Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: dougSF30 who wrote (133750)	9/24/2004 2:23:25 PM
From: pgerassi	Read Replies (1) \| Respond to of 275872

Dear Doug: Actually you are quite ignorant. The true heat sink is the air being blown past the fins (whether they are copper, aluminum or something else is immaterial). Still air is quite insulative. The fins of the HSF decrease the thermal resistance between the conduction of the HSF base and the air moved by the fan (in older times the movement due to convection and thermal radiation (likes black grainy surfaces) was enough to move heat into the air quick enough to stay cool). The problem with the heat spreader is that it is fixed and places a known resistance to the thermal flow. Another problem is by using thermal probes becomes more difficult. When there is a bare die, you could always place the probe at the edge of the die, assuming it was thin enough. With the IHS, that placement of the edge is far enough away from the die to give time delays and larger temperature differences with the true core temp. The best solution is the on die thermal diode which can be monitored with hardware by attaching to the pins on the bottom side of the socket, shutting off the system to calibrate it to a known thermal probe on the HSF (with no power to the die, the metal in the HSF has the same temperature as the die) and applying the known formula to get the accurate temperatures. But many of the 3rd party sites fail to do this, yet read the temperature to 0.1C resolution. Pity as they fail to know that the typical core die temp reading may be off by 5C or more to known standards. The best material to use in a HSF is a material that undergoes a phase transition at the desired temperature, has no electrical conductivity, high thermal capacity per unit volume, not corrosive to either the die or its surroundings and has high fluid flow rate across the bare die. Although a near second would be a small (not bigger than the die dimension cubed) copper thin finned heat spreader on the die and slower flow of the fluid above. How the heat is extracted from the resulting gas or fluid is irrelevant to the inlet fluid temperature. Of course if we ever get room temperature superconductors, they would be the best HSF material around. But that is currently only in the cards if we cool the chips to less than 120K or so (-153C). Pete

To: dougSF30 who wrote (133750)	9/24/2004 4:13:58 PM
From: Ali Chen	Read Replies (1) \| Respond to of 275872

Doug, earlier you made a bold statement, "the heat spreader doesn't really help the die temperature". And later you clarified your statement: "Actually, they don't. The real purpose is to protect the die". I merely pointed out that your first statement is not true or false, and the answer depends on the category of heatsinks employed. Your later "clarification" clearly clarifies that this principally-important distinction has escaped your attention. If some things has escaped your attention, it is ignorance in the subject, by definition. dictionary.reference.com I think the sub-definition #2 is directly applicable here. More, Pravin was talking about "a good heat spreader between die and heatsink", without any additional qualifier. If taken more generally, "a good heat spreader" may include industrially-available materials like diamond-metal composites, or "isoskin" type spreaders with even better conductance. In this case your first statement would be unconditionally false. I guess I can put more pgerassisms (like the above) on you if you are ready. I encourage you to make some effort to understand what I said about heat spreaders, and two different class of heatsinks. If you are still confused, then your "mere university degree" is clearly no match for my doctorate degree in the very same field, and 30+ years of research and industry experience, including the last decade in the U.S. semiconductor industry. Have a nice day, - Ali