You deleted the original post, so I can't see it. (Please don't do that). NEVER use epoxy on a CPU chip. AMD's and Intel's retail boxed CPU's come with those thermal pads. Although better than they used to be, they are still no substitute for GOOD TIM (thermal interface material). Don't use typical white silicone HS (heatsink) compound. You need to use something like AS3 or AS5 (AS = Arctic Silver), or Nanotherm Blue Ice, XTC, or PCM+. This is ESPECIALLY true for AMD CPU's since they run MUCH hotter than P4's. They also require a HSF unit (Heatsink & fan) quite a bit larger and more efficient that the retail unit. Why AMD insists on shipping retail chips with bad HSF units is beyond me. Their stock fan is only 18cfm and that would only work well for a HUGE HS. Every retail AMD chip I use, I replace the stock retail fan with a 60mm 40cfm 7200rpm YS-Tech unit, remove the thermal pad and use GOOD TIM, and it runs LOADS cooler.
Many times on AMD platform mobo's, in the BIOS is a feature called "CPU disconnect function". This can theoretically make a CPU run cooler by switching it "on and off" several times a second when it's not being used by the OS. A program called CPUIdle is similar. It used to be free, I don't know if it still is, but it works great and can reduce a CPU's temp by 20F. I've heard the BIOS version of this when enabled can cause some instability on some boards, so if you have it you can try it and see if it works. I found the original post in the archives and pasted it below. You don't mention any temperatures, what are they? That AMD CPU should have thermal spec of about 85C (185F) and you don't want to even get close to that. You need to post the temps from the BIOS, from the temp probe (if it's a separate one from the mobo's hardware monitor), and the temps from the hardware monitoring software. You also didn't mention the power supply. An AMD PC can spontaneously reboot for many reasons. They are more sensitive to things than a P4. They can reboot due to heat (not just CPU heat but "system" heat, PWM heat, etc.), voltages going out of spec, incorrect settings in the BIOS or in the hardware monitoring software, or a weak power supply. If you don't have a hardware monitoring utility installed, install one, then post ALL the temp and voltage readings, and also post your power supply and specs for it. Your temp and voltage specs could be fine and your problem may be with the threshold settings of your hardware monitoring software (if you have it installed and running). You also need to post which mobo this is and the BIOS brand. A high-low tone can mean different things on different boards. On AMD platforms, usually it's a heat alarm, or incorrect voltage alarm. There's also an outside chance the CPU could not be seated perfectly. AMD CPU's are also VERY VERY sensitive to how they are mounted. Even though you may think the HSF unit is seated OK, it may not be and that can cause some really oddball behavior. Since you disabled PC Heath in the BIOS and it still does it, it still could be temp's causing it but more likely voltage issues. You need to enable EVERYTHING that can be enabled in the PC Health area and watch that since that could tell you something. -Clint God Bless Clint Hamilton, Owner http://OrpheusComputing.com ----- Original Message ----- From: "Jason Arroyo" <[EMAIL PROTECTED]> Ahhh you know what? I bet that's the one thing I forgot to use when I built this thing last year (it';s been like this for awhile). OMG it makes perfect sense! I'll check it this weekend. Watch there be NO grease because I forgot or something equally dumb of me :P -J -----Original Message----- Keith Thompson Boy, that sounds like my first go round with my Athlon 1800+ chip. I had a custom machine made and first thing I had to do was add a couple more case fans. I still had heat problems and first they (the shop) told me that Athlons ALWAYS run hot. I still encountered occasional shutdowns and started monitoring the chip temp with a software program. When doing an extended search, I could watch the temp go up to a point where I would discontinue the search and let the temp come back down. It finally failed completely so taking it back to the shop (under warrenty) they dissasembled the chip from the heat sink and found that there was no thermal grease between the two. Your situation is sounding much like what I experienced. Even tho you have the extreme in coolers, etc, if there is not sufficent or no thermal grease or epoxy connection, the AMD chip will run with hot spots and cause the behavior that you are describing. Keith Thompson From: Jason Arroyo ----------------------------------------------------------------------------- --- I have a PC that seems to shut itself off. Gigabyte brand motherboard running an Athlon XP 2400+ processor. 512MB of memory (not generic, I don't remember what tho), sitting in a new Thermaltake Xaser III case. The case has 7 ball bearing fans, a temp sensor probe that sits under the CPU, and a temp readout on the front as well as alarm settings. The CPU fan I use is a CoolTek Aero V, which looks like a snailshell design (you car fanatics would see it and think "supercharger"), that evenly distributes air across the heatsink better than a helical fan can (helical fans leave a hotspot in the center of the fan, not extremely so but this is even flowing). The Aero V is knob adjustable. Between the Aero V and the seven ball bearing fans in the the Thermaltake Xaser III case, the case and CPU are COOL. Thermaltake makes excellent aircooling cases, it's where they get their name from. So why does my PC shut off? It can be sitting there, in a cold room or warm, and be fine for days. Or it can shut off at random times. Sometimes if I do a large file search (say images with thumbnails search) in XP (SP1), and scroll around, it locks up and sometimes shuts of. When it DOES shut off, the screen goes black, and a a two-note sound comes out of the PC speaker (sounds like a bios sound). A low note, then a higher one, then a low note, then it shuts off. For you music buffs, it's like a low C note, then an A, then it repeats (although generally it goes C, then A, then C, then shuts off before the next A). I had suspected that the bios had a shutoff setting for temp and that it was set too low from the factory, but all it had was warnings, and they were all disabled. In fact everything in the bios section called "PC Health" was disabled after I went through it. And yet this thing still shuts off! It can be sitting there, or I can be doing things. It seems tho that sitting there does it. Not always tho, some nights I wake up and the monitor is sleeping, I bump the mouse, and the thing wakes. Other times I wake up to find it completely off. Other times I see it do it and hear the bios beeps I described above. Any ideas? I've swapped out memory twice now, and that doesn't help, btw. -J ============= PCWorks Mailing List ================= Don't see your post? Check our posting guidelines & make sure you've followed proper posting procedures, http://pcworkers.com/rules.htm Contact list owner <[EMAIL PROTECTED]> Unsubscribing and other changes: http://pcworkers.com =====================================================
