1st crunching Casualty of the year...

[DrPop]

Just had the first crunching casualty of the year today.
Went to turn on Cheetah's monitor this morning when I got to the office, and had a fleeting glimpse of "shut down due to CPU1 overheat error" on my screen. Nice.

After a little sleuthing I found the culprit. The upper exhaust fan in the back of the case had died and was not spinning. With 6 cores 100% on PG LLR, it was generating so much heat that it was hitting 81C on one of the cores in CoreTemp without that fan working! Holy smokes...

So I backed off the O/C immediately, and crunching is paused on Cheetah until I can break away from the office and get a replacement at Fry's. This was a ThermalTake 140mm fan that came pre-installed in the case, so I suppose I should be happy it lasted this long, on 24/7.

[Al]

Ouch! I usually Tthrottle my amds at about 61c. Fortunately I haven't lost a case or cpu fan, so they usually max out about 5c below that, depending on the project. I hope Cheetah survived the baking.

[kmanley57]

Sorry to hear about the fan failure. I myself lost a 8-core AMD 3-4 days ago to a CPU (H100)cooler loss of coolant meltdown.

I have replacements on order that I hope fix the cruncher. Do not hold out much hope right now though!

[Al]

Maybe y'all should change you're screen names to "hotstuff1 and Hotstuff2" ...just a thought. Gotta go check my cooling. Bye.

[DrPop]

LOL! The goal is to be CoolDude when it comes to this stuff I think. So here's something I'm curious about. Since the H100 is supposedly a sealed unit, and you lost and 8 core CPU because it leaked fluid . . . will Corsair reimburse you for this failure of their part, or is sealed loop water cooling just as "on your own" as a DIY water cooling kit?

Update on Cheetah: Looking good so far, must have not been running that high of temp for too long before I walked in the office. Thankful about that! I put another fan in there, and have 5 of the 6 cores cranked up on PG LLR again, and it's stable at 40C. This is stock 3.3GHz right now. I'll let her run like that overnight and if no problems, we'll add the 6th core back in and then slowly take the O/C up again. Hopefully it was just a good scare and not out anything but a case fan.
PS - never realized just how much work that upper exhaust fan is doing!

[kmanley57]

LOL! The goal is to be CoolDude when it comes to this stuff I think. So here's something I'm curious about. Since the H100 is supposedly a sealed unit, and you lost and 8 core CPU because it leaked fluid . . . will Corsair reimburse you for this failure of their part, or is sealed loop water cooling just as "on your own" as a DIY water cooling kit?

Well, since I have heard nothing from them yet! You can guess what I think!

[Fire$torm]

LOL! The goal is to be CoolDude when it comes to this stuff I think. So here's something I'm curious about. Since the H100 is supposedly a sealed unit, and you lost and 8 core CPU because it leaked fluid . . . will Corsair reimburse you for this failure of their part, or is sealed loop water cooling just as "on your own" as a DIY water cooling kit?

Update on Cheetah: Looking good so far, must have not been running that high of temp for too long before I walked in the office. Thankful about that! I put another fan in there, and have 5 of the 6 cores cranked up on PG LLR again, and it's stable at 40C. This is stock 3.3GHz right now. I'll let her run like that overnight and if no problems, we'll add the 6th core back in and then slowly take the O/C up again. Hopefully it was just a good scare and not out anything but a case fan.
PS - never realized just how much work that upper exhaust fan is doing!

The answer is a big fat NO... Corsair's warranty excludes any damages to other components.

They wouldn't even replace the two Akasa 140mm fans that my H100 fried along with itself. That setup lasted all of 5 minutes of 1st install. The tech support guy on the other end of the phone refused to even acknowledge that I had mentioned the Akasas.... @#^*^@#^&^%&*^$#%@&#5

[DrPop]

Just a good news update on Cheetah - upped it to all 6 cores on PG LLR again, and holding steady at 40 to 41C temps. This is at stock 3.3GHz. If it holds that with all 6 cores for a few more hours, I'll slowly take the O/C back up when I have time. Thankfully, I think we may have avoided a serious meltdown.

[Slicker]

Just had the first crunching casualty of the year today.
Went to turn on Cheetah's monitor this morning when I got to the office, and had a fleeting glimpse of "shut down due to CPU1 overheat error" on my screen. Nice.

After a little sleuthing I found the culprit. The upper exhaust fan in the back of the case had died and was not spinning. With 6 cores 100% on PG LLR, it was generating so much heat that it was hitting 81C on one of the cores in CoreTemp without that fan working! Holy smokes...

So I backed off the O/C immediately, and crunching is paused on Cheetah until I can break away from the office and get a replacement at Fry's. This was a ThermalTake 140mm fan that came pre-installed in the case, so I suppose I should be happy it lasted this long, on 24/7.

Actually, it isn't the first of the year. My AMD 4000+ box with a HD 4830 died a month ago. I reinstalled Linux with a current version only to find it no longer supports the HD4830. So, now I either need to find an old Linux version that works with the current AMD APP SDK or just kiss it goodbye. That sucks because it was my Linux GPU development box.

[DrPop]

Actually, it isn't the first of the year. My AMD 4000+ box with a HD 4830 died a month ago. I reinstalled Linux with a current version only to find it no longer supports the HD4830. So, now I either need to find an old Linux version that works with the current AMD APP SDK or just kiss it goodbye. That sucks because it was my Linux GPU development box.

Oh man, sorry to hear about your older dev box there. Hope there is a good fix for it! Yeah, I realized now my title should have been "MY first crunching casualty of the year..." LOL!