How to repair the parts that explode in Lenovo yoga laptops
adammunich.com
adammunich.com
This wouldn't be my first guess based on that hole, which was formed more abruptly than a little overheating. Cooling the outside of the package is only going to buy a little more margin, not fix the root problem.
Looking at that schematic snippet, what is up with the capacitance on that gate drive circuit? They want a slower turn on to avoid a popping sound? Maybe it's a reference circuit that worked fine with a larger package or something, and a UF6 doesn't have enough thermal mass? If the turn on/off time is the problem, removing one of C489 or C530 could stop it from happening again.
Or it could be something downstream drawing too much power on that rail or shorting it out. Since you're replacing the part, you might as well just use a FET with a lower resistance and some hope to address this possibility.
FWIW some thinner flux (like a Kester 951) will generally let you fix solder bridges without starting all over.
Although, if it's a 5V0 or 3V3 rail they may just not be turning on the FET enough. It's Rds(on) is high at ~3 V, and they divide down the gate drive to get reasonably symmetric on/off times and slow speed.
Decreasing the turn on/off time might also be a good idea to prevent future repairs (nice spot!).
The first time for me was with their T43. This model was mostly a platform refresh of the T42. There's a chip called a southbridge that's located below the palmrest. Between the models this chip went from 400mhz to 533mhz. Lenovo decided it wasn't important to improve the cooling, no heatsink whatsoever included on the southbridge. However, they kept ample temp sensors in the case so the fans would pulse nonstop as it failed to cool this one chip.
I had two nicknames for that laptop. The 'nut roaster' and the 'vacuum cleaner'. I eventually fixed the problem by hammering out an old copper penny and sticking with thermal compound to bridge the gap between the heatsink and chip. It wasn't perfect but it dropped the temps like 10C. The rest could be controlled with an installed fan daemon.
Personally I believe the major source of heat of the X220 comes from the Sandy Bridge processor. The next-gen Ivy Bridge in X230 has almost no performance gain, but reduces heat dissipation significantly (ironically, Ivy Bridge is also the turning point when Intel started to use crappy thermal paste in the desktop CPU packaging, leading to thermal issues...).
I own both machines and the difference is great. Also, two of my X220s from different sources all suffer with random shutdown issues, I suspect the source is the cold solder joints under the power MOSFET or a defective one (saw a troubleshoot post of a similar issue in a fourm), but it's BGA, so...
Anyway, that's why I recommend everyone who wants a X220 to get a X230. You can even replace the motherboard and the bottom case to "upgrade" the X220 to X230.
> the heatsink from the x230 can be fitted to a x220.
I believe it's virtually identical. I never heard about the X230 heatsink performs better, perhaps it's the case, I may give it a try someday...
So it seems like the problem is a cooling issue, which doesn't surprise me. "Ultrabooks" often seem to sacrifice decent cooling for the sake of thickness, and I think it is because most people really do not tax their hardware all that much. If you do - and you do not have a proper workstation-class laptop - you're likely to run into similar issues.
My Thinkpad X280 suffers from poor cooling that quickly leads to throttling, even undervolted. My old work A485 wasn't that great, either.
We don't know the voltage on the rail, so we can't say exactly what it is, but the likely candidates are an SOA violation because they're switching it absurdly slowly, or the gate drive is below a reasonable threshold.
Not just a cooling issue. It's a MOSFET thermal runaway, as it is being driven to its limit, higher temperature => higher resistance => higher temperature. It can be stopped by cooling it, but the bigger picture is that it is either being driven improperly, or cheap, low power components are used to cut costs without adequate a safety margin.
What can be happening here:
- badly designed output Audio section shorting power in some rare circumstances, might be as weird as mechanically stressed audio jack touching traces underneath it.
- software glitch around audio power enable routine enabling/driving that transistor hundred/thousand of times per second (pwm) in some circumstances, keeping it in the linear region
- badly designed under powered mosfet driver, either voltage too low or not enough current keeping it in the linear region
You do not cool power rail switches like this one, they arent meant to dissipate any meaningful power when designet properly.
It could a whole slew of problems and it's impossible to tell without measuring. But, I would guess that gate voltage is not far enough above the threshold to drop the Rds(on) to it's low loss on-state.
You really should heat the whole board uniformly to be safe, but that is a danger to do when you aren't sure of the melting temperature of the solders used.
Edit: thank you for explanation. This was not thermal problem then. I am just curious why didn’t all laptops had this problem?
It is just bad design by Lenovo. You should not put a switching transistor in linear mode. That rc-dampener is just bullshit.
If AUDIO_PWR_EN is a pin that is controlled by software, and, if you keep switching that pin such that the MOSFET is kept in the linear region due to the RC on the gate, that MOSFET will be toast very quickly.
Yet it has a drastic crater-like dent in the IC enclosure. That tells me it was a voltage surge or current spike.
The hypothesis of entering a linear region caused by a random software glitch producing an unconscious PWM on a filter slope sounds plausible as well. But that crater-like dent says that the real cause might be way more intensive than that.
Overall though, the laptop feels well built. I like the key action (but personal preference). The hinge seems to be well made. Nothing flexes when opening or closing it. The screen, to my eyes, is fantastic, and with the Celicious anti-glare film on it, it works well in daylight.
It's wicked light. I don't love ultralight laptops and originally was trying to avoid them, but this one has almost all the ports I want (except for a mmc instead of sd card slot) and appears to be more serviceable than most modern non-ultralight laptops. One of the things I looked for was a user-replaceable battery, and surprisingly, this one should be. We'll see how much it actually sucks to replace in a couple of years.
If you try to run any Linux on the Gram 17, you will need the info at https://bugzilla.kernel.org/show_bug.cgi?id=203617
Regardless of the merits of the OP's problem, my point is that any large manufacturer is going to have duds (and/or dud models). People like to hate on Dell, but I've had some great machines from them over the years, for very good prices. There's more to brand quality than 'I had a bad experience once'.
I've done that before, managed to fix it with silver based conductive paint.
As I understand it they are working off of an older fab than AMD and are hitting its EOL, all while commanding a price premium. AMD has room to expand their lead and appears to be beginning to, despite being a 10x smaller company.
The amount of angry tweetstorms, memes and class action lawsuits would break the internet.