Another dead T23- done the usual

Message

seaweedsl · #1 Post by **seaweedsl** » Sun Apr 29, 2007 2:05 pm

Hi. I got me one of those dead T23s and am hoping for help. I've done a lot of research and believe that I've gone thru the usual steps, but maybe somebody has some insight for me.

Here's the run-down.

How it happened/what it does:

I'm prepping my T-23 to sell to a friend, cleaning it up, turn away for a minute, and it's froze up. Had it for over a year without problems.

But now, it's frozen- no keyboard, no pointer, no mouse. Ctrl-Alt-delete, nothing. Power off, nothing. Try hitting the function keys (any and all) to see if it will shake out of it. Nothing. Hold power off for a while, while holding escape key (I know, stupid, but should be harmless) and it makes an unusual error sound, like a Euro siren, up-down, up-down, up-down, a few times, but then I can't get it to repeat that. Never heard that sound in a pc before.
OK. Fine, it's time to pull the plug. I pull power, pull the battery and it finally goes off, of course. Though I was beginning to wonder!

I go to turn it back on and the power light comes on, the speaker make their dull click sound, the bank of indicators come on and the optical spins for a moment, just like it's booting, then it does nothing. All lights go out but the power light. No boot, no post, no error beeps nothing.

The fan is running though. Steadily.

I pull the CMOS battery as well, and hold down the power button for 20 seconds with it out, then reinstall and then try again. No good.

Background:

Lith-ion is practically dead, maybe runs the laptop for 5-10 minutes. CMOS gives 3 volts.

It came with a non-IBM power supply, an HP adapter that puts out 19 volts. The guy who sold it to me specializes in Thinkpads and said that this is fine and in fact it has been for a year. Is it possible that it has slowly degraded internal power components?

Laptop has been mostly sitting idle for the last 3 months. I took it out a week ago and ran it through it's paces to show it to the buyer. All was great.

What I've done so far:

Power/clearing capacitor etc: All variations of power via AC vs battery. Pulled batteries, CMOS etc. Let it sit for a week with no batteries including CMOS. Many variations of pulsing the power button with and without juice. Holding it down for minutes. Etc.
Flexing in every manner as per Ray's advice.

Tested CMOS battery for 3V, swapped in a known good IBM power supply (see above note about 19v power supply)

Pulled memory, mini pci, hard drive, optical and retested.

Disassembled laptop completely to check the inductors and reseat the CPU. After some light tapping, one of the inductors came loose and I re-soldered it with anticipation. Not the problem.

After all this, it's still the same, it lights up, acts very much like it's going to post, then doesn't. Hearing the click from the speakers really makes me think it's close, but no response after weeks of fooling around with it. One minute it was perfect, the next it was dead.
Is there anything else to try? I have not connected an ext monitor as I don't have one at hand, but there is no post, no HD activity, I doubt it's just the screen.....

Thanks,
Steve

#2 Post by **rkawakami** » Sun Apr 29, 2007 2:31 pm

Sounds like you did everything I would do!

Have you checked the inductor located just above the CPU? On my "map" of the top side of the T23 motherboard, it's located at A1/A2. You won't see it in my picture since I had to use it to fix another system. If that one and the two on the bottom side are firmly attached, then it's time to look at the entire board with a magnifying glass. I recently had a problem with a T23 that was due to a broken connection on a crystal, although that system could be powered up if twisted in just the right way. Yours may have a similar problem but the part might have detached itself too far for the board flexation to make contact.

edit: Corrected dsylexic spelling

Andersonjoe711 · #3 Post by **Andersonjoe711** » Mon Apr 30, 2007 12:19 am

*Tech Guys, close your eyes***

I'm setting up a T23 for my girlfriend, and Just got a new hard drive for it, and I was having that problem before and after i replaced the Hard drive. I'd power it on, and everything would go blank, and it wouldn't recognize that I had a drive in the ultrabay. I took it apart, and couldn't find anyting.

Well I did the same thing after putting a new hard drive in. and nothing!

So, I took out the new one, put mine in from my laptop, opened it up, powered it up, andf nothing. so I turned it off, and gave it a nice THWAP on the keyboard (

) and it turned on!

I was then able to repace the new drive in, and run it through recovery to set it up with XP.

Anyways, I know it's a bad thing to do, but it worked for me......for now. I'd like to find out what the actual problem is though.

#4 Post by **RealBlackStuff** » Mon Apr 30, 2007 2:23 am

Sounds very much like a loose keyboard connector.

seaweedsl · #5 Post by **seaweedsl** » Mon Apr 30, 2007 2:53 am

Ray, thanks for the suggestion. I guess I'll take one more shot at the motherboard, though I wonder how many components I can really expect to solder with a 30 watt iron and my skills. I had to file down the tip to do a hack job on that inductor and that was blessedly big. Also, for lack of experience with circuit boards, I have no idea how much heat these components can take.

I did check that inductor by the CPU, but did not check those crystals, even though I had read your post. Which side of the motherboard are they on? Thanks for the maps, by the way, great service to others.

Somehow, in this case, I don't think it's a solder joint. Why? Lack of response to flexing, instant problem - one minute good, next minute bad.

Before I start hacking up this motherboard, one of the two additional possiblities that I'm wondering about is voltage regulator components inside: Again I ask, is it possible that a year of 19volts has fried something at that level? I really don't know anything about this....

Steve

seaweedsl · #6 Post by **seaweedsl** » Mon Apr 30, 2007 3:12 am

RealBlackStuff wrote:Sounds very much like a loose keyboard connector.

Aha! Another lead! Thanks.

OK, this reminds me of two things I forgot/failed to disclose:

1) when I went to pull the battery after the initial freeze-up, I stupidly removed what turned out to be a keyboard screw. The battery was hard to take out, and I thought maybe there was a screw. Then I put it back in, but remember thinking something was odd when it went in. Can't say what, but my mechanic's sense of screws felt ...something.

Is it possible that first it froze for typical OS reasons, then when I went to pull the battery, I "screwed-up" the keyboard connection with that screw?

2) Once the keyboard was off, I turned it upside down for the next step and a few pieces fell out. One was a loose plastic piece that I identified as a bracket/frame or support for the keyboard connector. It was cracked in two. I glued it together and put it back the way it seemed to go and the keyboard seemed to install correctly without a hitch.

Also, I found that the keyboard ...bezel? was cracked in two weak spots as well when I went to pull it. I realized that somebody had previously been inside this and had cracked a few plastic pieces in the process. It was not a problem and I figured that the broken plastic keyboard connector support frame was related to this.

Knowing how problems go, it could be that they were unrelated and I caused the keyboad part to break when I went to pull the battery.

Hmm. I am going to look very carefully at that keyboard connector jack...maybe I broke it there...

#7 Post by **rkawakami** » Mon Apr 30, 2007 3:58 am

seaweedsl wrote:I did check that inductor by the CPU, but did not check those crystals, even though I had read your post. Which side of the motherboard are they on?

The three crystals that I found are on the "top" side of the motherboard at I2/I3 (marked 24.576), I4 (marked 10.000), and the one that was loose on my board, J5/J6 (marked 32.768k).

seaweedsl wrote:Again I ask, is it possible that a year of 19volts has fried something at that level?

It's a remote possibility. The IBM adapter provides about 16.8VDC (no load). I haven't looked too closely at the few datasheets on the power regulators that I found so I don't know what the maximum input voltage is. There's several fuses on the motherboard. It's possible that one of them has blown. It's more likely that one of the active components (IC, transistor, MOSFET, etc.) has given up its life to protect the fuse (!) if you stressed the system with 19VDC.

re: broken keyboard piece and bezel

The black plastic surrounding the keyboard connector on the motherboard is designed to slide right off the connector. My guess is that it's there to provide alignment with the matching connector on the keyboard. The keyboard bezel, if it's broken at the thinnest points on the right (next to Ultrabay opening) and left (just above PCMCIA slots) are known weak spots. The right side tends to crack if the mounting tab on the Ultrabay guide rail breaks off. That piece is attached to the back side of the laptop and generally separates from the rest of the Ultrabay guide rail because of the lid opening and closing. Tight hinges can put lots of stress at the back right corner. The same is somewhat true for the left side of the keyboard bezel, although from my experience, that side does not crack as often as the right.

edit: There is only one screw near the keyboard connector. It's one that's holding the ethernet daughter card onto the motherboard (closest to the MiniPCI mounting clip). There should be a standoff mounted in the hole of the motherboard. If it is missing, then you could be putting too much pressure on the ethernet card, causing it to wrap.

seaweedsl · #8 Post by **seaweedsl** » Mon Apr 30, 2007 8:52 pm

Thanks again Ray for being so generous with your time and knowledge !

I'm home and about to strip the laptop down again. Should I be checking for blown fuses? I gather that the lights wouldn't come on if they were blown, but maybe it's not so straightforward.

I didn't notice them before, where are they? Or should I not even bother....

Steve

#9 Post by **rkawakami** » Mon Apr 30, 2007 9:55 pm

There may be over a dozen, if the motherboard part legend is any guide. I've found one marked "F18" on the back side just "below" the audio jacks. The main one is marked "F2" and is next to the white connector that the DC jack plugs into. There are big silver-ended ones next to the two battery terminals (main battery and Ultrabay). They are marked "F11" and "F12". There are also two smaller ones near those spots. Those four are probably okay as long as you can power the system off a battery and the Ultrabay device(s) are working.

In addition, there are what appear to be thermal fuses in a couple of spots (thin, green-colored, rectangular "chips"). I would check them only since you are going to have the board out of the system again.

seaweedsl · #10 Post by **seaweedsl** » Tue May 01, 2007 1:09 pm

Well, I got it back together this morning and no luck. In fact, it's down for good.

I went over all the crystals and all the inductors with my 30 watt and a magnifying glass and put it back together.

Nothing. Now there is no lights coming on, nothing. I double checked all the little connectors and could find nothng disconnected, so I suspect that I overheated something in the process.

Oh well, 6 hours of trouble and still have a broken T23. I'll part it out on Ebay when I get to the US in the Fall. It will be worth most of the $250 I paid for it a year ago...

Thanks for the help anyway, I appreciate it!

Steve

#11 Post by **rkawakami** » Tue May 01, 2007 2:25 pm

Strange that you see NO indication of life (LED, sounds, display, etc.). Is this even with the IBM AC adpater plugged in, a main battery installed and the system not turned on? You should at least get a green/orange LED on the right hinge. If not, then something must have happened to the power controller chip(s), or the main fuse (F2) blew, or the DC power jack is not plugged into the motherboard.

If you have a port replicator/docking station, then you can try mounting the system in there and use the power button on the port rep/dock to turn on the system. That bypasses the DC jack AND the keyboard. In other words, if there's a problem with those two pieces, the port rep/dock will still be able to fire up the laptop.

seaweedsl · #12 Post by **seaweedsl** » Tue May 01, 2007 10:25 pm

I know, it's hard to believe nothing is lighting up now.

I tested that f2 main fuse when in there and already double-checked all the connections (from power supply,fan, keyboard, assted others) EXCEPT the motherboard's bottom connector, cause it's so far in. I did assemble carefully for that connection, but I COULD go back in and double-check that and the power jack.

It looks like the power jack has no soldered connections (good design) and thus the wire coming out of it into the motherboard is the only place to fail, right? That connector was the first thing I double checked when it wouldn't light up.

I hate to do a 3rd r&r on it, but maybe I should before writing it off. I suspect that the best case is that it will return to it's previous state of failure.

Maybe in a couple of days...

Steve

phool@round · #13 Post by **phool@round** » Wed May 09, 2007 11:08 pm

Try touching up the biggest coil with the iron even if it looks good. You mentioned that coil but didn't say if you reflowed it. You may have to go over the other coils again too.

Tip for tips;

This is what I did when I was a kid and didn't have the money for a real rework iron.

If you can find a mechical hand crank pencil sharpener you can shape your tips with it. Makes them long, very fine pointed and great for detail work. If your tip is copper, as most general use irons are, it's soft enough that it won't hurt the sharpener if your gentle. YMMV.

I use a 19v brick with several far lower rated laptops and haven't had any issues so far. Probably not the cause, more likely the power regulation of the board.

jamiphar · #14 Post by **jamiphar** » Thu May 10, 2007 2:02 am

phool@round wrote:...Tip for tips;

This is what I did when I was a kid and didn't have the money for a real rework iron.

If you can find a mechical hand crank pencil sharpener you can shape your tips with it. Makes them long, very fine pointed and great for detail work. If your tip is copper, as most general use irons are, it's soft enough that it won't hurt the sharpener if your gentle. YMMV....

Another way is to put your iron tip in a drill chuck and lathe it with a file, works great for me.

seaweedsl · #15 Post by **seaweedsl** » Thu May 10, 2007 10:58 am

"Try touching up the biggest coil with the iron even if it looks good. You mentioned that coil but didn't say if you reflowed it. You may have to go over the other coils again too."

Do you mean the big one by the processor? Tell the truth, I could never get that one to re-flow.

It seemed like the coil was just taking all the heat. It seemed to want a higher wattage iron in that case. I really wondered if I didn't damage it, melt something. I know it sounds like a hack job: it was.

Maybe I should have used one of those heat sink clips or what?

"Probably not the cause, more likely the power regulation of the board."

Exactly what I thought.

Only now I fried that inductor as well, and that compounded the problem. Or perhaps I fried a crystal, though they all re-flowed quite readily.

AbsoluteRaleigh · #16 Post by **AbsoluteRaleigh** » Thu May 10, 2007 8:59 pm

It is kind of rare but the CPU could have been bad- that would have caused the same initial symptoms that you described.

phool@round · #17 Post by **phool@round** » Wed May 16, 2007 9:43 pm

The big coil above the processor, yes even if it looks good. Yes, I always reflow it and all the others when I sell or just fix one for me. It's insurance against the future. I've repaired enough of them and it's a particular fault that is common. It's just my advice, it's completely up to you.

That big coil needs no more heat, 30W is more than plenty if not a little too hot actually. As a matter of fact it only needs a quick on-hold-off to reflow the solder. The mount/connection pads are not that big.

Well if you held the iron on the pads too long heat from the proximity of the iron could have easily transfered to the laquered wire.

I don't think you need a heat sink with the 30W.

Another dead T23- done the usual

Another dead T23- done the usual

Who is online