Note that parts like this are more easily replaced using a hot air station than a soldering iron.
Looks like the audio amp fed by the MOSFET is a Realtek ALC1304, which I can't find any data on, but searching for the pin names finds that it could be a clone/compatible to the TI TPA3131D2/TPA3132D2. This part has a power limiter, but if the Realtek one is configured the same way, then as implemented in the laptop, it has no power limit.
It is very likely that the reason the transistor failed was because of poor cooling, as it is a low cost, high-resistance transistor that is running near its design limits.
The MOSFET that blew up is rated for 2A continuous, and it's in the ~19-20V VIN circuit, so it begs the question of just how loud he was playing his music when it happened!
That said, not the first time laptops have come with sub-par audio circuitry:
Straight from the horse's mouth (mosfet datasheet):
Note: Using continuously under heavy loads (e.g. the application of high temperature/current/voltage and the
significant change in temperature, etc.) may cause this product to decrease in the reliability significantly even
if the operating conditions (i.e. operating temperature/current/voltage, etc.) are within the absolute maximum
ratings.
Please design the appropriate reliability upon reviewing the Toshiba Semiconductor Reliability Handbook
("Handling Precautions"/"Derating Concept and Methods") and individual reliability data (i.e. reliability test
report and estimated failure rate, etc).
Something like the SiA471DJ comes in the same pinout and package, has almost 1/10th the on-resistance, and over 10x the current rating. Moreover, its datasheet does not have such a lengthy disclaimer about its abilities:
If I were in the same situation I'd choose that part for a replacement --- sources online show it's basically the same price, and why replace it with the same failure-prone part just to risk it doing the same thing again? This isn't even a high-frequency application, it's just a power switch.
All mosfets are like that.
The marketing folksbL at Toshiba just were not able get it removed, or they have had one too many complaint from folks running them near the absolute max while not cooling them properly.
All datasheets for all parts contain a clause like this.
Have you ever wondered how manufacturers get lifetime estimates? They use an accelerated aging process which basically just means they run the part very hot and back-compute how long it would take to fail at normal temperatures. This is an industry standard practice that is (supposedly, I haven't verified) well backed by science. Running a part at elevated temperatures is somehow physically equivalent to running them longer at lower temperatures.
Heat and stress degrade silicon junctions. It's just physics. There's nothing particularly sketchy or unreliable about this specific part on its own. Probably whoever designed the amp circuit just made a mistake and under-spec'd this part.
>Note that parts like this are more easily replaced using a hot air station than a soldering iron.
Not to get into a soldering skills contest, but I'd rather use hot tweezers - hot air is the only viable option for components like QFNs and BGAs, but using it always carries the risk of accidentally disturbing neighbouring components, particularly if you're inexperienced. Affordable hot tweezers such as the Sequre HT140 are now readily available from China and make removing small SMD components almost effortless.
In my personal opinion using low melting point solder means you aren't qualified to do the repair and maybe should train on broken boards first. Sure, it makes desoldering easier but it's very hard to clean completely and mixing it with regular solder compromises the joints, which is especially important in a highly thermally cycled device like a laptop. Sure, use it to remove a through hole connector from an interface board but maybe keep it away from the mainboard or power delivery..
Just use a preheater for dogs sake, although I think this part can be removed just fine with a regular iron using a bent piece of copper wire that touches all its pins, then heating that wire. Not like you need to care not to overheat it - it's dead already.
I have become disillusioned with consumer laptops in this price range. In the past 1-2 years, I have had two different Asus Vivobooks die on me just outside the 24month warranty window. In both cases, I had to pay addiotional money to then have their repair department quote me
price 110% of a new specimen, for 'repair'.
That such pricy devices can blow up irrepairably in a window of 25 months, is not acceptable.
I've had screen issues with two MBP within warranty and Apple attempt to charge 1800 EUR to fix them. The laptops cost > 3000 EUR but still. These are issues which are fixable with either a 2 EUR part or a 50 EUR part.
We had a Lenovo IdeaPad die after 7 months. (An AMD model, in 2021.)
Their support was extremely painful... or deliberately slow. It took 4 phone calls (30-60 minutes each) over 3 weeks to finally get it escalated properly, to issue a replacement.
They would say, "we're escalating" then a week of silence. Then I call back, and they say, "It never got escalated", let me do that now. Repeat, repeat.
Eventually they shipped a replacement Yoga unit. (Motherboards w/AMD CPUs were on multi-month backlog during Covid)
Note that parts like this are more easily replaced using a hot air station than a soldering iron.
Looks like the audio amp fed by the MOSFET is a Realtek ALC1304, which I can't find any data on, but searching for the pin names finds that it could be a clone/compatible to the TI TPA3131D2/TPA3132D2. This part has a power limiter, but if the Realtek one is configured the same way, then as implemented in the laptop, it has no power limit.
It is very likely that the reason the transistor failed was because of poor cooling, as it is a low cost, high-resistance transistor that is running near its design limits.
The MOSFET that blew up is rated for 2A continuous, and it's in the ~19-20V VIN circuit, so it begs the question of just how loud he was playing his music when it happened!
That said, not the first time laptops have come with sub-par audio circuitry:
https://news.ycombinator.com/item?id=7205759
https://marcin.juszkiewicz.com.pl/2012/12/10/how-to-fry-spea...
https://apple.stackexchange.com/questions/100530/can-volume-...
Straight from the horse's mouth (mosfet datasheet):
Note: Using continuously under heavy loads (e.g. the application of high temperature/current/voltage and the significant change in temperature, etc.) may cause this product to decrease in the reliability significantly even if the operating conditions (i.e. operating temperature/current/voltage, etc.) are within the absolute maximum ratings. Please design the appropriate reliability upon reviewing the Toshiba Semiconductor Reliability Handbook ("Handling Precautions"/"Derating Concept and Methods") and individual reliability data (i.e. reliability test report and estimated failure rate, etc).
[0] https://www.mouser.com/datasheet/2/408/SSM6J402TU_datasheet_...
That's... not very confidence-inspiring.
Something like the SiA471DJ comes in the same pinout and package, has almost 1/10th the on-resistance, and over 10x the current rating. Moreover, its datasheet does not have such a lengthy disclaimer about its abilities:
https://www.vishay.com/doc/?76741=
If I were in the same situation I'd choose that part for a replacement --- sources online show it's basically the same price, and why replace it with the same failure-prone part just to risk it doing the same thing again? This isn't even a high-frequency application, it's just a power switch.
All mosfets are like that. The marketing folksbL at Toshiba just were not able get it removed, or they have had one too many complaint from folks running them near the absolute max while not cooling them properly.
All datasheets for all parts contain a clause like this.
Have you ever wondered how manufacturers get lifetime estimates? They use an accelerated aging process which basically just means they run the part very hot and back-compute how long it would take to fail at normal temperatures. This is an industry standard practice that is (supposedly, I haven't verified) well backed by science. Running a part at elevated temperatures is somehow physically equivalent to running them longer at lower temperatures.
Heat and stress degrade silicon junctions. It's just physics. There's nothing particularly sketchy or unreliable about this specific part on its own. Probably whoever designed the amp circuit just made a mistake and under-spec'd this part.
>Note that parts like this are more easily replaced using a hot air station than a soldering iron.
Not to get into a soldering skills contest, but I'd rather use hot tweezers - hot air is the only viable option for components like QFNs and BGAs, but using it always carries the risk of accidentally disturbing neighbouring components, particularly if you're inexperienced. Affordable hot tweezers such as the Sequre HT140 are now readily available from China and make removing small SMD components almost effortless.
None of the images load for me. The site is https but the srcset attributes all use http. Ironically the src attributes are correct.
Same problem for me on Firefox 137.0.1 and Chrome 135.0.7049.85. Having mixed HTTPS/HTTP content is a big no-no for webmasters.
To view all the images properly: https://archive.is/5mdgs
If you right click -> new tab, you can see them.
> Update: Sept 10, 2019
Comments are from 2019 and 2020. Date is not clearly visible, but this is very old article.
In my personal opinion using low melting point solder means you aren't qualified to do the repair and maybe should train on broken boards first. Sure, it makes desoldering easier but it's very hard to clean completely and mixing it with regular solder compromises the joints, which is especially important in a highly thermally cycled device like a laptop. Sure, use it to remove a through hole connector from an interface board but maybe keep it away from the mainboard or power delivery..
Just use a preheater for dogs sake, although I think this part can be removed just fine with a regular iron using a bent piece of copper wire that touches all its pins, then heating that wire. Not like you need to care not to overheat it - it's dead already.
Preheating is good advice if the rest of the components can handle it. Otherwise, shield them and hot air gun.
I clip the leads to dead chips then there is no heating issue when cleaning the pads.
IPC std certified to 0201. You can find many videos online.
Tin based solder will whisker over years time which can be a concern.
https://wikipedia.org/wiki/Whisker_(metallurgy)
I have become disillusioned with consumer laptops in this price range. In the past 1-2 years, I have had two different Asus Vivobooks die on me just outside the 24month warranty window. In both cases, I had to pay addiotional money to then have their repair department quote me price 110% of a new specimen, for 'repair'. That such pricy devices can blow up irrepairably in a window of 25 months, is not acceptable.
I've had screen issues with two MBP within warranty and Apple attempt to charge 1800 EUR to fix them. The laptops cost > 3000 EUR but still. These are issues which are fixable with either a 2 EUR part or a 50 EUR part.
I have a yoga from 2016 that had its backlight supply rail blow up. Very lame.
Title (2019)
We had a Lenovo IdeaPad die after 7 months. (An AMD model, in 2021.)
Their support was extremely painful... or deliberately slow. It took 4 phone calls (30-60 minutes each) over 3 weeks to finally get it escalated properly, to issue a replacement.
They would say, "we're escalating" then a week of silence. Then I call back, and they say, "It never got escalated", let me do that now. Repeat, repeat.
Eventually they shipped a replacement Yoga unit. (Motherboards w/AMD CPUs were on multi-month backlog during Covid)