Computing.Net > Forums > CPUs/Overclocking > New P4 too hot

Computer Problems? Computing.Net has over 1,000,000 posts about all things technology related! Over 90% answered within 24 hours! Click here to start participating now! Also, be sure to check out the New User Guide.

New P4 too hot

Reply to Message Icon

Name: 808p4
Date: April 5, 2003 at 20:13:46 Pacific
OS: Win2k SP3
CPU/Ram: P4 2.53/512 DDR PC2700
Comment:

Hi All,

I built a new P4 2.53 system with a P4PE Black Pearl MB about 5 days ago. I'm running Win2k SP3, 512 PC2700, Radeon 9000 PRO, 350W Antec Dual Fan PS.

The system ran for the first 3 days with a GlacialTech 4100 HSF, but the CPU temps would idle at 46-50C. After that I changed to the stock HSF, but the temps are still the same. My system temps, however, are at 36-38C. My old Athlon system in the same case would only have about a 6C difference and run much cooler.

Although the system is stable, I am worried about the high temps. Is there anything I can do? Will it decrease over time?

Any help is much appreciated. Thank you.



Sponsored Link
Ads by Google

Response Number 1
Name: Amourek
Date: April 5, 2003 at 20:21:06 Pacific
Reply:

Hmmm, are you using the thermal pad it came with? You could try some thermal compound or stick an exhaust fan in the back if you case allows for it.


0

Response Number 2
Name: n3985
Date: April 5, 2003 at 20:22:20 Pacific
Reply:

if you're using thermal compound, it will drop 2-4C when your system has been on for 72 hours total, also, if your temps stay that that range when under full load, it's fine. I have an Athlon XP 1900+ that idles at 47C and max at 51C. On the other hand, if you're willig to spend some money, I have a P4 3.06GHz with HT and a Thermalright SLK-900 and a Thermaltake Smart Fan II, which idles at 36C and max at 62C.


0

Response Number 3
Name: 808p4
Date: April 5, 2003 at 22:49:29 Pacific
Reply:

Thank you for the replies! I am using the stock thermal pad. Should I use the paste instead? I guess I'm uneasy about taking the processor out again and cleaning it again. Is it worth the risk, if any?

Thanks


0

Response Number 4
Name: gaz
Date: April 6, 2003 at 08:34:46 Pacific
Reply:

yes, the temps are a bit high, but unless you start to experience instability/decreased performance, i would leave the system as is. do you have case fans set up?


0

Response Number 5
Name: johnoh
Date: April 6, 2003 at 14:06:23 Pacific
Reply:

That's one cool board. What's with the rotating heatsink - how does that work?

I'd say your temps are a just a shade high, but certainly nothing to worry about. I doubt new thermal compound would make a difference. I think that 4100 hsf is temp controlled and may not be running at top speed if so. No idea if there's a way to turn off the temp controls and just run it at top speed.


0

Related Posts

See More



Response Number 6
Name: 808p4
Date: April 6, 2003 at 14:19:26 Pacific
Reply:

Thanks again! I was looking for a way to max out the fan speed, but I haven't found a program to do that yet. Does anyone know?

The board is pretty cool. The rotating heatsink isn't much...it's basicall a cylinder that you can spin around manually(for lack of a better analogy, it's like a toilet paper roll spinning around). The best bonus for me was the 32MB flash disk that came with it.

Thanks again everyone.


0

Response Number 7
Name: johnoh
Date: April 6, 2003 at 18:10:02 Pacific
Reply:

The boys at GlacialTech thought they would sell more 4100s if they kept wires out of the jpegs so as a result I cannot find a picture of a 4100 with its wires. Just a beautiful gold heatsink with no wires. So I have no idea how the temperature monitoring actually happens. But....

Usually a thermistor is how you temperature control a fan's speed. Here's how that works.

All cpu fans take a 12V voltage input and then produce rpms. All fans can be reduced in rpms by giving them less than 12V. All fans, given a 6V supply of voltage, will run at about half their normal rpms.

The 4100 has a temperature regulated fan which means that at max temp it must run at 12V and at lower temps it must run at a lower voltages. This is done by a thermistor, probably. No idea since they publish no doc or pics.

If you stick a resistor onto a power wire for a fan some of the dc voltage supplied to that wire gets used up by the resistor which leaves less for the fan and that makes the fan run at a lower speed.

A thermistor is a resistor which has its resistance varied by temperature. At max temp a thermistor has no resistance which means it acts like just a wire so the fan can run at max speed.

So if the 4100 has a thermistor you just cut the wires off on either side of it and then re-attach them. This takes away any temperature-created resistance and causes the wire to have no resistance and the fan will then run at max speed. So if you can find the thermistor, cut it off and re-attach the remaining wires and then your cpu fan will run at max speed. Pretty simple really.

This would be easier to explain if there were some pics online.


0

Response Number 8
Name: 808P4
Date: April 8, 2003 at 13:17:01 Pacific
Reply:

Thank you for everyone's help. I think I may have discovered the problem...

When I put on my first heatsink (the GlacialTech), it already had the white, zinc based thermal paste applied. I heard that this stuff tends to ooze when it gets hot, and I'm thinking that it may have seeped through the hole on the top of the P4 heat spreader. I read that you are not supposed to put thermal paste over this hole. As a result, no re-seating or cleaning I did after that first heatsink would matter since the stuff had already gone under the heat spreader.

Does this make sense or is it a possibility?

Thanks again...


0

Response Number 9
Name: johnoh
Date: April 8, 2003 at 16:16:00 Pacific
Reply:

The P4 hole has a created a lot of stories on the web. Its there for pressure relief during manufacturing. When the heat spreader is epoxied to the organic land grid array it releases gas as it cures. If there were no hole the pressure would build up and cause wet epoxy to blow out the sides resulting in an uneven bond between the ihs and the olga. Intel does not publish anything to indicate that paste should not be put over (and therefore into) this hole, and certainly if they did not want us to put paste over the entire ihs they would have said something to this effect. So I do not think that is your problem.

I put a 2.4g p4 into case that had had a 1.4g t-bird in it and the temps went up a little for me as well. Never did figure it out, but fan speed is the main thing that cools your hsf and that p4 fan is pretty quiet.

Did you try any wire modifications to make the 4100 run at max speed? I know of no way to make it run at max speed other than eliminating the thermistor. If you try it and find that its too loud, you can always splice it back in.



0

Response Number 10
Name: 808p4
Date: April 8, 2003 at 19:01:48 Pacific
Reply:

I found some pics at NewEgg that sort of show the wires on the GlacialTech. I don't think there is a thermistor on it, but I'll check when I get back home. You're right about the fan speed, since it was really only running at about 2600rpms max at any time.

Thanks again...this is my first post here and I appreciate the response.


0

Response Number 11
Name: johnoh
Date: April 9, 2003 at 07:18:19 Pacific
Reply:

I just looked at that pic too. The front wires have the 3-pin mobopower plug, the back wires (hidden in the pic) should have a thermistor on one of them.


0

Sponsored Link
Ads by Google
Reply to Message Icon






Post Locked

This post is quite old and has been locked from receiving new replies. Please create a new posting instead.


Go to CPUs/Overclocking Forum Home


Sponsored links

Ads by Google


Results for: New P4 too hot

P4 640 too hot and too slow www.computing.net/answers/cpus/p4-640-too-hot-and-too-slow/11451.html

Hot potato too hot? www.computing.net/answers/cpus/hot-potato-too-hot/13928.html

65C too hot for 3.06GHz P4 with HT? www.computing.net/answers/cpus/65c-too-hot-for-306ghz-p4-with-ht/5168.html