User Name
Password
AppleNova Forums » Speculation and Rumors »

G5 PowerBook Thermal Challenge Solved!


Register Members List Calendar Search FAQ Posting Guidelines
View Poll Results: What technology will new PB use to cool G5?
Nothing, G5 will be skipped and an entirely different chip will be used. 28 42.42%
Minor improvements in the existing cooling technology 15 22.73%
Vacuum-packed G5 1 1.52%
Immersion in a bath of cool fluid 6 9.09%
Peltier Joint modules 6 9.09%
Evaporation Cooling 2 3.03%
Radioactive materials 6 9.09%
Others 2 3.03%
Voters: 66. You may not vote on this poll

G5 PowerBook Thermal Challenge Solved!
Page 1 of 4 [1] 2 3 4  Next Thread Tools
Banana
is the next Chiquita
 
Join Date: Feb 2005
 
2005-02-07, 13:14

After thinking about Apple's comments that G5 is the "mother of all thermal challenges" and one poster's quip that Apple's hiring thermal engineers left and right... it occurred to me the problems is much simpler that, I with an outdated computer system and outdated engineering info can solve it! Singlehandedly!

Ok, my first solution:

To cool something rapidly, you can distort the volume so it's thinner and covers more surface area than volume. Whenever surface area is greater than volume, the heat disspaistes more rapidly. Therefore, for PB, G5 processors should be spread over the laptop in sub-compenent. It would mean that data would have to travel more from one subcomponent to next and so forth, resulting in lower clock speed, but at least it'll be significantly easier to cool.

Now if I'm told that G5 can't be laid out flat and must be essentialy intact, my second solution would be to insert it in a vacuum. Boyle's law, which can be stated in many ways, for our purpose, temperature is directly related to pressure and volume. So therefore when pressure is low, even better zero, what does it mean for temperature? It drops like a brick.

Now before you shake your thermos and shout, "Banana, you idiot! Thermos use vacuums to trap heat!" Remember, thermos uses vacuum to separate two space, outside and inside, and both are at atmospheric pressure. So when thermo is closed, the inside container applies pressure against vacuum and therefore holds the heat. But if you were to put a coffee in a container and vacuumed it of all air, bringing pressure to zero, then coffee would rapidly lose its heat as heat will move outward to the surrounding container. Besides, call the NASA and ask them what would happen if one were to be flung in outer space. They'll tell you that the blood will boil in your bloodstream.

Furthermore, what would happen if we put a fan in a vacuum? Because there is no air to move, fan would actually be more efficient at moving heat. So therefore, we insert G5 in a vacuum with a fan to move the heat toward a surface area of metal that conducts heat better than G5. I believe copper is the best, but am assuming that G5 uses copper and can't be changed, so I just hope they'll come up with an alloy that is much better. If the surface area of the alloy is 2:1 or ideally 3:1 to the G5's surface area, it means heat would not only dissipiate rapidly but also cool down enough so it wouldn't be a frying pan. Of course, whether pb has the surface area where it would not be inconvenient to the user is the only question. I'm thinking channeling the heat toward the back of LCD screen, but have no idea if heat do weird stuff to LCD.

If that's not a viable solution, the next solution would be to have a PB that's essentially a dumb terminal and cast the G5 into syncrhonizatic orbit around the earth, so it's always above your general geographical location and leave Apple to develop a high speed satelite connection that maintains its connectivity even during a blizzard, hurricane, and the like. :smokey:

At this point, thermal engineers are prolly laughing their ass off at my ludiricous ideas and my crappy understanding of psychics.. wait, I mean physics, but hey, hey, speculating is fun! :smokey:
  quote
Kickaha
Veteran Member
 
Join Date: May 2004
 
2005-02-07, 13:16

Yes, yes we are laughing our asses off...

Heat can only reside in matter. Remove matter (vacuum), and there's no way to dissipate heat. Spacecraft aren't heated so much as cooled. It's a bitch of a problem. People and electronics generate heat, and there's no way to cool the craft. Yeah, seems odd when you consider that space is just a few degrees above absolute zero, but heat needs a transfer mechanism, either radiation of conduction. Radiation only works for some applications (think infrared), while conduction requires non-vacuum. (It's called thermal mass - think of an oven. You can heat it to 400deg, and stick your hand in it just fine. But if you touch the metal, you get burned. They're both the same temperature, but one has more energy to add to your hand because it simply has more atoms to hold the heat.)

A Thermos doesn't work because of pressure (or it'd explode when you open it) but because by removing the efficient transfer of heat (by placing a vacuum in between), you can keep things at a constant temperature. Consider that you can put ice in a Thermos and keep it frozen for hours, as well as hot coffee hot. It *blocks* heat transfer. A chip in a vacuum would fry itself in minutes, if not seconds.

Now, the spreading things out idea would indeed work for helpful dissipation of heat, but it would also create *more* heat. The longer the traces (wires) on a chip, the more energy is needed to move the information from one section to another. More energy == more heat. So your flattening of the chip to increase surface area is also going to increase heat. One of the big wins of chip design has been *shrinking* the die size, and component size, and *reducing* that distance. Less distance = less energy needed = less heat.

So if what you want is a puddle of boiling silicon, then yes, flattening the chip and putting in a vacuum will work *beautifully*.

Last edited by Kickaha : 2005-02-07 at 13:25.
  quote
Banana
is the next Chiquita
 
Join Date: Feb 2005
 
2005-02-07, 13:20

At least enlighten me where I err and stumble.

May the blind lead the blind!
  quote
Banana
is the next Chiquita
 
Join Date: Feb 2005
 
2005-02-07, 13:24

Quote:
Originally Posted by Kickaha
Yes, yes we are laughing our asses off...

Heat can only reside in matter. Remove matter (vacuum), and there's no way to dissipate heat. Spacecraft aren't heated so much as cooled. It's a bitch of a problem. People and electronics generate heat, and there's no way to cool the craft. Yeah, seems odd when you consider that space is just a few degrees above absolute zero, but heat needs a transfer mechanism, either radiation of conduction. Radiation only works for some applications (think infrared), while conduction requires non-vacuum.
Makes sense. If we figured out how to convert convectional and conducting heat into radiating heat, then i suppose the vacuum thingy would work...

If not, then how does sun moves its heat from there to here (earth)?
  quote
Kickaha
Veteran Member
 
Join Date: May 2004
 
2005-02-07, 13:26

Radiation, the other energy transport mechanism. I have a flashlight that's about 2Mcd (megacandle). The beam feels *hot* when you shine it on yourself, even several feet away. It would also do so in a pure vacuum.
  quote
Banana
is the next Chiquita
 
Join Date: Feb 2005
 
2005-02-07, 13:28

Quote:
Originally Posted by Kickaha

Now, the spreading things out idea would indeed work for helpful dissipation of heat, but it would also create *more* heat. The longer the traces (wires) on a chip, the more energy is needed to move the information from one section to another. More energy == more heat. So your flattening of the chip to increase surface area is also going to increase heat. One of the big wins of chip design has been *shrinking* the die size, and component size, and *reducing* that distance. Less distance = less energy needed = less heat.

So if what you want is a puddle of boiling silicon, then yes, flattening the chip and putting in a vacuum will work *beautifully*.
Ah ha, don't forget the superconducting traces! No resistance=no heat=cool chip.

NOw only if I can keep that liquid nitrogen cool enough.....

A chip in a bath of liquid nitrogen in a thermo?
  quote
Banana
is the next Chiquita
 
Join Date: Feb 2005
 
2005-02-07, 13:29

Quote:
Originally Posted by Kickaha
Radiation, the other energy transport mechanism. I have a flashlight that's about 2Mcd (megacandle). The beam feels *hot* when you shine it on yourself, even several feet away. It would also do so in a pure vacuum.
Was that to imply that heat can be converted from one form to other (conduction/convection to radiation?)
  quote
ezkcdude
Veteran Member
 
Join Date: Jan 2005
 
2005-02-07, 14:08

I say we leave it for the Apple engineers. Obviously, there are solutions to the problem. It's a matter of cost that makes it difficult.
  quote
octavist13
Member
 
Join Date: Jan 2005
Location: IL
 
2005-02-07, 15:35

They need to check the specs on the endline rotarty girder, err, the kanuter valve, or something.
  quote
ashon3611
New Member
 
Join Date: Sep 2004
 
2005-02-07, 16:27

This reminds me of when my friends thought he developed a way to create energy (and thus electricity) by making a magnet in the shape of a donut that was hollowed out in the middle to create a circular inner tube where he would have another magnet ball in the inside that could continuously go around the in the inside of the donut forever.

To get it started he said he would have electricity running a magnetic field that would rotate the the poles of the sides of the donut and blah blah blah blah blah...

And after all that, the way the electricity to create the magnetic field would be powered would be from solar power... So instead of converting:
solar power -> electricity

he would have:
solar power -> electricity (to power the donut) -> a magnetic field (to rotate the ball) -> mechanical energy -> electricity

He's not majoring in engineering anymore.
  quote
Banana
is the next Chiquita
 
Join Date: Feb 2005
 
2005-02-07, 16:39

ashon3611,

sounds like an inefficient machine to me... not to mention the bitch of second law of thermodynamics which precludes any perpetual motion machine....

No wonder he's not in engineering...

Here's my next solution... Remmy, it's all in fun.

Seeing that Apple has everything wireless and bluetooth enabled, I figured, gee, why not just put G5 in a brick outside of the laptop and communicate wirelessly or bluetoothly or both or whatever.

Just think.

If you're at home and need to get the most out of the PB, just stick the brick in freezer. See dramatic improvements.

If you're on a plane, stick it in baggage compartment. Nobody'll notice the heat... at least until someone touchs the door.

and as a safety precaution, the G5 should be equipped with GPS and serial number identifying, so it's unique to only one PB and impossible to steal and lose.

There! Problem solved!

:smokey:
  quote
Ebby
Subdued and Medicated
 
Join Date: May 2004
Location: Over Yander
Send a message via AIM to Ebby  
2005-02-07, 19:45

Wasn't there a liquid cooling system developed for laptops that used some sort of alcohol and a vacuum (at least less pressure) to transfer the heat away from the processor and used the entire back of the screen as a heatsink? I remember reading this somewhere a long time ago.

^^ One more quality post from the desk of Ebby. ^^
SSBA | SmockBogger | SporkNET
  quote
billybobsky
BANNED
I am worthless beyond hope.
 
Join Date: May 2004
Location: Inner Swabia. If you have to ask twice, don't.
 
2005-02-07, 20:13

Actually there is a way to use liquid cooling in a laptop -- but it has to be a liquid with a high thermal expansivity (to allow the system to "pump" without creating more energy) hence the alcohol.

Banana's lack of understanding of basic science causes me no end of misery.
  quote
sith_lord
Member
 
Join Date: Jan 2005
 
2005-02-07, 23:54

I'm no expert at engineering but Banana's speculation on a vacuum solution sounds more like the processor entombed inside a light bulb. It'll burn bright for so many hours of operations and then it'll burn out and we'll have to replace it! I say no. There has to be a better way. Liquid coolant is the only feasable solution but I have my doubts to it's dependabilty in the long run. G5 or no G5 I'd be happy if Apple ever develops the first dual G4 processor laptop.

I didn't read the posting guidelines.
  quote
Ebby
Subdued and Medicated
 
Join Date: May 2004
Location: Over Yander
Send a message via AIM to Ebby  
2005-02-08, 06:20

Quote:
Originally Posted by Corpus_Callosum
Please read and post your thoughts on what you see on this page.
I see a new laptop in my near future.
  quote
octavist13
Member
 
Join Date: Jan 2005
Location: IL
 
2005-02-08, 12:17

Say it ain't so?!
  quote
JayReding
Member
 
Join Date: Oct 2004
Location: Minneapolis, MN
 
2005-02-08, 12:29

Quote:
Originally Posted by Corpus_Callosum
http://www.digitimes.com/news/a20050114A7040.html

Please read and post your thoughts on what you see on this page.
It's completely and utterly and totally untrue. There will not be any G5 laptops in Q2 of 2005, especially not since the PB line just got a speed bump.
  quote
kaupena
Member
 
Join Date: Jan 2005
 
2005-02-08, 12:35

Isn't this the same typo that we all went through in the monster PB discussion?
  quote
709
¡Damned!
 
Join Date: May 2004
Location: Purgatory
 
2005-02-08, 12:51

Quote:
Originally Posted by Corpus_Callosum
http://www.digitimes.com/news/a20050114A7040.html

Please read and post your thoughts on what you see on this page.
Christ, man. 1 post in 4 threads.

There's already a thread about this article here.
  quote
curiousuburb
Antimatter Man
 
Join Date: May 2004
Location: that interweb thing
 
2005-02-08, 14:37

Quote:
Originally Posted by 709
Christ, man. 1 post in 4 threads.

There's already a thread about this article here.
Equally unlikely from the underlying physics, but cool speculation nonetheless...


Aerogel
might seem likely as well... insulates so well they're now making winter wear out of it.

(Except for the minor detail that insulators aren't the only solution...
we'd prefer efficient dissipation to efficient insulation as a thermal fix.)
  quote
Banana
is the next Chiquita
 
Join Date: Feb 2005
 
2005-02-08, 17:47

Quote:
Originally Posted by billybobsky
Banana's lack of understanding of basic science causes me no end of misery.
Misery? You should be laughing at my buffornery. *put on a jester's hat* Better?

Seriously, I post them here just so I'd learn more how a computer works.

To rehash the flattening cpu thingy... Kickada's reasoning that smaller traces=smaller heat is intact. Yet, I wonder if there is somehow a balance; not to mention the fact that increasing cross section area of trace would also decrease resistance (heat)... So I'm just wondering if flattening CPU at least to a point is viable. But then there's only 127 cubic inches in a PB, as I'm told... Ah well.

Lemme see...

I know!

Let's put afterburners on the rear end! This would sell! Just think of the coolness factor..

Hey, guys look at this... a laptop that lightens up a cigarette! :smokey:
  quote
intlplby
Veteran Member
 
Join Date: Dec 2004
 
2005-02-08, 21:01

did you know aerogel is edible?
  quote
BlueRabbit
Veteran Member
 
Join Date: May 2004
Location: San Francisco, CA
 
2005-02-08, 21:25

Quote:
Originally Posted by intlplby
did you know aerogel is edible?
But does it taste good?

At this point, I can't see Apple introducing a laptop with lots of fans – that just wouldn't fit with their current portable strategy. Therefore, the most probable way to cool the thing will be something out of left field, like Aerogel. However, it would be pretty interesting if the new PB came with a tailpipe coming out of it.
  quote
billybobsky
BANNED
I am worthless beyond hope.
 
Join Date: May 2004
Location: Inner Swabia. If you have to ask twice, don't.
 
2005-02-08, 22:47

Quote:
Originally Posted by Banana
Misery? You should be laughing at my buffornery. *put on a jester's hat* Better?
Misery because I am a scientist.
  quote
Kickaha
Veteran Member
 
Join Date: May 2004
 
2005-02-08, 23:14

Quote:
Originally Posted by Banana
Was that to imply that heat can be converted from one form to other (conduction/convection to radiation?)
Um, yes?

Never heard the phrase 'red hot'? Something that hot gives off visible light... ie, radiation. This is how most electric heaters work. The electricity heats up the metal, which gets so hot it glows, and the infrared and red radiation spread out into the room, and heat anything they contact.

Jeez mon. This is like 3rd grade science here...
  quote
Banana
is the next Chiquita
 
Join Date: Feb 2005
 
2005-02-09, 13:02

Quote:
Originally Posted by Kickaha
Um, yes?

Never heard the phrase 'red hot'? Something that hot gives off visible light... ie, radiation. This is how most electric heaters work. The electricity heats up the metal, which gets so hot it glows, and the infrared and red radiation spread out into the room, and heat anything they contact.

Jeez mon. This is like 3rd grade science here...

But the metal itself is still hot... so there's going to be problems with the processor if it's glowing red hot... When I posted that question, I should have rephrased it-

can heat be converted directly to radiant form without having to conducting/ convecting heat...

(i.e. those new ovens that are cool to touch but cooks food, but i don't know the underlying physics behind, so it might be a entirely different prinicple.)


Also, I thought up another novel solution.... not that it'd be the efficient/best/cheap solution...

As explained above, Boyle's law describes the relationship between temparture, volume, pressure and density. Then there's second law of thermodynamics, which for our purpose, heat always moves from hotter to cooler region (and more faster when the inequality is greater).

So... liquid cooling works because of greater thermal mass which takes more heat rapidly.

However, I wonder if same thing would work if we were to put an air compressor next to the processor, and put the chamber into higher pressure (ideally without increasing temp- not easy to do as compressing air generates heat), and when the g5 get hot, air takes the heat, then the chamber expels the air, depressurizing, and therefore moving more heat....

I bought that up because liquid is incompressible and I wondered if it'd be possible to use air that has variable thermal mass (because it can be compressed).

*pea-sizd brain grows by a miniscule fraction, is progressing nicely toward graduating 1st grade science*
  quote
curiousuburb
Antimatter Man
 
Join Date: May 2004
Location: that interweb thing
 
2005-02-09, 13:25

Quote:
Originally Posted by Banana

Also, I thought up another novel solution.... not that it'd be the efficient/best/cheap solution...

However, I wonder if same thing would work if we were to put an air compressor next to the processor, and put the chamber into higher pressure (ideally without increasing temp- not easy to do as compressing air generates heat), and when the g5 get hot, air takes the heat, then the chamber expels the air, depressurizing, and therefore moving more heat....

I bought that up because liquid is incompressible and I wondered if it'd be possible to use air that has variable thermal mass (because it can be compressed).

*pea-sizd brain grows by a miniscule fraction, is progressing nicely toward graduating 1st grade science*
I think there's already an air-compressor PowerMac division, from Speed Hole to the 9-fan G5.

Although I must admit a macabre joy at reading your posts...
always a new laugh as the line "I've thought of a novel solution"
especially when followed by: "if we could only get the GPU to absolute zero and 80,000 Roentgens".

Comedy genius.
  quote
Banana
is the next Chiquita
 
Join Date: Feb 2005
 
2005-02-09, 13:29

Quote:
Originally Posted by curiousuburb
I think there's already an air-compressor PowerMac division, from Speed Hole to the 9-fan G5.
Weird. The specs didn't say anything about air compressing... I saw article on the plastic casing, thermal zones, and fans etc but no air compressing. Methinks need another look.


Quote:
Originally Posted by curiousuburb
Although I must admit a macabre joy at reading your posts...
always a new laugh as the line "I've thought of a novel solution"
especially when followed by: "if we could only get the GPU to absolute zero and 80,000 Roentgens".

Comedy genius.
Thank you thank you! My purpose in this life has been fulfilled! *delicious with joy*
  quote
Banana
is the next Chiquita
 
Join Date: Feb 2005
 
2005-02-10, 17:12

After reading about $100 laptop, it all suddenly occurs to me.... Apple actually can ship G5 powerbooks cheaper than G4 and fix the thermal problems as well.

See, if we put in material that radiate more enegry (read=heat), more readily, then it should be no problem keeping it cool. And what is this magic material?

Deleted uranium!!!

Seriously, there are million tons of those lying around in Benton and gov't would be more than happy to get rid of those... Apple can buy them 1 cent a ton and use it to make the chips.

and they can market it as "the first laptop ever with birth control built in!"



If that's no-go, then Apple can just start shipping G5 powerbook now, sales limited only to residents of Alaska, Northern Canada, Siberia, Scandinivan, Greenland and of course, Antarcia.
  quote
JayReding
Member
 
Join Date: Oct 2004
Location: Minneapolis, MN
 
2005-02-10, 17:22

Actually DU is only a light alpha emitter. Alpha particles don't penetrate matter much denser than air. A DU laptop would be a very bad form of birth control.

Plus, I think lugging around a laptop with a DU casing might be a bit heavy...
  quote
Posting Rules Navigation
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Page 1 of 4 [1] 2 3 4  Next

Post Reply

Forum Jump
Thread Tools

« Previous Thread | Next Thread »

All times are GMT -5. The time now is 04:11.


Powered by vBulletin®
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004 - 2024, AppleNova