What is this part on my H-Alvorix-RS880-uATX motherboard?

[Clay Aichin]

There's a part on H-Alvorix-RS880-uATX mobo that has metal fins that are attached by clips and looks like a heatsink for an integrated chip. It's right next to the CPU w/heatsink and fan (see photo link below). After unattaching the clips, it appears the adhesive glue has worn away as the "heatsink" can easily be removed from the chip. I've noticed that it feels hot to the touch after a few minutes of operation (the computer has been crashing witn BSOD's).

Anybody know what it is? Here's the photo (see white arrow). Notice that it's right next to the CPU which has its own heatsink and fan:

There's no manual for this motherboard that I could find. It's made by Foxconn for HP. It's the same layout as this one, which is the N-Alvorix-RS880-uATX. HP installed the BIOS for this board on my computer, and when I read the model # next to the ram slots, I realized it was the H-Alvorix, so I upgraded the Bios. Note the sparse diagram at the bottom:

http://h10025.www1.hp.com/ewfrf/wc/document?docname=c02560084&cc=us&dlc=en&lc=en#N125

I've been having loads of BSOD crashes and I've tried all kinds of tests. I even replaced the PSU and hardi drive, and reinstalled Windows, but it still crashes during Windows updates. It even crashed during a Darik's Boot and Nuke erase procedure For some reason, it runs well using a live Linux Mint DVD booted from the DVD drive.

 

[slag]

Probably the heatsink for your northbridge chipset. You can remove it, put a little Arctic silver on the chip, and put the heatsink back on. It should help lower temps.

 

[Ketchup]

Yep, Northbridge. I wouldn't mess with it unless you are worried about its temperature, but I don't think those old AMD's got too hot.

Here is a link if you haven't seen it yet:
http://h10025.www1.hp.com/ewfrf/wc/document?docname=c02560084&cc=us&dlc=en&lc=en

 

[Clay Aichin]

Probably the heatsink for your northbridge chipset. You can remove it, put a little Arctic silver on the chip, and put the heatsink back on. It should help lower temps.

I think I first have to remove the old adhesive that's sticking to the heatsink and the chipset. I can either use isopropyl alcohol or an Arctic Silver set that's on sale @ Newegg. Newegg is having a sale on the Arctic Silver line with free shipping. Any recommendations here?

http://www.newegg.com/Product/Produc...=-1&isNodeId=1

Could the loose heatsink and being hot to the touch be causing all the BSOD's
I've had the last 3 weeks? Is it possible that the heat could have caused the ram sticks to become bad, or the did damage to the CPU? It's a strange situation because I can run a Linux Mint Live DVD for an entire evening and view 1080p Youtube videos, but as soon as I run a live Linux based rescue disk, it crashes with no HDD activity light. I can't even boot into Bios unless the power is off for 15 minutes. I tried a Darik's Boot and Nuke disk to erase the HDD before another reinstall and that crashed after about 5%. The crashing has occurred before I replaced the HDD and after.

Yep, Northbridge. I wouldn't mess with it unless you are worried about its temperature, but I don't think those old AMD's got too hot.

Here is a link if you haven't seen it yet:
http://h10025.www1.hp.com/ewfrf/wc/document?docname=c02560084&cc=us&dlc=en&lc=en

Thanks for the link but it's the same motherboard that I linked to in my original post. It's not that old an AMD. The CPU is an AMD Phenom II X4 960Tquad core. It was released during the latter part of 2010 and still holds up nicely. Before I throw away another $100 on 2 new sticks or ram, I have to try something to stop the BSOD's, so maybe gluing the heatsink back to the northbridge will do it.

Another question I have is regarding the integrated ATI Radeon 4200 graphic card. Other than their Catalyst Center drivers causing alot of BSOD problems for many, should I be selecting "onboard" or PCI-E" in the Bios settings? The default is PCI-E.

 

[Ketchup]

Thanks for the link but it's the same motherboard that I linked to in my original post. It's not that old an AMD. The CPU is an AMD Phenom II X4 960Tquad core. It was released during the latter part of 2010 and still holds up nicely. Before I throw away another $100 on 2 new sticks or ram, I have to try something to stop the BSOD's, so maybe gluing the heatsink back to the Northbridge will do it.

Another question I have is regarding the integrated ATI Radeon 4200 graphic card. Other than their Catalyst Center drivers causing alot of BSOD problems for many, should I be selecting "onboard" or PCI-E" in the Bios settings? The default is PCI-E.

Gotcha. Have you tried running a program to see what the BSOD is pointing to?

 

[Clay Aichin]

Gotcha. Have you tried running a program to see what the BSOD is pointing to?

I'm going to upload a bunch of minidumps to the Sevenforums and see what they have to say. What I can't understand is why I can run a Live Linux disk for multiple hours, but
Window 7 in normal or safe mode, rescue disks, and Darik's Boot and Nuke (Dban) crashed in 5 to 10 minutes. Memtest86 ran for three hours plus with no problem.
If it were a ram problem, I probably wouldn't be able to sit for hours using a Live Linux DVD to browse the net. Either it's some conflicting drivers (probably AMD/ATI) or
some bad malware. I just hope it's not in the Bios somehow. I know that's rare, but there are several articles on the web that discuss the possibility.

 

[Ketchup]

Sounds like a driver, based on what you are describing. Why don't you run WhoCrashed and tell me (us) what you get:

http://www.resplendence.com/whocrashed

 

[JEDIYoda]

Probably the heatsink for your northbridge chipset. You can remove it, put a little Arctic silver on the chip, and put the heatsink back on. It should help lower temps.

The thermal paste that is already on it , is just fine!

 

[Clay Aichin]

Sounds like a driver, based on what you are describing. Why don't you run WhoCrashed and tell me (us) what you get:

http://www.resplendence.com/whocrashed

I remember Resplendence.com from the days of Registrar Lite. It was hard to figure out what was their freeware program and what the shareware one was. I would have run WhoCrashed, but I had already run a partial Darik's Boot and Nuke (Dban). That crashed with a BSOD, too.

The thermal paste that is already on it , is just fine!

I'm not sure what you mean. The paste that was on the chipset and heatsink no longer has a bond and the heatsink is no longer pasted to it.

 

[Fayd]

The thermal paste that is already on it , is just fine!

from the sound of it, he took the heatsink off already. therefore he needs to clean and reapply with *something*. whether or not it's arctic silver or alumina or ceramique (my personal favorite) or just some generic doesn't really matter.

 

[Fayd]

btw, from the sound of it, anything that involves significantly heating up the southbridge causes problems. (southbridge is the other heatsink covered microprocessor on the board). I suggest removing its heatsink, cleaning and reapplying paste, and reattaching heatsink. Do this for both north and south bridge.

(i'm guessing you may have either handled the board by the heatsink, or accidentally bumped it. either way, I'd guess you accidentally broke the dried thermal paste connection between the board and the heatsink.)

 

[Clay Aichin]

btw, from the sound of it, anything that involves significantly heating up the southbridge causes problems. (southbridge is the other heatsink covered microprocessor on the board). I suggest removing its heatsink, cleaning and reapplying paste, and reattaching heatsink. Do this for both north and south bridge.

I ordered the Arctic Silver Ceramique 2 as I've read that's best for the chipset. Many recommend the Arctic Silver 5 for the CPU and GPU's.

This is a photo of the N-Alvorix-RS880-uATX from the HP site. I have the H-Alvorix-RS880-uATX as that's what's written on the board. Naturally, there's no manuals on the HP site or any distinction between the two boards. HP doesn't even have the latest drivers listed under the computer. I think this may be the last time I buy an HP computer. I'm thinking of building my next one. At least I'll know the specs!

Would the arrow in this photo be pointing to the southbridge on the motherboard?

 

[Ketchup]

That's it. For future reference, whenever you take a heatsink off a chip, you need to re-apply the thermal material between the two. Depending on the chip, you can cause irreversible damage by not doing this. Like I said above, these chips don't get super hot, so hopefully everything is still ok.

 

[Clay Aichin]

That's it. For future reference, whenever you take a heatsink off a chip, you need to re-apply the thermal material between the two. Depending on the chip, you can cause irreversible damage by not doing this. Like I said above, these chips don't get super hot, so hopefully everything is still ok.

What's the best way to check if the southbridge adhesive is still solid without unattaching the clips, or should I disengage the clips and see if the heatsink moves at all by lightly touching it? Perhaps if it's not hot to the touch, I should just leave the southbridge alone for now, especially if applying the thermal Ceramique solves the BSOD crashes in Windows 7. For some reason (maybe different drivers?), I can run the computer for hours in Linux off a live DVD, but it would crash after minutes when booted into Windows.

 

[Fayd]

What's the best way to check if the southbridge adhesive is still solid without unattaching the clips, or should I disengage the clips and see if the heatsink moves at all by lightly touching it? Perhaps if it's not hot to the touch, I should just leave the southbridge alone for now, especially if applying the thermal Ceramique solves the BSOD crashes in Windows 7. For some reason (maybe different drivers?), I can run the computer for hours in Linux off a live DVD, but it would crash after minutes when booted into Windows.

a linux live CD doesn't interface heavily through the southbridge. it loads the session, then doesn't transmit any great amount of data. Windows is going to be constantly hitting the page file and application caches, which will increase the amount of throughput through the southbridge.

(southbridge contains the SATA controller, among other things)

btw, if the heatsink doesn't get hot to the touch, then either: (good)the chip doesn't get hot, or (bad) the thermal interface material connection is broken. You're best off just replacing the interface material.

I'd recommend doing it along with the northbridge. no sense in not doing so, especially when (imo) it's more likely that the southbridge is your problem.

 

[Clay Aichin]

a linux live CD doesn't interface heavily through the southbridge. it loads the session, then doesn't transmit any great amount of data....btw, if the heatsink doesn't get hot to the touch, then either: (good)the chip doesn't get hot, or (bad) the thermal interface material connection is broken. You're best off just replacing the interface material.

I'd recommend doing it along with the northbridge. no sense in not doing so, especially when (imo) it's more likely that the southbridge is your problem.

Isn't that the opposite of the problem I'm having now with the northbridge heatsink, which gets very hot to the touch as the thermal paste bond has broken?

I've been watching 1080p Youtube videos during Linux live DVD sessions and I've downloaded several files including a 3 gig + ISO version of Windows 7 from Digital River. That's quite a bit of data. Linux has also frozen the computer but after several hours. One session was over 5 hours before I shut down myself. Unlike Windows, the screen doesn't go black, although I have the same delays of 10 minutes or more before I can boot into Bios again. The HDD activity light also won't operate during this period, although all fans work and the power lights are on.

Even stranger, when I had Windows loaded, either I couldn't run Linux based antivirus rescue disks, or the system would crash while running the antivirus program via the rescue disk. Only once did I see any malware evidence. Dr. Web's Cureit said "amddriverdownloader[1].exe" was a trojan. When I selected Cureit, the system froze.

That's why my next install won't be from the HP Recovery Disk bloatware (burned to disk from my recovery partition). I'm going to try the Digital River Windows 7 download.

I'm also wondering if Foxconn (manufacturer of Alovrix mobo's) has a Bios upgrade for my motherboard, not listed on the HP site. HP had factory installed Bios for the N-Alvorix, and I flashed an upgrade for the H-Alvorix as that's the model that's clearly written on the motherboard.

I once think I saw an ATI insignia in the middle of the screen before one crash, and there's alot of posts about the ATI Catalyst Center upgrades causing problems for ATI Radeon graphics cards.

 

[Fayd]

Isn't that the opposite of the problem I'm having now with the northbridge heatsink, which gets very hot to the touch as the thermal paste bond has broken?

I've been watching 1080p Youtube videos during Linux live DVD sessions and I've downloaded several files including a 3 gig + ISO version of Windows 7 from Digital River. That's quite a bit of data. Linux has also frozen the computer but after several hours. One session was over 5 hours before I shut down myself. Unlike Windows, the screen doesn't go black, although I have the same delays of 10 minutes or more before I can boot into Bios again. The HDD activity light also won't operate during this period, although all fans work and the power lights are on.

Even stranger, when I had Windows loaded, either I couldn't run Linux based antivirus rescue disks, or the system would crash while running the antivirus program via the rescue disk. Only once did I see any malware evidence. Dr. Web's Cureit said "amddriverdownloader[1].exe" was a trojan. When I selected Cureit, the system froze.

That's why my next install won't be from the HP Recovery Disk bloatware (burned to disk from my recovery partition). I'm going to try the Digital River Windows 7 download.

I'm also wondering if Foxconn (manufacturer of Alovrix mobo's) has a Bios upgrade for my motherboard, not listed on the HP site. HP had factory installed Bios for the N-Alvorix, and I flashed an upgrade for the H-Alvorix as that's the model that's clearly written on the motherboard.

I once think I saw an ATI insignia in the middle of the screen before one crash, and there's alot of posts about the ATI Catalyst Center upgrades causing problems for ATI Radeon graphics cards.

i don't get it, why are you so resistant to something so simple?

the suggestion of changing the TIM will not brick your board. just do it.

DBAN writes at the maximum of drive speed. (~100 MB/s) downloading an ISO writes at whatever your download speed is (1-2 MB/s?) they're not comparable. data flowing through the southbridge corresponds with increased heat from the southbridge. if the thermal paste connection is broken, then the thermal paste that's on there is acting as a thermal insulator, not a thermal conductor.

I'm not saying for sure that this is the problem, but just fucking do it already to eliminate it as a possible cause.

you just said your system is not stable using the liveCD either. It's clearly not a problem with ATI CCC, or with windows.

 

[Ketchup]

What's the best way to check if the southbridge adhesive is still solid without unattaching the clips...

Can't do it. Take off the clips. If the heatsink falls off, replace the paste. If it doesn't, you're good.

 

[Clay Aichin]

I received the Arctic Silver Ceramique a few days ago and applied a small amount, but the northbridge heatsink is still running very hot. Speedfan and Hwinfo say the mobo temperature is 37° C or 38° C, but the heatsink feels much hotter. I can't even touch it for more than a few seconds. There was a foam shim on the bottom of the sink that cushioned the bottom from the rim of the chip, but it was wearing away and I had to remove it. Should the Ceramique thermal paste only be applied to the middle of the heatsink so as to make contact with the metallic chipset?

The southbridge heatsink was attached like cement to the chip so I didn't have to do anything there. It seems like the entire SB heatsink
couldn't move whereas the NB heatsink moves a bit as the Ceramique doesn't harden and the outer rims of the NB heatsink do not have any paste applied

My main problem however, is that even after a reinstall with a clean Digital River Window 7 disk, I'm still getting BSOD's. Whereas the two 4 gig ram chips were passing Memtest86 tests for over 3 hours two weeks ago, today they're failing when tested one at a time.

What's strange is that after each BSOD crash or a failed Memtest86 run, I can't boot into bios for over 20 minutes. How do I know if it's just the ram memory that's the problem or the motherboard has gone bad? What I can't be sure of is that the ram chips are failing on their own, or if the Memtest86 test crashes because the computer freezes. In the upper half of the screen during the last Memtest86 run, when the ram failed, I noticed that several of the zero's had smiley faces in them!

 

[crashtech]

If the old TIM was a thick pad, you need to modify the mounting as necessary to achieve firm pressure of the heatsink to the chip. You may need to remove the foam thing, and you may need to bend the attaching clip to that it develops more pressure. It also sounds like there should be a fan blowing across it, it's not good when the heatsink itself is too hot to touch.

It sounds like this obsolete board may be consuming more of your time and money than it is worth, other than perhaps being an educational experience. From your description, it may already have suffered irreparable damage from overheating.

The way you test the RAM sticks is to put them in another known good machine and run Memtest there.

 

[Clay Aichin]

It also sounds like there should be a fan blowing across it, it's not good when the heatsink itself is too hot to touch.
It sounds like this obsolete board may be consuming more of your time and money than it is worth, other than perhaps being an educational experience. From your description, it may already have suffered irreparable damage from overheating.

The only fan near the nb heatsink is the case fan and that blows air out. The new PSU I installed (Corsair CX500m) has its fan on the bottom of the PSU, facing the board. That seems to be an exhaust cooling fan which blows air in, towards the PSU. Most other PSU's I've seen have their exhaust cooling fans facing the back of the computer case.

Your comment about the PC consuming too much of my time is an uinderstatement. I had hoped that this computer could be saved, but a recent post on the Tom's Hardware forums mentioned similar problems with this board and stated that HP knew there were defective boards and extended the warranty by a year on some models (not mine).

You're also very perceptive about the "educational experience" I've had with the computer. I was experienced with IDE boards but this was my first foray into the world of SATA boards and I now feel comfortable knowing that I could put together a decent mid priced desktop of my own. I just need to find the right parts for a good, non-gaming necessary, desktop. I already have the PSU, hard drive, DVD drives and card reader. Now I just need a good case, motherboard, CPU, mid range graphics card, and maybe a card for extra USB 3.0 ports.

I'm going to need to purchase a cheap laptop in the interim so I always have an extra PC for research and something than can keep me going until I find the right parts and get a new desktop running. The 15 year old Dell I'm currently using (with Intel mobo and CPU_ is still running strong, but oh, so slow.

I'm not sure if my experience with HP is the norm, but I'm a little leery of continuing to go with AMD based motherboards and CPU's.

 

[Ketchup]

One problem I could see is that the air in your case being sucked up and around the northbridge, but nothing on it.

I have one of these in my parts cabinet, for occasions such as this.

http://www.sidewindercomputers.com/de40ef.html

I bet some cool air blowing down on that would put the temp question behind you, and we can see if heat is or is not causing your problems.

 

[tcsenter]

I think its probably time to consider that the BSODs have nothing to do with inadquate cooling on the NB/SB, but rather, that the excessive thermal dissipation from the chipset is an indication of a defect elsewhere, say, a dying or defective SMC resister, capacitor, voltage regulator, FET, etc.

 

[eegad]

In case it might help someone else with problematic Alvorix-RS880-uATX motherboard :

I have a pc with one that worked fine for a year and then started BSOD and randomly rebooting. I swapped RAM and power supplies to no avail. Then I removed the CPU & fan assembly and put it back together. The computer worked fine for another year or so. Then the same random reboots began again. The same cable checks, RAM swaps, etc didn't help. So once again I removed the CPU, this time cleaning and applying a fresh dab of arctic silver, put it back together and the machine has again been fine for the past 2 months. Based on my experiences, my guess was that there's a flaky connection in the CPU socket. So you may want to try removing and reseating to see if it solves your issues.

A second thing to try is to check the connector labelled J18 on the motherboard, near the SATA connectors. Someone posted a thread here

http://www.sevenforums.com/bsod-hel...elays-being-able-boot-into-bios-new-post.html

saying that they had found a loose connection there which they were able to fix and solve their BSOD issues.