T20 Blink Of Death Circuit Questions

Message

#31 Post by **Harryc** » Tue Sep 25, 2007 8:50 pm

I can see it just fine Ray. I emailed it to you.

#32 Post by **rkawakami** » Tue Sep 25, 2007 8:55 pm

Thanks Harry! Our internet connection here at work has been wonky today... Picture received.

edit: If I'd have to guess, I'd say it's something with the components in the center of the picture (inductors and caps). I'll have to tear apart the BoD T22 I have and see. As I said earlier, I've only looked at the T21 motherboard before. Not sure what that one lone capacitor next to the docking port connector is for (far upper left in the picture). The three caps or resistors below the CPU socket doesn't look all that important. I'd be very surprised if any of the connectors (main power input, battery, HD and audio ports; in clockwise order from 9 o'clock) had anything to do with the BoD.

The main reason why I suspect the parts in the center is that they appear to be around some Analog Devices and Maxim chips (below and to the right of the keyboard connector). Most likely they are some of the power conditioning controllers like I've found on the T23 motherboard.

phool@round · #33 Post by **phool@round** » Tue Sep 25, 2007 9:55 pm

Jumping in here to back your assumption Ray, I think it's the middle components as well, right in the middle where the greatest flex occurs. I don't have a BOD to test and isolate which component I think it is but I have a good guess which one it is by the photo.

I like the cap's theory as well and notice those reflowed are solid state.....hmmmm. I wish I had a board to test what I'm thinking right now.

Timoti · #34 Post by **Timoti** » Wed Sep 26, 2007 5:39 am

Count me in for the all-night one-by-one component resoldering party on my BoD T22!

Also i just ordered a BoD T20 board for a few bucks, maybe it can be cured too...

#35 Post by **Harryc** » Wed Sep 26, 2007 5:51 am

Timoti wrote:Count me in for the all-night one-by-one component resoldering party on my BoD T22!
Also i just ordered a BoD T20 board for a few bucks, maybe it can be cured too...

LOL....this one thread single handedly drove the price for BOD T2X's on EBay up 50% ...

phool@round · #36 Post by **phool@round** » Wed Sep 26, 2007 9:28 am

LOL.........Bill Gates had better look out, he is going to slip a notch down the list. Didn't he begin his sharp accent by being in the OS/2 Forum?......... LOL!

Timoti · #37 Post by **Timoti** » Wed Sep 26, 2007 6:36 pm

Turns out i'm stupid...
My T22 wasn't BoD, it starts up with a power led, just won't show a splash and go any further. So much for my long term memory...

Anyways, i removed the 2 blue stickers that should protect the HDD out of curiosity, and found out that everything under them is corroded. The traces on the PCB turned brown and stuff. Any ideas if this can cause our problem?

#38 Post by **Harryc** » Wed Sep 26, 2007 6:41 pm

Timoti wrote: Any ideas if this can cause our problem?

Yes a defective hard drive would cause the machine to stop at the initial splash screen. Please start a new thread for this problem if you require additional assistance.

tom_k · #39 Post by **tom_k** » Wed Sep 26, 2007 10:36 pm

instantly stripped the blue pads from the "Other one":
(referring to my posting in tuesday):

i would call it slight to medium corrosion, i`m not sure wether its enough to cause a total breakdown,
for i have seen other boards with more corrosion that still show function,
mainly caused by accident with fluids)

But as you can read in my last posting this is a mobo with "beginning" bod,
i`ll check some totally comatose next days asap

The board above has definitely no traces of fluid,
i think the corrosion maybe caused by a) condensation under the blue pads, this is a real hot spot,
or b) the glue or better said the solvent in the tape causes corrosion over a longer time

maybe i should give it a try and resolder those corroded legs
of u26 + u7

(meanwhile the last bod boards are sold on ebay for more $'s than functioning ones;)

tom_k

#40 Post by **rkawakami** » Thu Sep 27, 2007 12:06 am

The corrosion that appears on each hole in the circuit board won't be a problem unless the via is affected. This is the metal plating inside the hole which connects the top traces to the ones on the bottom (or even buried between). I'd pay more attention to the effects of the corrosion between adjacent pins/holes as that could cause a low resistance path (a "short") where there shouldn't be such a connection. You might be able to clean up any visible corrosion with some isopropyl alcohol on a cotton swab. Don't use too much force as that might damage portions of the copper trace which may have been weakened due to the contamination.

If the via has been damaged it can be verified by using an ohmmeter on opposite sides of the hole. It should read 0 ohms. If not, a repair can be made by placing a small wire in the hole and soldering it on both sides of the board, at the same time trying to flood the hole with solder. Not an easy fix to do on high-density board designs such as in these systems.

tom_k · #41 Post by **tom_k** » Thu Sep 27, 2007 2:40 pm

just a quick entry for the record:

3 motherboards with bod in the last stadium (nearly won't start anymore)
show more or less the same corrosion in and around the holes
but at u26+7 they look much better then the one you can see in my last posting.

i bet 90% of all T2_ mobos show that corrosion,
think checking my functioning ones will be the next to do...

Please tell me if you like to see Fotos of these 3

dogma420 · #42 Post by **dogma420** » Fri Sep 28, 2007 3:17 pm

In case everyone doesn't know this--and this won't fix the BoD but it is worth trying when you have initial problems...

System reset (and clearing capacitor related to startup of its charge)...

Pull out all accessories, including CMOS battery and normal battery. Disconnect AC.

Hold down power button 2 times for 20-30 seconds each time.

This has fixed some systems that appear to be broken.

But it isn't a miracle cure--just something from an IBM tech's point of view, is always worth doing as one of the first things you do when troubleshooting.

Congrats on fixing the BoD--kinda like fixing an XBox 360!!!

Robbyrobot · #43 Post by **Robbyrobot** » Sat Sep 29, 2007 3:42 am

Hi Jonathan...

If anyone else could try to solder the areas and post there results it would help to establish whether I was just lucky / it was a fluke / it works etc...

I would like to second that, and also ask whether the board you reworked is continuing to boot without problems. The nasty part about this BoD phenomenon is the irregularity, which can make you think a board is OK when it's not. I've had BoD boards exhibit no symptoms for as long as 7-10 days, and then suddenly revert to the BoD behavior. That makes it difficult to know exactly whether you've really solved the problem.

ashokc · #44 Post by **ashokc** » Sat Sep 29, 2007 4:33 pm

Johnathon

I have a T22 which started exhibiting the BOD problem last month. I saw the resolution and the diagram of the spots you did the soldering fix. Some questions...

What type fo soldering tools are you using for this fine type of work? Any special iron,flux etc.

Has your fix stood up till now?

What is your exact definition of BOD?

In my case I had the BOD for a few times but presently seem to have no power (whatsoever, all leds are off) whether I use battery or AC or both. When I first had BOD issue the machine went dead and after about 6 days came back to life ran for 15 minutes then had BOD. After that I am playing with things like CMOS battery etc. Once I disconnected the CMOS battery was not a single led would light up. I have reconnected the battery and also checked the voltage on it (2.76 v). That should be fine don't you think?

I am willing to be the second to try the soldering fix with some help from the forum.

Thanks for your help.

Rick the Green · #45 Post by **Rick the Green** » Sat Sep 29, 2007 7:20 pm

@ Ray : Jeep & Honda's (Alpine manufactured) Radio's had these common smd resistors problems.

This is why i think it's a good idea to look at large surface mount component.

From the t2X pic i see only one large surface mount capacitor.
Even with those i encountered some problems.
In doubt i unsolder them and check them with a capacitance meter, because they will loose capacitance overtime, especially if heat is involved.

@tom_k :

Corrosion : some dried up glue combined with heat will corode copper traces, but i don't think it is the case here, usualy the glue turn brown and flaky.

I don't see it well on the picture but brownish donuts dosen't scare me, if you scratch it a little and it's shinny underneat it's ok. But if the donut is half eaten, that's another story, like Ray said do ohmeter check between sides or from one end of the trace to another.

Question ? Are these board layered or just Double sided ?

Troubleshooting & folowing traces in a layered board is ... well you know ...

Rick

aditya1956 · #46 Post by **aditya1956** » Mon Oct 01, 2007 5:57 am

Sometime back I had removed my T20 mobo for a through inspection, and my observation had told me it to be a three-layered one. I may be wrong here though. As far as ckt tracing is concerned, it is a back-breaking task to say the least if not impossible.

Robbyrobot · #47 Post by **Robbyrobot** » Wed Oct 03, 2007 8:08 am

I don't see any further comments from Jonathan or from anyone else (Timoti) who wanted to try this. Was this another red herring in the never-ending search for the BoD cause?

Timoti · #48 Post by **Timoti** » Wed Oct 03, 2007 10:36 am

Sorry for the lack of update...

As i wrote, my T22 board's not BoD, anyways i tried resoldering all components i could (big enough for a soldering iron in my unsteady hands), in a controlled fashion. It still displays the same symptoms, so at least i didn't make it worse.

I lost the auction on the BoD T20 board i mentioned, so that's a no-go too, sorry... If i find a new one, i'll let you know.

In the meantime, i'm back on my crusade against the CRC1/0175 error (building eeprom reader circuit, extracting a known-good binary dump, reprogrammin a bad board...)

wireman · #49 Post by **wireman** » Thu Oct 04, 2007 6:06 am

From my experience of this BOD problem, until a circuit diagram is available then its near impossible to find the cause. The Analog devices and Maxim ICs in the middle of the board are integral to the board booting but they need a power on signal which comes from somewhere, wheteher it be an RC circuit or an IC.
If this could be found then maybe a solution to all of these broken mobos is possible.
I have since replaced my own T20 mobo and all is well again.

phool@round · #50 Post by **phool@round** » Thu Oct 04, 2007 6:21 am

What if that power on signal is being interrupted by a break near the middle? Large amount of flex occurs in that area, especially over time, it gets even weaker, loosening on the stand-offs, stretching.
Micro crack in a trace, maybe a component with cracked solder joint.

If I had a BOD board I'd put power to it and flex it to see what happens.......

Something is common, occurs over time to a large percentage of these boards and I'm not convinced that it's a component failure as much as a practical/physical one since these will boot up again and act as if nothing has happened, no artifacts remain while powered on.

Website · #51 Post by **tfflivemb2** » Thu Oct 04, 2007 8:38 am

OK, I guess I am going to have to pull out my dead T20 that has the same BOD problem, to see if I can see anything...

ashokc · #52 Post by **ashokc** » Thu Oct 04, 2007 2:04 pm

To phool@around... If you were to put power and flex it is it possible to do so with the board in your hands or does it have to be seated in the computer with keyboard, display and ports connections at the back all in place? Once seated there is limited flex one can provide. and it is difficult to observer the underside of the board where this micro-break could occur.

I think what you are suggesting is correct becuase my thinkpad went BOD way when my daughter started keeping stuff on top of the laptop while it was docked at her desk. The docking station makes the laptop sit at an incline(angle) so putiing stuff on it I am sure causes flex in the middle.

Also can anyone tell me the IC next to (immediate left of white display connector when facing the LCD display) the connector for laptop display is on the motherboard? It is a 8 pin IC and on my board there is nothing legible on it. It heats up quite a lot when I apply AC power to my BOD motherboard. any ideas to what its functionn is. Has anyone observed the same behavior on BOD motherboards?

Thanks.

Robbyrobot · #53 Post by **Robbyrobot** » Thu Oct 04, 2007 2:25 pm

It is a 8 pin IC and on my board there is nothing legible on it.

I doubt if it'll help much, but the chip is marked:

TPC8002
OK

And the "OK" is located just under the "02". The dot to indicate pin 1 is opposite the "T".

If you were to put power and flex it is it possible to do so with the board in your hands or does it have to be seated in the computer...

You have to put the board into the bottom housing, automatically connecting it to the I/O PCB below, in order to mount the CPU and the fan. No way I can see to hold the board in your hands under power even if you left off the fan, since the power connections run through the I/O card.

SMA · #54 Post by **SMA** » Thu Oct 04, 2007 2:51 pm

Wireman, the power signal you are missing is comming from a TB6807F chip.
That is, ADP3410 pin 2, the SD pin comes from a TB6807F pin 13 - again on a A2x board.
So far, I have not been able to find a datasheet for that chip.

And to my big surprise, I discovered that the MAX1632 chip is always on when the AC Adapter is in place.

The large surface mount capacitor, mentioned by Rick the Green, is not the problem either (on my t21 board). I have tried replacing it.

Concerning the brown stuff below the blue pads. My board has it too.
Most of the brown is hanging to the glue and it only occurs to holes not soldered.
After a little washing it is almost invisible - but I cannot exclude the possibility that it has left a problem.

For the time being, I have left the theory of the problem being caused by the max1632, adp3410, adp3421 chips. I have checked it as much as I can,
so I have started looking elsewhere.

ashokc, the TPC8002 is an N Channel MOSFET.

Kaervak · #55 Post by **Kaervak** » Thu Oct 04, 2007 4:20 pm

Timoti wrote: In the meantime, i'm back on my crusade against the CRC1/0175 error (building eeprom reader circuit, extracting a known-good binary dump, reprogrammin a bad board...)

If you do come across a good T21 EEPROM dump, would you mind sending it my way? I have two T21 boards, one with the CRC1 and one that doesn't POST but possibly has a good EEPROM. I'm close to desoldering the EEPROM from the non posting board and replacing the CRC1 EEPROM. But if I could get a good PROM dump to reflash I'd rather do that.

evo4jon · #56 Post by **evo4jon** » Fri Oct 05, 2007 9:37 am

Hi,

Sorry I have been a bit slow in responding but i have been away on my hols....

Just an update really... My T22 is working perfectly since I soldered the connections shown in the pic i posted. Not on that but it has just been flown to Lanzerote and back again... Dropped ! - from a plane seat onto the floor (me being clumsy) and used every day with 2 batteries and mains powered...

ALL IS 100%

It was dead (BOD) on 23rd Sept 07 and after soldering worked from 24thsept 07 - today (5th Oct07) which makes it 11 days of faultless performance

Jonathan

evo4jon · #57 Post by **evo4jon** » Sun Oct 07, 2007 6:53 pm

Update:

Still working fine (8th october 2007)

If anyone is near to Warrington, cheshire with a BOD T2* laptop then I would quite like to see if I could repair another....

I would charge 50 quid for a fix and if I can not repair it then i would not charge.... No Error 175's just BOD

Cheers

Jonathan

Msmax · #58 Post by **Msmax** » Mon Oct 08, 2007 5:45 pm

Hello,

Have bought a T22 with BoD some months ago.
Tried the power button trick without success.

Replacing the ADP3421 chip did solve the problem for this board. Still works fine.

Then bought a T21 with BoD problem.
I thought: easy, exchange the ADP3421 and I have a working machine... Did replace the ADP3421 but did not solve the problem.
Still not found the cause of the problem.

Then bought a Dell machine which also did not power up, totally dead. Reflowed the solder near all conductors, assembled and the machine booted. Still working fine for almost 3 months now.

Have some T23/T30 and R32 boards which seems to be dead. Will play with these in the next weeks. If any one gets alive then I will post my results here again.

Msmax

krewe · #59 Post by **krewe** » Tue Oct 30, 2007 3:55 pm

Hi there,

my T22 just started showing the infamous BoD problem, so I remembered registering to this forum, back then, when I head a dead T20. So I was reading through the posts with great interest, especially those of evo4jon.
Suffering from the BoD problem, of cause, I have to ask the most obvious question before I give it a shot and heat up my soldering iron:

evo4jon, is your board still running fine, after more than four weeks, now ?

Has anybody had the same luck with the soldering attempt ?
Also, has anybody probably isolated the faulty area, where soldering does the trick ? Otherwise I would just stick to the .jpeg evo provided and resolder everything.

I might rather have thought of some torn out capacitors, or the like.
Would be nice, if the fix would be really that simple.

chooch · #60 Post by **chooch** » Wed Nov 21, 2007 12:14 am

Might be an isolated case or just got lucky but I thought I'd share. BOD T20. Reading how a CPU fan can affect a machine, all I did was clean the fan contacts with sand paper and scraped the fan contact pads on the MB with a knife. 2 weeks now without any BOD symptoms. I can now leave battery or AC adapter or both in for any period of time.

T20 Blink Of Death Circuit Questions

corrosion

T2x-T6x system reset

T22 soldering fix for BOD issue

question : put power and flex the board

Who is online