Warranty shmarranty — toss the phone in the oven! There’s apparently a problem with the assembly of the Nexus 5X smartphones, and it looks like it is due to faulty BGA chip soldering. LG USA has had enough problems with the phone that they may not even have enough parts or new units to fix it, so they’re offering a refund. But we all know how it is to get attached to a device, right?
So [Alex] disassembled his beloved phone, pulled out the board in question, and gave it the XBox Red Ring Of Death treatment. He placed the board on some insulating aluminum foil, and baked it for six and a half minutes. Season with lemon and pepper, and serve! We’re honestly surprised that sticking the affected board into the oven at 195° C / 390° F for a few minutes would work at all. Isn’t that a low temperature for soldering, especially with a lead-free mix? Could it have been a problem with humidity after all?
We’d probably take the money and run, honestly. But if LG is offering refunds anyway, and you’re able to open the case cleanly, we might go for it — just maybe not in the (food) oven, because there are better places to get reflow done. Full disclosure: we’ve personally pulled a Thinkpad T41 back from the grave, and we’ve seen people fix everything from an HP Pavilion to an entire MacBook Pro, so it is possible to succeed. You might practice on something low-value first. (And why in the name of all that’s holy can’t anyone properly assemble boards with BGA parts?)
I had similar issues with the G3 about 3 months after its warranty expired, and did a similar treatment. This worked for a little while, but within a week, it was failing again.
Exact same issue for me. Baked it as suggested and it worked great for a week or so.
Same with me and PS3, after fix it worked for a month and died. Reballing doesn’t help if chip is damaged, for more information about reballing fix myth I suggest this video : https://www.youtube.com/watch?v=1AcEt073Uds .
I was going to mention Louis. Thanks! Don’t waste your time baking something in an oven.
Didn’t need to even click to know this is Louis’s video. At best, oven treatments are just giving you a temporary fix for a problem that’ll become permanent the more you thermal cycle the poor PCB and chips. If it ain’t hot enough to melt the solder, you’re fooling yourself.
I baked an old dead macbook in the oven about 4 years ago. Came back to life, still going strong!
The truth has been spoken.
Second one this week? I hope next we can get a pile of DV9000’s, a PS3, and maybe a few video cards from Core 2 era iMacs. I love the site, but my friend, you are becoming part of the problem.
I had an LG G3. It was one week out of warranty. The screen died and it is due to faulty soldering on the main board / digitizer. They wouldn’t repair it under warranty. Gave it the oven treatment and it’s now a backup pone. I’m glad to see they’re still struggling with making quality reliable phones and I hope it hurts their bottom line. In the mean time I bought a Nexus 6 and am very satisfied with Motorola / Lenovo’s support.
What temp and time Aaron? I’ve got the exact same phone and problem.
give the very low temperature my suspicions for the now working phone may lie with the heating causing some expansion and subsequent contraction in the joints enough to close the suspect joint, rather than reflowing as such.
good on him for giving it ago
I ran into the boot loop myself
Luckily, I paid for the 2yr warranty
LG support assured me the new ones don’t have this issue but we’ll see
Also, I used a sliding adjustable clamp (https://north40.com/media/catalog/product/cache/1/image/450×450/9df78eab33525d08d6e5fb8d27136e95/a/d/adjustableclamp_jorgensen_isdbar_581819_a.jpg) with a rubber bumper to apply pressure to the CPU/Mem package (offset from fingerprint sensor) so I could boot to recovery (TWRP) and back up my data.
I suspect micro cracks in the solder hence low temp might be sufficient or it maybe tin whiskers
Yeah, microscopic cracks FTW. Possibly tin whiskers though. RoHS is a real shame when it comes to lead-free solder. I’m pretty sure we end up tossing a lot more electronic devices, and probably causing at least a little more environmental damage, when we use solder that fails in places where it can’t be fixed. Heck, in my repair work I even notice very large joints, (on relays, inrush limiters, wires, and such), that are obviously failing, not because of bad soldering, but because of bad solder.
Much of RoHS is terrific, but we really need to find ways of safely using lead-based solder again, or this kind of thing is going to keep happening.
RoHS and it’s elimination of Leaded solder is a curse. The two largest contributors to lead in e-waste are already dealt with… CRTs have been obsoleted, and lead acid batteries already have an exceptional recycling program in place to reuse the materials! Lead-free solder just amounts to more e-waste than ever before.
I’d easily pay a premium for things like motherboards, GPUS, phones, tablets, etc made with leaded solder. Screw the EU and their RoHS regulations. Obviously did not do the research, and rushed an “environmental law” into place.
I worked with an engineer who claimed that RoHS had killed more people than lead solder.
I would not be surprised and I say it also have caused more pollution than lead solar as well as the mining of the lead alternatives like tin and bismuth is not exactly the cleanest operation esp in developing countries and the coatings they use to try and control corrosion and tin whisker can be worse than lead as far as toxicity go and same goes for what goes inside tantalum capacitors and many semiconductors which we now have more of ending up in the waste stream.
Yah I wonder if any of the people in the EU who came up with the RoHS ruling were engineers,or chemists as they should have known about the cracking and tin whisker problems.
I have a feeling the answer this question is probably no.
I think RoHS might be doing more harm to the environment than the problem it was supposed to solve and at a the very least it’s a useless regulation.
if a device doesn’t work, put it in the oven. if that doesn’t work put in the freezer. if that doesn’t work pay someone to put it in an ultrasonic cleaner. if all else fails put it in a bag of rice.
Don’t forget to sprinkle a couple sessions of percussive maintenance between steps to further encourage it.
I have a special anti-static grounded pillow case i use for data recovery. Every friday afternoon everything that comes in for the week, i toss in and bang it off the wall. 30% success rate every time.
Customers’ face in shock horror…. “But my data!” *Thump* *thud* *thump* *Click-hiss-zzzzzib* here is your data Sir!.
Customer goes home…. Next day brings a pile of broken machines… “But I saw you throw stuff at the wall to fix them”
Make sure you use extra insulating aluminum foil as well.
Ive considered wrapping it around my brain to insulate myself from otherwise intelligent people that think their PCBs weren’t baked long enough at the factory.
My brain read insulting aluminium foil because it refused to accept such a ridiculous sentence.
A reflow usually doesn’t last more than a few weeks, even when it’s done with proper tools.
The only way to go is reballing, but if the real cause is a design flaw, it won’t last no matter what.
Capacitors plague and BGA plague are the most common failures for years now.
If only there was something we could add to solder to make it more malleable and elastic, and also prevent tin whiskers.
Lead seems ideal, but I’m sure smarter people than I are already hard at work on this problem. /s
+1000
I have used a metallic epoxy to fix errant solder issues. This is not new technology by any standards. Probably comes from being a 60 year old techie with a manufacturing background.
Got some lead free low-tin high copper content solder I’ve long-term permanently borrowed from work. Now they only have the normal ROHS stuff that doesn’t solder well.
Even the low tin stuff needs a little encouragement to solder shiny joins. Most, especially the common cheaper solder just coat over during cooling… annoyingly. no matter what. Tried alsorts with varying results. Though the final product always outlasts their warranty by a huge margin.
Also seen an alarming number of fresh-from-manufacturer PCBs with the typical ROHS grey cold solder joints on every join (well over 80% from the lead-free era PCBs).
I have a lifetime supply of lead/tin solder.
I use only tin/lead solder. Even in the lab at the work place I insist on “the good stuff”. Hopefully we do not have to order it over the darknet at some drug dealer site or start to make it ourselves in the near future.
Fine, so if we promise not to eat our phones, can they please start using it again?
Yes please!
@Annie: I have the impression this happens anyway – at least some of the kids seem to be really stupid and can’t do much more than playing with this phones. Independent of the solder.
Between RoHS solder and BGA packaging I don’t know which is worse.
A mobo will try its best to be a parabola (or something?) for not being heated uniformly and will squish the balls after melting them so they run together, in the worst case. It takes an oven to heat it corner-to-corner but a phone… no problem, it’s easy enough to get by with a reflow station. I reflowed a G4 that way about 6 mos. ago, no idea how long it lasted. Too bad the guy tried a factory reset first so the freshest woulda-been-recoverable data was already gone :(
Nexus 5x phone is the worst mobile phone I ever had.i request don’t waste your money to buy such third class mobile which has no market value .and I regret myself to buy such mobile in my life time……
I’m designing automotive ECUs that use BGA microcontrollers.
Guess how many BGA soldering failures we had so far in the field?
None.
It’s just a matter of money, as always.
Well read my comment below. My experiences with BOSCH ME9.7 series (Merc M272, M273 and derivatives) are on the extreme opposite of what you say – the failure rates were so high that sometimes we had two or three of these a week, coming in for the same problem. The things were getting south after being in the field for 3 to 5 years (well, isn’t that just above the period when manufacturer actually gives damn about repairing it?) and in a 5-year car you had high chance of getting a related problem; they were ranging from signaling open circuit on some unimportant sensor (like a post-cat lambda, throws a CEL on you and people just tend to ignore it as long as the car drives OK) through opens on the injector control circuits to units not booting at all. If you have seen how the ECU is installed in the vehicle you might have a clue of the real source of the problem – they weren’t hardened enough against extreme temp changes; the ECU sits in the engine bay, bolted right at the top of the magnesium intake manifold, in a cold winter like we do, you are likely to get well over 100 Celsius deg amplitude. And I believe that’s why the BGAs were coming off the board – our temporary fix was a long extension cable that allowed to move the ECU further from the heat source.
I don’t know the company you working for, but hopefully you are emplying better design practices – not only about the ECU itself but also bout the way it’s integrated in a vehicle.
Trust me when I say no one wants that mode of failure. It’s a absolutely massive liability.
automotive sector is exempted from ROHS :-)
Nope
– Guy from automotive
If you are going to reflow, at least inject some no clean flux underneath the chip (or clean flux if you are able to clean it).
Just tried it on a bricked 5X it’s working fine for now. Not sure how long it’ll last though.
While indeed the temps are bit on the low side for RoHS solder to fully liquefy, the fix is only partially about reflowing the solder. It’s more about even heating and disposing of the tension “trapped” over the area of BGA chip. For that you do not need the solder to fully flow. Also, the heating creates a “cycle” of extreme expanding and shrinking and that by itself is often sufficient to get the things in contact. And also contributes to short lif e of some of these fixes, so I am curious how long the thing lived after the fix. I was in business of automotive ECU repair and we faced this problem with one of the gasoline ECUs made by Bosch for Merc. These things are made sturdy and have much higher temp ratings than the consumer devices so its not a problem to give it really thorough heating. All the components have automotive rating and the ECU itself is actually bolted to the engine so it gets really hot when operating. Back to the fix, we found that a full baking for a full reflow somehow yielded poorer result than a “partial” reflow. Sometimes full reflow worked, but we were getting high number of ECUs that were permanently disabled during the procedure. Therefore we sticked to lower temp baking combined with applying controlled, uniform pressure to the top surface of the suspect chip. Much better results and we stopped killing the modules in the process.
Never heard of this issue. Does it only affect the american model? Anyway, soldering issues apart I do absolutely love the Nexus 5X as a fine (cost/performance ratio) and I do really like the design with the single board, it is very easy to open and service even with very little tools. The replaceable TRS receptacle is something I would never expect in a product from the year 2015/2016(since everything is meant to be disposable nowadays).
My girlfriend recently dropped her in the water and it turned off by itself a few seconds later. Left it to dry for several days and tried turning it on. Nothing. Connected to the charger it would turn on and off repeatedly.
I ended up buying a phone with a broken screen on ebay to replace the board, but I should have known better and in the end the real issue was only the battery (€12). Apparently the phone will not turn on (and will keep resetting when connected to the charger if it isn’t able to identify the battery), I guess one of the IC’s on the board that’s part of the battery package went bad after the water go in.
Happy ending after all, the phone is back to life.
My 5x got prompt and complete repair by LG after a January boot looping cycle couldn’t be fixed via software.
Screw LG and Google. 5x cells are failing like crazy and neither company is doing a thing about it. Need more articles like this to let owners know what a POS the 5x cell is.
could we stop flooding the media with nexus 5x death stories? i’ve got enough sources of anxiety in my life as it is
Just fixed my iPhone wi-fi over the weekend using the Australian sun. Put it on the car dash for an hour in full sun. 37C outside, I’m guessing 70C on the dash – I couldn’t pick the phone up. Clearly not reflow but it worked.
Probably just the battery swelling until it press fit everything back together in there :P.
I’ve only had iPhones that colleagues at work brought up to me to resolder the WiFi chip. So far there were 4 of these boards on my desk. After reflowing the chip with our ERSA IR heatgun-station all reported back sucessfully, and so far only one failed again about 8 months after the “hot-fix”, and that was tossed to the trash at that point.
The problem with using a cooking oven is probably that you can’t really trust that temperature setting you put it on. If you set it to 195°C it may very well go over 220°C inside depending on how high the hysteresis of the regulation and how uniform the heat spread inside the oven is. Best idea would probably be to make a dry run on the oven without the board, stick a thermocouple probe in there and record the true temperature curve before you try to “resolder” something
100% agreed. That’s the primary reason I lowballed it. Depending on the location of the oven’s own temperature sensor is probably overshooting by some unknown margin so my plan was to start low and do various heats until it worked. However I got lucky first time. I probably should have documented this in the write up.
In this case of questionable repeatability I would sacrifice some more minutes of my time and stick a thermocouple in and watch and regulate the process of reflowing directly.
I have similar issue with my Nexus 5, where it’s not booting properly it’s stops at Google logo. Can some one help me out of this.
Thanks for sharing. This is really helpful. Keep up the good work
The phone probably stopped working because of the liquid damage as the LDI sticker has turned red/pink indicating it has got damp.
Baking it would drive the moisture from the PCB but wont stop the corrosion that is happening to the internal tracks within the PCB.
This is why it makes me laugh when I see people offer a fixing service for liquid damaged phones as it is impossible to fix them, all it can do is offer a potential short term repair.
Its also why I never buy a 2nd hand phone.
I tried a similarly approach with my much-loved but hard-to-replace small-profile Apple USB keyboard. The company only made them for a year, so they’re pricey used. I tried multiple times in the freezer. No luck. I tried a pass through my dishwasher. That only offered one benefit. A few bizarrely behaving keys became a keyboard was totally broken. Nothing left to fix.
Extreme cold worked much better with a travel pack that’d acquired a ‘old sock’ smell that no amount of cleaning could remove. When a cold wave sent temperatures plunging to 8 degree F, I left it outside for a couple of nights. No problem with that bad smell since. Cold may not work with electronics, but it works quite well with organic life.
I once revived a hard disc which did not want to startup with a hair drier. Perhaps there were some bad electrolytics, if you get them hot the internal resistance is (temporarily) reduced. It worked as long as I kept it running. Of course I did this just for data recovery and scrapped it afterwards.
This article totally saved my Nexus 5X. Thanks a lot man!!!
From complete brick to working machine! Thank you!
Just wanted to say that it worked for me as well. Followed the instructions on the “dev-with-alex” site and it revived a very dead Nexus 5X. My friend had a bunch of files on the phone which hadn’t been backed up and was willing to risk voiding the warranty to get them. Not sure how long the phone will work now but the solution was perfect for what I wanted to accomplish.
Many thanks for the instructions, managed to revived my dead Nexus 5x yesterday. So far it is till working. will update how long it will last.
This article totally saved my Nexus 5X. Thanks a lot man!!!
Flexispy, have you done the “Nexus hot-fix” with an oven or a heat gun? Thanks
I’ve just tried this method + straight to the freezer and it works for now. 5 min in the oven (even 3 did the trick) and then straight to the freezer for an hour (just to be on the safe side) and that’s it – I got my phone back!
Up and running! I´ll never put my phone int he backpocket again…Thanx!!!