Very Weird OS Windows Boot Crash (Thermal Error, Post Splash)

[LedHed]

Here is the error:

Log Name: System
Source: Microsoft-Windows-Kernel-Power
Date: 9/8/2012 12:22:40 PM
Event ID: 86
Task Category: (83)
Level: Error
Keywords: (32)
User: N/A
Computer: Keith-PC
Description:
The system was shut down due to a critical thermal event.
Shutdown Time = ‎2012‎-‎09‎-‎08T16:22:40.985351500Z
ACPI Thermal Zone = ACPI\ThermalZone\THRM
_CRT = 363K
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Microsoft-Windows-Kernel-Power" Guid="{331C3B3A-2005-44C2-AC5E-77220C37D6B4}" />
<EventID>86</EventID>
<Version>0</Version>
<Level>2</Level>
<Task>83</Task>
<Opcode>0</Opcode>
<Keywords>0x8000000000000020</Keywords>
<TimeCreated SystemTime="2012-09-08T16:22:40.985351500Z" />
<EventRecordID>150600</EventRecordID>
<Correlation />
<Execution ProcessID="4" ThreadID="60" />
<Channel>System</Channel>
<Computer>Keith-PC</Computer>
<Security />
</System>
<EventData>
<Data Name="ThermalZoneDeviceInstanceLength">21</Data>
<Data Name="ThermalZoneDeviceInstance">ACPI\ThermalZone\ THRM</Data>
<Data Name="ShutdownTime">2012-09-08T16:22:40.985351500Z</Data>
<Data Name="_CRT">363</Data>
</EventData>
</Event>

I am running:

CPU: Intel C2Q (EO) Q9550 @ 3.6ghz 1.336V (8.5x424mhz) 5:6
GPU: eVGA NVIDIA GTX 480 @ 840/1680/3920 (24/7) @ 1.113V
LCD: BenQ FP241W 24" P-MVA LCD (1920x1200)
Mobo: BioStar T-Power i45 AA512
Sound: Auzentech X-Fi Prelude + Onkyo TX-SR500 (LPCM) + Polk Audio Monitor Series II Speakers
RAM: OCZ Reaper 4x2Gb DDR2 @ 1012mhz 5-5-5-15
PSU: Corsair HX 750W 80+ Silver (62A)
UPS: Cyberpower CP1200AVR (720W)
OS: Microsoft Windows 7 Pro x64
Cooling: Antec 900 + Zalman VF3000F 480 + TT Water2.0 Pro + 4x Silverstone FM121 120mm (110CFM) w/ Controller

As you can see getting a thermal error is very odd considering the amount of cooling I have and the moderate OC is passing 20 passes in IBT 2.54. What happens is I will turn on my computer and recently on cold start ups I will sometimes come back to or see the computer is off. Even with the speakers on it makes no sound, it just kills the entire desktop. I recently put my TT Water2.0 Pro in with my 2x Silverstone FM121s (before this problem started) and I'm wondering, is my PWM or NB overheating. However how could this be possible in less than a minute of the computer being on? Especially when the rig passes IBT after 20 passes and nothing overheats then or crashes in the slightest.

Here you can see the PWM/NB area and the heatsinks used.


1215835800ivBn4B3xj6_1_12_l.jpg



Here is a shot of my computer setup:




You can see the room around the PWM is almost non-existent, the best reading I can get off of it (hard to hit any the PWM heatsink with a laser) is 38C while my NB is sitting at 45C, these are all idle temperatures. I will give load temperatures in the next post, but like I said, they shouldn't matter since this is happening before any load is really able to be applied. Can anyone give me any advice or have you seen this before yourself?

[LedHed]

Update: I managed to see what I think is the problem, when the computer booted a few minutes ago it showed over 100C across multiple CPU cores before it shut down. I know these temperatures are wrong obviously so I reboot and then everything is fine 30-40C across all 4 cores.

However I still don't know what to do to fix it, or what is causing it. Does anyone have any suggestions?

[CrazyNutz]

Sounds like dead, or intermittent dead pump on your TT water 2.0 pro. Your VRM would not get anywhere close to overheating that soon.

[LedHed]

the Water2.0 Pro is only 2 months old, do you really think that is what it is? Is there nothing else that could explain this, or do you know of a way to confirm this is the pump?

[LedHed]

I realized the fan header that powers the pump was set to smart fan settings in the BIOS, which I know wouldn't even work right with my GPU cooler (VF3000F). On the GPU cooler it would turn the cooler off and then back on randomly during OS loading, so this is probably the same thing that happened with the Water2.0 Pro. I turned smart fan off in the BIOS and I will let you know if this occurs again. If it does, I'll be calling TT.

[CrazyNutz]

Hope that works out

You might want to try hooking it up to a molex just to make sure.

[LedHed]

that must have been the issue, having the pump set to a smart fan header, I've had no problems since turning off Smart Fan.

thanks for the help though