T20 Blink Of Death Circuit Questions

[Voodle]

I recently got hold of a T22 motherboard and a T21 base with a BoD motherboard. Somehow the T22 motherboard has died in a different way, it turns off after a few seconds, but the T21 motherboard has the standard temperamental BoD, it worked for a day or two after the board had been removed and swapped over a couple of times, then today stopped working again.

I swapped the ADP3421 chip from the dead T22 to the T21 and now the T21 works this was after I'd tried to start it today and it didn't work, swapped the chip and it turned on straight away.

Thanks to whoever figured out that that might be the cause

edit: And swapping the bad ADP3421 chip back to the dead T22 motherboard has given it BoD too

[artic_squirrel]

hi,

that's a good news for us !
I'd like to try this on my laptop
but as I'm totally noob in soldering could someone tell me how to do this ?
I'm considering ebay to buy a solder and the lead which is needed but is there any tutorial where I can learn to this thing ?
is any soldering iron is ok to do that ?
like this:
http://tinyurl.com/5olqpv


furthermore, where is the chip on the mb ? (i've got a t21)
http://www.kawakami-ca.com/ibm_t2x/t21_ ... legend.jpg

[nikki605]

Check out this topic http://forum.thinkpads.com/viewtopic.php?t=34809

There are links to several photos in various posts, plus the story of my T21 BoD problem. I had the ADP3421 and several capacitors replaced on mine, but it didn't hold up for more than a couple of weeks.

I ended up replacing the motherboard.

Good luck!

[Msmax]

If you need to learn to solder then I would not start direct with a SMD chip.

The best is to use a hotgun so you do not have to touch the tiny pins which can bent easily, so it will become a horror story to get the chip fixed properly.

Msmax

[artic_squirrel]

thx

what's a hotgun ? (could you post a picture ?)

I've got a mb of another t21 (but I don't know if it has the bod pb), I'll try to replace the ADP3410 and ADP3421, of course if I will be able to desolder them

[RealBlackStuff]

Like this one on eBay: http://cgi.ebay.com/HEAT-GUN-1500W-PROF ... 0056320613

[Msmax]

Not exactly what I meant... but it produces hot air.

I meant a bit more sophisticated one with controllable air temperature.

Search Ebay for SMD Rework station.

Msmax

[Voodle]

To remove the one on mine, I used a 15w small tipped soldering iron and some solder wick.
First using solder wick on the pins to remove as much of the solder as possible, then very carefully prying up one side of the chip with a craft knife while heating as many pins as I could on the corner that I was prying up, which using a 15w soldering iron with a fine tip is rather hard to do but all the pads and the chip came away intact. It's really not something worth trying unless you're familiar with soldering though.

The chips you're looking for are both in D6 of that linked picture, by the keyboard connector.

SMD rework station would be nice but at that point you'd be better off with a new motherboard anyway. Using a heatgun will probably end up blowing other components away once the solder starts melting

[RealBlackStuff]

You can use the heatgun, if you cover everything with aluminum foil and make just an opening big enough for you to work in.

[artic_squirrel]

thanks for the info

I'll try this swap within a month, but I hope that's the only chip which is damaged

the two other chips MaX1632 and ADP3410 are listed on this topic as well

[paiwayne]

Hi everybody!
Where I did not use the Thinkpad t2x and t3x, but I use the ASUS S1300 series, It also use the MAX1632, and it also have the BOD problem, and I sovled it. So I provide my experience to this. Where
the MAX1632 has 2 current limit resisters 0.02 ohm, the reson cause the BOD problem is when the notebook used for long time(several years)
where the devices like cpu cool fan, lcd lighter,etc. make more comsumption larger then the first sale to the customer. So some time you remove the lcd connecter, or cpu cool fan connector, the
BOD problem will disappear ,so I make a judgement , sort the current limit resisters, and re-check BOD problem and the notebook boot normally.
So be sure safty, you can looking for 0.1ohm resisters to parallel
with the 0.02 ohm resister, or thin wire make some turns to paralle the resisters( because the 0.02 ohm is difficult to buy)
That'all hope can help you.

[rkawakami]

Welcome to thinkpads.com!

Thank you for this information. It appears that there's a very low ohms resistor in series with the output voltages of the MAX1632 power controller chip. They appear to be in place so that a current measurement can be taken. In the datasheet spec of the MAX1632:

http://www.alldatasheet.com/datasheet-p ... X1632.html

page 9 references resistors R1 and R2 as varying from 0.033 ohms to 0.012 ohms, depending upon the target load current. The MAX1632 block diagram on page 10 shows them to be connected as a divider which I believe is a mistake in labeling. The current sensing resistors are the ones that are on the left and right sides of the schematic, showing the +3.3V and +5V outputs. Page 8 shows a standard application circuit with resistors R1 and R2 correctly labeled (at least in my opinion). Connecting the CSH5 and CSL5 pins from the MAX1632 across the resistor means that it could detect the amount of current flowing out of the device (voltage drop across a resistor is directly related to the amount of current flowing; I=V/R).

Assuming that a 0.1 ohm resistor is placed in parallel with whatever is on the motherboard (let's assume 0.02 ohms), this will slightly decrease the total resistance (it will become 0.01666). However, if the resistor on the board is open, then you'll be putting a 0.1 ohm resistor in its place; quite a bit of difference given the standard design parameters.

I don't have my T22 BoD system taken apart but I will try to open up the system and see if I can locate these current sense resistors. Then it's a matter of ordering some from Mouser or Digikey. A BoD A21p that I just turned on yesterday (after several months with the battery still in), well, turned on. Several times in fact. It wouldn't not work (sorry for the double negative ).

Also from the wording of the current consumption being higher than when the laptop was new and disconnecting pieces to get it to boot, I'm not sure I tried this with any of my systems but it's a good thought.

[paiwayne]

Hi All ! ,
Where I don't have the t20 notebook, but i have the ASUS S1300 notebook, which is used the same chip of max1632 and it have the same BOD problem. Few days ago, I disassembly it, and I found it, if I remove the CPU coolfan and more time the BOD problem disappear, and if I remove the LCD pannel the BOD problem abslute disappear (no BOD). So I recheck the max1632 output I found when the full load condition the AC adpter plugin and the max1632 output have the correct output, but when I push the on/off sw the output turn to 0v. I suppose the max1632 getting into selfprotection condition. Everybody knows the max1632 have the current sensing resister 0.015 ohm, so I make a thin wire like air-coil and parallel with the 0.015 ohm resister, and it work OK, THE BOD FIXED. I think the BOD problem cause by the power comsumtion get more and more then the first buy,because of the LCD back light goto old, and CPU coolfan may turn slow and current increase ...etc. it result the max1632 sometime sense over current signal getin BOD status, Here I hope my experent have help for you!!!

[adiflorescu]

Paiwayne, I believe this is the best explanation! I met the same BOD on other devices based on the same MAX chipset (2 Compaq laptops and one portable DVD). I replaced the current sensing resistor and all was back to normality!
BTW, did anyone have a schematics or maybe only the identification of the audio amplifier circuit used in X61T? Maybe a part of a Service manual with the exploded view ?!?

Regards,
Adrian

[dragon]

Hi All ! ,
Where I don't have the t20 notebook, but i have the ASUS S1300 notebook, which is used the same chip of max1632 and it have the same BOD problem. Few days ago, I disassembly it, and I found it, if I remove the CPU coolfan and more time the BOD problem disappear, and if I remove the LCD pannel the BOD problem abslute disappear (no BOD). So I recheck the max1632 output I found when the full load condition the AC adpter plugin and the max1632 output have the correct output, but when I push the on/off sw the output turn to 0v. I suppose the max1632 getting into selfprotection condition. Everybody knows the max1632 have the current sensing resister 0.015 ohm, so I make a thin wire like air-coil and parallel with the 0.015 ohm resister, and it work OK, THE BOD FIXED. I think the BOD problem cause by the power comsumtion get more and more then the first buy,because of the LCD back light goto old, and CPU coolfan may turn slow and current increase ...etc. it result the max1632 sometime sense over current signal getin BOD status, Here I hope my experent have help for you!!!

I graduated in Computer Engineering in 1978 and have been working on ATM for 20 years. I have since moved into software doing C++ and Java. My electronics is rusty by now.

I have a T20 which exhibits BOD and a T22 that does not. I would like to repair the T20 as a hobby. I know something about soldering but haven't caught up with smt.

I tried pressing the power switch for 15 secs before putting the power cable but this trick don't seemed to work. Before I change the two offending chips.................

1. How do I put a wire as an air coil parallel to the resistor beside the two offending chips?

2. Where is the BIOS eeprom and how do update the bios eeprom when I can't even boot it up?

Thanks.

[RealBlackStuff]

Welcome to the forum!

If I interpret his English correctly, all he did was bridge the resistor.
As it has only a very small value, I don't think too much damage can be done by that, but I don't really know the long-term effect of doing this.

I don't know where the BIOS chip is located on the mobo, but T20 mobos are relatively cheap, and it may be a lot easier to replace the mobo, than try and swap out a multi-legged SMT chip.

[SMA]

Here is a link to a topic on this forum. In that topic you can find links to scans of a T20 board.
http://forum.thinkpads.com/viewtopic.php?t=55300

The resistors in question can be seen on the top side scan in coordinate C5-D5. There are 4 resistors sitting in parallel.
They are green. (On a T21 board there is only 1 black resistor.)

The bios chip is on the bottom side. Coordinate G7-G8, H7-H8. It is the long one.
Note that one can get to that chip through the memory slot door without having to disassemble the entire machine.

BIOS chips cannot be swapped between T20 and T22, without being re-programmed in-between.

[dragon]

Correct me if I am wrong.

To try to repair the T20 mobo, I have to solder a wire in parallel to the 4 resistors in question. That the only solution I can do before I change the two chips in question.

There is no way to reprogram the bios without an eeprom programmer, unless I can find a known working eeprom from another T20.

[SMA]

I have to correct my selves.

I was guiding you to the resistors that are related to the ADP3421, but I just realized, that what has been said is, that it is the resistors related to the MAX1632 chip that has to be dealt with.
That makes it a little bit more difficult, since the MAX1632 has 2 voltage outputs (3.3V and 5V) and therefore also 2 sets of resistors.
Both sets of resistors are at the bottom side of the board in coordinates E5-F5 and F5-G5.
There are 3 green resistors in each set.

For the BIOS update, the only alternative I can think off would be to find a condition in which the machine can start.
Often these machines do turn on after a weeks rest with no power and no battery.

[dragon]

I have to correct my selves.

I was guiding you to the resistors that are related to the ADP3421, but I just realized, that what has been said is, that it is the resistors related to the MAX1632 chip that has to be dealt with.
That makes it a little bit more difficult, since the MAX1632 has 2 voltage outputs (3.3V and 5V) and therefore also 2 sets of resistors.
Both sets of resistors are at the bottom side of the board in coordinates E5-F5 and F5-G5.
There are 3 green resistors in each set.

For the BIOS update, the only alternative I can think off would be to find a condition in which the machine can start.
Often these machines do turn on after a weeks rest with no power and no battery.

Thanks and I will keep you posted.

Correct me if I am wrong.

From electronics I learned almost 30 years ago, if I put a wire across the green resistor in parallel, am I not shorting it instead of increasing the resistance by 0.02 ohms or whatever. Shouldn't I solder the wire in series instead?

[dragon]

I have to correct my selves.

I was guiding you to the resistors that are related to the ADP3421, but I just realized, that what has been said is, that it is the resistors related to the MAX1632 chip that has to be dealt with.
That makes it a little bit more difficult, since the MAX1632 has 2 voltage outputs (3.3V and 5V) and therefore also 2 sets of resistors.
Both sets of resistors are at the bottom side of the board in coordinates E5-F5 and F5-G5.
There are 3 green resistors in each set.

For the BIOS update, the only alternative I can think off would be to find a condition in which the machine can start.
Often these machines do turn on after a weeks rest with no power and no battery.

I soldered wires across the two sets of green resistors as you suggested, it does not even power up now.

[dragon]

I have to correct my selves.

I was guiding you to the resistors that are related to the ADP3421, but I just realized, that what has been said is, that it is the resistors related to the MAX1632 chip that has to be dealt with.
That makes it a little bit more difficult, since the MAX1632 has 2 voltage outputs (3.3V and 5V) and therefore also 2 sets of resistors.
Both sets of resistors are at the bottom side of the board in coordinates E5-F5 and F5-G5.
There are 3 green resistors in each set.

For the BIOS update, the only alternative I can think off would be to find a condition in which the machine can start.
Often these machines do turn on after a weeks rest with no power and no battery.

I soldered wires across the two sets of green resistors as you suggested, it does not even power up now. There is even no power on light.

After I desoldered the wires off, it reverts back to BOD again.

Any other suggestions please?

[slaterlp]

First I would like to say hi to all. I am new here and so glad to have found such a FANTASTIC source of Thinkpad information.

OK, BOD. And HOW TO PREVENT IT. and recover from it. (sticking my neck out here, but here goes.

I have a T20 2647. Around 8 years old and still alive. Last year I started to experience the BOD. (only found that name for it via this forum). Exactly same symptons already described by others.

It started (the BOD I mean), when I started getting lazy and leaving the Laptop on for long periods, (many hours), or leaving it plugged into the charger when switched off. Previously I ALWAYS switched off and Unplugged the charger.

At first I was mystified, but I just had a hunch that it was not completely dead. Then I read the stuff about IBM techies referring to an over charged Cap on the System board. So I duly followed the instructions, and it still didn't work.

So I took it apart, and unplugged EVERYTHING. I even shorted out the pins where the DC adapter plugs goes onto the Mother Board for a few minutes.

After 3 days I got it going again.

So I reasoned that it HAD to be a stray charge held on the board somewhere and that there was no way to bleed this charge off. (Perhaps thats the nature of the fault). Perhps in the pre fault condition the system bleeds this charge off? Or perhaps due to heat, or a failing joint, or whatever, there is STRAY capacitance.

Anyway, since it happens to just about ALL Tx's eventually I figure it must be the same cause, whatever that is.

My "cure".

First if the BOD happens.
1) Remove AC Power, remove Main Battery, and remove Bios Battery.
2) Remove Keyboard.
3) Unplug DC power inlet plug and LCD screen connector.
4) Short out the 4 pins that connect the Mother board to the DC Coax plug. Leave for an hour or so.
5) REassemble, install Bios Bat, and Main Bat, but do NOT connect AC/DC power.
6) Switch on.

Works EVERYTIME for me.

PREVENTION.

Someone else here may have already suggested this I think, and I concur it works.

Assuming your Laptop is on, with both AC power and Battery, then BEFORE you shut down, ---- REMOVE the AC/DC power, so that its just running on Battery alone.

Leave the Lion Battery to discharge a little. I have found that 90% charge or less is good but maybe anything less that 100%, might do. (but how accurate is the measurement? Could it be that the Battery is already over charged, and 100% is really 110%?).

Anyway, I have a cooling pad (3 Fans) running on the USB port, and my Cisco 340 wireless card enabled, so it soon reduces the charge on the Lion Battery.

Then I shut down in the usual way.

Start up on Lion Battery only and then plug in the AC/DC charger.

If I ALWAYS follow this procedure, I NEVER get the BOD. When I forget, and leave the charger in, I OFTEN get it.

I also only get the BOD if I have been using the Laptop for an hour or so, during which time the Lion Bat has reached full charge, and I am wondering if then something OVER charges?

I welcome your comments, and hope the better skilled than me can perhaps find something that suggests the cause.

[Gatewave]

I have around 15 T20/T21/T22 motherboards here .... some with BOD some with 0175 and some with other issues.

I just ordered 4 ADP3421 from China and going to try to see if I can fix any of the motherboards. Will keep you posted.

I'll also see if I can get in contact with the designers who designed the T2x series and have a chat with them about this problem.... I'm guessing they work for Lenovo now.

If anyone needs any of these motherboards with issues let me know.

[chooch]

I posted this Nov 21st, 2007

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.

I see this is still a hot topic. I just got on here to find a source for a battery and thought I just check this thread.

Anyways, I must have been very lucky because I'm replying to this thread with the same T20. My hard drive failed last year and my AC supply hums.... Other than that, my T20 has and is still serving me well. Good luck.

[Mornsgrans]

It seems, that until these days the problems only could be solved ramdomly but no real solution .

[nikki605]

It seems, that until these days the problems only could be solved ramdomly but no real solution .

Sad, but true.

That's why when my T21 developed the problem several years ago, I just replaced the mobo and moved on.

[madrigal]

Hi, guys!
I have a T21 mobo which steadily showed BOD symptoms. Following Augusto Samame's instruction a couple days ago I replaced ADP3421 and MAX1632 with ICs from a dead T22 mobo and ... guess - BOD is gone!!!!! Now I'm testing the system and could say that everything is OK (at least for now) - with or without battery, with or without AC adapter every time laptop starts fine. May be it's too early for drums and trumpets but I hope that finally the problem is solved.

Admin edit: Added reference link to Augusto's post

[RealBlackStuff]

Are those ICs also on T23 mobos? I have a few dead soldiers here...

[madrigal]

I couldn't tell you because I've never had a T23, but in his post Augusto wrote that T23 uses different voltage regulation circuits.