Take a look at our
ThinkPads.com HOME PAGE
For those who might want to contribute to the blog, start here: Editors Alley Topic
Then contact Bill with a Private Message

Request for input on new heat related problem

Performance, hardware, software, general buying and gaming discussion..
Post Reply
Message
Author
Tõnis
Junior Member
Junior Member
Posts: 303
Joined: Mon Aug 11, 2008 2:43 pm
Location: Central Falls, RI

Request for input on new heat related problem

#1 Post by Tõnis » Mon May 19, 2014 3:51 pm

I've replaced the fan on my R61 many times, whenever it has gotten noisy or failed. A few weeks ago I changed it again, because it had gotten loud and immediately started experiencing a problem I haven't seen before.

This is the first time that I've re-used the original heat sink assembly and replaced the actual muffin fan only. The fan runs, but it seems to run a bit more erratically than usual, and the computer has started going into standby and/or shutting down at various times including at startup. I've also noticed that the temperature displayed by my temperature program running in the tray has been consistently higher than normal and sometimes erratic. It hasn't been radically higher. Instead of running around 41C degrees on both cores, it has been running 50 to 60 degrees, sometimes as high as 70-80 degrees. I still don't think this should be high enough to cause the machine to shut down, but maybe it is.

Here are my fan/heat related variables:

1. Possible bad or improper fan
2. Improper amount of Arctic silver (I tried to apply an amount that, in appearance, matched what usually came on a fan/heatsink assembly whenever I ordered the complete unit, but maybe I used a little too much?)

Also, I recently installed BlackBerry Link which gives me remote access to the files on my computer from my smartphone, and that sometimes seems to be resource intensive, but I think that was working with no noticeable problems for about a week before I swapped the fan. The problem seems to have started with the fan replacement, and I have a new one on order.

Here's what I'm hoping to get the community's thoughts on. Is this problem likely caused by a bad or improper fan or an improper amount of arctic silver, or am I looking in the wrong direction and possibly facing a GPU (or different) problem altogether?

TIA,

Tony
R61, Core 2 Duo T8300 at 2.40GHz, 15.4" WXGA, XP Pro

theterminator93
ThinkPadder
ThinkPadder
Posts: 1451
Joined: Fri Oct 11, 2013 7:45 pm
Location: North Ridgeville, Ohio, United States
Contact:

Re: Request for input on new heat related problem

#2 Post by theterminator93 » Mon May 19, 2014 4:31 pm

I'd start by looking at core usage. If your CPU usage is low (<25%) and your temps are that high, I'd suspect the heatsink isn't making proper contact with the core. Use thermal compound about the size of a grain of rice on the core and use a criss-cross pattern to tighten the mounting screws.

If core usage is high, try to narrow down which application/service/daemon is using the resources.

What software are you using to monitor temps? If you're still running XP per your sig, try TPFanControl to max out the fan at setting 64 over 65 or 70C and see if that keeps core temps at bay. Can you feel a good amount of airflow near the heatsink with the fan running at full speed? Does the fan ramp up in RPM as temps go up?

One other thought... the R61 was produced during the nVidia chip flaw period. What's the manufacture date of your machine? It's possible your GPU is what's causing the crashes, not the CPU. If you are using TPFC to monitor temps, it will show the hottest sensor; if it's showing 70-80+ and GPU, that would certainly lead me to believe your GPU is on its way out - assuming an affected chip...
T480 with T25 keyboard | T25 | W520 i7-2860QM·Quadro 2000m·IPS FHD | T601F T9900·NVS 140m·LED AFFS UXGA
T420 IPS FHD | X220 IPS FHD | T61p·T61·43·42p|X13 Yoga G3·220T·301·41T·24·23·22|G41|A31p·22m|i1200|TransNote
600|770Z|770|760XD|760EL|701C|755C

Tõnis
Junior Member
Junior Member
Posts: 303
Joined: Mon Aug 11, 2008 2:43 pm
Location: Central Falls, RI

Re: Request for input on new heat related problem

#3 Post by Tõnis » Mon May 19, 2014 6:47 pm

Thanks for the great info, theterminator93. It gives me a lot to go on.

The temperature problem can happen even with the core usage low. I'm starting to think I used too much of Arctic Silver; I used way more than a grain of rice. I think I'll start by taking it apart again and getting rid of some of that compound. I read somewhere too much can be bad as well as too little. I just didn't have a good way to visualize how much I should use. A grain of rice is an excellent guide.

I'll have to check which temp program I'm using. It shows the temp for both cores in the tray. I tend to think it's not the classic graphics card problem; I have the built-in graphics, not the separate card. But I think one of the two parts of the heat sink assembly that have the Arctic Silver rests on the GPU, the other on the CPU.

I'll give it a try. Thanks again. I hope to report back with good news.

Tony
R61, Core 2 Duo T8300 at 2.40GHz, 15.4" WXGA, XP Pro

Cigarguy
ThinkPadder
ThinkPadder
Posts: 1756
Joined: Thu Aug 09, 2012 3:08 pm
Location: Calgary, Alberta, Canada

Re: Request for input on new heat related problem

#4 Post by Cigarguy » Mon May 19, 2014 7:26 pm

Yep sounds like improper application of thermal compound is the problem. When it comes to thermal compound less is more. I use the small grain/pea size single dot in the middle on everything that I apply. Almost is important is properly cleaning off the old thermal compound. I use 1) kleenex to remove the majority of old thermal compound, 2) use lint free cloth dampened with Isoproply Alcohol to do final clean, then 3) use compressed air to blow dust off and dry off the alchohol on the CPU/GPU/Chipset. Do the same thing on the heatsink fan (HSF). At this time check to make sure that HSF have not been damaged or bent out of shape during removal.

Here's a good guide. (Lots of other guides google "how to properly apply tim")
http://www.hardwaresecrets.com/article/ ... -Paste/274

Some useful free tools from the overclocking world, CPUZ, HWMonitor and Prime95. CPUZ is great for showing system settings and status. HWMonitor is great for monitoring temps and battery capacity. Core temp is what counts not so much system temp. Prime95 is used to stress test the CPU and memory. In the old days, a system is considered stable if it will run Prime95 for 8+ hours. For laptops I only bother testing for 3 hours.

After you take everything apart, reapply thermal compound, reassemble everything, run HWMonitor. CPU core temps should be in the 40ish - 50ish deg C range idle. Check CPUZ that the CPU is throttling properly. If you are within this range, then go ahead and stress test the CPU using Prime 95. At this stage CPU usage should be 100% most of the time and temp will shoot up from idle to 80ish deg C within seconds. The fan should be running at 100%. Thinkpad fan control is great tool to check the fan speed. On a properly cooled, stable, Intel machine, the system should be able to run and remain stable in this state almost indefinitely. Anything not right, the system will crash in minutes.

On my Thinkpads, idle is around 40ish deg C, stress is 70-80ish deg C depending on the machine (higher on my X61T than W500 for example) but all will run Prime95 stable for at least 5 hours. I find this is common for Intel machines, even on the desktop side. My desktop overclocked machines run 15-20 deg C cooler due to better case, air flow and HSF. When picking up an older machine I've notice 5-10+ deg C drop in temp after cleaning and reapplying thermal compound. On an Intel machine, high eighties deg C and you are in the automatic down throttling to save the CPU zone, anything in the nineties and a crash is imminent.

theterminator93
ThinkPadder
ThinkPadder
Posts: 1451
Joined: Fri Oct 11, 2013 7:45 pm
Location: North Ridgeville, Ohio, United States
Contact:

Re: Request for input on new heat related problem

#5 Post by theterminator93 » Mon May 19, 2014 7:50 pm

I'd verify the source of the temps too. While too much thermal compound isn't good, I have a hard time believing any amount of overapplication could provide the kind of thermal isolation necessary to let the CPU cores get so hot under light utilization. Especially when the heat sink is clamped down on the core and forces almost all the excess paste out past the sides of the core.

I'm not super familiar with the R series; what GPU does yours have? The fact that you've got a dedicated heatpipe leading to it makes me think it's at least a Quadro NVS 140m - if it's a mid-2008 build or prior it's one of the bad ones.

My T61p was made in 10/07 and has one of the bad nVidia chips (a Quadro FX570m). I used TPFC to keep the peak temps under 80 if at all possible - usually temps would peak low 70s. The original board lasted me up until a month ago when I replaced it with a slightly newer board; and at that it hadn't completely failed when I replaced it (it would BSOD if I stressed the GPU or if enough heat from the CPU, when running hard, bled to the GPU). So it's (crashing when things get hot) a definite symptom that the GPU is on its way out.

FWIW, my T60 idles around 50, peaks high 60s (this is with custom fan schedules). The T61p idles low-mid 50s and peaks mid 70s. I know it's time to clean the heat sinks when the temps are 3-5 degrees hotter at idle and the fan is one setting higher. :)
T480 with T25 keyboard | T25 | W520 i7-2860QM·Quadro 2000m·IPS FHD | T601F T9900·NVS 140m·LED AFFS UXGA
T420 IPS FHD | X220 IPS FHD | T61p·T61·43·42p|X13 Yoga G3·220T·301·41T·24·23·22|G41|A31p·22m|i1200|TransNote
600|770Z|770|760XD|760EL|701C|755C

Tõnis
Junior Member
Junior Member
Posts: 303
Joined: Mon Aug 11, 2008 2:43 pm
Location: Central Falls, RI

Re: Request for input on new heat related problem

#6 Post by Tõnis » Tue May 20, 2014 5:05 pm

Thanks again to both of you for your help, theterminator93 and Cigarguy. Here's an update.

Last night, like a butcher, I took the unit apart and tried to remove some of the compound. To be honest, it didn't even look like there was enough there -- it was that thin -- but I still removed some, put it back together, and tried to start the machine. It would start and shut down instantaneously. I took it apart again and wanted to put more Arctic Silver back, but I was out, and Radio Shack was closed.

You brought up something else I had overlooked. I had gotten so used to successfully changing fans that I had become sloppy. Every other time I always cleaned the old compound off with qtips and alcohol. Well, I hadn't done that this time. I had just wiped it clean with a paper towel and qtips.

Today, I cleaned it all with alcohol, applied new Arctic Silver, started the machine, and it seems to be running perfectly, nice and cool at 38 degrees on one core and 41 on the other. The only thing that looks a bit strange is Core Temp is reporting that my processor is running at 1.2 when normally at the "Maximum Performance" setting it says 2.4 all the time. Any thoughts on that?

Tony
R61, Core 2 Duo T8300 at 2.40GHz, 15.4" WXGA, XP Pro

Tõnis
Junior Member
Junior Member
Posts: 303
Joined: Mon Aug 11, 2008 2:43 pm
Location: Central Falls, RI

Re: Request for input on new heat related problem

#7 Post by Tõnis » Tue May 20, 2014 5:15 pm

Edit/Update:

Before anyone wastes any time on this, I did a reboot, and the clock speed is back to 2.4.
R61, Core 2 Duo T8300 at 2.40GHz, 15.4" WXGA, XP Pro

Cigarguy
ThinkPadder
ThinkPadder
Posts: 1756
Joined: Thu Aug 09, 2012 3:08 pm
Location: Calgary, Alberta, Canada

Re: Request for input on new heat related problem

#8 Post by Cigarguy » Tue May 20, 2014 6:20 pm

It's amazing what a proper TIM install will do. Your idle temps is good but idle temps means nothing. Time to stress test it.

Just because it have a Nvidia GPU that is prone to failure doesn't always mean it will fail. I bought 2 different T61 that the previous owners swear died due to the Nvidia problem. On the assumption that the MB is toast, I bought it for $40 each for the T9300 CPU, 4 GB RAM and WSXGA+ screen in great shape. 1st one was turning off everytime the middle keys on the keyboard was lightly press. This was caused by an incorrect screw placement shorting the MB. 2nd one kept crashing after 1-2 min of use, fixed by reapplying TIM which dried out. Both are fully functional machines that gets a lot of abuse. For the price I don't care. In the winter I'd leave one in my truck which dips down to -20ish deg C then turn it on, use it for an hour and switch it back off again. Talk about temperature delta. So far so good.

Post Reply
  • Similar Topics
    Replies
    Views
    Last post

Return to “Thinkpad - General HARDWARE/SOFTWARE questions”

Who is online

Users browsing this forum: No registered users and 42 guests