|
|
|
|
| Author |
Message |
peter
Guest
|
 Posted: Wed Apr 04, 2007 8:14 am Post subject: solid heat sink? |
 |
|
I want to buy a heat sink to replace the intel one that comes with c2d
E4300.
Many fancy heat sinks hype "heat pipe" technology. But wouldn't a solid
block of copper conduct heat better than a few thin copper pipes?
Imagine a copper block that is 1.2 inch square x 3 inches tall with fins
radiating out, and a fan on top. That is what I'm looking for. Perhaps the
copper block should tapper off like a pyramid so it is not too heavy.
Is there such product? |
|
| Back to top |
|
 |
|
|
Paul
Guest
|
| Posted: Wed Apr 04, 2007 9:27 am Post subject: Re: solid heat sink? |
|
|
peter wrote:
| Quote: | I want to buy a heat sink to replace the intel one that comes with c2d
E4300.
Many fancy heat sinks hype "heat pipe" technology. But wouldn't a solid
block of copper conduct heat better than a few thin copper pipes?
Imagine a copper block that is 1.2 inch square x 3 inches tall with fins
radiating out, and a fan on top. That is what I'm looking for. Perhaps the
copper block should tapper off like a pyramid so it is not too heavy.
Is there such product?
|
Actually that is not true. The heat pipe is actually better than a
solid piece of copper. One web page mentioned it is better by a factor
of 500 (but that number is going to vary with construction details of
the pipe).
http://en.wikipedia.org/wiki/Heat_pipe
"The advantage of heat pipes is their great efficiency in transferring
heat. They are actually a vastly better heat conductor than an equivalent
cross-section of solid copper."
If you want to go back to the previous generation of products, you can shop
for stuff like this:
http://www.techbits.ca/images/reviews/kip/slk900.gif
Heat pipes are fine as long as they are not overwhelmed by the heat.
That means they can only handle so many watts of heat. If the "cool
end" is so hot that vapor cannot condense back to liquid again, the
thermal transport mechanism would be busted. And the temperature would
just climb if that happened. A system with heat pipes should have a
safety mechanism, that can detect if there is a problem with the
cooling ability of the assembly (whether due to fan failure, a
heat pipe leak, or going past the useful operating range). Fortunately,
the latest generation of processors, from both Intel and AMD, has
mechanisms for dealing with an overheat.
Paul |
|
| Back to top |
|
|
Phil Weldon
Guest
|
| Posted: Wed Apr 04, 2007 9:49 am Post subject: Re: solid heat sink? |
|
|
'peter' wrote, in part:
| I want to buy a heat sink to replace the intel one that comes with c2d
| E4300.
|
| Many fancy heat sinks hype "heat pipe" technology. But wouldn't a solid
| block of copper conduct heat better than a few thin copper pipes?
_____
Let's make some adjustments in how to look at heatsinks.
* The heatsink/fan that comes with the boxed retail E4300 is VERY good.
Unless you are going to overclock to above 3 GHz you likely don't need to
even consider changing heatsinks.
* Heatpipes can transfer heat MUCH better than solid copper, as much as
1000 times better. Heatpipes transfer heat by convection and phase change.
A heatpipe works best when the walls are thin because heat transfers INSIDE
the pipe is much better than through the copper walls. One way a heatpipe
can be made is to seal water and a partial vacuum inside. The partial
vacuum lowers the boiling point of the water to, say 45 C. The hot CPU
boils the water (phase change) which then travels (convection) as steam to
the cooler, lower pressure end of the heatpipe. This end is kept cooler by
convective cooling (air blowing through the fins of the heatsink at that
end. The steam is cooled below 45 C, water condenses, and travels as a
liquid through an internal wick to the hot end. The cycle repeats. As long
as the cool end is kept below 45 C, the hot end can never go much above 45
C.
* There are MANY heatsink/fan combos available that are made of aluminum,
copper, or aluminum + copper. Aluminum is lighter and cheaper. Copper
conducts heat about 45% better. The best compromise is to make the lower,
solid end of the heatsink out of copper, and to make the fins out of
aluminum. That's the construction of the Intel supplied heatsink/fan. If
it were completely copper, it would weigh several pounds, and the CPU
operating temperature would be only a few degrees C higher.
* The SHAPE of many after-market heatsink/fan combos is the opposite of
what you suggest. The BOTTOM of the heatsink is smaller than the top.
Motherboards for Core 2 Duo CPUs are very crowded; a heatsink with a bottom
even a small amount larger than the Intel supplied heatsink just would not
fit. The TOP of some after-market heatsink/fan combos is much larger than
the bottom. This allows the use of much larger fans that turn more slowly;
a greater volume of air moves through the heatsink and the noise level is
lower.
* Everything is a tradeoff. Larger fins have more surface for convection
cooling, but the heat must travel by conducting through a longer path in the
larger fins. Convective cooling is more efficient when the temperature
difference between the cooling fluid (air is a fluid, as is water) and the
heat sink is greater. Increasing the fin size helps reduce the hot end
temperature, but as the fin size increases the addition gains in cooling are
reduced. (Heatpipes are one way around this problem.
I started a thread on overclocking an Intel Core 2 Duo a few days earlier
'E4300 and 650i overclocking' (now 'E4300 and 680i overclocking'). CPU
cooling will probably come up quite often in the thread.
What I intend to do (and will report in the thread) is use the Intel E4300
boxed retail heatsink/fan and go for the highest overclock with that setup.
Only then will I begin to try out other methods (to include water cooling,
diamond filled thermal paste, butter, Peltier arrays, ...).
As I write this I am looking at the underside of the heatsink Intel supplies
with the E4300. I am amazed at the extremely small amount of thermal paste
on the heatsink contact area. Three almost translucent stripes about 1/4
inch wide. This indicates that Intel is supplying heatsinks with VERY
smooth, flat surfaces, that the heatspreader top of the E4300 CPU is VERY
smooth and flat, and that the retaining mechanism holds the heatsink and
heatspreader to VERY parallel. A lot of after-market heatsinks have contact
surfaces that are not nearly so flat and smooth. You could very well spend
$50 US on a heatsink/fan that would not keep the E4300 as cool as the Intel
supplied heatsink/fan. I would strongly recommend you use the Intel
supplied heatsink/fan with the Intel applied thermal compound before
considering alternative cooling. The overclock might be more limited by the
memory than by the CPU temperature.
Phil Weldon
"peter" <nospam@nospam.com> wrote in message
news:OKEQh.2893$gb6.2449@trndny07...
|I want to buy a heat sink to replace the intel one that comes with c2d
| E4300.
|
| Many fancy heat sinks hype "heat pipe" technology. But wouldn't a solid
| block of copper conduct heat better than a few thin copper pipes?
|
| Imagine a copper block that is 1.2 inch square x 3 inches tall with fins
| radiating out, and a fan on top. That is what I'm looking for. Perhaps the
| copper block should tapper off like a pyramid so it is not too heavy.
|
| Is there such product?
|
| |
|
| Back to top |
|
|
Venom
Guest
|
| Posted: Wed Apr 04, 2007 10:54 am Post subject: Re: solid heat sink? |
|
|
| Quote: | |
| Imagine a copper block that is 1.2 inch square x 3 inches tall with fins
| radiating out, and a fan on top. That is what I'm looking for. Perhaps
the
| copper block should tapper off like a pyramid so it is not too heavy.
|
| Is there such product?
Just a note on copper. While it will draw heat very quickly, the neat trick |
is to then get the heat out of the copper. |
|
| Back to top |
|
|
Phil Weldon
Guest
|
| Posted: Wed Apr 04, 2007 12:28 pm Post subject: Re: solid heat sink? |
|
|
'Venom' wrote:
| Just a note on copper. While it will draw heat very quickly, the neat
trick
| is to then get the heat out of the copper.
_____
Heat transfers in and out in the same ways. There is no 'draw heat'. Heat
goes in by conduction, convection, or radiation. Heat goes out by
conduction, convection, or radiation. What exactly do you mean by 'While it
will draw heat very quickly, the neat trick is to the get the heat out of
the copper.' If you mean that heat transfers by conduction from the CPU
spreader plate to the heatsink contact surface better than heat transfers by
convection using air, that's true. But it's the same for ANY material used
to construct a heatsink.
Phil Weldon
"Venom" <Venom@Mailhouse.com> wrote in message
news:H4HQh.9080$M.5389@news-server.bigpond.net.au...
|> |
| > | Imagine a copper block that is 1.2 inch square x 3 inches tall with
fins
| > | radiating out, and a fan on top. That is what I'm looking for. Perhaps
| the
| > | copper block should tapper off like a pyramid so it is not too heavy.
| > |
| > | Is there such product?
| >
| >
| Just a note on copper. While it will draw heat very quickly, the neat
trick
| is to then get the heat out of the copper.
|
| |
|
| Back to top |
|
|
|
|
Venom
Guest
|
| Posted: Wed Apr 04, 2007 12:48 pm Post subject: Re: solid heat sink? |
|
|
"Phil Weldon" <notdiscosed@example.com> wrote in message
news:IsIQh.134188$_73.95735@newsread2.news.pas.earthlink.net...
| Quote: | 'Venom' wrote:
| Just a note on copper. While it will draw heat very quickly, the neat
trick
| is to then get the heat out of the copper.
_____
Heat transfers in and out in the same ways. There is no 'draw heat'.
Heat
goes in by conduction, convection, or radiation. Heat goes out by
conduction, convection, or radiation. What exactly do you mean by 'While
it
will draw heat very quickly, the neat trick is to the get the heat out of
the copper.' If you mean that heat transfers by conduction from the CPU
spreader plate to the heatsink contact surface better than heat transfers
by
convection using air, that's true. But it's the same for ANY material
used
to construct a heatsink.
Phil Weldon
As I remember it, the copper slug in the bottom of an aluminium heat sink |
was deemed to be optimum for CPUs.
Then again, in the computer industry we are fed so much bullshit that it is
hard to know which way is up. I don't overclock and have always run the
stock cooler and fan that comes with AMD CPUs so I qualify for the warranty
and in the day of the AthlonXP Barton they got hot but never fried. The kids
have to have something to talk about and heat sinks are as good a subject as
any. My Athlon X2 is sitting on 30c and the ambient temp is 27.5c in here.
That will do me with a stock cooler. |
|
| Back to top |
|
|
Phil Weldon
Guest
|
| Posted: Wed Apr 04, 2007 12:59 pm Post subject: Re: solid heat sink? |
|
|
Correction for my post
news:g7GQh.18122$PL.3499@newsread4.news.pas.earthlink.net...
In the paragraph about the aluminum/copper heatsink tradeoff, the sentence
"If it were completely copper, it would weigh several pounds, and the CPU
operating temperature would be only a few degrees C higher."
should read
"If it were completely copper, it would weigh several pounds, and the CPU
operating temperature would be only a few degrees C LOWER."
(Emphasis added)
Phil Weldon
"Phil Weldon" <notdiscosed@example.com> wrote in message
news:g7GQh.18122$PL.3499@newsread4.news.pas.earthlink.net...
| 'peter' wrote, in part:
|| I want to buy a heat sink to replace the intel one that comes with c2d
|| E4300.
||
|| Many fancy heat sinks hype "heat pipe" technology. But wouldn't a solid
|| block of copper conduct heat better than a few thin copper pipes?
| _____
|
| Let's make some adjustments in how to look at heatsinks.
|
| * The heatsink/fan that comes with the boxed retail E4300 is VERY good.
| Unless you are going to overclock to above 3 GHz you likely don't need to
| even consider changing heatsinks.
|
| * Heatpipes can transfer heat MUCH better than solid copper, as much as
| 1000 times better. Heatpipes transfer heat by convection and phase
change.
| A heatpipe works best when the walls are thin because heat transfers
INSIDE
| the pipe is much better than through the copper walls. One way a heatpipe
| can be made is to seal water and a partial vacuum inside. The partial
| vacuum lowers the boiling point of the water to, say 45 C. The hot CPU
| boils the water (phase change) which then travels (convection) as steam to
| the cooler, lower pressure end of the heatpipe. This end is kept cooler
by
| convective cooling (air blowing through the fins of the heatsink at that
| end. The steam is cooled below 45 C, water condenses, and travels as a
| liquid through an internal wick to the hot end. The cycle repeats. As
long
| as the cool end is kept below 45 C, the hot end can never go much above 45
| C.
|
| * There are MANY heatsink/fan combos available that are made of aluminum,
| copper, or aluminum + copper. Aluminum is lighter and cheaper. Copper
| conducts heat about 45% better. The best compromise is to make the lower,
| solid end of the heatsink out of copper, and to make the fins out of
| aluminum. That's the construction of the Intel supplied heatsink/fan. If
| it were completely copper, it would weigh several pounds, and the CPU
| operating temperature would be only a few degrees C higher.
|
| * The SHAPE of many after-market heatsink/fan combos is the opposite of
| what you suggest. The BOTTOM of the heatsink is smaller than the top.
| Motherboards for Core 2 Duo CPUs are very crowded; a heatsink with a
bottom
| even a small amount larger than the Intel supplied heatsink just would not
| fit. The TOP of some after-market heatsink/fan combos is much larger than
| the bottom. This allows the use of much larger fans that turn more
slowly;
| a greater volume of air moves through the heatsink and the noise level is
| lower.
|
| * Everything is a tradeoff. Larger fins have more surface for convection
| cooling, but the heat must travel by conducting through a longer path in
the
| larger fins. Convective cooling is more efficient when the temperature
| difference between the cooling fluid (air is a fluid, as is water) and the
| heat sink is greater. Increasing the fin size helps reduce the hot end
| temperature, but as the fin size increases the addition gains in cooling
are
| reduced. (Heatpipes are one way around this problem.
|
| I started a thread on overclocking an Intel Core 2 Duo a few days earlier
| 'E4300 and 650i overclocking' (now 'E4300 and 680i overclocking'). CPU
| cooling will probably come up quite often in the thread.
|
| What I intend to do (and will report in the thread) is use the Intel E4300
| boxed retail heatsink/fan and go for the highest overclock with that
setup.
| Only then will I begin to try out other methods (to include water cooling,
| diamond filled thermal paste, butter, Peltier arrays, ...).
|
| As I write this I am looking at the underside of the heatsink Intel
supplies
| with the E4300. I am amazed at the extremely small amount of thermal
paste
| on the heatsink contact area. Three almost translucent stripes about 1/4
| inch wide. This indicates that Intel is supplying heatsinks with VERY
| smooth, flat surfaces, that the heatspreader top of the E4300 CPU is VERY
| smooth and flat, and that the retaining mechanism holds the heatsink and
| heatspreader to VERY parallel. A lot of after-market heatsinks have
contact
| surfaces that are not nearly so flat and smooth. You could very well
spend
| $50 US on a heatsink/fan that would not keep the E4300 as cool as the
Intel
| supplied heatsink/fan. I would strongly recommend you use the Intel
| supplied heatsink/fan with the Intel applied thermal compound before
| considering alternative cooling. The overclock might be more limited by
the
| memory than by the CPU temperature.
|
| Phil Weldon
|
|
| "peter" <nospam@nospam.com> wrote in message
| news:OKEQh.2893$gb6.2449@trndny07...
||I want to buy a heat sink to replace the intel one that comes with c2d
|| E4300.
||
|| Many fancy heat sinks hype "heat pipe" technology. But wouldn't a solid
|| block of copper conduct heat better than a few thin copper pipes?
||
|| Imagine a copper block that is 1.2 inch square x 3 inches tall with fins
|| radiating out, and a fan on top. That is what I'm looking for. Perhaps
the
|| copper block should tapper off like a pyramid so it is not too heavy.
||
|| Is there such product?
||
||
|
| |
|
| Back to top |
|
|
Bob Fry
Guest
|
| Posted: Wed Apr 25, 2007 12:23 am Post subject: Re: solid heat sink? |
|
|
| Quote: | "p" == peter <nospam@nospam.com> writes:
|
p> Many fancy heat sinks hype "heat pipe" technology. But wouldn't
p> a solid block of copper conduct heat better than a few thin
p> copper pipes?
Not when the thin copper pipes contain a phase-change fluid...changing
phase from liquid to vapor absorbs a lot of heat energy.
--
If you ever reach total enlightenment while drinking beer, I bet
you could shoot beer out of your nose.
- Jack Handey |
|
| Back to top |
|
|
Wayne Carr
Guest
|
| Posted: Wed Apr 25, 2007 12:23 am Post subject: Re: solid heat sink? |
|
|
"peter" <nospam@nospam.com> wrote in message
news:OKEQh.2893$gb6.2449@trndny07...
| Quote: | I want to buy a heat sink to replace the intel one that comes with c2d
E4300.
Many fancy heat sinks hype "heat pipe" technology. But wouldn't a solid
block of copper conduct heat better than a few thin copper pipes?
Imagine a copper block that is 1.2 inch square x 3 inches tall with fins
radiating out, and a fan on top. That is what I'm looking for. Perhaps the
copper block should tapper off like a pyramid so it is not too heavy.
Is there such product?
|
Whe did you do your course in thermodynamics? |
|
| Back to top |
|
|
Guest
|
| Posted: Fri May 04, 2007 12:12 am Post subject: Re: solid heat sink? |
|
|
On Tue, 24 Apr 2007 14:55:25 -0700, Bob Fry <bobfry@mailinator.com>
wrote:
| Quote: | "p" == peter <nospam@nospam.com> writes:
p> Many fancy heat sinks hype "heat pipe" technology. But wouldn't
p> a solid block of copper conduct heat better than a few thin
p> copper pipes?
Not when the thin copper pipes contain a phase-change fluid...changing
phase from liquid to vapor absorbs a lot of heat energy.
|
Yep. They're *** great as long as they work. I recall some
early sony heatpipes had leaks or something. It was ugly 
One thing about a block of copper, it doesn't leak. |
|
| Back to top |
|
|
Fix your Windows Problems - FAST.
FREE Safe Scan Registry Check. Locate & Fix Errors in Minutes!
|
|
Guest
|
| Posted: Fri May 04, 2007 12:13 am Post subject: Re: solid heat sink? |
|
|
On Wed, 04 Apr 2007 00:27:08 -0400, Paul <nospam@needed.com> wrote:
| Quote: | peter wrote:
I want to buy a heat sink to replace the intel one that comes with c2d
E4300.
Many fancy heat sinks hype "heat pipe" technology. But wouldn't a solid
block of copper conduct heat better than a few thin copper pipes?
Imagine a copper block that is 1.2 inch square x 3 inches tall with fins
radiating out, and a fan on top. That is what I'm looking for. Perhaps the
copper block should tapper off like a pyramid so it is not too heavy.
Is there such product?
Actually that is not true. The heat pipe is actually better than a
solid piece of copper. One web page mentioned it is better by a factor
of 500 (but that number is going to vary with construction details of
the pipe).
If you want to go back to the previous generation of products, you can shop
for stuff like this:
http://www.techbits.ca/images/reviews/kip/slk900.gif
|
Actually I've got one of those in one box and a heatpipe AL heatsink
in the other. I've had the same cpu in both and temps were about
the same. The heatpipe was quieter though with a slower running
120mm fan to the noiser 92mm on the CU heatsink. Those copper
chunks were pretty good performers. Not as good as today's best
heatsink models but much better than the stock heatsinks. |
|
| Back to top |
|
|
Venom
Guest
|
| Posted: Fri May 04, 2007 10:22 am Post subject: Re: solid heat sink? |
|
|
<nobody@nospam.com> wrote in message
news:vksk331j4lk4m1ib0njabvhsktvcpjsn60@4ax.com...
| Quote: | On Tue, 24 Apr 2007 14:55:25 -0700, Bob Fry <bobfry@mailinator.com
wrote:
"p" == peter <nospam@nospam.com> writes:
p> Many fancy heat sinks hype "heat pipe" technology. But wouldn't
p> a solid block of copper conduct heat better than a few thin
p> copper pipes?
Not when the thin copper pipes contain a phase-change fluid...changing
phase from liquid to vapor absorbs a lot of heat energy.
Yep. They're *** great as long as they work. I recall some
early sony heatpipes had leaks or something. It was ugly
One thing about a block of copper, it doesn't leak.
Yair, it is just so heavy it rips the cpu and socket right out of the |
motherboard. |
|
| Back to top |
|
|
Bob Fry
Guest
|
| Posted: Fri May 04, 2007 3:26 pm Post subject: Re: solid heat sink? |
|
|
| Quote: | "nobod" == nobody <nobody@nospam.com> writes:
Not when the thin copper pipes contain a phase-change
fluid...changing phase from liquid to vapor absorbs a lot of
heat energy.
|
nobod> Yep. They're *** great as long as they work. I recall
nobod> some early sony heatpipes had leaks or something. It was
nobod> ugly One thing about a block of copper, it doesn't leak.
I recall some early horseless carriages wouldn't run well! One thing
about a nag, they don't get flat tires.
--
Go, and never darken my towels again.
Groucho Marx |
|
| Back to top |
|
|
Guest
|
| Posted: Sat May 05, 2007 9:10 am Post subject: Re: solid heat sink? |
|
|
On Fri, 04 May 2007 05:22:50 GMT, "Venom" <Venom@Mailhouse.com>
wrote:
| Quote: |
nobody@nospam.com> wrote in message
news:vksk331j4lk4m1ib0njabvhsktvcpjsn60@4ax.com...
On Tue, 24 Apr 2007 14:55:25 -0700, Bob Fry <bobfry@mailinator.com
wrote:
"p" == peter <nospam@nospam.com> writes:
p> Many fancy heat sinks hype "heat pipe" technology. But wouldn't
p> a solid block of copper conduct heat better than a few thin
p> copper pipes?
Not when the thin copper pipes contain a phase-change fluid...changing
phase from liquid to vapor absorbs a lot of heat energy.
Yep. They're *** great as long as they work. I recall some
early sony heatpipes had leaks or something. It was ugly
One thing about a block of copper, it doesn't leak.
Yair, it is just so heavy it rips the cpu and socket right out of the
motherboard.
|
They were heavy, but about the same as some of the gigantic heatpipe
stuff that's the rage today. |
|
| Back to top |
|
|
Guest
|
| Posted: Sat May 05, 2007 9:12 am Post subject: Re: solid heat sink? |
|
|
On Fri, 04 May 2007 07:14:27 -0700, Bob Fry <bobfry@mailinator.com>
wrote:
| Quote: | "nobod" == nobody <nobody@nospam.com> writes:
Not when the thin copper pipes contain a phase-change
fluid...changing phase from liquid to vapor absorbs a lot of
heat energy.
nobod> Yep. They're *** great as long as they work. I recall
nobod> some early sony heatpipes had leaks or something. It was
nobod> ugly One thing about a block of copper, it doesn't leak.
I recall some early horseless carriages wouldn't run well! One thing
about a nag, they don't get flat tires.
|
yeah, but i think someone did the math and we'd be up to our ears in
horseshit if we all used horses
Flat tires don't destroy your engine either. |
|
| Back to top |
|
|
Fix your Windows Problems - FAST.
FREE Safe Scan Registry Check. Locate & Fix Errors in Minutes!
|
|
|
|
| |
Watched Topics
Search
Register
Log in to check your private messages
Profile
Log in
Memberlist
Usergroups