29 Dec
2001
29 Dec
'01
5:32 a.m.
From: William Donzelli <aw288(a)osfn.org>
I would think the thermal shock of the chips hitting
the hot water would
be a failure mechanism. Hot air and hot water are different things, even
at the same temperature. The thermal resistance of a water to ceramic
(or
Not an issue as the temps are well blow boiling (nominal 145f).
plastic) junction is much, much lower, than one with
air, so during the
first seconds of the wash cycle, the chips go from ambient
temperature to something rather high. Lots of stress results, especially
if only some of the chip's package gets wet. Preheating the boards would
help reduce the shock greatly, but home dishwashers do not do that (not
a
good idea to cook the food onto the plates before
trying to wash them!).
Not required.
The above is one of the reasons why liquid cooled
electronics are a bit
tricky. The cooling units *never* just start pumping cold water when the
power is applied. There is always a stabilization period, so the shock
is
reduced.
Some do, other run the cooling first and let the coolant temp climb to
operating temp gradually. Others preheat the coolant to working temp
so the system can come on line faster.
Allison
29 Dec
29 Dec
5:49 a.m.
Not an issue as the temps are well blow boiling
(nominal 145f).
You missed the whole point. A sudden change (as in a second or so) from
65 F to 145 F will shock a chip far more than a gradual change (15
seconds) from 65 F to 212 F. It is the rate of change, and not the
change itself, that matters. With hot water hitting the chips instantly,
the rate of change is going to be *really* fast. It may also be uneven -
if a large chip only gets half soaked with the hot water at startup.
William Donzelli
aw288(a)osfn.org
4:14 a.m.
William Donzelli wrote:
But is not that change the same as going from COLD to HOT when you apply
power to the chip under normal operation of the computer?
--
Ben Franchuk - Dawn * 12/24 bit cpu *
www.jetnet.ab.ca/users/bfranchuk/index.html
Not an issue as the temps are well blow boiling
(nominal 145f).
You missed the whole point. A sudden change (as in a second or so) from
65 F to 145 F will shock a chip far more than a gradual change (15
seconds) from 65 F to 212 F. It is the rate of change, and not the
change itself, that matters. With hot water hitting the chips instantly,
the rate of change is going to be *really* fast. It may also be uneven -
if a large chip only gets half soaked with the hot water at startup.
William Donzelli
aw288(a)osfn.org
8:07 a.m.
But is not that change the same as going from COLD to
HOT when you apply
power to the chip under normal operation of the computer?
For the big stuff, sure. Not for the small stuff (glue, I/O, memory, and
other stuff that does not suck much power).
Here is a test. Boil some water in a big pot, as if you were to make
pasta. Turn the gas off, let it cool just a few seconds, then dump all your
favorite chips in. I am sure one will not comeout alive (or maybe crippled).
William Donzelli
aw288(a)osfn.org
5:31 a.m.
William Donzelli wrote:
Now who would want soggy french fries ... My favorite chip!
I expect there are better ways to clean a board rather
than in the dishwasher.
While you hear a lot about 'classic' common old CPU's -- apple, radio
shack,
commodore do people find any homebrewed computers did that all stop when
the S-100 bus came out?
--
Ben Franchuk - Dawn * 12/24 bit cpu *
www.jetnet.ab.ca/users/bfranchuk/index.html
But is not that change the same as going from
COLD to HOT when you apply
power to the chip under normal operation of the computer?
For the big stuff, sure. Not for the small stuff (glue, I/O, memory, and
other stuff that does not suck much power).
Here is a test. Boil some water in a big pot, as if you were to make
pasta. Turn the gas off, let it cool just a few seconds, then dump all your
favorite chips in. I am sure one will not comeout alive (or maybe crippled).
William Donzelli
aw288(a)osfn.org 
8:32 a.m.
On Sat, 29 Dec 2001, William Donzelli wrote:
Scarcely a valid test. In the first place, the water temperature is
dramaticly higher than 145-160 degrees F and, in the second place,
dishwashers spray the hot water on their contents rather than submerge
them.
- don
But is not
that change the same as going from COLD to HOT when you apply
power to the chip under normal operation of the computer?
For the big stuff, sure. Not for the small stuff (glue, I/O, memory, and
other stuff that does not suck much power).
Here is a test. Boil some water in a big pot, as if you were to make
pasta. Turn the gas off, let it cool just a few seconds, then dump all your
favorite chips in. I am sure one will not comeout alive (or maybe crippled). William Donzelli
aw288(a)osfn.org
9:39 a.m.
> Here is a test. Boil some water in a big pot, as
if you were to make
On Sat, 29 Dec 2001, Don Maslin wrote:
Scarcely a valid test. In the first place, the water
temperature is
dramaticly higher than 145-160 degrees F and, in the second place,
dishwashers spray the hot water on their contents rather than submerge
them.
And where can I get a dish^H^H^H^Hboard washer that STARTS at 150 degrees?
Mine takes a while to flush the cold water through the system before it
starts spraying hot. When I turn on a "HOT" water faucet, in MY house, it
takes it close to a minute to come up to temperature. One of the
consequences of long pipes and not sharing plumbing with a lot of other
people.
But MOST OF ALL:
Always shut down the machine before running it through the
dishwasher! Park the hard-drive and turn off the monitor.
Besides, if the computer is plugged in, the cord keeps the door from
sealing completely, and some water leaks out. A water-tight bulkhead
fitting would solve that, but I'd still want a GFCI for it.
Q: will a Pentium chip put out enough heat to boil water? It takes it a
long time to toast bread.
4:56 a.m.
Hey,
Q: will a Pentium chip put out enough heat to boil
water? It takes it a
long time to toast bread.
16W max typical for 166-200, same for MMX as well. I suggest bunch
of cheap durons (35W to 50W) and seal it up w/ small cylinder that
water is in contact with die itself and bring that water to boil.
:-)
Cheers,
Wizard
6:50 a.m.
"Fred Cisin (XenoSoft)" wrote:
Q: will a Pentium chip put out enough heat to boil
water? It takes it a
long time to toast bread.
A: No use several Z80's. They have the same thermal characteristics as
standard stove elements, in watts per mm^2. The Pentium chips run
hotter internally but ever since the Three Mile Island* meltdown there
as been laws against more than 1 cpu per motherboard.
* The meltdown was from the control computers -- not the reactor! :)
--
Ben Franchuk - Dawn * 12/24 bit cpu *
www.jetnet.ab.ca/users/bfranchuk/index.html
12:40 p.m.
On Sat, 29 Dec 2001, Ben Franchuk wrote:
A: No use several Z80's. They have the same
thermal characteristics as
standard stove elements, in watts per mm^2. The Pentium chips run
hotter internally but ever since the Three Mile Island* meltdown there
as been laws against more than 1 cpu per motherboard.
* The meltdown was from the control computers -- not the reactor! :)
SNL (Saturday Night Live) postulated that Three Mile Island was a case of
Pepsi Syndrome - somebody spilled a Pepsi into one of the control computer
keyboards.
Soon after that, the US and CCCP (USSR) normalized relations enough to
exchange vodka for Pepsi, introducing Pepsi into Russia. Right after that
was Chernobyl.
Apple's ride onto the rocks was with a Pepsi executive at the helm.
Unrelated incidents? maybe
12:46 p.m.
Scarcely a valid test. In the first place, the water
temperature is
dramaticly higher than 145-160 degrees F and, in the second place,
dishwashers spray the hot water on their contents rather than submerge
them.
Wow, I did not think anyone would take the test so seriously. Lighten up
a bit. It is just an example.
William Donzelli
aw288(a)osfn.org
8346
days inactive
8346
days old
10 comments
6 participants
participants (6)
-
ajp166＠bellatlantic.net -
aw288＠osfn.org -
bfranchuk＠jetnet.ab.ca -
cisin＠xenosoft.com -
donm＠cts.com -
jpero＠sympatico.ca