Cheap fire suppression system for a PC?

ShadowTek wrote:
On 2009-10-20, Al Dykes wrote:
In article ,
ShadowTek wrote:
On 2009-10-20, kony wrote:

Swap the PSU fan for a quieter one. Done.

Send me the money to do it, and it'll get done.

I'm not going to spend $70 of *my* money on another PS when I can simple isolate the
sound that my current one is making.

*You* are the one worried about fire.

Without the PS vibrations, the noise from everything else still needs to
be dealt with, which means I will still be using some kind of sound/vibration
insulation, which means I'm still concerned with fire.

Change fans or mount the fans on shock absorbers:

http://www.endpcnoise.com/cgi-bin/e/...u=Fanisolators

Rubber grommets with thick-shaft screws can be used for isolating hard
drives. Special screws are needed to allow tightening (so they don't
eventually loosen from vibration) without crushing the grommets.

wrote in message
...


Ken Maltby wrote:
wrote in message
...

You can also use watercooling to remove all the
noise producing elements to another room. That
leaves just the hard drive "noise" to manage.

I thought the purpose was to improve safety, not worsen it by creating
a potential electric shock hazard with a cooling system that hasn't
been UL, VDE, or CSA certified. The cooling systems I've seen use
vinyl hoses that can melt if the processor's heatsink gets hot enough,
and I haven't seen any leak detection devices, like those made for
preventing washroom flooding. A better solution may be a heatpipe
cooling system

You obviously have no understanding of how a watercooling
system works, or basic physics for that matter.

Obviously.

The processor
and any other component needing active cooling, would have a
"waterblock" not a heatsink.

And you're not splitting hairs.


No I'm not. A waterblock by design never gets hot.
A passive heatsink needs to heat up to function. That
you do not understand this basic operational factor,
just proves my point. That you are unaware of the
difference and view my statement as "splitting hairs"
further demonstrates your ignorance.


The distilled water being pumped
through the waterblock(s) and the "hoses" is at most only a few
degrees above the ambient temp., of the room with the radiator.
The hoses are at the same temp. as the water.

Even under worst-case conditions?

Any properly designed watercooling system well not
allow the computer to operate unless there is power to
the pump and radiator fan. You can make up anything,
but there are no real world conditions that would lead
to hoses filled with water melting, from any heat source
within a healthy computer.



Most any water pump you can buy will have all the required
certifications.

You're under the false impression that a device consisting exclusively
of UL approved components is itself UL approved. If that's wrong,
then explain how a 120VAC device with dozens of UL Y-certified 5nF
capacitors soldered in parallel between hot and ground could pass UL
standards for electrical leakage.


I make no such claim, nor would I expect your misuse of
components have any but predictably poor results. The
parts nor their certification status can make up for your
incompetence. As with most owner built and operated
equipment, (like the computer that I am watercooling)
the watercooling system is assembled from a number
of parts, meeting any number of specifications, but the
assembled result (like the rest of the computer) bears no
overall certification or stamp of approval. By the way, both
the watercooling and the computer function very well and
very safely, despite the lack of an overall certification.

There is no more "shock hazard" with water cooling than with
air cooling.

But if I had to choose, I'd rather have a big air leak than almost any
water leak into a high voltage supply. What about you?

Let's see almost 15 years of watercooling and no leaks "into a
high voltage supply", ever. No hot tubes, just slightly warm to the
touch. I guess I must be half way competent at plumbing my
watercooling systems, just like the many thousands of others all
around the world who watercool their systems. It's not that hard
to do it right.

Luck;
Ken

wrote in message
...


There were some photos of a failed homemade water cooling system,
probably for a Pentium III or original Athlon, where the vinyl tubing
had melted, and the loss of water led to the soldered copper water
block falling apart and spilling molten solder over a large area of
the motherboard. I doubt any other tubing would have failed nearly
as easily.


Total BS. Post a citation or link to these photos so
we can get the whole story.

On 2009-10-20, wrote:


ShadowTek wrote:

Without the PS vibrations, the noise from everything else still needs to
be dealt with, which means I will still be using some kind of sound/vibration
insulation, which means I'm still concerned with fire.

Change fans or mount the fans on shock absorbers:

http://www.endpcnoise.com/cgi-bin/e/...u=Fanisolators

Rubber grommets with thick-shaft screws can be used for isolating hard
drives. Special screws are needed to allow tightening (so they don't
eventually loosen from vibration) without crushing the grommets.

Actually, there don't seem to be any annoying vibrations that are noticible from anything other
than my power supply. My fans are pretty quiet, and the hard drive cages
that came with my case have vibration padding along each side.

Every other irriatation is pretty much high-frequency noise.

"ST" == ShadowTek writes:
ST but
ST I've got to make space to set one up, as my house is *very*
ST tight on space.

The one I use is 5.5" high x 7" deep (including the two USB
flashdrives sticking out the back) x 1" wide. Should fit in anybody's
house.
--
I spent thirty-three years in the marines, most of my time being a
high-class muscle man for big business, for Wall Street and the
bankers. In short, I was a racketeer for capitalism.
~ General Smedley Butler, Marine

On Tue, 20 Oct 2009 12:35:06 -0700 (PDT),
wrote:


They were plug-in Woods Wire brand GFIs. The first was 1-3 years old
when I pressed the test button and noticed it stayed off (power
transistor or triac had ruptured and sprayed copper inside the case).
The second GFI - the replacement for the first and a very different
design - failed in a few days and kept the power on. The
replacement for it has been fine.

That's why I don't trust GFIs.

Can't speak for all of them, but generally the plastic gets
brittle and a little tab or post breaks, meaning it would
have worked to interrupt power fine as needed, or from your
test, but won't reset after that failure or test... so in
other words, up to you how often to test vs replace, but I
would think about a surge protector for the premises if
parts rupture.



There were some photos of a failed homemade water cooling system,
probably for a Pentium III or original Athlon, where the vinyl tubing
had melted, and the loss of water led to the soldered copper water
block falling apart and spilling molten solder over a large area of
the motherboard. I doubt any other tubing would have failed nearly
as easily.


Fortunately that is not possible by any normal failure
method.

1) P3 had thermal shutdown, maybe it was an Athlon prior to
the motherboards having thermal shutdown.

2) Vinyl tubing doesn't melt at a very low temp, something
else was going on that was already a severe system failure
independent of the tubing. Might have been PSU, or
motherboard shorting out, CPU alone cannot create enough
heat to melt the solder on the heatsink, it would have far
sooner ruptured and PSU shut off.

3) PSU being crude /defective is the likely problem, any
normal PSU would shut down from overcurrent condition long
before delivering enough heat to melt solder on a heatsink
or melt vinyl tubing. It just isn't a realistic failure
scenario with any properly built PC, water-cooled or not.

4) As I've mentioned previously, the electrical wiring
insulation itself has no higher temp rating than the melting
point of vinyl tubing. If excessive temps are the problem
then we start drifting into the topic of lowest melting
point materials and rewiring everything with teflon wiring
but does anyone do that for a PC?

Regardless of the above I'd love to see those pics.

Ken Maltby wrote:
wrote in message
...


There were some photos of a failed homemade water cooling system,
probably for a Pentium III or original Athlon, where the vinyl tubing
had melted, and the loss of water led to the soldered copper water
block falling apart and spilling molten solder over a large area of
the motherboard. I doubt any other tubing would have failed nearly
as easily.


Total BS. Post a citation or link to these photos so
we can get the whole story.

It was probably 10 years ago, and you shouldn't call people liars,
little boy.

kony wrote:
On Tue, 20 Oct 2009 12:35:06 -0700 (PDT),
wrote:


They were plug-in Woods Wire brand GFIs. The first was 1-3 years old
when I pressed the test button and noticed it stayed off (power
transistor or triac had ruptured and sprayed copper inside the case).
The second GFI - the replacement for the first and a very different
design - failed in a few days and kept the power on. The
replacement for it has been fine.

That's why I don't trust GFIs.

Can't speak for all of them, but generally the plastic gets
brittle and a little tab or post breaks, meaning it would
have worked to interrupt power fine as needed, or from your
test, but won't reset after that failure or test... so in
other words, up to you how often to test vs replace, but I
would think about a surge protector for the premises if
parts rupture.

The first Woods GFI had been subject of a recall by the company, but
they merely told customers to press the test button before each use,
not return the unit.




There were some photos of a failed homemade water cooling system,
probably for a Pentium III or original Athlon, where the vinyl tubing
had melted, and the loss of water led to the soldered copper water
block falling apart and spilling molten solder over a large area of
the motherboard. I doubt any other tubing would have failed nearly
as easily.


Fortunately that is not possible by any normal failure
method.

1) P3 had thermal shutdown, maybe it was an Athlon prior to
the motherboards having thermal shutdown.

2) Vinyl tubing doesn't melt at a very low temp, something
else was going on that was already a severe system failure
independent of the tubing. Might have been PSU, or
motherboard shorting out, CPU alone cannot create enough
heat to melt the solder on the heatsink, it would have far
sooner ruptured and PSU shut off.

3) PSU being crude /defective is the likely problem, any
normal PSU would shut down from overcurrent condition long
before delivering enough heat to melt solder on a heatsink
or melt vinyl tubing. It just isn't a realistic failure
scenario with any properly built PC, water-cooled or not.

4) As I've mentioned previously, the electrical wiring
insulation itself has no higher temp rating than the melting
point of vinyl tubing. If excessive temps are the problem
then we start drifting into the topic of lowest melting
point materials and rewiring everything with teflon wiring
but does anyone do that for a PC?

Regardless of the above I'd love to see those pics.

I wish I could find them. The water block was just a box of flat
copper pieces soldered together, and a sheet of solder spanning at
least 3" completely coated a part of the motherboard. I'm guessing a
CPU can generate enough power to melt solder, if it doesn't shut down
first, as 20C/W isn't unreasonable for a small water block running dry.

wrote in message
...


Ken Maltby wrote:
wrote in message
...


There were some photos of a failed homemade water cooling system,
probably for a Pentium III or original Athlon, where the vinyl tubing
had melted, and the loss of water led to the soldered copper water
block falling apart and spilling molten solder over a large area of
the motherboard. I doubt any other tubing would have failed nearly
as easily.


Total BS. Post a citation or link to these photos so
we can get the whole story.

It was probably 10 years ago, and you shouldn't call people liars,
little boy.

In other words, you can't supply any valid reference to the
photos you claim to have seen. Stop making up phony and
implausible "evidence" in support of your claims. Most CPU
manufacturers expect a total thermal breakdown for the
devices at ~120c. Lead-Tin based solder melts between
183c and 250c. The current Leadfree solder at higher temp.

All modern Waterblocks are not soldered. Back when it
was very rarely done, silver soldering/ brazing was used.
(500c melting points)

That would be 16yrs retired Plt Sgt "little boy", to you.

Luck;
Ken

On Thu, 22 Oct 2009 15:35:58 -0700 (PDT),
wrote:


Regardless of the above I'd love to see those pics.

I wish I could find them. The water block was just a box of flat
copper pieces soldered together, and a sheet of solder spanning at
least 3" completely coated a part of the motherboard. I'm guessing a
CPU can generate enough power to melt solder, if it doesn't shut down
first, as 20C/W isn't unreasonable for a small water block running dry.


For the heatsink to be at the melting point of solder, the
CPU die would have had to be hotter than that. It could not
be, it would fracture, short out, the load would immediately
shut down the system unless the PSU is faulty, in which case
the extreme current through the switching regulation circuit
transistors would short them, or at the very least they
would get so hot that they could melt solder.

I don't think the CPU did it, I think the fets did because
of a faulty PSU. Water cooling wasn't the cause then, only
the cascade result of a prior failure so it didn't really
matter whether it was water cooled.