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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Time Traveler who wrote (33818)	7/1/1998 5:48:00 PM
From: Maxwell	Read Replies (1) \| Respond to of 1574002

Time Traveller: <<Question: Is a K6 ice-cold while running?>> My K6-233 which I overclock to 250MHz is hot as hell. It is like a radiator. My K6-300 (built on 0.25um) is very cool. When I touch my hand to the heat sink I can't even tell if it is warm or room temperature. Maxwell

To: Time Traveler who wrote (33818)	7/1/1998 6:56:00 PM
From: Ali Chen	Read Replies (1) \| Respond to of 1574002

John, do you think the more you write, the more you are right? <With that IR gun, you can measure the temperature of the case. By measuring the supply voltages and currents into the CPU, you can calculate how much power is dissipated. Multiplying this wattage by the thermal coefficient of junction to case, you can figure out how much higher the temperature of the die is than the case. There you have it --- the junction temperature. Time Traveler hopes this is all common sense to you, Ali.> I am sorry to tell you but you are ridiculously ignorant in any field you are touching. For your education: your proposed method is highly impractical for several severe reasons. Try to follow me. The junction temperature is the quantity one would like to measure most accurately to avoid failures. Therefore the more direct the method is, the better results are. Using your approach I can calculate the temperature without ANY measurements just assuming room temperature of 23C as a standard room temperature in America. But can I trust the result? No. 1. In your approach the IR gun will measure the temperature of walls or components in the enclosure. To use the thermal coefficient of the heat sink you must know the inlet air temperature which does not practically emit radiation and will not show up in your IR detector. So, one assumption needs to be made about the air temperature. 2. Even more, all heat sinks with attached fans are subject to so-called "bypass recirculation" around the fan - it sucks the hot air coming out of the sink back, so the actual air is hotter. Another assumption. 3. The heat sink thermal coefficient can be specified only at certain air flow rate. You do not know this rate. Do you? Assumption? 4. There must be a layer of heat compound between the sink and CPU surface. You do not know it's thermal resistance for sure. 5. There are secondary ways to conduct heat away from the chip - via pins to the socket and motherboard. You do not know this resistance either. 6. PC boards usually have no provisions to measure current consumed by the CPU. To measure it it would require a special adapter that will skew up results further. 7. Oops! Only now I see you are talking about CPU case temperature! With IR gun! Sorry, this means that all my lecture is way above your head. You just missed one thing - the CPU case is covered by heat a sink and a fan. Have you ever looked inside a computer, John? Now get lost, home-grown PC-theorist.