Heat from heatpipes probably damaged a memory chip.
in [Asus]
|
From: Archimedes' Lever on 31 Oct 2009 11:11 On Sat, 31 Oct 2009 01:01:10 -0700 (PDT), "larry moe 'n curly" <larrymoencurly(a)my-deja.com> wrote: > > >Skybuck wrote: >> >> Accoring to memtest86+ at least one of the memory chips in Skybuck's >> Dream PC from 2006 is damaged and has a bit error. >> >> It turns out the memory chip was *new* from corsair which means it was >> not touched by the plastic sticker tape incident. >> >> This probably means that the bit error could have been caused by heat >> expelled from the heatpipes of the top down cpu cooler (Scythe Zipang >> CPU Cooler). >> >> I have taken some pictures to show you guys and to get some response >> from what you think killed it... (I am pretty convinced it was heat >> from heatpipe and will probably change the orientation from the cpu >> cooler again ?! hopefully for the last time... ;) I am not sure if >> it's possible though but I might give it a try.) >> >> Long ago somebody recommended I place the heatpipes near the big fan >> at the top... now I kinda wish I had done that from the start it might >> have saved a lot of damage...or maybe not ?!? >> But now I think it could have... maybe sucking away air around >> heatpipes is a good thing... >> >> Anyway here are the pictures: > >> Picture2: >> http://members.home.nl/hbthouppermans/MemoryCorruption/Pic2HeatpipesAsSuspect.JPG > >Doesn't indicate heatpipes at fault; view doesn't show whether or not >heatpipes are touching chips or module circuit boards. > >> Picture3: >> http://members.home.nl/hbthouppermans/MemoryCorruption/Pic3HeatpipesCloseUp.JPG > >It looks like they're not touching, but I can't be sure. If they are, >you can protect them by placing cardboard between the pieces. If you >want to be fancy, get fish paper from an electronics supply. > >> Picture4: >> http://members.home.nl/hbthouppermans/MemoryCorruption/Pic4PlentyOfFans.JPG > >Too many fans. Silly cooling setup. Memory doesn't mind heat -- >notice Micron and Samsung specify 85-95 Celcius max. operating temp. >for their DDR2 chips. > >> Picture8: >> http://members.home.nl/hbthouppermans/MemoryCorruption/Pic8FailedChipWasNew.JPG > >Where are the chips? All I see are the modules, which have heatsinks >over the chips. > >The only ways a heatpipe could have caused damage: > >1. It pressed against a heatsink and made a chip glued to it pull away >from the circuit board. > >2. It touched the edge of the circuit board and just happened to short >some copper plane or traces right at the edge. But I thought module >circuit boards were designed to leave a tiny bit of space between the >copper and the edges. > >But it couldn't have been the heat because the contact area was too >small, even if the contact was against a heatsink. > >You're not using the world's best memory because the chips are covered >by heatsinks, and Corsair admits that it uses UTT (UnTesTed) RAM >chips, rather than prime quality chips (will have actual >manufacturer's ID markings on chip packages). > >> I was also wondering if it's maybe possible to not use dim1 which is >> closest to the cpu ?!? > >Almost always. > >> I doubt it though...anyway I think it's best to get those hot >> heatpipes away from the memory chips... > >They probably don't matter. From 2k6? The AMD CPUs then were very good at thermal management, and if he did NOT overclock, the whole damned set-up is HUGE overkill. The best way to manage your PCs thermals are to MONITOR them, and observe what temperatures things are being raised to. DUH!
From: Archimedes' Lever on 31 Oct 2009 11:19 On Sun, 1 Nov 2009 00:41:26 +1300, "Craig Sutton" <xcxcnxxnxxx(a)ihug.co.nz> wrote: > >"Skybuck" <skybuck2000(a)hotmail.com> wrote in message >news:4db23453-1ee7-4cd9-a745-f31614a4e67b(a)g27g2000yqn.googlegroups.com... > >> >> For me this is becoming a drag... I have fixed this pc so many times >> it's starting to bum-me-out. >> >> Feels like I am fixing a formula 1 car or dragracing car all the >> time ! LOL I am kinda fed up with it ! ;) >> > >You still don't get it? > >The problem is YOU messing with it you don't have a CLUE! I bought my >similar setup to yours at around the same time and havn't had single >problem. > >Same ram,same m.b, same hd's Like I said, he is so clueless about ESD, that he probably *thinks* his gear is protected while he futzes around with it. I have been swapping hard drives, optical drives, sound cards, video cards, etc. in my PC since I bought it in '04. I keep the dust off the blades and tines. I use several SATA drives as backups, and archives, and they get swapped in and out ALL the time. Never a problem from XP, through Vista, and now Windows 7. HE is the problem, and ALL his posts he EVER made about his machines are the absolute proof.
From: GMAN on 1 Nov 2009 00:24 In article <4db23453-1ee7-4cd9-a745-f31614a4e67b(a)g27g2000yqn.googlegroups.com>, Skybuck <skybuck2000(a)hotmail.com> wrote: >Hello, > >Accoring to memtest86+ at least one of the memory chips in Skybuck's >Dream PC from 2006 is damaged and has a bit error. > >It turns out the memory chip was *new* from corsair which means it was >not touched by the plastic sticker tape incident. > >This probably means that the bit error could have been caused by heat >expelled from the heatpipes of the top down cpu cooler (Scythe Zipang >CPU Cooler). > >I have taken some pictures to show you guys and to get some response >from what you think killed it... (I am pretty convinced it was heat >from heatpipe and will probably change the orientation from the cpu >cooler again ?! hopefully for the last time... ;) I am not sure if >it's possible though but I might give it a try.) > >Long ago somebody recommended I place the heatpipes near the big fan >at the top... now I kinda wish I had done that from the start it might >have saved a lot of damage...or maybe not ?!? >But now I think it could have... maybe sucking away air around >heatpipes is a good thing... > >Anyway here are the pictures: > >Main folder: >http://members.home.nl/hbthouppermans/MemoryCorruption/ > >Picture 1: >http://members.home.nl/hbthouppermans/MemoryCorruption/Pic1BitErrorsDetected.JP >G > >Picture2: >http://members.home.nl/hbthouppermans/MemoryCorruption/Pic2HeatpipesAsSuspect.J >PG > >Picture3: >http://members.home.nl/hbthouppermans/MemoryCorruption/Pic3HeatpipesCloseUp.JPG > >Picture4: >http://members.home.nl/hbthouppermans/MemoryCorruption/Pic4PlentyOfFans.JPG > >Picture5: >http://members.home.nl/hbthouppermans/MemoryCorruption/Pic5LittleRoomForMemoryC >hips.JPG > >Picture6: >http://members.home.nl/hbthouppermans/MemoryCorruption/Pic6SomeDustOnTopDownCPU >Cooler.JPG > >Picture 7: >http://members.home.nl/hbthouppermans/MemoryCorruption/Pic7Overview.JPG > >Picture8: >http://members.home.nl/hbthouppermans/MemoryCorruption/Pic8FailedChipWasNew.JPG > > >I was also wondering if it's maybe possible to not use dim1 which is >closest to the cpu ?!? >I doubt it though...anyway I think it's best to get those hot >heatpipes away from the memory chips... > >I guess this was also kinda a experiment to see what would happen and >yup after many months... a memory chip got damaged ! So that pretty >much confirms the experiment for me: "heatpipes near memory chips not >a good idea ?!" > >Let me know what you think and peace out ! > >For me this is becoming a drag... I have fixed this pc so many times >it's starting to bum-me-out. > >Feels like I am fixing a formula 1 car or dragracing car all the >time ! LOL I am kinda fed up with it ! ;) > >Heat is a killer that's for sure ! My next PC must be a "cool" and >"low heat" and "low power pc" otherwise I will go nuts ! ;) :) (And >hopefully have still somewhat decent performance ;)) > >Bye, > Skybuck. From looking at the picture of how you have the CPU heatsink attached, you should have the heatpipes facting towards the back of the case close to the exaust fans. in that Antec 1200 case
From: Skybuck on 1 Nov 2009 03:24 On Oct 31, 10:24 am, "SteveH" <steve.houghREM...(a)THISblueyonder.co.uk> wrote: > Skybuckwrote: > > Hello, > > > I have taken some pictures to show you guys and to get some response > > from what you think killed it... (I am pretty convinced it washeat > > All those pictures show is that you live in a dusty environment and either > don't have air filters on your PC, or don't clean it enough. > > -- > SteveH Which gives me a new interesting theory: The Scythe Zipang is a top down cooler... which could mean it blew dust into the memory chips from above... and somehow that tiny little bit of dust caused a short circuit in the memory chip ! ;) Bye, Skybuck.
From: Skybuck on 1 Nov 2009 04:11
Another theory/hypothesis which I am slighty worried about is the following: "Water vapor from the dust filters was blown towards the memory chips and was blown into the memory chips which caused a short circuit". Not so long ago I cleaned the pc-case dust filters while the pc was running... I decided not to dry the dust filters thoroughly because a kid on youtube also didn't try them thoroughly... I just made sure most of the moisture was gone... just little bit of moisture was left. So I wonder if it's possible for tiny little micro drupplets of water to float through the air and kill a chip ?!? To me this seems unlikely but reality is usually more strange than fantasy ! ;) One last theory/hypothesis which I already wrote about in the past is: "pc is not grounded, and me touching pc causes static electricic feedback into the pc, since motherboard(s) is/are known to have some leakage, so reversed path seems plausible". I will probably call an electrician to check out and change the power sockets with "grounding power sockets"... I am curious if he will discovery anything about the power sockets... like maybe no grounding wire present :) Bye, Skybuck. |