• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Need some help over here! =(

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

yatiac

New Member
Joined
May 15, 2012
Hello all,

My configuration is:
- AMD Athlon 64 X2 5000+ 2.7 GHz
- Cooler Master Hyper 212 Plus (CPU Cooler)
- Thermaltake V4 Black Edition Case (With 4 120mm Fans)
- 3Gb RAM:
* 2GB Kingston PC2-6400 (400 MHz)​
* 1GB Kingston PC2-6400 (400 MHz)​
- Saphire AMD Radeon HD 6850 (1GB GDDR5)
- MoBo ALiveNF6P-VSTA (CPUSocket) Socket AM2
- Cooler Master Silent Pro M 600W
- 1 x 160GB SATA Seagate HDD
- 1 X 500GB SATA Seagate HDD


My problem is that after I get the Radeon HD 6850 My pc started to randomly shut down when I am playing games like Battlefield or Sniper Elite V2...

I though it may be the CPU overheating so I get a Cooler Master Hyper 212 Plus and the Thermaltake V4 Black Edition Case (With 4 120mm Fans) but the problem persist, so today I bought the Cooler Master Silent Pro M 600W and started to play BF3 and again my PC randomly shut down.

If any one has any idea or any solution I will be very Thankful.

Regards,
Yatiac
 
Thats right, I am using Windows 7 Ultimate - 32bit,

but I can play for like 20 - 30 minutes and then the PC just shuts down. =S
 
The first thing I would suggest is to do some "controlled" stress testing. Don't assume high temps are not a problem simply because you have installed a nice aftermarket cooler and have a good case with several fans. For instance, your cooler may not be seated properly or your TIM may be applied incorrectly. Let's find out for sure.

Download and install these free programs that are standard stress testing, monitoring and reporting tools: HWMonitor, Prime95 and CPU-z.

First, open up HWMonitor on the desktop. This will give you lots of information about system voltages and temps. Leave it open while you stress test.

Second, stress test for 20 minutes with the Prime95 blend test. This will load your CPU cores to virtually 100% and really stress your ram as well.

Post back with pics of the HWMonitor interface so we can have a look at temps. To post with pics attached, do a cropped screen capture of the desire image and save to disk. Snipping Tool in Windows Accessories is great for this. Then click on the "Go Advanced" button at the bottom of any new post window. This will bring up the Advanced Post window. Then click on the little paperclip tool at the top which will bring up the file browser/uploader tool. The rest will be obvious.
 
Thanks for your response @trents.

As indicated, I ran the prime95 on blend mode and after like 6 minutes my PC shuts down. Here is the Last SS I made (hope it helps)
CPUID.PNG

Regards,
Yatiac
 
Last edited:
Well, I think we have the answer to your problem. Your processor core temps are extremely high. Dangerously high in fact, if the calibration of the core temp sensors is anywhere close to accurate. AMD says max safe temps for those CPUs is around 70c. We also know in the overclocking community that instability typically sets in at temps somewhat lower than the max safe temp threshold.

If I were you I would not continue to use that computer for anything remotely stressful until you get this temp problem solved. We'll try to help you with that.

It is likely that something is wrong with the way the cooler is fastened or with how the TIM (Thermal Interface Material; the paste between the cooler base and the CPU) is applied.

1. Did you apply any TIM at all? Hope you aren't insulted by this question. We have actually seen this mistake made on the forum before.
2. How much TIM did you apply and what technique did you use to apply it? The most common mistake noobs make is to apply way too much TIM which actually has an insulating effect.
3. Is the cooler making good contact with the CPU? Is there good pressure there? If you grab the top of the cooler and gently try to move it back and forth do you feel rocking at the base? There are screws and nuts that hold the cooler on and that need to be tightened if contact/pressure is poor.

We also need to look at the case ventilation and how the four fans are turned with respect to each other for intake/exhaust and overall exchange of air in the case. But first, please interact with my questions above. While we are on the subject of case ventilation, that Theraltake V4 case is only 7.5" wide. Does your Coolermaster Hyper 212 fit in it with the side panel in place? Usually, to fit a 120mm fan-based tower style cooler in a case the case needs to be about 8" wide.
 
Last edited:
Ouch, don't be discouraged, but those core temps are cringe-worthy!

For now I'd say try and rearrange your fans to get as much air as possible on the CPU until you can take apart the CPU Cooler and examine it. If you have side fan spaces and there's enough room inside to mount fans there, do so. Even if you have to pull the side panel off and point a room fan at the insides, that would be better than nothing. That will give you at least a bit of a buffer against burning things out, but as trents said DON'T do anything stressful until you can get this solved.
 
@trents:

I have just removed the CPU Cooler, cleaned the CPU and the Heat-sink, applied TIM again (the one that comes with the Hyper 212) as recommended in HERE (Small Dot) and re installed.

Also I have added a 2nd fan to the CPU Cooler.

Here is a pic of my setup:
MyPC.PNG

you can see I have:
-2x120mm fan as Intake: 1 in front, 1 in the bottom.
-2x120mm fan as Exhaust: 1 in top, 1 in the rear.
-2x120mm fan in the Cooler Master Hyper 212, both blowing towards the rear exhaust.

I dont know what else I can try to fix this issue. Any idea is welcome.

and thanks for your response.

Regards,
Yatiac
 
If I were you I would flip the top fan so that it's an intake fan, right now it looks like the top fan is pulling air away from the cooler. If you make it an intake, it will feed cool air from outside the case right into the cooler.

Either way, now that you have re-seated the cooler, you are gonna want to run the stress test again, but keep a close eye on it, if you hit 70C or so, shut off the test. Ideally, you want to run it for at least 20 minutes, more if possible, and you want the core temps to never go above 65C, but more ideally, you want the temps to stay between 55C-60C or lower.
 
OH MY GOD YOUR CPU IS ON FIRE!!!! I HAVE NEVER SEEN SUCH INSANE TEMPS!!!! I CAN'T BELIEVE YOUR MOTHERBOARD ISN'T BLACK AND MELTED!
 
yatiac, that air flow scheme you show in post #9 is correct. But one question I asked earlier I don't believe you answered. Is your case wide enough to allow you to put the side panel on with that tall cooler? If you are running an open-sided case your air flow will not be effective, especially for cooling the motherboard components. And if your side panel is putting pressure on the top of the cooler it is likely deflecting it to the side so that it pulls the heat sink base partially off of the CPU such that there is poor contact.

I also suggest a different method of applying the TIM than what the cooler manufacturers suggest and I find it works better with many of the exposed heat pipe coolers like yours because the ridges and valleys don't block the spread of the TIM as often happens with the "pea-sized blob in the center" method. I think spreading two very thin lines of paste across the base at right angles to the pipes works better for that reason.

And as someone else said, you need to rerun the Prime95 blend 20 minute test and recheck core temps.
 

Attachments

  • TwoLineMethod.JPG
    TwoLineMethod.JPG
    46.4 KB · Views: 301
Ah, sorry, missed the bottom fan, your fan layout is just fine, no need to switch the top fan.

As trents said make sure you can put the side panel on without affecting the cooler, otherwise you may be in the market for a wider case or a smaller cooler.
 
I would also suggest finding something to do with that mess of wires over your bottom fan. I'd say it's probably blocking 90% of your air intake.
 
@trents:

I have just removed the CPU Cooler, cleaned the CPU and the Heat-sink, applied TIM again (the one that comes with the Hyper 212) as recommended in HERE (Small Dot) and re installed.

Also I have added a 2nd fan to the CPU Cooler.

Here is a pic of my setup:
View attachment 110012

you can see I have:
-2x120mm fan as Intake: 1 in front, 1 in the bottom.
-2x120mm fan as Exhaust: 1 in top, 1 in the rear.
-2x120mm fan in the Cooler Master Hyper 212, both blowing towards the rear exhaust.

I dont know what else I can try to fix this issue. Any idea is welcome.

and thanks for your response.

Regards,
Yatiac
Man, You have a motherboard tray with a SPACE behind it. I'd be willing to bet you have a removable side panel on that side too. You need to clean up that garbage clutter all over your case; It's HORRIBLE! it also, to pertain to this thread, would improve airflow. A LOT. Now, if someone decides to go look at pics of MY case, My excuse is that my case is Dell and therefore Sucks by way of any cable management and what you see is literally utilizing all of the cable management the case offers.
 
There's something fishy about those temps ! according to hwmonitor his skt temp is over 50c different than the core, plus I don't think a amd chip could hold up to almost 120c of heat, I have never seen such a high temp before. I vote for a busted temp sensor on the die itself.
 
There's something fishy about those temps ! according to hwmonitor his skt temp is over 50c different than the core, plus I don't think a amd chip could hold up to almost 120c of heat, I have never seen such a high temp before. I vote for a busted temp sensor on the die itself.

You're probably correct but even so with that cooler his socket temps should not reach 63 and have shutdown after only 6 minutes. Those temps are still way too high.
 
If the pipes on the cooler itself feel hot and the air getting blown out of the cooler is like hair dryer then it's time to worry, but yeah youre right that skt temp should be below 50c really, may be a load of static air round the skt area :shrug:
 
Thank you all for your responses.

taking the keny post, I do some research and read in toms hardware that SpeedFan is more accurate for AM2 sockets, so, as WoodenKiwi and trents said, I run again prime95 with both, speedfan and HWMonitor and It run for 11 minutes without shutting down, then I stoped the test because HWMonitor saids that my core#1 was at 110°.

Here's the last SS I took before stoping the test:
1.png

As you can see SpeedFan said one thing an HWMonitor another totally different. Now Idk what to believe.

Also I changed some settings at the bios setup like the "Cool n Quiet" feature and it looks like it help.


Please let me know your thoughts. And again, thank you for taking your time to help me with this issue.

Regards,
Yatiac.
 
Socket temps between SpeedFan and HWMonitor are identical. SpeedFan still shows the core temp as 89C and still way too high. So no matter the use of SpeedFan or HWMonitor they both show the core temp as too high for my feeling of safety long-term.
 
Back