Dead T30-no fan, no LEDs ("Strange Patch On Mainboard") PICS

[singbad]

Hi,
I got a dead T30 with a strange "patch" on the mainboard / system board:


patched 2366-DG3, FRU 26P8828 ........ original 2366-EG9, FRU 91P7698

Pins 2 and 3 of the upper IC are unsoldered and lifted from the board. The green wire connects them to pin 4 and 5 as well as to pin 4 of the lower IC. An additional capacitor has been inserted between pin 6 and 5.

What's puzzling me most is that compared to the original photo (from my good T30) the lower IC has been soldered upside down, I mean in the wrong direction.

Does anybody know what the 55AG or 55AC is? From what I measured with a multimeter it's not a resistor grid.

EDIT: with today's state of knowledge (see SMA's post below) things begin making sense...

/EDIT

The patch was covered by a small piece of amber colored transparent tape. From the dust on it and underneath I guess this patch must have worked for a certain time at least.

Has anybody seen similar patch on a T30 board?

Best regards,
singbad

[RealBlackStuff]

Never seen that, and no idea what they are for.
That they are upside down is probably coincidence, as it probably does not matter which way they go in.

[SMA]

Just because both boards comes from two T30's they doesn't have to be identical.
If the two boards has different fru numbers, then I'd say that all the pictured chips are sitting correctly.

[singbad]

Yes, you're right, I forgot to mention the FRUs:

Bad board: Model number: 2366-DG3, FRU 26P8828

Good board: Model number: 2366-EG9, FRU 91P7698

But before I go too much into detail: there is a guy in Langhorne, Pennsylvania selling working T30 FRU 26P8422 motherboards for 16$ a piece on the Electronic Bay.

http://cgi.ebay.com/IBM-THINKPAD-T30-LA ... 5d26f05a45

If the URL tag doesn't work: the item number is 400085244485

Unfortunately he isn't willing to ship internationally. Would anybody in the US mind to score one for me and send it to Germany?

Best regards,
singbad

ps
I promise I'll sing better

[RealBlackStuff]

Thanks for the link.
I just bought 3 of them. I had to pay 6% tax, because I also live in PA, so price each (rounded) = $17.-
I'm willing to ship one over to you in Germany at cost. Can calculate that when they arrive.
Payment via PayPal but add 3% handling charge.
PM me if interested.

PS: only if you can reach the High C note

[singbad]

GREAT!!! You've got a pm.

Thank you very much!
singbad

[RealBlackStuff]

OK, let you know sometime next week.

[SMA]

The chips pictured are mosfets.
Why there are 2 of those sitting in series I cannot explain.
The purpose of these mosfet's are to switch the supply voltage to the ultrabay device on/off.
I have a T30 board - fru: 26P8328 - where these mosfet's has a top marking of S2D, which is TPC6003.
TPC6003 mosfets are used a lot on several T4x boards.

[singbad]

Great, thank you very much, SMA!

Now it makes me think the patch (maybe initially made by IBM/Lenovo?) is probably not related to the board's failure. I'll change the Title of this thread.

But where should I start debugging a dead T30 board showing no activity at LEDs or fan? The only good voltages are at the 16 V supply, the 10.8V Li-ion and the CMOS battery.

Any hints?

Thanks,
singbad

[UnixNut]

I don't have a T30, but I have a T23, so I can't give you exact positions, but I suspect they share the same components. As such, on my T23 I have a chip with "MAX1632EAI" written on it, This is apparently a notebook power supply controller, the datasheet is here: http://pdf1.alldatasheet.com/datasheet- ... 32EAI.html

It's a good place to start, tracing with help from the datasheet to the rest of the power circuit should be useful. On my T23 the power (both for LCD and general DC-DC conversion) is in the centre of the body (near inductors and chokes), but it could be somewhere different for you.

Also look for a chip with "ADP3421" written on it, this is the DC-DC converter that powers the CPU, datasheet here: http://pdf1.alldatasheet.com/datasheet- ... P3421.html

I hope this is of any use to you, I'm working on the pretext that to save costs, IBM probably used the same chips in multiple versions of their laptops. Hopefully this will be the case

[singbad]

Thanks for your hints. I was able to fix my T20 this way. Someone in a German forum found a solution for the T20 BOD syndrome by soldering a pull-up resistor to the POWER-GOOD input of the ADP3421. This made my day and saved my T20

This is the Google translation of the German post

If the link doesn't work, here's the original:
http://www.thinkpad-forum.de/thinkpad-h ... post645703

EDIT: the German poster soldered a 10k ohms resistor from pin 16 to pin 24 of the ADP3421. However, 10k didn't work for me but 1.8k does the trick.



I'll soon have a working T30 board for comparison of voltages. I'll have to separate the board from the lower shell since the coils and capacitors are underneath. Someone suggested to re-solder the coils first.

[Harryc]

If you can email the Pics to me I can host them.

[singbad]

Thank you. I managed to post my own image, see above.

I wish I had a similar solution for the T30-No-Startup syndrome too.

[rkawakami]

I would note that if you do connect a pull-up resistor to the PWRGD pin of the ADP3421 in this manner, you will be overriding the IC's normal "power is good" signal and telling the rest of the power circuitry that the ADP3421 is operating "correctly", irregardless of the voltages that it is really supplying. That could lead to other consequences (i.e., improper operation or actual damage) to other systems on the motherboard.

[singbad]

Hi Ray,
thanks for your advice. I'm well aware that this fix rather circumvents than solves the problem. But what's the problem anyway? Is the circuit designed poorly and does the ADP3421 get damaged by overcurrent/overheating?

Meanwhile my new ADP3421 arrived from China and I could replace the old one. But why should I take the working T20 away from my girlfriend? She wouldn't understand it. Why fix it if it's not "broke"?

Best regards,
singbad