|HELP| Computer crashes minutes after startup {WinXP} {SysTweak ASO}

[i3lueHorneT]

[Apologies should this not be the correct forum {originally 'General Hardware'} AFAIK as it stands it may be a HW overheating issue or WinXP,]

NOTE: At this time (see edit date for this post) due to the sheer mass of my posts it may be best to follow this thread post up with post 4th post of mine from this post, otherwise its my '03/02/2006 07:25 PM' GMT-7 post.



This is an absolute first for me (outside of a fauty heatsink fan).

Issue Start:
Upon startup today, once logged in for no more then 10 minutes (the computer was on and wating at the 'login' screen at between5-15 minutes, then 'logged in' for another 'no more then 10') I returned (logged in then walked away) to find the monitor in power save mode and the computer off entirely. Restarted, loaded and logged in, within one minute the mouse froze and seconds later crashed. Restarted again and it crashed before the windows loading splash (green 3-piece bar progressing across black screen with windows logo etc), restarted once more, this time a windows 'comp not properly shutdown last time, we're sorry for the inconvience, how would you like to startup this time: If you're not sure what caused the issue choose normally otherwise use last good config: Options -safe, -"other options", -last good config, -normally' I went with normally figuring if it crashed again I would start to narrow it down and go with last good config next time, after my selection crashed even prior to displaying the loading splash. No error reports/dialog pop-ups, just 'insta-off' everytime.

Issue Start Recap:
1. Computer on and waiting at 'windows login' ~5-15mins
2. Logged in and computer wating <10mins
3. Return to find computer off, immediately restart
4. Logged in and crash once looking at desktop <1min
5. Paused for moment to think of any obvious explanation, nothing, so restart.
6. Crash at windows loading splash
7. Opened comp case side panel (expecting the heatsink fan to be the issue -overheating cpu and my ASUS MOBO forcing at shutdown) and restarted watching the fan for issues -> were non, fan kicked on fine.
8. Windows 'bad shutdown how to load windows this session: '-safe, ...etc' screen, first time its shown up during this issue. Chose 'normally'
9. Crash previous/earlier to #6 crash (in less time).
10. EW Thread on alternate machine [began]

Issue Next:
After beginnng the notes in this thread (so I have facts right etc.) I went back to computer now 10 minutes since last crash, started up, was not given windows bad shutdown screen instead it loaded as normal, logged in (began checking last restore point info <-notes later) after 5 minutes crash just as before.

What do I do:
Should I just keep resetting until I get the 'windows bad shutdown, how should windows load now?' screen again, should I keep resetting until I can make it to restore point in fully loaded windows (gotten to desktop etc) and manually load last restore? Should I wait a good amount of time as it seems the longer I wait the longer it gives me (which leads me to believe overheating to be the issue, no?) is there anything (dxdiag.txt copied to floppy etc) I can export/get off the machine that I should post/analyize through a website/utility?

Also: what other info do I need to collect and post? Where else on the net/other forums should I go?

System Restore & Other Info:

-Systweak: Advanced System Optimizer & what computer/I did previous to last shutdown:
Last night I had been working with 'Systweak: Advanced System Optimizer [2.0.1?/2.10], for the first time after each process I restarted the machine, it was running a good 3+ hours with at least 1 restart from the last time it made a single change/I used it. It gave me no issues throughout. I ran the first 3 processes under the first ordered selection once the program is loaded -> left panel first option 'System Cleaners' -> right panel all 3: 'System & Disk Cleaner', 'Registry Cleaner & Fixer', "Startup Cleaner & Autorun Manager' again with restarts after each change whether prompted to do so or otherwise, with no noted issues. I also ran, in the left panel second option 'System Optimizers' -> right panel 1st options: 'Memory Optimizer' & 'Registry Defragger & Optimizer' again with no issues and restarts post each. FYI I had been following along with the help guide (one loaded in-program from help manual) which is a bit anemic but nonetheless I wanted to know specifics of everything it was going to do previous to doing it.

-after ASO's run:
Again the machine was running another +3 hours without incident, played a movie, ran language learning software, iTUNES, typical things to open/close etc. Shutdown w/o incident as well.

-System Restore:
ASO opens upon first run to a dialog requesting you create restore points as well as enabling/installing 'Windows Recovery Console' (of which I was presented with an error at the end of install ~something like: 'can't access/write to disk drive' [I took screens shots of it and made notes but haven't been able to look into it further since] -machine was from ABS Computer Technologies Inc. www.abspc.com [ http://www.resellerratings.com/seller1431.html for their info] with WinXP preinstalled and an OEM WinXP disk, which is what the console needed to be installed from as it wasn't finding anything on HDD. Nonetheless the rest of the program ran fine and restores were created, I removed/deleted the old ones. All this is to say there is a restore from that point just over 3 hours from the last time ASO did anything (Registry Optimizer/Defragger) on 2-28.


System Specifications:
(Been awhile since I've needed to list them, bear with me on the proper version labelling)
Allied ATX400P4 R (400W) PS
AMD Athlon XP 2600+ (~2.1GHz) CPU
ASUS A7N8X Deluxe (nFORCE2) MOBO
Corsair XMS 3200C2 (Could be 2.5?) 2x512MB RAM
High Point RocketRaid133 Array Controller (Array 0 -stripping, not redundancy)
Western Digital 80GB 7200RPM 8MB 2x(160GB) HDD
Radeon 9800Pro 128MB VPU
WinXP SP1 OS


Any advice, comments, applicable info, past experiences, links/FAQs, consolation, cookies appreciated. Hopefully this is a minor issue but I would rather spend the +1/2 hour writting this out and having it available on the net now then be even more inconvienced later on.


*UPDATE:
Alright, I left it some time to make sure I'd would load to the desktop, from there I manually selected the last restore point. Loaded it. System restarted and roughly 20mins later (I logged in and got to desktop and left it alone) it crashed again. I haven't any restores previous to that one, or at least windows system restore didn't display any listings previous to 2-28's using the back [<-] arrow.

*UPDATE #2: Edit to '-Systweak: Advanced System Optimizer & what computer/I did previous to last shutdown: ' ASO's utility names entered correctly.

*UPDATE #3: [Format] I didn't realize this at first, but formatting isn't going to be an option whatsoever, as I only have 10-20 minutes with which to work on the machine prior to a crash, there's no viable way to export vital data, reports, documents etc. regardless of the GBs of my retail CDs I've ripped to HDD. On top of which, I'd have to wait upwards another hour with the system off to be granted another full 10-20 minutes with which to work, otherwise it shuts down after only just reaching my desktop. It would litterally takes days upon days of dedicated around-the-clock effort (should I wish for a full manual/by hand backup, and that's just time enough to queue files for NERO to burn, actually creating some kind of backup from a backup program -I don't know if that could be processed in enough time before the next crash as I've never done that kind of thing), I just can't make this commitment. Plus I need it up and running ASAP, I suppose I do have other options for machines to work on, but it would come as a major inconvience, time wise especially.
/My sincere thanks.


peace,

 

[i3lueHorneT]

MSCONFIG: Boot.ini -> /Safe & /Bootlog. I set both and restarted, indeed I logged into desktop and waited, crashed within 10 minutes.

I'll watch BIOS temps now. What kind of numbers am I looking for? Actually now that I think of it, unless ASUS has temp gauges pre-set in the BIOS (which since it auto-shutdown when the PC was initially shipped me and I first ran it, it was due to that faulty heatsink fan I mentioned, so for that to happen would mean the gauges are all setup then wouldn't it?), I haven't set anything regarding it, they may not even be fully active, then again I may be thinking of fan speed controllers...we'll see.


*EDIT:
Alright, I watched the Hardware Monitor inside BIOS till it crashed within 5 or so minutes and then a second time after a 10 minutes delay where it crashed again within 5 or so minutes, crashing both time at the same temp readings:

CPU Temp: 90C/194F

-it takes a bit to arrive at this, fluctuating from ~135F when I first get to the monitor after its been off for over a half-hour, jumps around 160's & 170's, slowly & steadily increasing to 194F where it instantly shutsdown.

The CPU fan is running the whole time as I see it through a windowed case, however the CPU Fan Speed [xx] RPM reading displays a '0' after sometime, where previously I note it reading ~4000RPM seconds before.

The CPU reading the same temperature upon both crashes would leave it to be the issue no? What is the range it should be in, and if the fan is working then how could this happen, could something be forcing the CPU into more and more calculations speeding it up to temps a simple heatsink+fan can't dissapate quickly enough? Could the temperature monitor itself be faulty? Could the fan just be dying/slowing down -the first fan this desktop reseller (again, ABS PC Technologies Inc.) was DOA, but this one has been working faithfully the whole time (again, just under 3yrs).

The 'Q-Fan Control' is enabled with 'Fan Speed Ratio' set to the minimum of '10/15'. 'Speed Up/Down Response Time' is set at '4/8', the max (for anyone familiar with these BIOS options). Should I disable Q-Fan Control (I believe the fan will then run at max full-time, either that or it'll let me set the '15/15' option)?

Is this going to cause damage to the CPU, running the system over and over to diagnose it (should overheating be the issue) at these temperatures, or is this still a safe range to be in even if it's reaching the range each time I run it till the issue is resolved? That is to say, at 90C/194F am I melting anything by running it over and over/any precautions I should take (should 90C/194F be the correct reading)?

*EDIT: Made '*UPDATE #2: Edit to '-Systweak'
*EDIT: Made '*UPDATE #3: [Format]' to original post in regards to the option of formatting WinXP.
/Thanks.


peace,

 

[pelikan]

I would do a repair install. This will get everything working again (repair the damage done by that system optimizer program). You could leave it at that, but I would then back everything up, reformat and clean install XP.

 

[Snooper]

Maybe you should fix the CPU fan issue BEFORE worrying about reinstalling the OS... 90C is WAY too hot for that chip. There is an issue somewhere. The heat sink hold down could be the issue, the fan could be failing, etc. I would take that CPU fan off, inspect it CLOSELY and see what you can see. If the fan looks good, clean the CPU and heat sink up real good (heatsink compoud off the bottom and the dust and crap off the top) and reinstall it with good heatsink compound (like many here, I am very partial to Artic Silver 5). Make sure you plug the fan back into the proper connector on the MB!

Leave the case side off and reboot into the BIOS and go to your hardware monitoring section and watch the CPU temps while watching the fan speed (both in the BIOS and physically verifying that it is spinning correctly). If your temps stay down (I'm not all that familiar with the XP 2600, but I would guess 55C or lower idle at least!) and stable for 15 minutes or so, go ahead and reboot the computer into Windows and see what happens. Just keep your eye on that fan! Make sure it keeps spinning at speed and doesn't stop for some reason. If it does, it will be very apparent as your computer will lock up pretty quickly afterwards.

If the heat sink is on good but you have a fan issue, you will have to do a bit of experimenting to find out if the fan itself is bad or if the power connector is bad on the MB (physical connector or power control circuit that feeds it the juice). I'll leave you to figure out those tests (hint: 12V DC is available all over the inside of those hard drive cables...)

 

[i3lueHorneT]

Originally posted by: Snooper
Maybe you should fix the CPU fan issue BEFORE worrying about reinstalling the OS... 90C is WAY too hot for that chip. There is an issue somewhere. The heat sink hold down could be the issue, the fan could be failing, etc. I would take that CPU fan off, inspect it CLOSELY and see what you can see. If the fan looks good, clean the CPU and heat sink up real good (heatsink compoud off the bottom and the dust and crap off the top) and reinstall it with good heatsink compound (like many here, I am very partial to Artic Silver 5). Make sure you plug the fan back into the proper connector on the MB!

Leave the case side off and reboot into the BIOS and go to your hardware monitoring section and watch the CPU temps while watching the fan speed (both in the BIOS and physically verifying that it is spinning correctly). If your temps stay down (I'm not all that familiar with the XP 2600, but I would guess 55C or lower idle at least!) and stable for 15 minutes or so, go ahead and reboot the computer into Windows and see what happens. Just keep your eye on that fan! Make sure it keeps spinning at speed and doesn't stop for some reason. If it does, it will be very apparent as your computer will lock up pretty quickly afterwards.

If the heat sink is on good but you have a fan issue, you will have to do a bit of experimenting to find out if the fan itself is bad or if the power connector is bad on the MB (physical connector or power control circuit that feeds it the juice). I'll leave you to figure out those tests (hint: 12V DC is available all over the inside of those hard drive cables...)

Ahh thank you, quickly here as I'm heading out the door:

The temps I mentioned, the first one I was able to read (at the BIOS and H/W Monitoring tab) was ~57/134, so spot-on the idle temps you mention (55). Artic Silver I read as being one of the best years back when I first researched my PC, so I guess it may be time for me to expand a bit out of complete H/W novice and look into CPU/VPU cleaning/chasis dust cleaning etc. as I've never toyed around in the guts, only to wipe down the chasis and do connector test when I first got it, move it, and replace the video card -stuff of this nature, I've never touched anything to do with heatsink-to-processor cleaning or maintainence.

Reformat is always the last option for me, though I'm gratefull for any advice I get, its just after a reformat there's a lot that needs to be configured and beyond that I get a headache trying to figure out what I'm going to do should I want to backup, be it partition or otherwsie.
/Thanks for all the help thus far, I'm not much into PC H/W as the reader may be able to tell, the last time I did any research regarding anything PC H/W was on & off for some months prevous to this PC (crashing PC not this alt. PC I type from here) purchase, and its 3rd B-day will be the end of this month.


peace,

 

[i3lueHorneT]

Dusting Case/Fans:
Alright, since its been sometime since I've gone into the case to dust it, that and I wasn't really sure how to go about it in the first place, I'm going to ask for some advice on cleaning the inside of the case. I don't think it too smart to do anything like have a desk fan blowing on the CPU to test for (and possibly temporarily alleviate) heating issues in its current state. Dust isn't thick or in balls or anything, but there is a distinct dusty film over all surfaces, therefore its definately think enough to warrant concern.

NOTE: Working w/o any computer dusting equipment -no specific cleaning utilities, left to use typical household items. Though I suppose if its advisable I could pick up some things now (like at a Fry's Electronics etc), but I really need just a quick fix for the next few days and in the coming weeks I can return with a bit more research and do a more thorough job, once the machine has been up and running again.

1. What can I use to to pick up dust around the case itslef, is just some lightly wetted papertowels best for this?
2. Going over the video card and mobo, I'm going to want to remove them for proper cleaning, at least the video card as I'm not too comfortable with taking out the motherboard just yet, what can I use to grab the dust off these two?
3. Wipping down the fans -anything I should know about ahead of time?
4. General do's and don'ts. Obvisously I'm not going to want to get anything wet, so then what do I use, etc. Typical novice mistakes/assumptions to avoid.
5. Basically what do you do that you have looked into/researched and are confident in.

Some of these may be a bit daft to ask but I'd rather be sure not to make the one mistake, overlook the one thing that's going to come back and bite me.

Some interent guide/tip page on going about internal maintainence like this, something you've read through and trust, would be great.


Once Dusted:
My next step once I'm confident going about it won't further aggrevate the issue, would be to add a small desk fan to the side of the case (open panel) and see if that allows it to continue running, from there at least its a temporary fix and I can still use it. My next step would be to invest in some compound gel (Artic Silver has come up, most likely I'll be going with that) and look into how to clean and re-set the heatsink to the mobo -as I've never done this and I figure it worthwhile either way, afterwhich I may use a power cable hookup for the cpu fan instead of using a mobo outlet as its set now.
/Thanks.


peace,

 

[evilsaint]

The fan blowing into the case (preferably directly onto the fan on top of the processor HSU) is a good start... you're not as much the novice as you might think You probably don't wan't to pull the processor out until you have some thermal compound of sorts around, because if you just used the OEM thermal 'patty' POS that came with the original heatsink, then your CPU might very well heat up even more when you put the HSU back on.

The 57Celcius temp that you saw in BIOS isn't bad, but for example, on my XP2500+ Barton with Arctic Silver and a huge TTake fan/sink, I only get temps that high running stress tests on a fresh OC... my idle is around 40C usually, mostly lower.

On the dust front, you can try using paper towels (dry!) to knock some dust off the non-fragile parts, but TBH, you should wait until you can get some compressed air to just blow the dust out of *everything*. I doubt that is what's causing the big overheating problems though, unless there are, like, pelts of dust in there

 

[i3lueHorneT]

Brief of the following:
Today I went about dusting the inside, wipping down the fan blade, made an interesting discovery regarding the top of the heatsink sans fan and after positioned a small desk fan (READ: Duraraft small 2-setting) to blow on the HSU itself, sitting just outside where the panel would be had I not removed it for this purpose (so pretty close IOW). After all this I started up, BIOS temperature monitoring reported numbers akin to what's been suggested as normal (after some time ~55C for the CPU). Eventually I reverted/restored the system point back to the day prior to any issue (2-28's post ASO's last changes made -which I've noted as having restarted at least once thereafter and continuing to run fine for another +3 hours before shutting down, the next startup being 3-1 and the crashing issue beggining) and the system has run fine w/o issue at desktop (without any use but to open ASUS probe I should note) and at stable CPU temperatures (55C) for nearly two hours as of this writing. What follows is the breakdown, realize there is still an issue here, possibly at least, I've yet to test without the deskfan blowing into the CPU HSU and so on (wanted to get this written up first, may not have time until tomorrow to continue testing as there's a fair amount of work I'm best to do on the rig in question), there are detailed notes on all that here following should the reader skim through using the headings.


Dusting:
NOTE: I either used dry papertowels or a 'misted' papertowel, I never shouldn't have referred to my intention as 'wetted', what I use is a small fine misting water spray bottle, with just a squirt or two to lightly dampen the towel -within a few minutes its completely dry again. As well as q-tips (cotton swabs) dry & again with the same type of mist. Without misting these items its near to impossible to get some of the dust out, I don't have a compressed air can available but this exercise has convinced me (among a regular cleaning and software maintainence program) to invest in one.
1. Wipped down the chasis/case insides, panels, floor.
2. Removed HSU fan, cleaned thoroughly via q-tips. -was quite dusty. Noted a considerable buildup of dust on the top of the heatsink 'veins/slits' at least 1/4 had a continuous 'plate' of dust -unable to see down the slits. The four corners where the screws and triangular plates that make the screws possible, the undersides (between the topmost of the slits and matal plates that the fan sits on) each had a good buildup of dust, Q-tips and caution made sort work of most of it.
NOTE: It may be hard not being in-person and without any visual recording but couldn't this alone cause of difference of an idle temp normally being ~55C to go to ~90C or would such a jump require more aggrevators, say a HSU fan not running well enough (READ: speed wise) due to simply dying out?
3. Wipped/blotted PSU, RAM, VPU card, cables, fan blades. Again used commonsense and caution, I say 'blotted/padded' meaning with the VPU for instance, can't exactly drag a paper towel (dry mind you) across it, but the top side did have a thin layer, so I lightly 'stamped' the towel on the surface a few times over.
4. Replaced fans, anything moved/changed. Checked all connectors/power etc. Reset/connected rig -outer cables etc.
5. Set desk fan just outside removed side panel. I let it run on high for a few minutes previous to boot to blow out all the newly moved/jolted dust, no sense in driving any back into the CPU/VPU heatsinks.
6. Booted into BIOS and watched temperatures whilst taking recordings for 40 minutes. These are to follow.


Temperature Recordings:
NOTE: I made a time, MOBO, CPU, and CPU fan RPM note every so often, count 16 for the 40 minute period, here's the important ones (by my account at leat). The first number is the recording number, should the reader be curious to see a number not presented ask and I'll produce it, the rest of the readings are self explanatory.
1st: 0 mins MOBO 20C/68F, CPU 36C/98F, RPM (N/A = forgot to record)
4th: 4 mins MOBO 28C/80F, CPU 44C/11F, RPM 2700
7th: 7 mins MOBO 28C/80F, CPU 49C/120F, RPM 2500
9th: 13 mins MOBO 32C/89F, CPU 55C/131F, RPM (N/A = '0' monitor stops/errors or for whatever reason begins displaying a reading of '0' after some time, realize that the fan was indeed still spinning on its own power)
12th: 24 mins MOBO 32C/89F, CPU 52C/125F, (N/A = '0' again, note that it will often 'flicker' a RPM recording every so often -these flickers fluctuate around 2500)
13th: 33 mins MOBO 32C/89F, CPU 56C/132F, RPM (N/A = '0' but spinning fine -not due to desktop fine as I've tested it by finger interuption upon the first '0' and again every so often)
16th: 40mins MOBO 32C/89F, CPU 55C/131F, RPM (N/A = '0') LAST RECORDING.


At this point I exit bios and continue to desktop, it boots to safe mode as that's how I last had the win.ini setup, I change the win.ini for normal startup and restart. Back at the desktop after a normal restart (as a note of interest -took ~4 minutes time from the 16th(Last) temp recording to be at this point which is consistent with a usual boot followed by restart time for my rig as I last had it) where I let it sit untouched for another 10 minutes after which I loaded and began watching ASUS probe (which can display same temperature monitoring as BIOS) which hovered around the same temps as last noted (16th: 40mins MOBO 32C/89F, CPU 55C/131F, RPM (N/A = '0')) for another few. Finally I ran a system restore confident my last change made with the ASO program were completely safe once a total of 20 minutes had passed during this last restart session.


Now: (6:20 PM)
System has been at desktop and running the previous system restore (2-28's post ASO's last adjustments) with the same temps as the 16th recording (16th: 40mins MOBO 32C/89F, CPU 55C/131F, RPM (N/A = '0')) steadily for over an hour.


What's next:
I'm confident the issue that presented itself initially upon (3-1) first boot was a CPU overheating issue. Granted there may be more errors/issues waiting for me with such poor attention to maintainence (HW & SW) over the years since I've had it. Well its been sporadic at best lets say, besides my greater attention to these details in the past few weeks with my running and installation of BitDefender, Spy Sweeper, and Outpost Firewall -upgrades/updates/virus definations/etc. that were long overdue for pretty much similiar software (Nortorn Sysworks '04 & ZoneAlarm Pro) but all that is a bit off topic. As it stands its an HSU issue I'm dealing with now, things like having a partition ready (windows on its own) and a more thorough & regular restore point/maintainece/cleaing regimen/etc. can be set into place in the upcoming days & weeks, slowly and with my full understanding of just what I'm doing (which explains my great hesitancy to any sort of reformat -I don't know the full extent of just where it will put me or just how/what to do about backing up and I'm not about to just 'leap into' such a thing and learn after the fact either).

Overheating (though its getting late for tonight) -next up are tests w/o the desk fan added and with the side panel replaced, regardless of the results from those I would like to reset the HSU (possibly even get a new one if its advisable, fan or combo) with some new compound. Figure out why the RPM monitor dies out after awhile, and see if changing BIOS Q-Fan control to force a fulltime 15/15 works to speed it up a bit. Then I think that's about it, unless I'm forgetting something?


What I ask of the reader:
My posts are quite long, I know, just trying to be thorough, at this point I should add a quick note to the top of the original thread post should it still be editable to ease any new readers, anyway here's what I'm looking for now and of course anything else the reader would think to add is most welcome as well.

HSU: Replacement advice -needed/not needed, worth it to a rig w/o any overclooking, who and what to go with, price range I'm looking to spend regarding the reader's particular advice? Was the collection of dust atop it (noted in cleaning procedure in this post) a likely culprit and if so to what extent could it drive the temperatures up?

BIOS: Anything I should know/do/change/tweak particular to heat monitoring/fan controls and speed? Down clocking (again its never been overclocked to begin with) would serve as a temp fix should I need it, any links to what temps a default/vanilla working of my system (see orig. post specs) should be as it is possible I mistakenly overset (overclocked without intending to) a BIOS option here or there as I've have gone though and done things like 'Aggressive' over 'Optimial' RAM settings. What regarding this do you recommend reading over (links, guides, etc. -something digestable for me at this point, all learn on my own accord another time, not looking to do anything else right now other then to get the rig back in working order on the same level it was when I left it)?

READER: Whatever you feel you should add, again I've written quite alot over the two days, I'll forgive you should your question/comment already be gone over no worries, any advice you feel applicable is GREATLY appreciated.


/Thanks ever so much, I would like to, with a bit of time, reconstruct this issue/thread into something of a resource once its resolved in-full, hopefully serving to help anyone else who should come across this issue and, something of a FAQ guide would be in my scope of curiousity and initiative.


peace,

 

[evilsaint]

Umm... well... you sorta answered your own questions there, so voila! I present to you the way to properly apply thermal compound, for when you get around to it : Nice Tuturial from AS

 

[i3lueHorneT]

Brief of the following:
Today (Fri, 3-3) I went about testing the CPU without having a desk fan lined up and blowing on it and watched in BIOS to see what would happen. It ran to 90C and safety-shutdown in just under 10 minutes, so clearly something more needs to be down ASAP as the desk fan is only feasible as a very temp solution. At this point I'm desperate for some HSU specific feedback, just what to get (solely fan replacement or go for HSU+Fan combo) and what to go with (brand, make etc).


Boot up w/o deskfan test:
NOTE: Not really necessary to write out any more then what's in the brief that opens this post other then to say CPU temp went from 56C at 6 minutes to 71C at 8 minutes to 90C less then two minutes later (forcing an auto shutdown) whilst the MOBO remained a steady 24C. This is with the case closed up again, HSU and its fan still intact and untouched from the dusting it recieved on 3-2.


HSU Fan: I figure it has an issue for two reasons:
1. ASUS Probe and BIOS aren't able to display a steady reading of it, often drops to '0' every so often (even though it continues to spin) (SEE: 3-2's Temp Recordings for specifics on '0' RPM) only to return to reading <1000RPM ranges -today's especially whereas the previous days testing showed that when it did display a RPM recording it hovered around 2500RPM.
2. Within the past few days everyso often, and usually it seemed at least most promenient at initial boot, the HSU fan has been giving an awkward sound, as if the corner of some loose-leaf paper was clipping it, only not nearly as loud as I imagine a sound paper would produce. It will go away for a bit, only to return -on top of the great roar ABS PC's installed fans are known to make (I read a system review earlier today out of curiousity to see how the company stood up after all these years since I spent any time researching them, and it seems they've become somewhat common in this regard).



What's next:
Thermal compound -on a new HSU or not? Which is the next feasible step? Socket A (462) (SEE: orig post for full sys. specs.) after that I've no clue what to look for in a HSU and would rather not just grab one at random, I clearly have access to the net and buying offline (NewEgg or equivalent trusted site, which BTW I recently found out is owned by the PC assemblers I bought the rig from: ABS PC Tech Inc /Computer Technologies -again orig. thread post if curious) isn't an issue, and other then a return (RMA) issue, should one arise, I'd rather go with over going somewhere local to pick it up as I figure I could save some cash between the two (or more?) items: compund & fan assembly.

1. Thermal Compound: Is Artic Silver 5 all I should look at for buying?
2. HSU(+/- Fan)? Who, what, why, when?
3. Is there a 3rd, what else is there -my previous post above this one, the very end of it 'What I ask of the reader:' I'm still looking for and remain relevant, so please take a look through that part as well, no sense in reposting it all.
/Thanks as always.


peace,

 

[pkrush]

The best heatsink you can get for Socket A is a Thermalright SI-97 (you'll also need to get a fan to put on it). If you don't want to spend that much, then anything with at least a copper base and an 80mm fan should cool fine. Arctic Silver is still the best thermal paste, but it's only about a 3-5C difference from the OEM white stuff.

 

[i3lueHorneT]

Originally posted by: pkrush
The best heatsink you can get for Socket A is a Thermalright SI-97 (you'll also need to get a fan to put on it). If you don't want to spend that much, then anything with at least a copper base and an 80mm fan should cool fine. Arctic Silver is still the best thermal paste, but it's only about a 3-5C difference from the OEM white stuff.

That's exactly what I've gathered so far (been searching around a bit since my last post) both in regards to the SI-97 as well as the the AS-5, of which I may go with Ceramique http://www.arcticsilver.com/ceramique_instructions.htm as I've heard its easier to work with and non-capacitive (though slightly lower #C results).

The HSU however good it may be is out of what I'm wanting to spend (more like 1/2 that for $15 max on a complete solution, not $35 on solely the HSU), should I want to upgrade down the road then fine. Right now I'm leaning on either just grabbing a fan that fits the demensions (which I don't want to do as I'd rather get something that I've had specific reviews for dBA considerations esp) to slap it on for now. I already have a second HSU when ABS had to replace the DOA one the sent me originally on first PC order they send it out packaged together, so its sitting here brand new should the current HSU be faulty as well.
/Thanks.


peace,