T21 not turning on

Message

dave111 · #1 Post by **dave111** » Wed Oct 04, 2006 5:20 pm

Gave my dad a T21 that was in good working order a few weeks ago.

Now he tells me that it won't turn on.

I have told him to take battery out and check power supply.

The Green power light shows when the power lead is connected but the power button does not do anything.

He is not using this with a docking station.

Any ideas would be most appreciated.

Thanks.

ajaslow · #2 Post by **ajaslow** » Thu Oct 05, 2006 12:46 pm

- Disconnect from power outlet
- Remove batteries
- Hold down power button for 60 sec
- Press and release power button about 10 times after that
- Put batteries back in
- Press power

dave111 · #3 Post by **dave111** » Thu Oct 05, 2006 2:43 pm

have passed the msg on.
will let u know if it works

thanks for trying

dave111 · #4 Post by **dave111** » Fri Oct 06, 2006 9:08 am

Amazing - that actually works.
Anyone tell me how?

Website · #5 Post by **tfflivemb2** » Fri Oct 06, 2006 10:09 am

It releases the residual charge in the capacitors, and works like a reset.

dave111 · #6 Post by **dave111** » Sat Oct 07, 2006 4:19 am

a big thanks.

dave111 · #7 Post by **dave111** » Mon Oct 09, 2006 4:05 am

Its Stopped Working Again!!!

Tried the above technique without success.

What else can we do?

#8 Post by **rkawakami** » Mon Oct 09, 2006 12:02 pm

Is the nature of the problem such that when you press the power button, the hard disk drive LED (the one on the far left, away from the power button) flashes? No other signs of operation? No beeps, no fan, no display? If so, join the crowd as this is a known issue with the motherboard. I am currently trying to pinpoint the source of this problem on a T22 of mine. The laptop sometimes also powers up if you just leave it alone (battery removed) for a day or two. If you search the board for a problem like this you will find references to an "overcharged capacitor". I do not totally subscribe to that theory. At the moment I'm looking at a Maxim power controller chip as the root cause of this problem.

jruschme · #9 Post by **jruschme** » Mon Oct 09, 2006 3:06 pm

rkawakami wrote:Is the nature of the problem such that when you press the power button, the hard disk drive LED (the one on the far left, away from the power button) flashes? No other signs of operation? No beeps, no fan, no display? If so, join the crowd as this is a known issue with the motherboard. I am currently trying to pinpoint the source of this problem on a T22 of mine.

If anyone else wants to experiment, I have a just-pulled T20 board which exhibits this problem and is free for the taking.

vlyne · #10 Post by **vlyne** » Tue Oct 10, 2006 2:17 am

I've read a number of these posts in search for a cure and hopes turn to despair when one reads of someone trying to sell 60 dead T20 motherboards on Ebay for $1 each! Obviously not an easy problem. One potential problem that a colleague has come across is with fading firmware memory (in cameras and other equipment with supposedly non-volatile memory). Those who can get their T2* to work, after sorting out the usual suspects, should consider a firmware update/refresh as a first order priority.

Good Luck!

dave111 · #11 Post by **dave111** » Tue Oct 10, 2006 3:42 am

Vlyne,

Do you mean update the bios?

vlyne wrote:I've read a number of these posts in search for a cure and hopes turn to despair when one reads of someone trying to sell 60 dead T20 motherboards on Ebay for $1 each! Obviously not an easy problem. One potential problem that a colleague has come across is with fading firmware memory (in cameras and other equipment with supposedly non-volatile memory). Those who can get their T2* to work, after sorting out the usual suspects, should consider a firmware update/refresh as a first order priority.

Good Luck!

#12 Post by **rkawakami** » Tue Oct 10, 2006 4:11 am

If Vlyne is referring to a BIOS/embedded controller update, then, no, I don't think that will work.

The problem with those 60 dead T20 motherboards and my T22 is that it simply won't power on. You cannot update the BIOS/controller on a system that will not stay on for more than 1/100th of a second. And even when the finicky systems finally do decide to power up, I don't believe that it was "fixed" by a bit that changed states.

I'm aware of non-volatile memory being volatile. Flash memory, and it's precursor, EEPROM have known lifespans of data storage. Kept in ideal conditions data can remain unchanged for decades in both flash and EEPROM cells. However, subject them to extreme temperature and that storage time can degrade into days or hours. Sure there can be "weak" cells, ones that don't have the similar storage profile as the rest of their neighbors, but that's true for any kind of memory. In order to correct a situation like that you would have to physically remove the chip from the motherboard and replace it with a new one. That requires tools that most people don't have (SMT re-work station to be precise), thus the "easier" solution of replacing the entire motherboard.

My gut feeling for this power on problem is that it is something that can be simply explained: a bad solder joint heads my short list. With almost every component on the motherboard using surface mount technology, preparation before the solder re-flow process is critical. Any contamination on the leads of the components or not enough solder paste put down by assembly equipment can mean problems later on in the life of the motherboard.

I am still studying the workings of that Maxim chip I referenced earlier, in between actually trying to do the job I'm paid to do at work

.

vlyne · #13 Post by **vlyne** » Tue Oct 10, 2006 7:16 am

"dave111" Vlyne,

Do you mean update the bios?

Yes. And, of course, as Ray points out you have to get your laptop to work first! I thought I said that in my post? If the fading bios also controls startup functions then your laptop may not start (or else, if you're lucky, return a CRC error). Your options are then: replace the motherboard (easiest option), replace the EEPROM (difficult - been there, done that, and destroyed a motherboard in the process), or reprogram it in-situ using a clip-on tool that fits over the top of the chip. I suspect that a "fading" EEPROM will behave somewhat erratically (will start sometimes) perhaps in an analogous way as a fading CMOS battery loses the setup info.

I agree with Ray that the non-start problem could be a dry joint. It could also be a heat-affected component located close to the CPU. There are a number of fuses as well, including what looks like thermal fuses, but they don't appear to be the culprit in a couple of non-start T2's I have (interestingly, some power does seem to be going to the battery terminals, despite the non-start). Capacitors could be a problem but they're not easy to test in-situ - and there's lots of them. The fact that some have managed to get their machines to work for a while would suggest dry joints or overheated components as the most likely culprits. I'm not as certain as Ray that fading firmware is not a potential issue, although I'm quite certain your laptop won't start if the firmware has indeed lost its marbles. I'm certainly losing mine trying to figure out what's going on!!
Cheers

dave111 · #14 Post by **dave111** » Tue Oct 10, 2006 11:03 am

I think my dad kept the laptop plugged in for about 1 month with only minimal usage.

thinking about changing the cmos battery as it could be a quick win.

#15 Post by **rkawakami** » Tue Oct 10, 2006 3:52 pm

vlyne wrote:'m certainly losing mine trying to figure out what's going on!!
Cheers

LOL, agreed! This issue seems to have been driving a lot of people buggy from the research I've done.

With the limited example I have personally got my hands on (one T22), and the postings on this site and others, the symptom of just a flashing hard drive light (no matter how long the power button is held down), seems to me to be a basic issue of something in the laptop not delivering power to the rest of its buddies. If, for example, the hard drive spun up, but the fan didn't, nor did the LCD backlight fire, then, yes, it could be an issue of some firmware on the board not sending out a command or two.

I'm about at the point of borrowing one of the extra portable Tek o'scopes here at work and putting it to some good use. Wish I could take the logic analyzer home, but I think that someone would notice it missing

.

vlyne · #16 Post by **vlyne** » Tue Oct 10, 2006 8:32 pm

dave111 wrote:I think my dad kept the laptop plugged in for about 1 month with only minimal usage.

thinking about changing the cmos battery as it could be a quick win.

Certainly worth checking the voltage and connection. Replace only if necessary otherwise you'll be chasing fairies forever.

vlyne · #17 Post by **vlyne** » Tue Oct 10, 2006 9:14 pm

the symptom of just a flashing hard drive light (no matter how long the power button is held down), seems to me to be a basic issue of something in the laptop not delivering power to the rest of its buddies. If, for example, the hard drive spun up, but the fan didn't, nor did the LCD backlight fire, then, yes, it could be an issue of some firmware on the board not sending out a command or two.

In my case the LED flash occurs only when connecting up the AC power supply. Nothing visible happens when pressing the power button. A high pitch sound comes from the vicinity of the CPU.

I'm about at the point of borrowing one of the extra portable Tek o'scopes here at work and putting it to some good use. Wish I could take the logic analyzer home, but I think that someone would notice it missing .

I'm curious about your plan. I'm assuming you'll have a dead and a good machine opened up on the operating table? Are you planning to test voltages/signals on the dead and the good before power-up? Once the good one is powered up there's really little hope of tracing back to the source. Tracing forward from the power button may also be difficult? The usual procedure is to swap components but in the case of the motherboard there's only the CPU and the rest of the board. As a start, I'd probably at least do the CPU swapping just to verfify that it is the MB by a process of elimination.

I can just picture the scene of Dr Ray in his dungeon sticking probes into the innards of a couple of thinkpads. "Tell those men in white coats I'll be with them in a minute! [censored] it where was I!? Is this the good one or the bad? Ahha! Hey Lurch, I found it! I found it! A speck of dust! Yes, that MUST be it! Try the powerbutton Lurch. [censored]....!!" And so, that was the last operation performed by Dr Ray who is now confined never to see another thinkpad again, or flashing LEDs. Lurch however continues his quest...

dave111 · #18 Post by **dave111** » Mon Oct 16, 2006 4:01 am

Thinking of changing the cmos battery.

on ebay.co.uk they are selling for about £8 (which converts to $15) seems a bit pricey.

any ideas where i can 1 cheaper.

dave111 · #19 Post by **dave111** » Fri Oct 20, 2006 6:40 am

Tried a new power supply with no success.

vlyne · #20 Post by **vlyne** » Mon Oct 30, 2006 9:01 am

Ok, it's time to get back to basics on this problem. The recommended procedure is to first strip the machine bare and see if that get's it going with just A/C power. Assuming it doesn't, test the keyboard by subtituting with a known good one, or else via a docking station. Still no go? Check the lid switch. Given that you get a power-on LED with the AC adapter plugged in I'm assuming the power pack is OK.

Still no go?? It's time to enter the dungeon and break out the witchcraft tools. Is the brain screwed on OK? Extract the CPU, smear a bit of that magic Arctic Silver, carefully refit and screw back on. Still no go??? This is where the marbles start falling off. Welcome to the world of the dead!

Lots of fairies flying here!! Check for signs of blown capacitors, and test all fuses (flat ones, temperature fuses, ceramics, small ones with single characters...) and dry joints around major power components especially on the underside around CPU (large round inductors). If you have a steady soldering hand you can try to resolder components located near the CPU - guaranteed to make your hands unsteady! OK, enough is enough! If you got this far you need serious counselling help! Time for some economic rationalism. What is your time worth in relation to replacing the MB? Still not giving up? All I can say is I hope Dr Ray does come back with some good news. It's past time for me to take my medication...
Cheers

Ag99JYD · #21 Post by **Ag99JYD** » Tue Oct 31, 2006 7:06 pm

Ok, so I have a T-20 that is acting up in the same manner - just random shut downs. For lack of a better analogy, it acts like a car with a bad starter - after it dies, it wants to come back, but just can't. Alot of clicking sounds, and some really wierd screen activity, then BAM - right back.

If I leave it for a few days, i can get on for a few minutes, then right back to shutting down. I tried the capacitor thing - no luck, at least not long term.

It seems that this is a mystery to everyone - is that a safe assumption? What I'm wondering is if it is still hapening if the motherboard is changed, and perhaps some re-soddering as mentioned before, then maybe some firmware updates. Has anyone tried all?

It was asked if all the trouble is worth the time. Well, for me, the laptop was free...so, I have very little invested. With that said, would changing the motherboard be any help whatsoever? Possibly that, adding some thermal paste (I think it's called artic something or other - form Radioshack) and updating the firmware?

PS - I tried it again after posting the above. Interestingly, I noticed that it acts very different depending on if plugged in or not. The latest trial was unplugged. It started to boot, showed the initial screen (where you can choose to go to BIOS set-up...) then died. After that it tried to re-boot, and the lights just went bezerk. In contrast, when plugged in, the lights do not blink as much - it just keeps restarting. I'm not sure if this makes sense or is any help - but here it is.

vlyne · #22 Post by **vlyne** » Wed Nov 01, 2006 12:41 am

It's a classic symptom of a T-2* daemon but not so bad in your case as the machine does boot up.

A few things to check:

1. Did you trying booting up from just the A/C alone - any difference?
2. Wriggle the switch for the top cover/LCD (RHS above keyboard from memory) - any difference?
3. Does it make any difference if you boot up when the laptop is cold or not? In other words, does the erractic behaviour occur only after the machine has been up for a while (from a cold start)? So, once it's erractic it remains erratic? If so, it could be a thermal problem. Time to get out the CPU and apply that paste.
4. While you're at it check the CMOS connection and voltage.

If all the above fails, you either have to give in to the deamon or else go through the standard ritual - stripping off all cards etc. In this case, most likely a cold solder joint or failing joint.
Good Luck!

Ag99JYD · #23 Post by **Ag99JYD** » Fri Nov 10, 2006 1:38 am

Ok - I know at this point the horse is dead...

I just can't seem to get this thing working. It's not a thermal issue, as now it won't boot at all.

So here's the deal - I really, REALLY need this thing working - quick. So, what if I just replaced the motherboard? Like this one on ebay - link

My thoughts...since the thing is really old, replacing the mobo can't be that bad of an idea, but I don't want to spend the time and money (mostly money) if this will happen with a new mobo. (And of course, while I have it open, check the connections)

Thoughts...opinions...?

Thanks.

MODERATOR EDIT: Fixed link for better viewing.

#24 Post by **rkawakami** » Fri Nov 10, 2006 2:01 am

1) When inserting a long URL it's better to use the {url=http://www.blahblah.com}Short description{/url} construction as this creates a link which does not cause people to have to scroll left-to-right. Replace the "{" with "[" and "}" with "]" (without the double quotes)

2) The motherboard appears to be exactly what you need. You will of course have to transfer your existing LCD and other components onto the new board. Best you have a copy of the Hardware Maintenance Manual handy.

3) As it appears that eBay seller offers a very short DOA guarantee (24 hours), you should be prepared to get your pieces onto that base ASAP in case there is problems.

4) For a good view into the negative/neutral feedback profile of that eBay seller, go to this site:

http://www.toolhaus.org/

and plug in their eBay ID into the first box, then click the "Received By" button.

5) Good Luck!

vlyne · #25 Post by **vlyne** » Sat Nov 11, 2006 12:44 am

A number of recent posts on problems with T30's are suggesting a possible issue with the CMOS. To see if your problem is related, try the solution offered previously by ajaslow:

ajaslow wrote:
Disconnect from power outlet
- Remove batteries
- Hold down power button for 60 sec
- Press and release power button about 10 times after that
- Put batteries back in
- Press power

But, this time take out the CMOS battery as well in the remove batteries task and don't put it back in.

Let us know what you find.
Cheers

Ag99JYD · #26 Post by **Ag99JYD** » Mon Nov 13, 2006 12:58 am

Interestingly enough, something different actually happened...for a while.

1st time, it started to spin up, gave me three error messages (one about DATE/TIME, one about CMOS something) then went straight to BIOS config. I checked the Date and time, realized it was decidedly NOT 1980 and reset, F10 to exit. Then it did the restart thing again, where it just blinks on and off, went back to BIOS config one more time, then nothing.

I tried it again, this time not replacing either battery but plugging in the AC adapter, and after a second I got three beeps - a "D" in morse code - best way to explain with typed words. I put the battery back (not CMOS) and unplugged from adapter, got the same beeps again, then it just sat there.

Not sure if that can help diagnose, but happy to try. BTW, thanks for all the help and patience!

#27 Post by **rkawakami** » Mon Nov 13, 2006 1:48 am

Ag99JYD wrote:1st time, it started to spin up, gave me three error messages (one about DATE/TIME, one about CMOS something) then went straight to BIOS config.

That sounds like your CMOS (backup) battery was not connected properly or is dying/dead.

Ag99JYD wrote:I tried it again, this time not replacing either battery but plugging in the AC adapter, and after a second I got three beeps - a "D" in morse code - best way to explain with typed words.

That would be ONE long and TWO short beeps... (Had to look that up! I used to know my code back in the early '70s) From the Hardware Maintenance Manual that's a system board, LCD or memory problem. Usually a memory problem would cause an "S-S-E" beep sequence upon boot or a BSOD (blue screen of death) when running Windows. If those things have not happened, then I would say there's a problem on your motherboard. Could be a loose component or a general system power problem. However, it will cost you nothing to re-seat the memory module(s) and/or pull out one (if you have two), and try booting with just a single module. If that doesn't work, move that module to the other slot and try again. If THAT doesn't work, try the second module the same way.

vlyne · #28 Post by **vlyne** » Mon Nov 13, 2006 2:06 am

rkawakami wrote:That sounds like your CMOS (backup) battery was not connected properly or is dying/dead.

That was a planned disconnection Ray.

Yes, reseating/swapping the memory chips in the DIMM slots is worth doing. Clean the chip contacts with a rubber while you're at it. If that fails take off all the other removable devices (network/modem/CD/DVD/...) and try again.

Ray, why taking off the CMOS battery works (gets the system attempting to boot up) in some cases is still not clear but it seems to get a dead system sort of going again and hopefully there will be some symptoms that help diagnose what the real problem is.

#29 Post by **rkawakami** » Tue Nov 14, 2006 1:29 am

I don't exactly know... just some conjecture, but removing the CMOS battery could be enough to force a total reset of the static RAM (SRAM) or real-time clock (RTC). Assume that for some reason one or more bits in the SRAM has flipped. Removing power to that memory would cause the sytem to load defaults upon the next power-up. Same could be true for the RTC.

T21 not turning on

T21 not turning on

Fading Firmware?

Re: Fading Firmware?

Re: Fading Firmware?

Re: Fading Firmware?

Re: Fading Firmware?

Who is online