Thinkpad laptop flooded with liquid while powered off, won't turn on anymore [not trivial]

2

A classic IBM Thinkpad T30 (MT2367) laptop has been flooded with a salty water for a period of ~2 hours while powered off, having AC cord unplugged & both batteries (main battery and a tiny backup battery) removed.

Took these steps immediately:

  1. Removed all the non-vital components (HDD, miniPCI card);
  2. Flushed the whole system with a hot, deionised water;
  3. Flushed gently with an isopropyl alcohol;
  4. Let all the liquids evaporate for 5 days in a dry room;
  5. Reinstalled the backup battery and tried to power on.

The laptop is dead now and won't respond (no LED, no fan, etc.) to a Power Button on AC nor on a battery (adequately charged).

Followed this advice: 9 times * 10 secs + 1 time * 30 secs Power Button pressed, AC reconnect -- no avail.

I'm O.K. with inspecting, replacing, resoldering, reflowing elements on the mainboard PCB to revive this unfortunate casualty.

[edit] The helpful Web suggests, there can be IBM's own moisture sensing device(?) in place, which could remain moist or go defective.

Tomiš

Posted 2014-02-11T04:11:21.830

Reputation: 21

1.... ow. and a T30 is pretty old. I suppose it was being used for something a modern system wouldn't be useful for? – Journeyman Geek – 2014-02-11T04:43:32.273

How long before you managed to get the salty water off it? There may be some corrosion, which could mean component replacement. – Bob – 2014-02-11T05:03:43.927

If you get no response from your system whatsoever after 5 days I think you might have to accept its finished. You would have been better simply draining and leaving it for a couple days rather than pour other various things inside it. – Matthew Williams – 2014-02-11T09:18:43.940

@Bob, thanks, OQ clarified, it has spent something like 2 hours underwater. I flushed the salty water off with a deionised one immediately. Agree, there may be still corrosions -- will inspect the mainboard, once get access to a decent ohmmeter and some magnifying glass. (Looking for a specific advice, though.) – Tomiš – 2014-02-11T09:29:12.573

@JourneymanGeek, right, AFAICT, RS-232 port, battery in good condition and a reliable construction makes this old laptop useful e.g. as a service computer. It may no longer be considered guaranteed after this accident, but I retain some hopes. – Tomiš – 2014-02-11T09:35:16.983

@MatthewWilliams No. The general recommendation is to get the water (especially salty water) off quickly, and wiping with isopropanol is a good way of doing so (though I'd rather not flush/soak it, personally - some components don't really like that). Corrosion is one of the most damaging issues, just after a short circuit while it's on. Putting random crap on is bad, but deionised water followed by isopropanol is a pretty standard technique. Heck, most people can't even get their hands on deionised, or don't bother. – Bob – 2014-02-11T09:36:51.990

@Tomiš Unfortunately, this kind of issue is difficult to provide specific advice for. You've already followed the standard method (water flush, alcohol flush). I'm not sure what is left short of taking it to a service centre. Maybe someone will come and answer. If you can get a little bit more rep, feel free to join us in chat - troubleshooting like this tends to work a little better there (then we can compile an answer and post it here).

– Bob – 2014-02-11T09:39:03.137

@Bob | Sorry I should have made the point more clear. I was referring specifically to "flushing" the system. I agree the substances used where correct, but the implied method of implementation I was raising an issue with. I just had images of him upending his laptop and pouring it through the vent from a jug.... – Matthew Williams – 2014-02-11T09:43:58.640

@MatthewWilliams That's a good point, actually. Tomis, just how did you flush it - there may still be residual minerals on the board. Have you tried a soft brush (toothbrush?) or cloth along with the alcohol? – Bob – 2014-02-11T10:18:10.273

@Bob, @MatthewWilliams: Thanks, that's the advice to follow -- `cause no, once drives and covers were all removed, I had it only submerged and shaken in a basin of hot deionised water, then I drained H2O, pried the keyboard a bit, so I could pour the isopropanol onto the visible elements, and proceeded with that from different directions, and into any other open slots (I was rather concerned about moisture entering the LCD). (Also please see the OQ's most recent edit about the alleged IBM's on-board moisture sensor[?]). – Tomiš – 2014-02-11T11:15:53.003

[SOLVED] Latest update: after additional 5 days of curing in the dry room, the machine resurrected -- powers on under AC and on a battery, completes Diagnostic POST, boots Hiren's Boot USB. Well done, IBM. Assessing any damage ATM -- as for now, BIOS loses date and sporadically reports checksum error (suspecting backup battery's fault) & --aesthetics: LCD indeed displays some awful stains on the substrate and 2 bad columns, cover strips off paint. Will update the Q/A when further data is acquired. \m\ ^_^ /m/ – Tomiš – 2014-02-16T22:32:08.633

@Tomiš You should post your comment as an answer. – Ƭᴇcʜιᴇ007 – 2014-03-26T14:29:07.120

Answers

0

I believe the salty water has already affected the motherboard, so you need to do the full motherboard troubleshooting. Looks like there is short circuit in some of these power sockets that leads to the fact that it doesn't react to the PW-ON signal.

You'll need to study a scheme diagram carefully. I've found it for you: http://ldasystem.com/files/ibm/IBM_Thinkpad_T30.pdf

Take of MB, disconnect from all peripherals and other boards and flexible cables. First you need to connect AC adapter and check on the power jack pins on the PCB if 16V come to the jack. Then this voltage is supplied to the voltage convertors and power switches to power up SIO/MIO controller, CPU power, north bridge (PCH), south bridge (ICH), firmware hub (FWH) and battery charging scheme. There should be stand-by and power-on +5V and 3.3V voltages on these circuits.

Signals:

  • power on logic (page 52 of the PDF). Check fuse, TC7W66FK. They generate VCC3SW for the ICH to enable and wake from sleep mode;
  • DC IN and charge: ADP3806 (U42) is responsible for +16 V charging of the battery, but it requires VCC3SW (page 61). +16 V should be on the controller input;
  • SIO PC87392 is powered on by VCC3B (page 53)
  • VCC5M and VСС3M should come to H8S/2169A controller (page 48);
  • VCC5M and VСС3M are generated by MAX1631 controller. It is enabled by V3ON signal and V5ON (page 65)
  • CPU is powered by ADP3203 (U19), which in turn required PWRGD (power good signal) that is enabled by VCC3M and VCC3SW. VCC3M also enables -V_CORE_ON signal. VCC3M also feeds ADP3203 on VCC input. CPU is powered on as the result by signals coming from ADP3203 and VINT16 (a signal from AC adapter). Page 65. You need to check ADP3342 (U10) also that generates -V_CORE_ON signal based in VCC3M and VCC3SW.
  • Very often there is a problem with generation of 1.25 and 2.5 V. This is controlled by MAX1715EEI (U56) controller. The input signals for it are VINT16, +PWRON (which is generated by PMH). Another MAX1715 (U54) is responsible for Video core VCC generation. Page 70 Signals V3ON and V5ON as well -PM_SLP_ENABLE are also generated by PMH and come to TB6808F (U76). All VCC3 and ММС5 power signals come to TB6808F, and it, in turn outputs Power Good signals. Page 72.

So, in general, you need to check input fuse, all transistors and DC switches in the circuits above. Please also carefully inspect PCB for possible putrefied elements around these power circuits. Sometimes even one resistor is enough that the required stand-by voltage is lower than required and it results in no power on. In your case I think the failure should be in the input power circuits.

Enjoy!

Hardoman

Posted 2014-02-11T04:11:21.830

Reputation: 751

Asked: 2014-02-11T04:11:21.830

Viewed: 318 times

Active: 2015-12-11T14:55:13.060