Asus X50gl keeps crashing with ACPI THERMAL ZONE event
*long story* !
Good day folks. Today I'd like to open up a topic which I only mentioned a couple of times in some other threads but never actually got into as deep as I would've liked, so hopefully some of our wonderful members here who may be more familiar with laptops than I am can provide some tips. Please bare with me, as there's a lot to cover...
So: my old Asus laptop which I've used for the past 10 years and I dearly love
started acting up some time ago...sounds boring, but it's actually more complicated than you might think. A while back, it would randomly shut down whenever I was watching something on Youtube in Firefox - yes, it's pretty specific, which further adds to the issue..aand it did it again....and again...and again...I thought for sure it's a heat issue (which it might be), so I checked and triple checked my heatsink, paste, fan all that jazz, even went out of the way to order a second fan and heatsink, thinking perhaps the heatpipe lost some of its properties and no longer dissipates properly...never solved the issue. It delayed the crashes slightly but it would still do it nonetheless.
I then began monitoring my temps more closely, so I grabbed HWMonitor and ran some Prime95...nothing too out of the ordinary to report here: around 40C on the CPU and 57C on the GPU when idle at an ambient temperature of 27C and 70C on both at full load for 5 minutes. It's worth noting that it never once crashed when I stress tested it - THAT's the main issue here: you'd think it's a heat issue, but I'm skeptical about that, as it only seems to happen with Firefox and Youtube, never under clearly heavier loads - not even a different online video platform, say Dailymotion, causes the issue...it's ALWAYS YT. To confirm, I started using Opera for the past weeks or so and it solved the problem - no crashes at all. I thought for sure it's some sort of bug and I'd have to live with Opera, which is not terrible, but it all turned to sh!t today when it crashed once again and inspired me to write about it. It was on YT too, of course >_> The log indeed returns an "ACPI Thermal Zone has been enumerated" event every time this happens (don't have one on hand right now since I cleared it), but it always reports some trip point of 90+C (converted from K) which is clearly impossible for it to reach even at full load...don't know where these numbers come from.
RAM ? Tested that as well several times with no apparent issues. Of course RAM testing is not exhaustive and it could very well have issues despite managing to work enough that one time to pass MemTest.
Getting down at hardware level, I couldn't find a schematic for the F5GL model, but I did find a boardview (ironic, since it's usually the other way around !). If you open it up in your favourite boardviewer and search for U5000 (a G780P11U) you'll see it's a temp sensor. Unfortunately with no schematics, I have no idea how it's all wired up exactly, although I did use the boardviewer to track down some of its pins in that general area, which ultimately leads to the Nvidia GPU/Northbridge on CPU_THRM_DA/DC pins (SDA/SCL maybe ?). I doubt U500 ITSELF is faulty, since the temperature readouts appear to function. Usually a bad temp sensor swings negative or is stuck...IMO...
So many variables here I'm not even sure where to start: failing BGA on the Nvidia chip ? Software issue (did a clean reinstall on a separate HDD BTW) ? What other tests should I perform to at least pinpoint the cause ? It's mostly about the sentimental value of the device and about learning stuff rather than the material worth of this laptop in its current state, which is close to zilch at this point really....Cheers guys. Any idea or tip is welcome as always
Good day folks. Today I'd like to open up a topic which I only mentioned a couple of times in some other threads but never actually got into as deep as I would've liked, so hopefully some of our wonderful members here who may be more familiar with laptops than I am can provide some tips. Please bare with me, as there's a lot to cover...
So: my old Asus laptop which I've used for the past 10 years and I dearly love
started acting up some time ago...sounds boring, but it's actually more complicated than you might think. A while back, it would randomly shut down whenever I was watching something on Youtube in Firefox - yes, it's pretty specific, which further adds to the issue..aand it did it again....and again...and again...I thought for sure it's a heat issue (which it might be), so I checked and triple checked my heatsink, paste, fan all that jazz, even went out of the way to order a second fan and heatsink, thinking perhaps the heatpipe lost some of its properties and no longer dissipates properly...never solved the issue. It delayed the crashes slightly but it would still do it nonetheless.I then began monitoring my temps more closely, so I grabbed HWMonitor and ran some Prime95...nothing too out of the ordinary to report here: around 40C on the CPU and 57C on the GPU when idle at an ambient temperature of 27C and 70C on both at full load for 5 minutes. It's worth noting that it never once crashed when I stress tested it - THAT's the main issue here: you'd think it's a heat issue, but I'm skeptical about that, as it only seems to happen with Firefox and Youtube, never under clearly heavier loads - not even a different online video platform, say Dailymotion, causes the issue...it's ALWAYS YT. To confirm, I started using Opera for the past weeks or so and it solved the problem - no crashes at all. I thought for sure it's some sort of bug and I'd have to live with Opera, which is not terrible, but it all turned to sh!t today when it crashed once again and inspired me to write about it. It was on YT too, of course >_> The log indeed returns an "ACPI Thermal Zone has been enumerated" event every time this happens (don't have one on hand right now since I cleared it), but it always reports some trip point of 90+C (converted from K) which is clearly impossible for it to reach even at full load...don't know where these numbers come from.
RAM ? Tested that as well several times with no apparent issues. Of course RAM testing is not exhaustive and it could very well have issues despite managing to work enough that one time to pass MemTest.
Getting down at hardware level, I couldn't find a schematic for the F5GL model, but I did find a boardview (ironic, since it's usually the other way around !). If you open it up in your favourite boardviewer and search for U5000 (a G780P11U) you'll see it's a temp sensor. Unfortunately with no schematics, I have no idea how it's all wired up exactly, although I did use the boardviewer to track down some of its pins in that general area, which ultimately leads to the Nvidia GPU/Northbridge on CPU_THRM_DA/DC pins (SDA/SCL maybe ?). I doubt U500 ITSELF is faulty, since the temperature readouts appear to function. Usually a bad temp sensor swings negative or is stuck...IMO...
So many variables here I'm not even sure where to start: failing BGA on the Nvidia chip ? Software issue (did a clean reinstall on a separate HDD BTW) ? What other tests should I perform to at least pinpoint the cause ? It's mostly about the sentimental value of the device and about learning stuff rather than the material worth of this laptop in its current state, which is close to zilch at this point really....Cheers guys. Any idea or tip is welcome as always
Attached Files
Comment