View Full Version : For Geniuses ONLY
To those who know...
This is my first build and I've seemed to take all the necessary precautionary measures in building my system.
The board is insulated from the case's mounting screws by rubber washers, all hardware was mounted with power disconnected, heatsink was mounted per instructions found at arcticsilver.com, and the appropriate amount of blood, sweat and tears included.
Now the carnage: 2 motherboard and 2 processors.
First mb and processor combo would always stick at various points during the posting process. Never completed posting.
Replaced mb and got nothing at all. I did hear some clicking followed by that peculiar electrical odor of something possibly gone wrong....disconnect.
Replaced processor and motherboard and finally success! But only for a moment. Installed Win2k fine but noticed 80C readings in BIOS. Unsure of accuracy so downloaded MBM and verified with same readings. Ran for a total of two days doing minor installs and actually played Quake awhile but DIDN't notice any rise in temps (?). Only had one spontaneous reboot. Shutdown...
Next day I removed and cleaned heatsink and reapplied thermal grease then reinstalled and tried to boot and NOTHING. Only splash screen and freeze on initial BIOS post screen. Judging by my mb diagnosis lights it seems to stop every time at assigning IRQ to PCI devices.
My setup is as follows;
MSI turbo limited edition
Athlon 1.2 ghz AXIA
Antec 400W PSU
Fong Kai fk-603 case
Thermalright SK-6 w/ sunon 5300 rpm fan
2 WD600BB drives
Pioneer slot load DVD + Plex 16x
3COM 905C net card + Firewire PCI card
Midiman delta66
SBLive 5.1 sound card
Possible guesses at what's wrong: PSU, another bad processor (unlikely), faulty temp. readings, or software prob.
Any educated guesses out there?
Thanks for any help at all....:D
Heydre
well... i'd strip out every piece of hardware.
remove mobo battery for 3mins.
now you have,
cpu w/Hsk/Fan
RAM
mobo
vid card
and default bios settings.
if you boot that alone... it should say O/S not found.
everything seem ok?
ok, try now with HDD... change default bios to "boot from C"
that being your only change from default, k
should boot into windoze...
does it? let us know.
r0ckstarbob
09-07-01, 05:59 PM
hard to say from here but it sounds like a bad powersupply.
you've already eliminated the possibility of the mobo and the cpu by replacing them both already and frying them both times... this covers ALOT of ground in the troubleshooting field of things. i'm assuming you know what a metal on metal connection looks like so am assuming your case isn't the problem and the Fong Kai is a good case generally speaking. (just to make sure, i'd go get some nylon screws and spacers just to utterly remove that possiblity.) the same could be said for the heat sink and your ability to recognize a metal on metal connection. your hard drives/disk drives aren't going to fry your system almost no-matter-what and seeing as you've already replaced the mobo the only thing there we'd have to worry about is the units themselves. i've never heard of a corrupted hard drive eating a processor and a motherboard, have you?
so what problems have we eliminated already?
Case
Heatsink
Drives
**Processor <-------- these two are the
**Motherboard <------- big ones to worry about
whats left?
Powersupply
Vid Card
RAM
i've never heard of a Vid Card (or any other card for that matter) eating an entire system without there being some OBVIOUS damage to the card... (and i mean you'd have to be Rainman to miss the kind of damage that would be required to frag both your mobo and your cpu). i'm assuming you've inspected your cards multiple times and would have seen any kind of damage to them.
if your RAM is corrupt, i wouldn't know how to recognize it to be honest or know if it's even recognizable. i don't know if it's even possible for your RAM to eat your system or not to be honest.
mr man is correct. strip your system down to bare minimums and try to get it to boot that way, including using only one stick of RAM. slowly add cards/components/perhipials one by one until the system stalls out.
if you continue to experience problems with no improvements at all, go swap your Powersupply if you can. i bet it'll make a difference and solve your problem. if you continue to experience problems, let us know and we can go from there.
RSB
PS... there is one last thing, as unlikely as it may sound... do you have any super powerful fans plugged into your mobo headers? i just read something about seriously not being able to plug an 80mm Delta screamer into your mobo headers because the draw is just WAY too strong and it will DEFINATELY eat your entire system.
so i lied, theres two real things to consider. Powersupply and Fans. Swap your Delta (and make sure it's able to run off your fan headers before you do it, if you have an 80mm delta, DO NOT PLUG IT INTO YOUR MOBO HEADERS) and swap your PS making sure it's compatable too and you should be good to go. Good luck.
80oC!!! woh, I think this may be your problem. Are you using a temp probe touching the core to get the temp?
Something is definately wrong if you are getting that high a temp.
Does the sk6 have a protective cover on the base and if so did you take it off?
It definately sounds like a heat issue to me. Do you have any friends with experience in this area that could come have a look over the system for you as it's hard to diagnose accurately without actually having contact with the system
TRANCER24
09-07-01, 06:39 PM
i would try putting your finger on the heatsink itself if it is too hot to touch or is real warm where you have to remove your fingers after awhile it is too hot thats for sure. On my heatsink i could leave my fingers on it without it bothering me it is not even that hot at all that is how yours should be ..
what case cooling do you have?
I like how you bump your thread, but dont answer any of the questions people put. I'll not bother I guess :D
killem1x1
09-10-01, 01:51 PM
The only two things I could recommend would be:
1. Do everything that has been asked previously in this post and answer with the results.
2. I would do these things "out of the case" Hook up my PSU, mobo with just my chip, and nothing else, then proceed to add RAM, etc...
strip her down to the bare bones. Use the smallest amount of RAM possible, underclock it and see what she does.
Bare Bones:
CPU, mobo, RAM, FDD,graphics(use a pants 4Meg one or something).
{X-H}KnownKiller
09-10-01, 02:22 PM
Bad power supply is my guess...
I have lost 2 mother boards and a cpu once 2 a bad power supply...
Buy a new 1 and test it... If its all good then call up Antec and scream to them about it and see what they will do..
I've got a Enhance 350 WATT thats on AMDs recomended list. Works for all chips.. and is stable at 1650 at 1.95 volts..
Not saying that Antex is bad but that p/s is prob bad..
AZZKICKER
09-10-01, 10:26 PM
"The board is insulated from the case's mounting screws by rubber washers,"
WHY???????????????????????????????????????????????
ummm.....the board needs a earth ground ( as in the CASE!!!!!!!!!!!!!!!!!)
till you do something differnt.....................keep on FRYING
Originally posted by AZZKICKER
"The board is insulated from the case's mounting screws by rubber washers,"
WHY???????????????????????????????????????????????
ummm.....the board needs a earth ground ( as in the CASE!!!!!!!!!!!!!!!!!)
till you do something differnt.....................keep on FRYING
i do believe there is a ground wire on the power supply plug to mobo...i got all the rubber washers etc too...as far as your prob...im not computer engineer just electronic engineer and i would guess it is power supply...
r0ckstarbob
09-11-01, 02:17 PM
Originally posted by AZZKICKER
"The board is insulated from the case's mounting screws by rubber washers,"
WHY???????????????????????????????????????????????
ummm.....the board needs a earth ground ( as in the CASE!!!!!!!!!!!!!!!!!)
till you do something differnt.....................keep on FRYING
don't be a jerk. do you ground your speakers to the frame of your car? don't ground your board to your case. thats just dumb. that powersupply has more then adequate ground. the electrical circuit is already being created with the powersupply when you plug it in. when you plug the powersupply into your board you've now extended that circuit. you could put your computer in a totally plastic/rubber housing and it would be fine. if you're counting on those little nubbins on the back of your board to make "ground" the only thing you're going to to is create a short in the system.
go slap your mobo against a good piece of metal and let us know how well your board does. even if you have to do it from a public terminal afterwards.
Originally posted by r0ckstarbob
don't be a jerk. do you ground your speakers to the frame of your car? don't ground your board to your case. thats just dumb. that powersupply has more then adequate ground. the electrical circuit is already being created with the powersupply when you plug it in. when you plug the powersupply into your board you've now extended that circuit. you could put your computer in a totally plastic/rubber housing and it would be fine. if you're counting on those little nubbins on the back of your board to make "ground" the only thing you're going to to is create a short in the system.
go slap your mobo against a good piece of metal and let us know how well your board does. even if you have to do it from a public terminal afterwards.
ya what he said..with out the offences of course
r0ckstarbob
09-11-01, 02:37 PM
heydre
do you have ANYTHING TO SAY IN RESPOSE TO ALL THE FEEDBACK YOU"VE GOTTEN ???
you gotta keep talkin to us brotha if you want us to help you.
Thanks to all who have responded helpfully to my unusual situation as you all know this is getting to be a very expensive problem.
I am sorry I haven't posted sooner to your responses but I wanted to get a broad array of suggestions before I undertook any actions to fix it.
To those who have responded with ridicule or snide remarks your inability to be of help has been noted.
I will begin troubleshooting today...
In response to specific posts
Warlok :Thanks for pointing out someone's lethal combo of arrogance and ignorance.
RSB : Ditto above and no I am only using a Sunon 5300rpm and a YSTech 80mm, all with quite reasonable current draw. Thanks for the anaylysis.
KnownKiller :I am interested to know the details of what happened with your PSU probs.
Phil :You read my mind...I've been building a temp sensor mod from a RadioShack indoor/outdoor therm. for that very purpose. As for heatsink pad...NO WAY...chucked it.
Trancer :The heatsink is just warm. I can leave my fingers on there all day long.
N2 :Right on the money. Thanks man.
Stay tuned...I'll keep you posted..
Thanks all. :D:D:D:D:D
Doesn't look good...with a little free time I've been able to try a few things and none of the results are encouraging :confused:
First: I stripped system down to RAM, MOBO, PSU and VGA and pulled MOBO battery to reset. Still freezed during POST process. Usually during PCI bus init or IDE controller init. Therefore couldn't access BIOS to check temps. But I noticed the processor was building up a lot of heat around it but the heatsink was still not too hot to touch.
Second: I pulled the PSU and replaced with a brand new Enermax 350w. Booted and same problem. Freeze during POST process.
Third: I replaced processor with another AXIA 1.2 ghz 266, prepped/cleaned heatsink and applied new coat of ArcticSilver to CPU and prayed. No problem booted right into BIOS. Immediately went to check temps. Watched it rise from 60c to 68c in a minute....Shut down.
New Theory: I am now thinking that the highly praised Thermalright SK6 heatsink is the problem. Since it has been very hard to eliminate the mobo or the processor as the problem I believe what has been happening is that the processor has been burning up due to faulty cooling.
Tomorrow I am going to try 2 other heatsinks I've got (Globalwin WBK38 and FOP32) to get a comparative view of the performance of my HSF. Then hopefully I can put this nightmare to rest....ZZZZZZZZZZZ.
We'll see....:rolleyes:
Heydre
Originally posted by heydre
Doesn't look good...with a little free time I've been able to try a few things and none of the results are encouraging :confused:
First: I stripped system down to RAM, MOBO, PSU and VGA and pulled MOBO battery to reset. Still freezed during POST process. Usually during PCI bus init or IDE controller init. Therefore couldn't access BIOS to check temps. But I noticed the processor was building up a lot of heat around it but the heatsink was still not too hot to touch.
Second: I pulled the PSU and replaced with a brand new Enermax 350w. Booted and same problem. Freeze during POST process.
Third: I replaced processor with another AXIA 1.2 ghz 266, prepped/cleaned heatsink and applied new coat of ArcticSilver to CPU and prayed. No problem booted right into BIOS. Immediately went to check temps. Watched it rise from 60c to 68c in a minute....Shut down.
New Theory: I am now thinking that the highly praised Thermalright SK6 heatsink is the problem. Since it has been very hard to eliminate the mobo or the processor as the problem I believe what has been happening is that the processor has been burning up due to faulty cooling.
Tomorrow I am going to try 2 other heatsinks I've got (Globalwin WBK38 and FOP32) to get a comparative view of the performance of my HSF. Then hopefully I can put this nightmare to rest....ZZZZZZZZZZZ.
We'll see....:rolleyes:
Heydre
well good luck i use a volcano 6 on 1.4... i clock 135/11 temp = 34 *c/30*c system
!-=sky=-!
09-12-01, 04:14 AM
maybe it's the motherboard crapping out?
rma it if possbile
r_stripe
09-12-01, 02:55 PM
just dropping a line,
a little tension over here about the use of washers on the mobo..... well everyone is right. the ground wire from the power supply "IS A GROUND" but you will not fry your mother board if you didnt happen to use rubber, cork, plastic, gum, or socks as an insulator. For one, the mounting holes in the motherboard are issolated from any connection on the board, thats why you can use a screw directly to any motherboard....take a look, that little silver ring around each hole isn't connected to anything, and if they are, its "100% GROUND" beef, Theres no real advantage or disadvantage to using insulated washers. you could make a wood box and the mobo would work fine off the PSU ground. The only time a short can occur is if there is a stand off under the mobo that does not align with the holes in the mobo.....this could cause the board to shortout pins on the back side. If your changing mobo's be sure that you only have standoffs that align with your mounting holes, other wise, *sniff* *sniff* a nice electrical smell magically fills the air, or in a not so serious case, random errors, or no operation at all.........just a word or two.
not trying to drag this out anymore than necessary......:)
Originally posted by r_stripe
just dropping a line,
a little tension over here about the use of washers on the mobo..... well everyone is right. the ground wire from the power supply "IS A GROUND" but you will not fry your mother board if you didnt happen to use rubber, cork, plastic, gum, or socks as an insulator. For one, the mounting holes in the motherboard are issolated from any connection on the board, thats why you can use a screw directly to any motherboard....take a look, that little silver ring around each hole isn't connected to anything, and if they are, its "100% GROUND" beef, Theres no real advantage or disadvantage to using insulated washers. you could make a wood box and the mobo would work fine off the PSU ground. The only time a short can occur is if there is a stand off under the mobo that does not align with the holes in the mobo.....this could cause the board to shortout pins on the back side. If your changing mobo's be sure that you only have standoffs that align with your mounting holes, other wise, *sniff* *sniff* a nice electrical smell magically fills the air, or in a not so serious case, random errors, or no operation at all.........just a word or two.
not trying to drag this out anymore than necessary......:)
r_stripe is right about the shorts etc on the mobo
thanks for the clearing the air Stripe ;)
Update just finished my mod of a RadioShack thermometer to get an official reading of CPU back temps. Mod article found at:
http://www.benchtest.com/thermometer-2.html
Will let all know tonight as I try different heatsink combos.
Heydre
This is unbelievable...:mad:
After finishing my mod on a RadioShack thermometer I placed the sensor next to the CPU as it was flat enough to not interfere with the seating of the heatsink.
Upon booting of the machine the temp IMMEDIATELY exceeded its reading capacity of 157 degrees. When I went into the BIOS it said the temp was over 200 degrees fahrenheit. I was surprised that I could even access the BIOS without it freezing. Shut down.
Switched the heatsink from the Thermalright SK6 to the Globalwin WBK38/Denki fan combo. Booted WITHOUT RadioShack sensor in place and got the same reading. Hmmmmm....
Now for further confusion: the processor I was using for these tests was the one I spoke of in an earlier message that froze EVERY time during the POST process (6 attempts). I was never able to get into the BIOS.
Now after a full day turned off I was able to get into the BIOS with the original setup that froze before (SK6 heatsink) as well as with the WBK38/Denki combo.
Now for the geniuses
Why would a processor freeze every time during POST process one day then on the next day POST without problems at an even HIGHER temp reading than before (170F vs. 200F)?
How could my processor be reaching such temperature extremes so quickly in basically an idle state?
What now?
Unbelievably frustrated....
Heydre
Power-On-Self-Test. Who told you that CPUs are at idle during boot up? There is so much juice at start up & this is what causes a temp jump. W/O seeing how everything is done or maybe NOT done would be really difficult to answer your "For Geniuses ONLY" questions. Calm down & check your notes & figure out whats causing the inconsistent results.
Thanks for the input Sonny :).
By saying "idle" I meant it comparatively. The work a CPU is performing during POST process is minimal compared to what it performs otherwise.
As for the idea that the initial surge of electricity at start up causes this drastic of a temperature rise, it doesn't seem likely.
As for what is being done correctly:
-CPU and HSF are cleaned and ArcticSilver reapplied for each testing (according to instructions on Arctic website).
-HSF is mounted firmly on top of CPU and no adjustment is made after sink contacts CPU. So as to not smear or displace the thin coating of paste on the CPU.
-Power supplies used are either a Enermax 330w or Antec 400w.
-Testing is done with only the CPU, vid card and RAM.
-Have received extremely high temp. readings with 3 different CPU. All AXIA1200 (266).
What'd I miss?
Heydre
I disagree with your comment on boot up temps. I see a large difference between my idle temps & right after boot up halt commands are not issued during post or when your in the BIOS. As for what your missing well I reckon enough to get you in a bind. Ability to ask questions is different from humility that you need in order to accept answers even if it's not what your looking for. Hear what nobody is telling you.
CrystalMethod
09-13-01, 06:36 PM
Which CPU do you have? The regular 1200, or the 1200/266?
Can anyone interested in this give me temp readings from your computer after initial POST versus temps after OS load at an idle state?
Please list temps and source of temp readings.
Thanks to any and all...
Heydre
Crystal it's a 1200/266...
thanks
Heydre
CrazyDave
09-13-01, 08:29 PM
Truthfully, this sounds like a faulty motherboard. You may be getting false readings from the bios. Since you've pretty much cleared up the possibility of the heatsink not being properly seated I would definately recommend changing the motherboard again. I have personally had bad motherboards twice in a row, (although I will never buy PC-Chips mobos again). Good luck to ya.
ok it would seem to me you got too much current,obviously...the only way you are going to get current like that is with a short...more than likely a resistor on the mobo...every resistor has a voltage drop from one side to another...throw a multi meter on the mobo and start checking voltage drops..if you get to a point where there is no voltage drop...that is the one...if you have a logic probe you could test the cpu itself as well...have you tried to put the cpu in a different mobo ???if you got one lying around you might want to try that after all that is less work..of course more money though
Thanks for the input guys...
CrazyDave: So far I have used 3 separate mobos (Asus, MSI) and same temp readings from all. The curious thing is that system temps are always about 26-28C even when the CPU temps have read over 100C. As for faulty BIOS readings, I have cross checked the readings with Motherboard Monitor and a home-built therm and got the same readings.
Warlok: That seems like quite a possibility. But I wonder if there is a short shouldn't it affect my system temps as well?
Thanks
Heydre
Get a new power supply. Sparkle or antec. 300W is plenty.
Athlon 1.4s run consistently at 46C with no special treatment. AMD heatsink, fan, and minitower with 2 aux fans.
Those temps are way out of range. The CPU may be getting too much power. Like running 220 through a lightbulb that needs 110. It works for a couple of seconds, then see-ya.
Seems you've done everything but replace the power supply.
The regulator may be out.
1) Some power supplies have a switch for 110/220 volts. Make sure it's set to match your conditions.
2) If you have a multimeter, check the power supply output. See if it's right.
-=UR=- Ranger
09-15-01, 04:01 PM
Originally posted by heydre
Can anyone interested in this give me temp readings from your computer after initial POST versus temps after OS load at an idle state?
Please list temps and source of temp readings.
Thanks to any and all...
Heydre
Here your temps
Bios temp after cold boot: 26 Degress raising till it reaches 36
Windows idle: 35-36 Degrees (no cooling software)
Full Load: 39 Degrees (Folding@home]
For system specs see sig.
I know itīs no T-Bird, but at 2.05 V the Slot A also produces a huge amount of heat...
Hope that helps!
:) thats an interesting question...hmmmnn...i would have to say that depends on what side of the "thermastat"it is placed in conjuction with the rest of the the system component sections on your mother board...and the order of which the rest of the components are placed on the mobo...ie...(hypothetic design)...power comes in ...to system sensor...then to cpu...then to cpu sensor...if this being your mobo design...then if the short is before system sensor yes system will over heat...if it is after but before the cpu then system remains unchanged...there are a whole lot of other factors involved...paralell/series/etc but i hope this gives you an idea...your mobo manual might have a block diagram/or even schematic which you would have to look at to get those kind of details :)as far as the power supply checking that is a good idea...im not sure weather such components have a voltage regulator on them or not...most power supplys are basically inductively produced...ie upstage/downstage transformer
...keep poundin...sooner or later the nail has to go in...
To respond quickly to your posts:
Ranger: thanks for contributing....:cool:
ibuild: good idea but have already replaced an Antec 400w with an Enermax 330w= no difference in temps.
Warlok: Thanks so much for taking so much time to respond to my posts. It helps a lot to share the load. I guess I'll have to bust out the multimeter. BTW I am going to try taking the motherboard out of the case for the hell of it and see if that makes a difference. It's the only part I haven't replaced besides my brain (sigh).
Heydre
Update:
Booted with mobo outside of case with vid, psu, ram. No luck same temps..WHAT the F? :mad:
Heydre
An update for all y'all:
After MUCH digging through an MSI newsgroup I've found >>>http://groups.google.com/groups?hl=en&group=alt.comp.periphs.mainboard.msi-microstar >>>> repeated remarks of higher than normal operating temps for MSI boards. It seems from most posts that they are operating in the 50-60C range.
They also mention that a Vcore voltage adjustment downwards of .2v will compensate for a tendency of MSI boards to run at a higher than expected default Vcore and thus temps.
All of this is merely owner commentary and not verifiable but lacking any other sources of info I'll take it. I just wish it could explain my exorbitant temps of 80-100C. Hope this helps any MSI owners out there.
Heydre
dimmreaper
09-20-01, 10:19 PM
Originally posted by warlok
i do believe there is a ground wire on the power supply plug to mobo...Actually with ATX there are 7 of them, and with ATX 2.03(P4 standards) there are 9 of them. But still I would discourage insolating the motherboard mounts, they should be grounded, just incase the MoBo designer got lazy and used one of the mounts for a particular components ground . . . .
PanteraGSTK
09-20-01, 10:31 PM
I'm a genius so i decided to post here.
sorry it took so long to reply i had a bad power supply....it fried mobo etc...lol it is working now tho...still heavy on the heat huh ??? say what is the ambient temp there your comp is locateded???
Warlok,
Ambient temp is generally about 25c and comp. temps are with the case off...
Heydre
heydre
here is something which nobody mentioned as far as i noticed...the heatsink clamp holding it to the mobo...try to replace that...if you havent...it may be just "bent" from shipping or something...and if the conection isnt tight your not gonna get heat released...and it is a cheap test too... :)
remember if it dont fit... force it if it breaks..it needed replacing anyway :)
Eureka!
After taking my rig in to have it examined at the local PC store the tech told me some interesting news. In his experience he's had way too many temperature issues with MSI boards. He went over my setup and was getting temps of 80C. He switched out the board with another MSI and got the same temps.
Sooooooooo I got on the phone with Newegg and they not only took back my 31 day old CPU but also my 70 day old MB and let me switch it for the Iwill KK266+ for a little extra cash.
I was almost too nervous to even set it all up again. Luckily different result..............
AXIA o'ced from 1.2 >>>>1.42
ambient temp: 29c/84f
cpu fan intake temps: 30c/86f
cpu bios reading on cold boot: 40c/104f
system bios reading on cold boot: 30c/86f
No OS installed yet but win2k coming. Not real happy with unloaded CPU temps :confused:. Have swapped out CPU fan with 60mm ystech and 80mm panaflo/denki. Only a degree or two difference. BTW this is with the case OPEN. I noticed with case closed that temps were:
cpu bios: 44c/111f
cpu sys: 30c/86f
Will keep u posted as put Humpty Dumpty together again.
Heydre
foxmulder
10-04-01, 06:26 AM
Ah a question for geniuses eh? Well as you can see from my avatar I am one:D
Let me be the first to congradulate you on getting your problem solved. Way to go man. As far as temps go I have noticed that alot of people state very low temps. Donn't worry about it too much. If your OC is stable at 65 degrees celsius under load take it and be happy. The onboard thermistors are so bad anyway that all they will give you is a general idea of when you're getting dangerous.
I'm running a watercooled system and my BIOS reads
CPU 48 Idle -- 50 load
System 34 Idle -- 35 load
Comparetivly speaking that's pretty high but since it works who cares?
Hi, i always have between 30-40C with AMD CPU into the BIOS after a cold start.
I had a stability problem once with a system - vidcard combination. CPU temp was always way to high. Switch the vidcard for a cheapest one and it solved the problem.
Did you tried with a different vidcard?
Do you still work with the case connections to the mobo (speaker, reset, power led)?
If so, remove everything that you can.
And by the way, i never used any kind of washer between my mobo and the case.
sciss
Pitspawn
10-04-01, 07:49 AM
Your prolly gonna kick yourself after this. I have had a problem like this. Its almost certainly your SK6.
I have a 1.4Ghz o/c'd and when i set it up with the SK6, I noticed bios temps rocketing up to 80C, instantly I would power pc and look for problem.
The problem is the clip! If the clip is facing the wrong way (Screwdriver push bit away from psu or pointing towards agp) then pressure is not on cpu. I switched the clip to the right way round and temps went from 80C down to 30C!!!!!!
I bet this is the problem. The SK6 rox if you get the clip on right.
well pitspawn it looks like i had the right idea about the heatsink clip eh...
heydre...im glad to see you got it runnin...it is a good accomplishment for all of us who try to give a hand at the immediate problems...it is a good challenge...keep on clockin
r0ckstarbob
10-05-01, 02:19 AM
nice to see you got yer rig up and runnin mr man. congrats.
krakerman
10-05-01, 02:45 AM
im confused. did you or did you not try the setup w/ the other heatsinks yet? if so, im just dumb and didnt pick that up. wait, yeah you did i think, and still got bad temps... bear with me its late here ;-) anyways, are you sure the heatsink is making good contact? i mean positive. might wanna put it on, then take it off and make sure the whole rectangle thingy shows in thermal paste on the bottom of the heatsink. you probably already did that though, judging by the thoroughness of what youve done. this shouldnt help, but have you tried underclocking? or can you even get into bios far enough. also, what type of case cooling do you have? and how hot is the room you are in? i know my comp would definetely not like being in a case w/ just a heatsink and in a hot room... though i dont see that making the temps that high. have you tried runnin w/ case open and all? also, where are you gettin the parts from, maybe they are just a lousy place that sells lousy parts and all just start out broken, :-(
Warlok, RSB: Thanks for sticking it out with me and helping me wade through this....sometimes I felt like I was either crazy or the unluckiest guy in the world. You guys rock...this forum rocks....
Krakerman: I will answer your questions shortly.
As for the clip issue that wasn't the cause of the temps as unbeknownst to me it was installed in the correct direction. As for my MSI board/CPU combo that one is still a UFO (Unidentified Frying Object). It still remains a mystery though we did place that overheating CPU on another MSI board and got 80C temps again. So go figure...it still irks me not to really know. Although I think the tech's opinion of MSI boards running hotter may be correct.
BTW my temps seem to have dropped a little with room temps. I noticed that with ambient temps of 23C I was stabilized at:
bios temps sys: 26C
bios temps cpu:35C :D
Yet as the saga continues on my new Iwill KK266+ I face the daunting win2k install hangs. So I am still the owner of an inoperable computer as I've begun my research on a new pain in my ass :mad:
Anyone up for another 1000 view thread? ....;)
My temps "idling" in bios are about the same as when running a benckmark. I posted this asking why and Hoot basically said this is normal. Idling in windows is much cooler. There is a program on the front page under utilities that runs in bios and it heats your cpu much more than any program run in windows.
Derek
bsblues
10-11-01, 01:41 AM
My guess is your puting to much thermopast inbetween
your heatsink and cpu.
How much pressure is there inbetween your heatsink and cpu?
I've recently run into basically the same problem with a MSI MB. You probably have a bad voltage regulator on the board. This is a VERY common problem with MSI boards about 12 - 16 months old. Try checking out alt.comp.periphs.mainboard.msi-microstar on USENET to get more input.
paupton
06-15-02, 10:27 PM
Now for my 2 cents worth.
i know this may sound incredible but..... I purchased an xp 2000 with new mobo in March. Put it in and it booted up and ran fine. When I shut down the machine to install other parts it would not boot up again. I RMA'ed and got a new one. This one wouldnt even boot up once. Sent it and mobo back and got third one. Same thing. Sent them both back and got new mobo and new cpu. This time they worked first try.
It was hard to believe but I got three bad cpu's in a row.
good luck
allan
Caffinehog
06-16-02, 12:45 AM
If win2k installation hangs, make sure virus protection is disabled in the bios.
Or, if you can get win2k installed on another computer, use norton ghost to copy that hard drive to this one.
kevmarks
06-16-02, 05:56 PM
There is a ton of advice here to sift through, some of it good, and some ermmm not so good.
It sounds like a defective heatsink to me. Your processor will be cold when first powered up. The POST is a very small program stored in EPROM. Takes very little processor power to execute this. (even 8086 handled it pretty well)
I would take the HSF off and clean any AS off both the chip and the HS. Re-apply and re-mount the hsf. Take the HSF back off again and check your inprint. You should have a nice square where the AS has been pushed out towards the edges. (FYI Always unplug the power cord when the HSF is off) Otherwise an accidental push of the power button = new proc and board.
While the power cord is out, reset the bios with the jumper. Allways do this while the power cord is out.
If you have a perfect square. Take a good look at the board. Look at the mem slots and the PCI/AGP slots. Look for anything that looks dodgy. (little bent contacts etc..)
Hanging while installing could be a DMA problem. But unlikly.
I am sure you have done this before but pull everything out from the board. If you have an old PCI graphics card laying around use that.
Boot up DOS or linux from a floppy. Leave the machine for about half an hour and see what happens. Disable as much as you can in the BIOS. USB, IDE controller etc. (Keep the floppy contoller :D )
let us know how this works out.
Like I said, I bet this is a dodgy heatsink
What are your vcore readings?
sangram
06-17-02, 08:38 AM
Since you've replaced everything in the chain except your video card, I'm beginning to think somehow this is the culprit.
I hope you're setting up Win2K with conservative / default settings in BIOS. It's very easy to first tweak up your BIOS to give you aggressive mem timings and such and have the machine hang repeatedly on installs.
Also when installing you might want to disconnect/disable hardware except essentials: Vid card, mobo, Keys, mouse...
I find that the reverse is true, actually: I like to enable everything while installing 2K as nothing manages to install later. I'm working on this problem of mine, should have a fix by this weekend.
Unfortunately when Windows setup is running your CPU is effectively at full load, always. So for BIOS. If you turn on the system and leave the BIOS loaded, you will see the CPU temp gradually rise till it reaches its full load temp (or close to it) very quickly. The Setup does not care how fast/slow your CPU is but will eat up all CPU cycles while running. Windows is where the CPU is actually put at rest for a while, before that no Halt commands are issued. It's common knowledge that you boot up, let the system idle in Windows for a while to bring the temp down, and then apply juice to OC it.
Check irrelevant hardware like the keyboard and mouse. Hey you never know. I got caught with my pants down (literally) when a new system I was building refused to start in anything but safe mode, and would not start Windows install. Turned out it was a defective keyboard, which had its <shift> key shorted out. Go figure! I spent three hours trying to fix that problem.
Set PnP/PCi to 'No' for PnP OS. I find installs work better that way.
Assign all IRQs to slots manually if possible. Try switching ACPI off while Windows installs. Sometimes this works wonders. Sometimes it does not.
You may just have got yourself another lemon, in spite of all this.
bigben2k
06-19-02, 02:00 PM
Heydre,
I don't understand why you insist on using a 330W PSU, especially from Enermax. If you're going for an overclock, I think you should either get an antec 400, or just about anything else that's rated higher. That way, it'll keep all your voltages stable. Also, when you're ready to upgrade, you'll have a decent PSU.
I'll be using an Antec 460W myself.
If you have a voltmeter (or MBM), take a close look at those voltages. If they fall more than 5% outside the rated voltage, then you found your problem (then and now).
Malakai
06-19-02, 02:30 PM
this thread is old and the member long gone, he only osted for this thread, but the only prob i see is he was mountiing the HSF wrong. thats the ONLY way u could overheat multiple CPU's on different mobo's w/ the same HSF
Flamed_Chip
06-20-02, 04:06 AM
I think he was mounting the HSF wrong, but was too arrogant to see or accept it.
Just a note about bios temps, In bios, I get temp readings the same as I do after running a couple of hours of P95 if I leave th bios open until the temp maxes out. It would apperar the bios loads the CPU nearly 100%
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.