help with generic Core 2 Duo laptop with power on issues

[MacGyver's paperclip]

Hi everyone. Hope everyone's doing OK and thanks for reading.

TL,DNR; Laptop would charge, but not power on. Replaced dead CMOS/BIOS battery and identified flimsy power button daughtercard connector; got it to POST fine outside chassis. After reassembling, still would not power on. Disassembled and after more troubleshooting, now in a state where it *will* power on, but won't POST or charge. Passive and active components all visually OK, capacitors check out unshorted, voltage rails seem proper. Suspect SMBus IC damage, but need a second set of eyes to look and comment before I delve into detailed photos, measurements, etc.

[MacGyver's paperclip]

Re: help with generic Core 2 Duo laptop with power on issues

Device I'm dealing with is an old Core 2 Duo laptop, allegedly in working condition and was given to me for use in testing, random tinkering, etc. Will not turn on *completely* (explained below). I'm an experienced engineer and geek who has built and fixed hundreds of computers, only not many of them at the component level, with decent soldering skills and proper equipment for micro-soldering jobs. I'm basically using troubleshooting this as a good opportunity to teach myself more about working on board/component level fixes in my spare time. I would appreciate any and all suggestions.

This is a VESTEL "ONYX1 D141NG-T52-ZC5S" laptop PC. You won't find much (if anything) about it online; search results on the motherboard markings or numbers on the labels within produce few if any hits, mostly for replacement parts like keyboards. Now, VESTEL is a Turkish appliances manufacturer which may be familiar to those of you in Europe; it occasionally also manufactures and sells domestically various rebranded OEM laptop kits from Chinese companies and this is probably one of those. It's more or less a generic Intel 945GM chipset/Centrino laptop with Core 2 Duo Merom/Socket M CPU, DDR2 RAM, SATA and the usual accoutrements per its era.

As received, plugging in a known-good, properly rated (18.5 - 19 V x 3.4 A) AC adapter with the battery inserted would result in the charging indicator LED turning orange. Pressing the power button, however, would do nothing. Same result with the battery removed.

Suspecting the CMOS/BIOS battery or the power button being faulty, I carefully disassembled the laptop as I have done hundreds of times before with others. CMOS/BIOS battery was indeed dead; I replaced it with a good one. The power button, on the other hand, is on a daughterboard with the power connector and dual USB ports. It connects to the motherboard using a fairly flimsy plastic connector that has a lot of give and take and easily wiggles. As in the photos below, it's like the kind with 2 rows of inner/outer pins that you would find in older laptops for connecting CPU, modem, or network card daughterboards. I checked the connector pins on both boards were straight, clean, and in place. I was then able to assemble the motherboard, power/USB daughterboard, LCD panel and keyboard out in the open and able to get the machine to POST consistently enough that I concluded it was the finnicky daughterboard connector that was the problem.

I reassembled everything back together in the laptop chassis. Once again, the charging LED would turn on orange upon plugging in the AC adapter, but pressing power would again do nothing. Carefully disassembling a second time, I was able to once again get the motherboard/daughterboard/etc. to power on in the open. While playing around with the motherboard/daughterboard connector, I noticed that if inserted halfway but not fully in, the power button would consistently work. So my plan was to insert some sort of an insulating "shim" in between when putting them back together to achieve the same exact placement that made it work.

It's around this time the motherboard behavior suddenly changed. I wanted to do one final test in the open and I'm not sure if I did anything different in particular this time around, but plugging in the AC adapter will no longer trigger the charging indicator LED. The board still powers on (halfway), assembled in the open, though nothing on the screen and no POST, no CPU fan spinning, etc. A few times, the motherboard powered on with the charging indicator red (as opposed to orange), indicating a bad/missing battery or failure to charge, but I can no longer replicate this. At this point, the power button turns the machine on, and the power LED inside the button lights up and stays on; the separate motherboard power LED and/or charging LED does not, unlike before. CPU fan does not spin, and no display or POST.

[MacGyver's paperclip]

Re: help with generic Core 2 Duo laptop with power on issues

Now, before I go in and provide a whole pile of photos, component IDs, voltages, measurements — I have very carefully inspected the board over and over for any visibly burnt or damaged component, but I cannot find any. I tested all large and most of the smaller capacitors for shorts with a multimeter, but all seem fine. There are some capacitors who trigger the multimeter to beep in diode/continuity mode, some as expected like those around the CPU, but measuring in resistance mode indicates these are not actual shorts, just low ohms in the 30 range. MOSFET leads read the expected 19 V adapter voltage or 12.5 V charging voltage around the battery connector; and I read the expected 3.3 V and 5 V around most of the other active components. I therefore suspect the charging IC and/or the SMBus IC. The charging IC, an ICS 954226AGLF, looks fine visually and the capacitors around it all check out. The SMBus IC, an SMSC KBC112, on the other hand — well, it looks like it has a very, very shallow "bump" on its topside. It's hard to take a photo of it and I can barely feel it with my finger, but I tried to show it below anyway. So, as I wrote, before I go deep with measurements and trace hunting, does this look like a burnt/failed IC to anyone? Could it be a manufacturing defect, though I have never seen anything but a smooth surface on any undamaged ICs I have come across?

BTW, this motherboard has probably the most unpopulated passive component locations (from the factory) I have ever seen in a laptop — I don't know if they produced different models with wildly different component sets, but it's bizarre.

I can post detailed pictures and some measurements, but would first appreciate your comments on the SMBus IC, so as to save time if that's likely the issue. Thanks in advance and for reading this far!

[todenkopf]

Re: help with generic Core 2 Duo laptop with power on issues

Looks burnt as for me, would you kindly measure resistances to ground at bigger smd capacitors surrounding that ic?
Have any attempts been made to connect the LVDS cable to powered board?

[MacGyver's paperclip]

Re: help with generic Core 2 Duo laptop with power on issues

Looks burnt as for me, would you kindly measure resistances to ground at bigger smd capacitors surrounding that ic?
Have any attempts been made to connect the LVDS cable to powered board?

Thanks for responding. Of course; there is one (relatively) bigger SMD capacitor at around 1 x 2 mm square near the SMSC chip and a bunch of smaller ones. With no power or AC adapter/battery connected, large and small, all close ones measure roughly 30 Ohms on the non-grounded ends. Other SMD capacitors close by but grouped with other circuitry (like a PC Card slot controller) from their traces all show rising resistance starting from hundreds/thousands of Ohms when measured until my meter goes off-scale, I assume as the capacitors charge as expected from the probes.

I have NOT connected the LVDS cable to the powered board since my first testing with the motherboard/daughterboard assembled outside the chassis, and have been using the VGA port to an external monitor to test. I realize there is a chance the BIOS may not redirect video out to that port automatically, but I didn't want to introduce additional complexity and since I've seen backlight inverters blown before by prematurely connecting them to a motherboard with unknown power issues. I was able to observe normal POST behavior outside of what's displayed on the screen when testing out of chassis the first time and before it went haywire — power LED and charging LED coming on, CPU fan spinning, etc. BTW, this machine has a single connector for the display on the motherboard side, combining LVDS signal line and the inverter feed like most laptops of its era.

[MacGyver's paperclip]

Re: help with generic Core 2 Duo laptop with power on issues

Now, before I go in and provide a whole pile of photos, component IDs, voltages, measurements — I have very carefully inspected the board over and over for any visibly burnt or damaged component, but I cannot find any. I tested all large and most of the smaller capacitors for shorts with a multimeter, but all seem fine. There are some capacitors who trigger the multimeter to beep in diode/continuity mode, some as expected like those around the CPU, but measuring in resistance mode indicates these are not actual shorts, just low ohms in the 30 range. MOSFET leads read the expected 19 V adapter voltage or 12.5 V charging voltage around the battery connector; and I read the expected 3.3 V and 5 V around most of the other active components. I therefore suspect the charging IC and/or the SMBus IC. The charging IC, an ICS 954226AGLF, looks fine visually and the capacitors around it all check out. The SMBus IC, an SMSC KBC112, on the other hand — well, it looks like it has a very, very shallow "bump" on its topside. It's hard to take a photo of it and I can barely feel it with my finger, but I tried to show it below anyway. So, as I wrote, before I go deep with measurements and trace hunting, does this look like a burnt/failed IC to anyone? Could it be a manufacturing defect, though I have never seen anything but a smooth surface on any undamaged ICs I have come across?

BTW, this motherboard has probably the most unpopulated passive component locations (from the factory) I have ever seen in a laptop — I don't know if they produced different models with wildly different component sets, but it's bizarre.

I can post detailed pictures and some measurements, but would first appreciate your comments on the SMBus IC, so as to save time if that's likely the issue. Thanks in advance and for reading this far!

Incidentally; as a moderator (privately) commented, the SMSC chip is of course a Super I/O chip. I referred to it as the "SMBus IC" mainly because it — among other things like keyboard control, legacy connections, etc. — does some managing of power and charging using the SMBus (protocol). The moderator also noted it looks damaged to them as well.

[todenkopf]

Re: help with generic Core 2 Duo laptop with power on issues

Thanks for responding. Of course; there is one (relatively) bigger SMD capacitor at around 1 x 2 mm square near the SMSC chip and a bunch of smaller ones. With no power or AC adapter/battery connected, large and small, all close ones measure roughly 30 Ohms on the non-grounded ends. Other SMD capacitors close by but grouped with other circuitry (like a PC Card slot controller) from their traces all show rising resistance starting from hundreds/thousands of Ohms when measured until my meter goes off-scale, I assume as the capacitors charge as expected from the probes.

I have NOT connected the LVDS cable to the powered board since my first testing with the motherboard/daughterboard assembled outside the chassis, and have been using the VGA port to an external monitor to test. I realize there is a chance the BIOS may not redirect video out to that port automatically, but I didn't want to introduce additional complexity and since I've seen backlight inverters blown before by prematurely connecting them to a motherboard with unknown power issues. I was able to observe normal POST behavior outside of what's displayed on the screen when testing out of chassis the first time and before it went haywire — power LED and charging LED coming on, CPU fan spinning, etc. BTW, this machine has a single connector for the display on the motherboard side, combining LVDS signal line and the inverter feed like most laptops of its era.

Then try to inject 1V to these capacitors, observing the heating of SMSC. IMO 30 ohms on some 3V_EC is far too low to be normal.
Lots of laptops got killed by connecting lvds/edp cable while main power is present.

[MacGyver's paperclip]

Re: help with generic Core 2 Duo laptop with power on issues

Then try to inject 1V to these capacitors, observing the heating of SMSC. IMO 30 ohms on some 3V_EC is far too low to be normal.
Lots of laptops got killed by connecting lvds/edp cable while main power is present.

Yeah, I had the same feeling, just from watching similar repairs but not from experience, so I appreciate the input.

I don't have one of those special-purpose power injection tools, but I presume using a bench power supply set to 1 V and current initially capped very low and slightly increasing should achieve the same result, with the bench power negative to board ground and bench power positive to the non-grounded ends of the caps in question? That's how I've seen it done in some of the videos I have seen by those without a power injection tool.

Thankfully I knew about the LVDS cable connecting while on live mains killing motherboards and certainly didn't do that. My best guess, given the flimsy and finnicky motherboard/daughterboard connector, it might have slipped and shorted something while testing, resulting in this.

[todenkopf]

Re: help with generic Core 2 Duo laptop with power on issues

Yeah, I had the same feeling, just from watching similar repairs but not from experience, so I appreciate the input.

I don't have one of those special-purpose power injection tools, but I presume using a bench power supply set to 1 V and current initially capped very low and slightly increasing should achieve the same result, with the bench power negative to board ground and bench power positive to the non-grounded ends of the caps in question? That's how I've seen it done in some of the videos I have seen by those without a power injection tool.

Thankfully I knew about the LVDS cable connecting while on live mains killing motherboards and certainly didn't do that. My best guess, given the flimsy and finnicky motherboard/daughterboard connector, it might have slipped and shorted something while testing, resulting in this.

Yes, bench power supply will do, limiting current to 1A in CC mode will be good. That IC probably have heat dissipation pad, so give it some time to heat up finely.

[MacGyver's paperclip]

Re: help with generic Core 2 Duo laptop with power on issues

Yes, bench power supply will do, limiting current to 1A in CC mode will be good. That IC probably have heat dissipation pad, so give it some time to heat up finely.

Great, will do — don't have the bench supply in front of me right now, but will get it and test that way. I'll update the thread as I make progress.

BTW, I assume replacing this IC would involve the standard fare of steps — flux and heat gun with nearby components protected by Kapton tape, clean up pads with solder braid after chip removal, re-tin with flux and some solder in case the new IC isn't pre-balled, place new chip and apply heat gun? From the datasheet I can see it has an inner and outer row (square) of pins, but hopefully it's not as complicated as BGA rework.

[todenkopf]

Re: help with generic Core 2 Duo laptop with power on issues

Somehow its even worse than BGA.. You'll need to apply some amount of solder as equal as possible on it's pads and pray.