -
such as MOVs are designed to protect by vaporizing. Which
plug-in manufacturer do you work for, Ron Reaugh?

Vaporizing...are you gonna bring in Klingons now as we seem to be having a
bit to drink?

Ron Reaugh wrote:
"w_tom" wrote in message
...
Was UPS between AC mains and computer? No.

Wrong. Topologically the UPS is "between".

UPS and computer both connect to AC mains just like
light bulbs. In fact it would be same protection if both
computer and UPS shared same wall receptacle.

WRONG! About a critical 10 nanoseconds WRONG nevermind the
impedances and common mode condiderations!

Any transient from the receptacle confronts UPS and
computer equally.

Not in the critical time domain.

However protection inside a UPS is often so grossly
undersized that a surge too small to damage a computer
might still damage the UPS. Furthermore, some computers
can even act as surge protectors - shunt a destructive
surge so that it does not seek earth ground via other
computers.

Until you define specific circuits - including how
every wall receptacle is wired, then I cannot provide
more information.

I cannot say exactly why that particular event happened.
But above is one reason why a UPS may be damaged and
computer is not.

NO, the first component with surge suppression topologically
is usually the one that takes the HIT. Do you suppose that's
by design?

Computer power supplies have internal protection.
Protection so sufficient that there is little adjacent to a
power supply that can enhance protection. But computer
internal protection can be overwhelmed if destructive
transients are not earthed before entering the building.

OH, you mean unless the building is a heavily constructed
Faraday cage and all wiring has feedthru bypass and surge
suppression, then a destructive transient could get through
and that has NOTHING to do with you high transient impedance
ground wire.. Why is it that we all knew that?

Bottom line is this. You had UPS failure. Therefore you
have no effective surge protection. Even surge protectors
must not be damaged due to a surge.

HUH, frequently good surge protectors are destroyed by big
surges just as they are designed to do. The good one FAIL
closed circuit where protection is even better!

To provide a better answer, do as I do - autopsy the dead
body. Replace the defective part to learn what has actually
been damaged. Autopsy only complete when the failed unit is
fully functional.
...
From Sun Microsystems planning guide:
http://www.sun.com/servers/white-pap...ning-guide.pdf
Lightning surges cannot be stopped, but they can be diverted.
The plans for the data center should be thoroughly reviewed
to identify any paths for surge entry into the data center.
Surge arrestors can be designed into the system to help
mitigate the potential for lightning damage within the data
center. These should divert the power of the surge by
providing a path to ground for the surge energy. Protection
should be placed on both the primary and secondary side of
the service transformer. It is also necessary to protect
against surges through the communications lines. The
specific design of the lightning protection system for the
data center will be dependent on the design of the building
and utilities and existing protection measures.

"w_tom" wrote in message
...
Ron. Did you read your citation before posting it? Where
is the reference to common mode protection?


Right where I said it was as anyone who reads it can see for themselves.
Gain some technical background before you tackle such technical issues.

"w_tom" wrote in message
...
Well lets see. The Boeing 707 was an entire 'faraday cage'
of aluminum. And inside that aluminum 'faraday cage' was a
completely enclosed fuel tank - another faraday cage. How did
lightning get through two faraday cages to explode the fuel
tank on that Boeing 707? And why was the correction to
install more grounds inside that 'faraday cage'? Could it be
that no sufficient 'faraday cage' exists? Yep.

If you'd read my previous post I already said exactly that.

Ron, when you get some real world experience with 'faraday
cages', then come back and share your experiences. In the
meantime, grounding inside that 'faraday cage' is essential
for safe airline operations - so that lightning will pass
through the inside of that 'faraday cage' without doing
damage. Lightning caused damage inside that 'faraday cage'
over Elkton MD.

No, inside a discontinuous Faraday cage....get a clue.

Ron Reaugh wrote:
"w_tom" wrote in message
...
Transmitted an incomplete post. Sorry for the mistake.
Now for that airplane. Notice earth ground is the tail
section:
http://bm6aak.myweb.hinet.net/file/456.gif

OH, cool but of course it simply proves my points.

Of course this airplane ground is completely beyond the
scope of the current discussion.

You mean beyond you.

Airplanes are more difficult to ground.

No, airplanes are impossible to proactively ground while in
flight(save a high energy beam) and much more importantly the
is no need to ground an airplane in flight. There is just a
need to have a good continuous Faraday Cage. Damn, how did
anyone ever survive when planes were made of wood or paper(or
are they non-conductors)?

A Pan Am 707 was destroyed by lightning over
Elkland MD because internal grounding was not sufficient.

Oh, you mean the Faraday Cage was discontinuous or flawed
or maybe it was a super bolt of the kind that has punched
holes in heavy gauge steel petroleum tanks. In any case I'll
bet that the cockpit radio was undamaged at least until
impact.

An airplane must be grounded so that any part can become
an earth ground; making airplane design more challenging.

HUH?

We, on the other hand, are having enough trouble discussing
simple structural earthing - a well proven 1930 technology.
Why then complicate it with airplanes and other irerelevant
questions?

Because device/PC protection design has little to do with
earth grounding.

Put an impulse down a wire. Where on that wire will the
impulse voltage first be seen? On the near end where the
impulse starts? Obviously not. Impulse first appears at the
far end of wire. How can this be when it takes the impulse 1
foot per second to get to that far end? Topology as defined
by Ron would erroneously conclude impulse first appears where
impulse was first applied to wire.

Why do we know that topology not relevant? Its a simple
second year course called E-M fields. One first learned basic
concepts before promoting rubbish such as topology and
'faraday cage'.

Notice that Ron Reugh also cannot provide MOV datasheets to
demonstrate protection by vaporization. He is typical of
those who would recommend plug-in protectors. Facts remain
unchallenged: a surge protector that vaporizes during a surge
is ineffective and even violates the MOV manufacturer's own
specifications. Ron's best technical response:
Gibber ignored.

Ron Reaugh wrote:
"w_tom" wrote in message
...
Please feel free to show us an MOV datasheet that says
... good surge protectors are destroyed by big surges just
as they are designed to do. The good one FAIL closed
circuit where protection is even better!

That claim is classic urban myth. MOV data sheets define
normal operation. MOV is at end of life typically when it
degrades by about 5%. How can it degrade 5% and yet
vaporize? It cannot. Bottom line remains - a properly sized
protector shunts the transient and remains fully operational.
Eventually MOV degrades; does not vaporize. Vaporizing is
when the MOV grossly exceeds manufacturer specification - is
grossly undersized for the task. But purveyors of undersized
and ineffective protectors want consumers to believe their
overpriced protector should vaporize on every surge. Scam is
the better word.

Gibber ignored.

In the meantime, there is no topology in electronic
circuits.

And this wacko nonsense is from the guy who brought up DA
design! At high frequency it is most ALL about topology!

Electrically, a shunt mode protector is not
"between" the appliance and a surge no matter how much junk
science topology is rationalized.

Did you ever hear about the speed of light or about 1 foot
per nanosecond?

But then this thread is full of myth purveyors promoting
such junk science reasoning - such as MOVs are designed
to protect by vaporizing. Which plug-in manufacturer do
you work for, Ron Reaugh?
...

IOW there is no claim in that APC sales brochure for common
mode transient protection. Even worse, they no longer make the
specific claim for normal mode protection. Ron again
demonstrates propaganda used to promote plug-in UPSes for
ineffective surge protection.

Ron Reaugh wrote:
"w_tom" wrote in message
...
Ron. Did you read your citation before posting it? Where
is the reference to common mode protection?

Right where I said it was as anyone who reads it can see
for themselves. Gain some technical background before you
tackle such technical issues.

Just another reason why FAT was obsoleted by HPFS which in
turn was obsoleted by NTFS.

I'l bite:

I think I understant FAT systems.

But the only thing I "think" I know about the NTFS is that, effectively, the
system first makes a record of what it is about to do, then it does it, and
then it either erases the original record or somehow marks it.

SO: can someone "explain" the NTFS to me. (Please don't tell me to "look
it up.")

In article , J.J. wrote:

HDD are frequently corrupted due to power events.

Does anyone have a reference to HDDs getting corrupted by power
events on the mains power supply?

Reference Microsoft

"Because you didn't shut down Windows properly, Scandisk is now trashing
your disk to complete the job.

In future, always shut down Windows properly"

Scandisk is not *always* going to pull your nuts out of the fire after a
power interruption. In fact, it might make things worse

Wouldn't you *expect* data corruption, if the system was writing/about to
write cached data, and the mains power went off?

You don't need references to figure out that it's a bad idea to just
lose power in an uncontrolled way.
--
--------------------------------------+------------------------------------
Mike Brown: mjb[at]pootle.demon.co.uk | http://www.pootle.demon.co.uk/

In sci.physics, w_tom

wrote
on Sat, 10 Jul 2004 02:06:11 -0400
:
Put an impulse down a wire. Where on that wire will the
impulse voltage first be seen? On the near end where the
impulse starts? Obviously not. Impulse first appears at the
far end of wire. How can this be when it takes the impulse 1
foot per second to get to that far end?

Erm, 1 foot per *nano*second. Light speed.

Perhaps it's somewhat less than that; AIUI electric current
is along the lines of 2/3 c, but it's still pretty darned
fast.

[rest snipped]

--
#191,
It's still legal to go .sigless.

"w_tom" wrote in message
...
Put an impulse down a wire.

You can't even define "impulse".

Where on that wire will the
impulse voltage first be seen? On the near end where the
impulse starts? Obviously not. Impulse first appears at the
far end of wire.

Now you've self contradicted and imploded. Ask Albert.

"w_tom" wrote in message
...
IOW there is no claim in that APC sales brochure for common
mode transient protection. Even worse, they no longer make the
specific claim for normal mode protection. Ron again
demonstrates propaganda used to promote plug-in UPSes for
ineffective surge protection.

Your wacko claims have already been refuted by my citation earlier in this
thread.
APC does include "common mode" as I cited.

In article ,
Ron Reaugh writes

Vaporizing...are you gonna bring in Klingons now as we seem to be having a
bit to drink?

I suggest you consult the thread with the same title crossposted to the
following groups before wasting any more time on w_tom:

uk.comp.vendors,uk.comp.homebuilt,alt.comp.hardwar e,alt.comp.hardware.pc
-homebuilt

--
A. Top posters.
Q. What's the most annoying thing on Usenet?

w_tom wrote:
Hard drives can be corrupted for various reasons based also
upon what the filesystem is. For example, if using FATxx
filesystems, then a loss of electrical power at the right time
can even erase files from that drive. Just another reason why
the technically informed want NTFS filesystems on drives; not
FAT.


Or RieserFS. But that is used by OS's other than Gate$
Lemingware! :-)

In sci.physics, John Gilmer

wrote
on Sat, 10 Jul 2004 10:53:27 -0400
:



Just another reason why FAT was obsoleted by HPFS which in
turn was obsoleted by NTFS.

I'l bite:

I think I understant FAT systems.

But the only thing I "think" I know about the NTFS is that, effectively, the
system first makes a record of what it is about to do, then it does it, and
then it either erases the original record or somehow marks it.

SO: can someone "explain" the NTFS to me. (Please don't tell me to "look
it up.")


NTFS stands for NT File System, presumably. (I've no idea what NT
stands for. Certain jokesters have their own opinions, mine among
them.)

A file system is a method by which the unorganized data in
a disk partition -- basically, a very long chain of blocks,
or perhaps a mapping from an integer (the logical block
address) to a fixed-size chunk of data (the block) -- can
be organized into something more appetizing to humans:
files, directories, symbolic links, or in Microsoft
parlance (perhaps), documents, folders, and shortcuts.
DOS 1.0's FAT filesystem didn't even have directories
(that was added in 1.1 or 2.0; I forget which). NTFS is
fairly sophisticated; it has, among other things:

- per-file locking (to the intense annoyance of UNIX and Linux
programmers, this appears to be on by default)
- resource streams a la Macintosh (which aren't apparently used yet?)
- Access Control Lists
- Unicode support
- Case-preserving filenames
- a master file table, which is where the small files live
- sparse files (files with "holes" in their blocklists -- a
useful capability in some contexts related to databases, AIUI)
- short file name capability for DOS backwards compatibility
- hidden files
- support for running a defragmenter while the volume is mounted.
(Don't ask.)

There are a few other capabilities but I'd have to look.

If you really do want to look it up, you can try
the Linux source code -- an NTFS implementation
is in the kernel under /usr/src/linux/fs/ntfs or
/usr/src/linux/Documentation/filesystems/ntfs.txt. It is
definitely not for the faint of heart. There should
be some documentation somewhere on Microsoft's website,
of course; again, I'd have to look.

HTH

--
#191,
It's still legal to go .sigless.

quoting:
w_tom wrote:
Hard drives can be corrupted for various reasons based also
upon what the filesystem is. For example, if using FATxx
filesystems, then a loss of electrical power at the right time
can even erase files from that drive. Just another reason why
the technically informed want NTFS filesystems on drives; not
FAT.


The files aren't actually gone forever. If the power fails during the
writing of the file, but the FAT have not been updated yet, the data will be
found as "lost clusters" by Scandisk. It'll probably be in a bunch of
pieces though, due to fragmentation.

There are little backup programs that back up the disk's FAT's, boot sector,
etc. in the event that that any of the disk's reserved sectors get
obliterated by some other means.

NTFS stands for NT File System, presumably. (I've no idea what NT
stands for. Certain jokesters have their own opinions, mine among
them.)

Yeah, yeah.

And NT stands for New Technology. It was written as a "Windows Like"
operating system to run on hardware stuff by Sun Micro and the old DEC which
used UNIX. At some point Micro$oft is to make the NT and Windows
essentially the same operating system.



A file system is a method by which the unorganized data in
a disk partition -- basically, a very long chain of blocks,
or perhaps a mapping from an integer (the logical block
address) to a fixed-size chunk of data (the block) -- can
be organized into something more appetizing to humans:
files, directories, symbolic links, or in Microsoft
parlance (perhaps), documents, folders, and shortcuts.
DOS 1.0's FAT filesystem didn't even have directories
(that was added in 1.1 or 2.0; I forget which).

Well, you lost me again. The directory is supposed to point to the entry
in the FAT corresponding to the first "allocation unit" of the file. From
my old memoery, the Intel development system just had a fixed directory.
Directory entry #1 was the first allocation unit, etc. Longer files were
accomodated by "chaining" directories.


NTFS is
fairly sophisticated; it has, among other things:

- per-file locking (to the intense annoyance of UNIX and Linux
programmers, this appears to be on by default)

Well, WTF does it "lock?'

- resource streams a la Macintosh (which aren't apparently used yet?)

That doesn't help.

- Access Control Lists

OK.

- Unicode support

Which means ...

- Case-preserving filenames

OK

- a master file table, which is where the small files live

Huh?

- sparse files (files with "holes" in their blocklists -- a
useful capability in some contexts related to databases, AIUI)

Neat!

- short file name capability for DOS backwards compatibility

OK

- hidden files

Old stuff.

- support for running a defragmenter while the volume is mounted.
(Don't ask.)

Well, I understand was the defragmenter does in a FAT system but since I
still don't understand how files are stored I can't understand how that are
either fragmented or defragmented.


There are a few other capabilities but I'd have to look.

OK



If you really do want to look it up, you can try
the Linux source code -- an NTFS implementation
is in the kernel under /usr/src/linux/fs/ntfs or
/usr/src/linux/Documentation/filesystems/ntfs.txt. It is
definitely not for the faint of heart. There should
be some documentation somewhere on Microsoft's website,
of course; again, I'd have to look.

Sorry, you have just asked me to think and work harder than I care to.

w_tom wrote:

Ron. Did you read your citation before posting it? Where
is the reference to common mode protection? Where are any
numbers that apply to common mode protection?


snip

Where is the common mode protection? Not in that citation.
Not in what Ron describes.

snip

Yes, it IS in that citation, exactly where he said it is.
It is the line below the two you quoted from the page
he cited.

It says: "Surge response time: 0 ns (instantaneous)
normal mode, 5ns common mode."

In article ,
"John Gilmer" wrote:
snip-valiantly refraining from comment

- support for running a defragmenter while the volume is mounted.
(Don't ask.)

Well, I understand was the defragmenter does in a FAT system but since I
still don't understand how files are stored I can't
understand how that are
either fragmented or defragmented.

Consider a file system that writes empty blocks in numerical sequential
order. Now think of a file that's deleted. This leaves an empty
"hole" in the filled blocks. Now make a file whose size is less
than the "hole". Now you have a smaller hole that will be filled
with the next file that is written. That file is larger than the
hole so the hole gets filled, then the next block that isn't filled
is found and written into. Over time, all files, when viewed from
the geometry of the physical disk look like swiss cheese.

A defragmenter takes the whole file system and rewrites each file
such that all its block numbers are monotonically increasing.

Now, where this gets really, really ****ed up is when the defragger
program "forgets" which should be the next block (real easy to do
with off-by-one bugs) or has to call its error handling when it
can't do a fit or the block chain pointers become broken. The
last one is a feature of all Misoft OSes because of memory
management problems--but that's another nightmare in the not-an-OS
biz.


snip

/BAH

Subtract a hundred and four for e-mail.

"Mike Tomlinson" wrote in message
...
Q. What's the most annoying thing on Usenet?

A. Cross posting. Second only to people who bitch about top posters .

Leonard

In sci.physics,

wrote
on Sun, 11 Jul 04 09:29:22 GMT
:
In article ,
"John Gilmer" wrote:
snip-valiantly refraining from comment

- support for running a defragmenter while the volume is mounted.
(Don't ask.)

Well, I understand was the defragmenter does in a FAT system but since I
still don't understand how files are stored I can't
understand how that are
either fragmented or defragmented.

Consider a file system that writes empty blocks in numerical sequential
order. Now think of a file that's deleted. This leaves an empty
"hole" in the filled blocks. Now make a file whose size is less
than the "hole". Now you have a smaller hole that will be filled
with the next file that is written. That file is larger than the
hole so the hole gets filled, then the next block that isn't filled
is found and written into. Over time, all files, when viewed from
the geometry of the physical disk look like swiss cheese.

A defragmenter takes the whole file system and rewrites each file
such that all its block numbers are monotonically increasing.

Now, where this gets really, really ****ed up is when the defragger
program "forgets" which should be the next block (real easy to do
with off-by-one bugs) or has to call its error handling when it
can't do a fit or the block chain pointers become broken. The
last one is a feature of all Misoft OSes because of memory
management problems--but that's another nightmare in the not-an-OS
biz.

Indeed. In Linux, there's no defragger[*], because the file
code in Linux is a little smarter. I'd admittedly have
to look for the details though, and ext2's organization
is quite different from FAT's or NTFS. FAT in particular
is terrible, basically every file is a single chain --
but you probably knew that already. NTFS is more or less
as I've described it in my prior post, at a high level,
and it feels like an engineered solution, whereas Linux's
ext2 is more elegant, even if it's still engineered.
But there's no perfect solution anyway; as you've described
the problem, there's always going to be a hole or two,
and a determined program can probably fragment any file
system if it does something like the following:

open big file
write block to big file
open little file
write block to little file
close it
write block to big file
open little file
write block to little file
close it
write block to big file
....

(It's a good thing the trend is towards centralized dedicated-machine
syslog-type logging. :-) )

I'll admit to wondering whether NT had the rather interesting
capability or not of "let's just write it here". I base
this hypothesis on observations using DiskKeeper Lite, which
copy I had at the time on a machine at my then-employer.
Basically, the notion is to simply write the new block at
an open sector in the cylinder over which the head is
flying.

Of course this would fragment things terribly, and I have no proof.
But things did fragment pretty badly when I used such tools
as Visual C++.



snip

/BAH

Subtract a hundred and four for e-mail.
[*] actually, there is, but it's very rarely used.

--
#191,
It's still legal to go .sigless.

Common mode to what? To the safety ground? How much? Does
it conduct 1 microamp in 5 ns to the safety ground? What kind
of protection is that? Based upon facts and numbers provided,
then my digital multimeter is even a better surge protector -
a claim I can make because specs are better called an
'executive summary'.

At least that manufacturer once provided insufficient specs
for Normal mode protection that the manufacturer does claim to
provide:
Normal mode surge voltage let through 5% of test peak
voltage when subjected to IEEE 587 Cat. A 6kVA test
Now manufacturer cannot bother to provide even that
insufficient information. After all, they are not trying to
sell a 'common mode' claim to the informed. They dumb down
the numbers into rubbish so that one who wishes MOVs absorb
the energy of a surge will see what he wants.

My car tires have a common mode response time AND that
proves those tires are effective protection? Common mode
what? Does not matter. That tiny phrse would be enough even
for a poet to believe what he wants to know.

How much common mode current in less than 5 ns? From what
or which one wire is that common mode response? Is that
common mode response really just a response inside the UPS
controller circuit? Or is that a common mode response on the
serial port. RS-232 is a common mode communication ports. So
does the serial port haves a less than 5 ns response? Wow.
That means the UPS must provide massive lightning protection -
if living in the world of Harry Potter.

IOW they mention 'common mode response' but give not one
indication that the UPS provides common mode surge
protection. It only does something - and they don't even say
what or how much. That woefully insufficient and deceptive
information is enough for some to loudly declare that a UPS
provides lightning protection. IOW another urban myth has
been promoted.

There are no claims of common mode transient protection on
the incoming AC input. Provided are words without relevance so
that a poet might hope for common mode response to something -
which therefore must be a direct lightning strike? It's
called wild speculation on your part - the same person who
foolishly believes shunt mode devices (such as wire) are
designed to absorb energy. But an engineer says, "What is
this crap. There is no numerical information to work with."

That UPS does not claim common mode protection. It simply
claims some undefined of response to common mode noise from or
to an undefined location. It does not even say those 160
joules are involved in such protection. Furthermore it admits
to being grossly undersized - only 160 joules. A poet then
can assume the response time means the UPS will conduct 50,000
amps? A poet can. So can Harry Potter. Those who must deal
in reality cannot.

There is nothing in those specs beyond gobbledygook. Using
ehsjr and Ron Reaugh reasoning, should we assume the UPS is
sufficient even for aeronautical environments? After all,
they do
claim 'something' that myth purveyors can distort into a real
world miracle.

ehsjr - when will you claim that a faraday cage also makes
that UPS so effective?

I have this 741 op amp (a semiconductor amplifier). It also
has a common mode rating. So that operational amplifier (that
little IC) is also a lightning protector? Yes according to
how ehsjr reasons. Give me a break. That UPS does not even
claim to provide common mode protection - which is why they
must all but encrypt their specifications. Its called name
dropping. They dropped the phrase "common mode". That
without any numbers is enough for ehsjr to loudly claim the
UPS provides common mode protection. It is called Junk
Science reasoning.

wrote:

w_tom wrote:

Ron. Did you read your citation before posting it? Where
is the reference to common mode protection? Where are any
numbers that apply to common mode protection?


snip

Where is the common mode protection? Not in that citation.
Not in what Ron describes.

snip

Yes, it IS in that citation, exactly where he said it is.
It is the line below the two you quoted from the page
he cited.

It says: "Surge response time: 0 ns (instantaneous)
normal mode, 5ns common mode."

"w_tom" wrote in message
...
Common mode to what?

Nice try liar.

Power supply damages motherboard when a computer assembler
purchases power supplies without consulting specifications.
Intel specs for ATX power supplies demand that PSU not damage
motherboard and other components:
Section 3.5.1 Overvoltage Protection
The overvoltage sense circuitry and reference shall reside
in packages that are separate and distinct from the
regulator control circuitry and reference. No single
point fault shall be able to cause a sustained overvoltage
condition on any or all outputs. The supply shall provide
latch-mode overvoltage protection as defined below.
Table 11: Overvoltage Protection
Output
+12 VDC Max is 15.6 volts
+5 VDC Nominal is 6.3 volts
+3.3 VDC Nominal is 4.2 volts

Too many computer assemblers don't have necessary technical
knowledge and therefore don't even know that overvoltage
protection (OVP) has been a defacto standard for 30+ years.
That motherboard damage probably may be traceable to the
ill-informed computer assembler (who does not demand specs) or
a power supply manufacturer who outrightly lies on his
specifications.

There is nothing in a UPS that will accomplish the necessary
OVP.

Other essential functions that should be found in the power
supply specification, but that many 'bean-counter' selected
supplies may be missing:
Specification compliance: ATX 2.03 & ATX12V v1.1
Short circuit protection on all outputs
Over voltage protection
Over power protection
EMI/RFI compliance: CE, CISPR22 & FCC part 15 class B
Safety compliance: VDE, TUV, D, N, S, Fi, UL, C-UL & CB
Hold up time, full load: 16ms. typical
Efficiency; 100-120VAC and full range: 65%
Dielectric withstand, input to frame/ground: 1800VAC, 1sec.
Dielectric withstand, input to output: 1800VAC, 1sec.
Ripple/noise: 1%
MTBF, full load @ 25°C amb.: 100k hrs

Power supplies missing these and other functions are sold at
good profit in the North American computer clone market. OVP
must be in all computer supplies but is often missing in clone
computers.


"J.J." wrote:
If a computer PSU fails then I have heard that it may (or may
not) blow the mainboard and perhaps various other components
with a power surge or soemthing like that.

It seems that better PSUs are designed so that when they fail
they have some circuitry which protects the other components
in the PC.

Is this failsafe feature of the PSU I am referring to pretty
much the same feature you are referring to? Or are they
separate features?

Does anyone know how common it is to get this failsafe
feature in a PSU?

"w_tom" wrote in message
...
snip
There is nothing in a UPS that will accomplish the necessary
OVP.


You need to specify which kind of UPS. Some UPS will provide excellant over
voltage protection.

Charles Perry P.E.

Overvoltage protection being discussed is on the 3.3, 5, and
12 volt outputs. Table 11 from Intel specs even defines where
and what that OVP must do. There is nothing in a plug-in UPS
- outputting 120 VAC or 230 VAC to power supply - that is
going to over voltage protect those DC outputs. Nothing.
Only OVP that a UPS can provide - limit 120 VAC or 230 VAC.
That will not solve an overvoltage problem on DC output of a
'defective by design' power supply.

J.J. asked:
If a computer PSU fails then I have heard that it may
(or may not) blow the mainboard and perhaps various
other components ...

Yes, if power supply does not have OVP. No if supply does
have the required OVP. No UPS will solve this missing OVP
problem on DC outputs of power supply.

Charles Perry wrote:
"w_tom" wrote in message
...
snip
There is nothing in a UPS that will accomplish the
necessary OVP.

You need to specify which kind of UPS. Some UPS will
provide excellant over voltage protection.

Charles Perry P.E.

He has no knowledge, education or experience. He has not
one technical fact to post in response. At least a junk
scientist would try to invent a fact. Ron Reaugh is even
worse than a junk scientist. He insults. Some claim that a
plug-in UPS provided hardware protection. They can insult.
That alone proves they must be right. Hey Ron. Is god on
your side? No wonder you just 'know' these things.

Ron Reaugh wrote:
"w_tom" wrote in message
...
Common mode to what?

Nice try liar.

On Fri, 09 Jul 2004 19:27:13 -0400, w_tom wrote:

Again, the trashed filesystem is a problems of FAT and other
simplistic file systems. It is not a problem to superior
(journalizing) filesystems.

Which would be?

Will a disk drive write to the platter as voltage drops?

Certainly, at least to some point.

Of course not.

You're *ONCE AGAIN* yalking through your ass.

The disk drive controller is a complete computer
that also monitors voltage.

Really? I'm not from Missouri, but close enough. An IDE port monitors
its supply voltage? You're simply talking out your ass, since it's been
shot off repeatedly.


It does not matter to disk
hardware when power is turned off. But it does matter to some
'simplistic' disk filesystems that power is not removed during a write
operation.

Just another reason why FAT was obsoleted by HPFS which in
turn was obsoleted by NTFS.

You haven't a clue (as usual). NTFS is a slight modification to HPFS
(written by the same SB, AFAIK) to make sure that OS/2 couldn't access NT
systems. Neither is a JFS, nor is either less corrruptable than FAT.
Indeed NT systems are far more susceptable to corruption than other
similar OSs because of the agressive write buffering. Even (non-JFS) OS/2
systems are better at self-healing than NT. Of course JFS is a standard
part of OS/2 now. Windows? YMBK!

--
Keith

In article ,
w_tom wrote:

People might take you a tad more seriously if you exhibited
a scintilla of trainability. The next line is a hint...

snip topposting

/BAH

Subtract a hundred and four for e-mail.

In article ,
Keith wrote:
On Fri, 09 Jul 2004 19:27:13 -0400, w_tom wrote:

Again, the trashed filesystem is a problems of FAT and other
simplistic file systems. It is not a problem to superior
(journalizing) filesystems.

Which would be?

Will a disk drive write to the platter as voltage drops?

Certainly, at least to some point.

Of course not.

You're *ONCE AGAIN* yalking through your ass.

The disk drive controller is a complete computer
that also monitors voltage.

Really? I'm not from Missouri, but close enough. An IDE port monitors
its supply voltage? You're simply talking out your ass, since it's been
shot off repeatedly.


It does not matter to disk
hardware when power is turned off. But it does matter to some
'simplistic' disk filesystems that power is not removed during a write
operation.

Just another reason why FAT was obsoleted by HPFS which in
turn was obsoleted by NTFS.

You haven't a clue (as usual). NTFS is a slight modification to HPFS
(written by the same SB, AFAIK) to make sure that OS/2 couldn't access NT
systems. Neither is a JFS, nor is either less corrruptable than FAT.

How in the world does he think that FAT has anything to do with
physical disk specs?

Indeed NT systems are far more susceptable to corruption than other
similar OSs because of the agressive write buffering.

That's really, really, really too bad. It used to know how.

snip

/BAH

Subtract a hundred and four for e-mail.

In article ,
(Mike) wrote:
In article , J.J.
wrote:

HDD are frequently corrupted due to power events.

Does anyone have a reference to HDDs getting corrupted by power
events on the mains power supply?

Reference Microsoft

"Because you didn't shut down Windows properly, Scandisk is now trashing
your disk to complete the job.

In future, always shut down Windows properly"

Scandisk is not *always* going to pull your nuts out of the fire after a
power interruption. In fact, it might make things worse

Wouldn't you *expect* data corruption, if the system was writing/about to
write cached data, and the mains power went off?

Not if you have an OS that knows what it's doing. A great part
of file system code is supposed to anticipate such things. For
instance, one approach is to not put the file entry in the user
file directory file until after the last bit is written and the
user's prog has issued a close.


You don't need references to figure out that it's a bad idea to just
lose power in an uncontrolled way.

Of course it's a bad idea :-) but most systems, that went through
a design, allocate quite a bit of their code to Murphy anticipation.

/BAH

Subtract a hundred and four for e-mail.

You knew that an IDE port is not a computer; has no
intelligent functions? A disk drive is an embedded computer
complete with voltage monitoring circuits. This controller is
not the same as an IDE port.

Long before voltage drops below what digital circuits on
disk drive require, the disk drive has detected the falling
voltage and stopped writing. Disk drive hardware protects
itself from voltage drop which is also why a power down will
not interfere even with normal disk drive housekeeping.

The idea that an IDE port monitors voltage is an erroneous
assumption and was not made in any previous post. An IDE port
is nothing more than a bidirectional repeater. An IDE port
has no intelligent functions and does not monitor voltage.
Where did you get this idea that IDE port functions were even
being discussed? Are you confusing IDE with some other
hardware interface? Or did you just fail to read that
previous post will sufficient care?

Posted previously was:
The disk drive controller is a complete computer that
also monitors voltage.

Where is an IDE interface even implied here?

Keith wrote:
On Fri, 09 Jul 2004 19:27:13 -0400, w_tom wrote:
Will a disk drive write to the platter as voltage drops?

Certainly, at least to some point.

Of course not.

You're *ONCE AGAIN* yalking through your ass.

The disk drive controller is a complete computer
that also monitors voltage.

Really? I'm not from Missouri, but close enough. An IDE port
monitors its supply voltage? You're simply talking out your
ass, since it's been shot off repeatedly.
...

In article ,
Richard Herring ] wrote:
In message ,
writes
In article ,
Keith wrote:

Indeed NT systems are far more susceptable to corruption than other
similar OSs because of the agressive write buffering.

That's really, really, really too bad. It used to know how.

That objection misses the point. You may lose a lot of data because of
the buffering, but the commit/rollback transaction processing supposedly
means that what you lose is a complete transaction and what you have
left will always be consistent.

Sigh! Those problems were solved (one product was JMF's) in a variety
of ways in an OS that smells like NT. Consistency wasn't the major
problem with the airline reservation system.

At least, that's the theory...

Yea. And then somebody tries to program the thing. Then you
find out what the real theory is ;-).

/BAH


Subtract a hundred and four for e-mail.

In article ,
w_tom wrote:
You knew that an IDE port is not a computer; has no
intelligent functions? A disk drive is an embedded computer
complete with voltage monitoring circuits.

Oh, my! You are young.


pins

/BAH

Subtract a hundred and four for e-mail.

In message ,
writes
In article ,
Keith wrote:

Indeed NT systems are far more susceptable to corruption than other
similar OSs because of the agressive write buffering.

That's really, really, really too bad. It used to know how.

That objection misses the point. You may lose a lot of data because of
the buffering, but the commit/rollback transaction processing supposedly
means that what you lose is a complete transaction and what you have
left will always be consistent.

At least, that's the theory...

--
Richard Herring

"Keith" wrote in message news
On Fri, 09 Jul 2004 19:27:13 -0400, w_tom wrote:

Again, the trashed filesystem is a problems of FAT and other
simplistic file systems. It is not a problem to superior
(journalizing) filesystems.

Which would be?

Will a disk drive write to the platter as voltage drops?

Certainly, at least to some point.

Of course not.

You're *ONCE AGAIN* yalking through your ass.

The disk drive controller is a complete computer
that also monitors voltage.

Really? I'm not from Missouri, but close enough. An IDE port monitors
its supply voltage? You're simply talking out your ass, since it's been
shot off repeatedly.

And yours blew-up just now when you can't make the distinction between
an IDE Disk Controller and an IDE Hostbus Adapter.



It does not matter to disk hardware when power is turned off. But it
does matter to some 'simplistic' disk filesystems that power is not
removed during a write operation.

*And* to the drive when it may encounter a bad sector afterwards,
nomatter what filesystem is in use, though the 'damage' is temporary.


Just another reason why FAT was obsoleted by HPFS which in
turn was obsoleted by NTFS.

You haven't a clue (as usual). NTFS is a slight modification to HPFS
(written by the same SB, AFAIK) to make sure that OS/2 couldn't access NT
systems. Neither is a JFS, nor is either less corrruptable than FAT.
Indeed NT systems are far more susceptable to corruption than other
similar OSs because of the agressive write buffering. Even (non-JFS) OS/2
systems are better at self-healing than NT. Of course JFS is a standard
part of OS/2 now. Windows? YMBK!

--
Keith

wrote in message ...
In article , w_tom wrote:
You knew that an IDE port is not a computer; has no
intelligent functions? A disk drive is an embedded computer
complete with voltage monitoring circuits.

Oh, my! You are young.

Well, in that case you are probably old as methusalem.
Not a working braincell left in your cranium.



pins

/BAH

Subtract a hundred and four for e-mail.