Alain escribió:
>> * Impure water is an excellent conductor, while pure water is a very 
>> good insulator. Yet both substances have almost identical optical 
>> properties.
> ABSOLUTELY pure water is a prety good insulator, but just the tyniest 
> impurety will change it into a conductor.
>

I read on a website about an interesting CPU watercooling technique: 
submerging the motherboard. They still had trouble because of ions on 
the water. It's hard to get *totally* pure water...

Post a reply to this message

Nicolas Alvarez wrote:

> I read on a website about an interesting CPU watercooling technique: 
> submerging the motherboard. They still had trouble because of ions on 
> the water. It's hard to get *totally* pure water...

We have a machine at work that produces "20 mega-Ohm water".

Of course, water reacts with *air* to absorb various ions. (For example, 
CO2 disolves in water to yield carbonic acid.)

But seriously - computers work at 12V. The water's resistance would have 
to plummet to yield any meaningful current.

I would imagine a far bigger problem is water's viscosity; I'd think 
fans wouldn't like that...

Post a reply to this message

Orchid XP v7 escribió:
> I would imagine a far bigger problem is water's viscosity; I'd think 
> fans wouldn't like that...

Not really. They then had much better results with oil.

Post a reply to this message

Nicolas Alvarez nous apporta ses lumieres en ce 2007/11/09 21:28:
> Alain escribió:
>>> * Impure water is an excellent conductor, while pure water is a very 
>>> good insulator. Yet both substances have almost identical optical 
>>> properties.
>> ABSOLUTELY pure water is a prety good insulator, but just the tyniest 
>> impurety will change it into a conductor.
>>
> 
> I read on a website about an interesting CPU watercooling technique: 
> submerging the motherboard. They still had trouble because of ions on 
> the water. It's hard to get *totally* pure water...
Even if you have totally pure, desionised, degazed water, it will never stay 
that way. When you submerge your components, the water comes in contact with 
bare metal parts. Not all of those parts are of the same metal. Some of that 
metal will dissolve in the water, and, while doing so, will produce some 
current, whitch in turn will accelerate the process. Turn the power on and this 
will accelerate 100 folds.
In a few minutes, your inert water will become to conductive.

You need a non-ionic fluid like oils or other organic, non-aquous, fluid.

Alchool could be used, as well as glycerin.

-- 
Alain
-------------------------------------------------
You know you've been raytracing too long when you call in sick in order to render.
David Kraics

Post a reply to this message

Orchid XP v7 nous apporta ses lumieres en ce 2007/11/10 04:19:
> Nicolas Alvarez wrote:
> 
>> I read on a website about an interesting CPU watercooling technique: 
>> submerging the motherboard. They still had trouble because of ions on 
>> the water. It's hard to get *totally* pure water...
> 
> We have a machine at work that produces "20 mega-Ohm water".
> 
> Of course, water reacts with *air* to absorb various ions. (For example, 
> CO2 disolves in water to yield carbonic acid.)
> 
> But seriously - computers work at 12V. The water's resistance would have 
> to plummet to yield any meaningful current.
> 
> I would imagine a far bigger problem is water's viscosity; I'd think 
> fans wouldn't like that...
Main power is 5V for all the electronics. The 12V is for the fans and drives motors.

A good deal of the currents are counted in miliamps or less. In particular, the 
data transmition use very low curents at 5V. A leak of less than a 1 miliamp can 
easily cause malfunctions. And then, those 12V going to various motors can push 
or pull some curents into the data and address paths...

You don't need fans on a submerged system, if you keep them, it's only for the 
show with a clear case. Convection does all of the work.

-- 
Alain
-------------------------------------------------
Keep your eyes wide open before marriage, half shut afterwards.
Benjamin Franklin

Post a reply to this message

Alain wrote:
> Orchid XP v7 nous apporta ses lumieres en ce 2007/11/10 04:19:
>> Nicolas Alvarez wrote:
>>
>>> I read on a website about an interesting CPU watercooling technique: 
>>> submerging the motherboard. They still had trouble because of ions on 
>>> the water. It's hard to get *totally* pure water...
>>
>> We have a machine at work that produces "20 mega-Ohm water".
>>
>> Of course, water reacts with *air* to absorb various ions. (For 
>> example, CO2 disolves in water to yield carbonic acid.)
>>
>> But seriously - computers work at 12V. The water's resistance would 
>> have to plummet to yield any meaningful current.
>>
>> I would imagine a far bigger problem is water's viscosity; I'd think 
>> fans wouldn't like that...
> Main power is 5V for all the electronics. The 12V is for the fans and 
> drives motors.
> 
> A good deal of the currents are counted in miliamps or less. In 
> particular, the data transmition use very low curents at 5V. A leak of 
> less than a 1 miliamp can easily cause malfunctions. And then, those 12V 
> going to various motors can push or pull some curents into the data and 
> address paths...

My point was more that at a mere 12V, the resistance has to drop *very* 
low indeed for any noticable current to happen. (Ohm's law, and all that...)

Post a reply to this message

Orchid XP v7 nous apporta ses lumieres en ce 2007/11/10 17:17:
> Alain wrote:
>> Orchid XP v7 nous apporta ses lumieres en ce 2007/11/10 04:19:
>>> Nicolas Alvarez wrote:
>>>
>>>> I read on a website about an interesting CPU watercooling technique: 
>>>> submerging the motherboard. They still had trouble because of ions 
>>>> on the water. It's hard to get *totally* pure water...
>>>
>>> We have a machine at work that produces "20 mega-Ohm water".
>>>
>>> Of course, water reacts with *air* to absorb various ions. (For 
>>> example, CO2 disolves in water to yield carbonic acid.)
>>>
>>> But seriously - computers work at 12V. The water's resistance would 
>>> have to plummet to yield any meaningful current.
>>>
>>> I would imagine a far bigger problem is water's viscosity; I'd think 
>>> fans wouldn't like that...
>> Main power is 5V for all the electronics. The 12V is for the fans and 
>> drives motors.
>>
>> A good deal of the currents are counted in miliamps or less. In 
>> particular, the data transmition use very low curents at 5V. A leak of 
>> less than a 1 miliamp can easily cause malfunctions. And then, those 
>> 12V going to various motors can push or pull some curents into the 
>> data and address paths...
> 
> My point was more that at a mere 12V, the resistance has to drop *very* 
> low indeed for any noticable current to happen. (Ohm's law, and all 
> that...)
Not that much. When the distance is short and the surface relatively large, even 
10K ohm can let pass to much durrent.
On a typical computer mother board, there are 1000's of places where the 
distances are less than 0.5mm, with neigoing surfaces in the 2~3 mm square 
range. Have some 200 MHz signal passing there and you can easily get currents in 
the 0.1 miliamp range, and a parasitic current of about 0.01 miliamp is 
sometimes enough to cause malfunctions.
So, no, you don't have to have a big drop in resistivity at all.
The 20 Mohm water quickly drop to the 100Kohm range. And that's for a 1cm cube 
of water. A 1 cm travel trough a 1 cm square section. In a case where it's 0.5mm 
with a section of 2 mm square, you divide the resistance by 4! Add some 
contaminant buildup to the mixture, whose concentration is the greatest near 
those conductors, and you easily get a HUGE resistivity drop.

-- 
Alain
-------------------------------------------------
You know you've been raytracing too long when you stop using a protractor to 
measure angles because you can do it just by looking.
Taps a.k.a. Tapio Vocadlo

Post a reply to this message

Alain wrote:
> A wave guide ior tend to go higher as the wave length go shorter. There 
> is a wave length rhere it's ior is = 1, and at shorter wave length, it 
> goes > 1.

Huh. Very cool. Thanks for explaining.

-- 
   Darren New / San Diego, CA, USA (PST)
     Remember the good old days, when we
     used to complain about cryptography
     being export-restricted?

Post a reply to this message

Invisible wrote:
> Well, you know, there are conductive materials that are reflective, and 
> ones that aren't. There are insulators that are reflective, and ones 
> that aren't. There seems to be little correlation here.

I suppose if you don't understand what makes something a conductor or 
what makes it reflective, that could be the case, yes.

>>> * Electricity does not, under any remotely "normal" conditions, 
>>> produce light or affect it in any way. (E.g., you can't bend light 
>>> using electricity.) The same goes for magnetism.
>>
>> Except for photoelectric effects, LEDs, solar cells, florescent light 
>> bulbs, all that sort of thing.
> 
> Solar cells work by using strange chemistry rather than directly turning 
> light into electricity. (Presumably that's why they're so inefficient.)

Depends on the solar cell, but in general, I think when you say "strange 
chemistry", you don't understand that "chemistry" is an electromagnetic 
effect.

> Florescent light bulbs work by stimulating atoms to release photons, not 
> by directly turning electric oscilations into light.

And what is it that stimulates atoms to release photons?

> I have no clue why LEDs work. But apparently they do. ;-)

You have a crystal of semiconductor (i.e., 4 valence electrons), one 
side of which is doped with a mild concentration of atoms with 5 valence 
electrons, one with a mild concentration of 3 valence. (Say, gallium and 
arsinide.) You put negative voltage (i.e., an excess of electrons) on 
the side with 5 electrons, and it pushes them across the gap to the side 
with waiting "holes", into which they fall. As they fall, they release a 
photon of a color specific to how far they fall.

Originally, the cheapest chemicals to do this with had holes deep enough 
to emit red light. Hence the traditional red LEDs.

For bonus: A light sensitive diode is the same thing, basically, with 
less electricity. When light hits it, it knocks an electron free, which 
can then migrate across the junction and into the + side of the voltage, 
causing current that can be detected.

So LEDs both turn electricity *directly* into photons and vice versa.

>> Don't you use a computer? What do you think you're looking at?
> 
> Electricity can be used to excite atoms in such a way that they release 
> photons. So can heat energy, chemical energy, and all kinds of other 
> energy.

Nope. Heat doesn't stimulate atoms to release photons. The electrons in 
atoms banging into electrons in other atoms generates photons. When you 
heat atoms up, they bang into each other much more.

But take a bunch of atoms that are very hot but going all the same way, 
and you won't get much light out, until they hit something.  You know, 
like a cosmic ray.

And, as I said, "chemical energy" is energy in the valence electrons of 
atoms. Hence, it's electromagnetic.

> So it seems that unless you can find a material with an index of 
> refraction of several thousand, you aren't going to make light with an 
> electric oscilator.

Well, no, not like that sort of thing. You'd have to use a quantum-scale 
ossilator.

-- 
   Darren New / San Diego, CA, USA (PST)
     Remember the good old days, when we
     used to complain about cryptography
     being export-restricted?

Post a reply to this message

Phil Cook wrote:
> A stopwatch and an astronaut with really quick reflexes.

You're not joking. The first evidence that light had a speed was someone 
(galileo?) noticing that the shadows cast by jupiter's moons on 
jupiter's clouds were not where they should have been.

-- 
   Darren New / San Diego, CA, USA (PST)
     Remember the good old days, when we
     used to complain about cryptography
     being export-restricted?

Post a reply to this message