T20 mystery shutdowns, cmos?

T2x/T3x series specific matters only
Post Reply
Message
Author
tokyo_x40
Posts: 35
Joined: Sun Sep 04, 2005 1:31 am

T20 mystery shutdowns, cmos?

#1 Post by tokyo_x40 » Fri Aug 14, 2009 1:16 am

Inherited a T20, delighted at the build quality, but sadly it won't stay on very long..

After a period of time between 20-60 min, screen goes black (even for memtest86+) and machine is unresponsive. Happens whether main battery is inserted or when under AC only. Two LEDs remain illuminated, the main power one, and the far left "tray area" indicator, best described as "right-arrow-X". System unresponsive to any keys, including the power button. Except: hitting the power button multiple times leads to what sounds like a "too many keystrokes" beep.

I noticed that the date was set to 1999; are the random shutdowns the result of a bad CMOS battery? (I always thought they got power/charged from the AC) Since screws near the CMOS battery bracket are missing and the CMOS battery is visible when ejecting the main battery, perhaps I'm not the first person to wonder that.

Any cheap DIY ways to fix this, or are IBM parts strictly required?

Thanks

sjthinkpader
Senior ThinkPadder
Senior ThinkPadder
Posts: 2908
Joined: Tue Jan 22, 2008 8:29 pm
Location: San Jose, CA

Re: T20 mystery shutdowns, cmos?

#2 Post by sjthinkpader » Fri Aug 14, 2009 1:36 am

Remove the fan and reapply heatsink compound. You may need CMOS battery as well.
T60p 2623-DDU/UXGA IPS/ATI V5200
T60 2623-DCU/SXGA+ IPS/ATI X1400
T43p 2668-H8U/UXGA IPS/ATI V3200
R50p 1832-NU1/UXGA IPS/ATI FireGL T2
X61t 7762-B6U dual touch IPS/64GB SSD
X32 2673-BU6/32GB SSD
755CDV 9545-GBK Transmissive Projection LCD

RealBlackStuff
Admin
Admin
Posts: 17485
Joined: Mon Sep 18, 2006 5:17 am
Location: Mt. Cobb, PA USA
Contact:

Re: T20 mystery shutdowns, cmos?

#3 Post by RealBlackStuff » Fri Aug 14, 2009 6:39 am

I agree with sjthinkpader on both accounts.
The T20 is most likely overheating, and has a near-dead CMOS battery (CR2032).
Lovely day for a Guinness! (The Real Black Stuff)

Check out The Boardroom for Parts, Mods and Other Services.

tokyo_x40
Posts: 35
Joined: Sun Sep 04, 2005 1:31 am

Re: T20 mystery shutdowns, cmos?

#4 Post by tokyo_x40 » Fri Aug 14, 2009 10:32 am

Thanks, sounds like I should try the heatsink compound first. My first time doing this, will google for instructions/brands
RealBlackStuff wrote:The T20 is most likely overheating, and has a near-dead CMOS battery (CR2032).
I read that battery leads are welded to the T20 oem cmos battery unit. Can anyone confirm this, or would a drugstore CR2032 be possible? I'd have to strip the yellow plastic heat-shrink to check mine, prefer to wait.

Neil
Senior ThinkPadder
Senior ThinkPadder
Posts: 2914
Joined: Sun Aug 07, 2005 5:41 pm
Location: Paragould AR USA

Re: T20 mystery shutdowns, cmos?

#5 Post by Neil » Fri Aug 14, 2009 10:40 am

Yes, the leads are soldered onto the battery under the yellow sleeve. A CR2032 will work if you can solder the leads to it. Or, you could take the leads off the old battery and tape them to the new one (contact will not be good though, best to solder).

A proper replacement battery won't cost you any more than the drug store one, it will just take a few more days to get it.
Collection = T500 - R400 - X300 - X200 - T61 (14" WXGA+) - T61 (14.1" SXGA+) - T60 (15" SXGA+) - X40 - T43p - T43 - T42p - A30P - 600E

RealBlackStuff
Admin
Admin
Posts: 17485
Joined: Mon Sep 18, 2006 5:17 am
Location: Mt. Cobb, PA USA
Contact:

Re: T20 mystery shutdowns, cmos?

#6 Post by RealBlackStuff » Fri Aug 14, 2009 12:38 pm

If you set the date, and at the next cold boot the date is back to 1996, then your CMOS battery is definitely dead.
You can NOT solder contacts onto those CR2032 batteries, but taping will do if you can't get an original. Like Neil said, get a complete one (off e.g. eBay).
Lovely day for a Guinness! (The Real Black Stuff)

Check out The Boardroom for Parts, Mods and Other Services.

Robbyrobot
Senior Member
Senior Member
Posts: 573
Joined: Tue Jun 26, 2007 7:46 am
Location: Berlin, Germany

Re: T20 mystery shutdowns, cmos?

#7 Post by Robbyrobot » Thu Aug 20, 2009 2:46 am

@Neil:
A proper replacement battery won't cost you any more than the drug store one
In that case, your drugstore must be a very expensive place indeed. I generally replace the CMOS batteries by pulling off the welded - not soldered - contacts and taping them (with plastic electrical tape, pulling it tight) on a new battery. No problems to date with that, although some advise against it.

It's inadvisable to solder contacts on a lithium battery, since this often leads to nasty explosions. The original contacts are spot welded, and the heat is restricted to the contact and the surface of the battery. However, you need special equipment to do this yourself.

tokyo_x40
Posts: 35
Joined: Sun Sep 04, 2005 1:31 am

Re: T20 mystery shutdowns, cmos?

#8 Post by tokyo_x40 » Thu Aug 20, 2009 9:13 am

sjthinkpader wrote:Remove the fan and reapply heatsink compound.
New compound had almost no effect.

What does "almost" mean? Now after 40 minutes the machine powers off completely--it no longer hangs with the power LED and "arrow->X" LED stuck on. In case it matters, the previous heatsink compound was a grey solid rubber-like square pad, and I applied one diagonal line of standard cheap white "silicone base" heat sink compound. I kept the grey rubber thing, in case it would be better.

(I also noticed that the chassis screws for securing the keyboard are missing. I mention this in the off chance they are used for grounding/etc)

Is there a way to test the fan, in case it is bad? Is memtest86+ really that CPU intensive?

PS Thanks for the discussion on the CMOS battery, still haven't touched that yet. I'll likely try tape

Post Reply
  • Similar Topics
    Replies
    Views
    Last post

Return to “ThinkPad T2x & T3x Series”

Who is online

Users browsing this forum: No registered users and 1 guest