Does this copper shim really help dissipate the heat well? Also, is there no worry of the shim damaging the L1-L7 bridges?
SocketA Copper Shim
- Thread starter gimps
- Start date
haha ok nevermind. I never actually looked at the pic too well. I see that they cut out a spot for the bridges. Stupid me lol.
But do they really help in the heat dissipation?
But do they really help in the heat dissipation?
the potentially help heat dissappation by making sure the heatsink sits flat on the core, but other than that, no.
the shim rests on idle substrate, and that has very little if any heat.
Mike
the shim rests on idle substrate, and that has very little if any heat.
Mike
Actually it's not worth the $12, you dont really need it and it doesn't do too much of a good job anyways
Peace,
Nore
Peace,
Nore
Here's another question about the shim. What if, after you put the shim on, and fill in the small holes around the CPU die (where the cutout is), with some thermal paste. Would that help spread the heat onto the shim? Or does the paste not work alone, and need a surface to put it on?
I hope that made sense lol.
Thanks.
I hope that made sense lol.
Thanks.
NO NO , don't fill any holes with thermal paste. Some of the stuff is conductive and will give you fits.
AMD strongly recommends against the shims. They can actually reflect heat off the hs. The ceramic substrate is designed to obsorb heat. I know this mat appear wierd but AMD told all hsf manufacturers to stop using anodized material because of reflection. Another concern is that as everything heats up the copper may expand in such a way as to take pressure off the die/hsf connection.
AMD strongly recommends against the shims. They can actually reflect heat off the hs. The ceramic substrate is designed to obsorb heat. I know this mat appear wierd but AMD told all hsf manufacturers to stop using anodized material because of reflection. Another concern is that as everything heats up the copper may expand in such a way as to take pressure off the die/hsf connection.
reflection?
give it a rest dude
you don't know what you are talking about
Heat is energy, there is NO way possible that it will "reflect" back into its source. Energy travels in the path with the least resistance... what has less resistance, the heatsink, or "reflecting" back into the cpu??
Anyone with any physics or electricity background will tell you it will go out into the heatsink.
I make the ones at 3DFXcool. They will not assist in cooling, I did the testing, there is no temp gain by using one. However, unless you are TRYING to crush your core with one on, you won't. So, is $12 bucks worth that peice of mind? That is a question you must answer yourself.
If you have any questions for me... email me at cuspacer@swbell.net Thanks.
give it a rest dude
you don't know what you are talking about
Heat is energy, there is NO way possible that it will "reflect" back into its source. Energy travels in the path with the least resistance... what has less resistance, the heatsink, or "reflecting" back into the cpu??
Anyone with any physics or electricity background will tell you it will go out into the heatsink.
I make the ones at 3DFXcool. They will not assist in cooling, I did the testing, there is no temp gain by using one. However, unless you are TRYING to crush your core with one on, you won't. So, is $12 bucks worth that peice of mind? That is a question you must answer yourself.
If you have any questions for me... email me at cuspacer@swbell.net Thanks.
So, you've never heard of a heat reflector ??. It doesn't reflect to the source, the core is the source. Copper heat reflectors have been used for years. Before you open your mouth, try E-Mailing AMD, they will give you the straight scoop. If you had taken the time to read my post it says the heat is reflected OFF the heatsink NOT the core.
I second DaddyG.
tgillitzr,
You make the shims. IMO, that makes you almost totally biased when responding to DaddyG. Are you worried that this will hurt your sales or something? Because I doubt it will, because a lot of useful information on this bbs goes vastly unnoticed, even when the useful information is repeated.
Mike
tgillitzr,
You make the shims. IMO, that makes you almost totally biased when responding to DaddyG. Are you worried that this will hurt your sales or something? Because I doubt it will, because a lot of useful information on this bbs goes vastly unnoticed, even when the useful information is repeated.
Mike
So what you mean by reflect off the hs, is that the copper may absorb the heat, after the hs, and since the shim has no cooling attached to it, that the heat will not get transferred off the copper shim?
Here's the issue that I have with the shims. Heat from the core does not act like a heat seeking missile and head for the path of least thermal resistance. As pointed out in other threads, quite well by John Car, MikeW and others, multiple thermal pathways are present.
The primary pathway is between the cpu die and the heatsink, via the thermal copound. A secondary pathway, which actually passes a substantial amount of heat, exists between the cpu die and the ceramic substrate. AMD didn't pick ceramic just for fun. The high heat output of TBIRDS pretty much demanded it. Now we come and cover the ceramic with a copper shim, which touches or comes close to touching the hsf, the primary thermal path. The air gap between the hsf and the ceramic substrate is gone, and new thermal paths are created. Its entirely probable that heat from the hsf heats up the copper shim, instead of the heat in the hsf being dissipated to the air.
As to the idea thats its cheap insurance, considering that you have to remove the foam pads to install the shim, this cheap insurance voids your warranty, both Retail and OEM.
I am not trying to stop anybody making a buck or two, but there is more to the issue than simply punching out copper shims.
The primary pathway is between the cpu die and the heatsink, via the thermal copound. A secondary pathway, which actually passes a substantial amount of heat, exists between the cpu die and the ceramic substrate. AMD didn't pick ceramic just for fun. The high heat output of TBIRDS pretty much demanded it. Now we come and cover the ceramic with a copper shim, which touches or comes close to touching the hsf, the primary thermal path. The air gap between the hsf and the ceramic substrate is gone, and new thermal paths are created. Its entirely probable that heat from the hsf heats up the copper shim, instead of the heat in the hsf being dissipated to the air.
As to the idea thats its cheap insurance, considering that you have to remove the foam pads to install the shim, this cheap insurance voids your warranty, both Retail and OEM.
I am not trying to stop anybody making a buck or two, but there is more to the issue than simply punching out copper shims.
Actually, I've seen copper shims that do not require you to take the rubber pads on the CPU. They are cut out so that the pads also have a hole in the shim. Therefore, those would not coid a warranty.
IN that case, why bother with a shim at all? As long as you're careful in mounting your heatsink, with a proper socket A approved clip, the chances of chipping/cracking a duron/t-bird are significantly reduced.
Mike
Mike
All I was wondering is if they help keep the CPU cool! JEESH! and look what you guys did to this thread! 
bias has nothing to do with it
its basic physics
heat IS NOT going to travel back to its source when it can take an EASIER path out into the air
That is my opinion. AMD says what they say because most of the spacers require you to remove the pads, and they bring another chance for a screwup.
If you want to prove me wrong, do it. I've tested and tested, there is no gain or loss in heat transfer, thats not what I designed them for
Actually, its kinda funny. There is NO reason whatsoever to use Copper, NONE. I only used copper because thats what I started with on the Slot A Athlon/P3 spacers(it helped heat transfer a Slot A Athlon) and continued to use it on these. Why would the other places that borrowed my idea use copper as well? Strange.
its basic physics
heat IS NOT going to travel back to its source when it can take an EASIER path out into the air
That is my opinion. AMD says what they say because most of the spacers require you to remove the pads, and they bring another chance for a screwup.
If you want to prove me wrong, do it. I've tested and tested, there is no gain or loss in heat transfer, thats not what I designed them for
Actually, its kinda funny. There is NO reason whatsoever to use Copper, NONE. I only used copper because thats what I started with on the Slot A Athlon/P3 spacers(it helped heat transfer a Slot A Athlon) and continued to use it on these. Why would the other places that borrowed my idea use copper as well? Strange.
Quote tgillitzr :
<< Heat IS NOT going to travel back to its source when it can take an EASIER path out into the air >>
Try reading DaddyG's post again, what he's trying to say in easier terms is that the COPPERSHIM would actually heat up because its flush with the HEATSINK. The non-core surface of the CPU gives off virtually no heat, the COPPERSHIM would actually be warmer than that surface; therefore reflecting heat off the heatsink. In other words, its actually heating up the area around the core of the CPU, which AMD completely dissapproves.
<< Heat IS NOT going to travel back to its source when it can take an EASIER path out into the air >>
Try reading DaddyG's post again, what he's trying to say in easier terms is that the COPPERSHIM would actually heat up because its flush with the HEATSINK. The non-core surface of the CPU gives off virtually no heat, the COPPERSHIM would actually be warmer than that surface; therefore reflecting heat off the heatsink. In other words, its actually heating up the area around the core of the CPU, which AMD completely dissapproves.
thats not exactly what I would reflecting heat back to the core...
however, it doesn't seem to make the cpu run any hotter. I say seem because there is no good way to measure cpu temp on the socket a systems.
sorry for the flames guys, just kinda get aggitated sometimes
however, it doesn't seem to make the cpu run any hotter. I say seem because there is no good way to measure cpu temp on the socket a systems.
sorry for the flames guys, just kinda get aggitated sometimes
Dexion,
Thanks for the help. I NEVER said that heat could travel back to its source. If Tgillitzer's theory were correct, the under die thermistor would read ambient temperature because all the heat would travel to the pathway with the lowest thermal resistance. Obviously this is not the case. The issue is how the heat actually dissipates. The primary pathway is the heatsink, but the ceramic substrate is a considerable secondary path. The air gap between the actual core and the heatsink is also a path. The addition of the copper shim effectively connects the primary and the secondary paths and eliminates the air gap.
The effects of a copper shim on P3's is quite different. First off, they produce significantly less heat than TBIRDS. Secondly, the substrate material is the green fiber/epoxy board stuff (not sure of the precise material). What I am sure of is that it does not offer the same type of pathway as the ceramic.
Thanks for the help. I NEVER said that heat could travel back to its source. If Tgillitzer's theory were correct, the under die thermistor would read ambient temperature because all the heat would travel to the pathway with the lowest thermal resistance. Obviously this is not the case. The issue is how the heat actually dissipates. The primary pathway is the heatsink, but the ceramic substrate is a considerable secondary path. The air gap between the actual core and the heatsink is also a path. The addition of the copper shim effectively connects the primary and the secondary paths and eliminates the air gap.
The effects of a copper shim on P3's is quite different. First off, they produce significantly less heat than TBIRDS. Secondly, the substrate material is the green fiber/epoxy board stuff (not sure of the precise material). What I am sure of is that it does not offer the same type of pathway as the ceramic.
hehe
we can argue this into the ground
In theory, all the heat will go through the heatsink... however that is not realistic.
In reality, there will be some "leakage" where heat will go out through other paths.
As far as I can tell, it does not raise or lower the cpu temp.
we can argue this into the ground
In theory, all the heat will go through the heatsink... however that is not realistic.
In reality, there will be some "leakage" where heat will go out through other paths.
As far as I can tell, it does not raise or lower the cpu temp.
well, based on how socket A mb's read cpu temp, the difference could be up to 1C and not register with the thermsitor.
And for hte p3 shims, it may not produce enough heat to effect the temp of the shim.
Mike
And for hte p3 shims, it may not produce enough heat to effect the temp of the shim.
Mike
TG,
One thing I do agree on, you made a great effort to help correct an AMD screw-up. AMD should NEVER have released a cpu, that puts out so much heat, with such a fragile core. Gotta give Intel the 'nod' here for protecting the P4.
One thing I do agree on, you made a great effort to help correct an AMD screw-up. AMD should NEVER have released a cpu, that puts out so much heat, with such a fragile core. Gotta give Intel the 'nod' here for protecting the P4.
well, off the topic of SocketA Copper Shim, I got my heat probs figured out. The TaiSol that came with my CPU is MUCH better, without the thermal pad and AS Thermal Compound instead, than it was with the Vantec SocketA Cooler.
Took off the Vantec, put the TaiSol back on, put a bit more AS on the die, and now, I know it's inaccurate, but the temp reading at idle for about 30 minutes, is 28 Celsius. Yeah I know it's incorrect, but still a big drop from roughly 40-45 C with the Vantec using AS.
Took off the Vantec, put the TaiSol back on, put a bit more AS on the die, and now, I know it's inaccurate, but the temp reading at idle for about 30 minutes, is 28 Celsius. Yeah I know it's incorrect, but still a big drop from roughly 40-45 C with the Vantec using AS.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 23K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter