Why I Drilled Holes in My MacBook Pro and Put It in the Oven
ifixit.org
ifixit.org
A few months ago I disassembled and cleaned a grape juice damaged MBP. Tried it afterward and no dice. Tried it yesterday on a bored whim and I now have a working MBP with a sticky keyboard. Why? I have no idea.
My guess would be that grape juice shorted some of those connectors. After few months it dried sufficiently, such that when you reseated them short went away.
[0] And by damage I mean mostly short and/or corrode the metal.
Six months from the first incident, and I randomly plugged it in and it booted instantly - I was ecstatic. It lasted two weeks before it wouldn't boot up again. A week later, it booted up again, and has been fine ever since.
Beats me man. Apple "Geniuses" wouldn't touch it and wanted me to buy a new one.
Regarding the "Apple Geniuses", I doubt there would of been anything they could do for you, other than sit there with a cuetip and attempt to wipe down your entire mainboard after a full disassembly. Not to mention water damage probably isn't covered...
And no, water damage wasn't covered by warranty, but it was out of warranty anyhow. I just didn't want to pay another $2K for a MacBook Air (maxed out specs at the time).
But I did buy the new one and let them have the cracked one.
Use an actual dessicant. Which might also not work, but then your problem is less likely leftover wetness.
And that works :)
Also "The speed holes worked: The boot chime rang. The screen glowed. The fans blew." doesn't make sense to me. The extra holes may help and may even fix his problem, but I find it hard to believe that successful booting shows that. Yes, it is a 'little' early for a final verdict.
From what I know about heat treat ageing of metals something similar may be happening. Because he is pushing his system through above average temperature cycles, this causes microcrystalline changes in the solder. Ageing at even a temperature below reflow/melt temperatures can help restructure the grain.
If he didn't run his system at such a high temp and then eventually turn it off to cool so much then he might not have the problem to begin with. As a side note, personal experience building pcs for many years has shown me that pcs that sleep/get powered off don't last as long as ones that stay powered on 24/7.
When I solder electronics, I'm usually somewhere between 500F to 650F, but that's so things get hot enough quickly enough to not burn. I don't doubt a slow bake at 340F would do the same thing... some online googling shows people doing reflows with toaster ovens in the same temperature range (340F-450F).
It's likely the solder joints were not completely disconnected, but rather simply cracked. So heating them a bit would allow the crack to reconnect, at least for a short while (seems to explain his symptoms of repeated failures after extreme system temperatures during operation).
If it fails again, one possible next step would be to grab a multimeter and test the resistance across each component on the board. If the resistance is infinite, that might be your problem component.
Also, the problem might as well be one of the numerous chips in BGA packages whose connection points you can't even see, let alone- probe.
I'm not sure whether or not cracks can heal at temperatures below the melting point, but I would be surprised if that were possible.
The whole procedure is wild, though, and I wouldn't recommend trying it out. He was lucky that the board didn't get destroyed. Most SMD components can withstand temperatures above 250C for tens of seconds, not necessarily for nine minutes.
I baked a Nvidia GTX470 at 390f for 10 minutes and it eliminated the artifacts/crashing and has been working fine for months.
It probably won't work forever, but it is at least a temporary fix.
https://www.change.org/p/timothy-d-cook-replace-or-fix-all-2...
It's ludicrous that to attempt to fix this problem, you would have to go so far as to bake your laptop.
It cost me over $US 500 to get this fixed at an Apple Store.
When a friend was buying her mother a christmas present iPad, I went with her to examine which models handle the tablet workload better. We ended up picking an older generation, non-retina model. A few previous generation retina models, presumably running latest iOS updates, had ridiculously high temperatures just idling on the shelf.
It's not a BGA solder issue like we've seen on numerous laptop GPUs over the years, but a problem with the internal interconnects. Heating it, as if to reflow a BGA does temporarily solve the problem, but in my experience, the period of time it keeps working after heating gets shorter each time.
[0] http://www.tomshardware.com/news/nvidia-g84-g86-chips-overhe...
http://europa.eu/youreurope/citizens/shopping/shopping-abroa...
Finally, just a few days ago -- right before Christmas -- it just up and died. Seeing how I didn't want to pay for a new logic board with the same design defect, I just ended up buying a new laptop (and no, it is not a Mac).
My early 2011 MBP was probably the most expensive laptop I've bought in a long time (the most expensive was the Powerbook 170 - oh how I loved the trackball on that thing), and I loved the form factor and expandability. It really bugs me that it died while it was still more than usable in terms of performance.
I ended up selling it for whatever I could get for it a month ago and just cutting my losses before the Applecare ran out.
If the balls were factory tested for solidity this would become far less of an issue. Industry papers on the manufacture of solder balls indicate they are not; when those balls were lead, it didn't matter because lead's not thermally brittle.
I'm not saying it is a good fit for everyone, but it works out well for me. I do recommend squarespace for individual parts of a self-built tower pc. I spent 120 on a motherboard and on the third replacement spquarespace upgraded me because it turned out that model had an almost 50% failure rate in the first year.
The insurer will calculate the risk of whatever happening, and obviously charge more than likelihood * cost.
So unless your risk is higher than average, you're losing money. And keep in mind that the average insurance taker probably has higher risk than the overall population.
By this logic, nobody should buy life insurance because the insurance companies have calculated their premiums so that they come out ahead. At least on average.
If you flip the statement, it's "do insure against things that can ruin you".
Whether life insurance is worth it depends on what you mean by that exactly. You can insure your untimely death with a benefit for your family, but there are also contracts that are more of a form of investment, and combinations thereof.
While arguably once you're dead, you're, well, dead, so you personally might not care that much anymore, your family might. And for them the loss of one of the primary earners of the family is probably ruinous, so it's reasonable to take insurance against that.
The investment case is different, it's essentially just an investment contract with associated cost. In that case, you're just buying a service (managing your investment, and to some degree insuring against investment risk). The two forms are commonly mixed together, and then it depends a lot on the structure and cost of the contract. YMMV, but I think over here these contracts have very intransparent cost structures and are commonly more expensive than getting a plain life insurance and a separate investment contract (or investing yourself, for that matter).
It is also about making insurance companies fabulously wealthy.
So everyone definitely needs insurance for healthcare.
Losing a smartphone isn't going to ruin me, but it sure would piss me off to have to fork out £500 or whatever because it gets jacked. Similarly, someone could break into my house and clean it out. Not going to ruin me, but I'd rather not have to replace everything.
The whole point of insurance is collective risk-spreading. Yes, this means that the majority (by value) of people with insurance will have 'wasted' money, in the sense that they purchase insurance without claiming. But that's not what they're purchasing – they're trading off a small, certain cost for the elimination of a large, uncertain cost. That's a purchase of 'certainty' or 'stability' – and it's what they get.
In other words, since you have no reliable way to understand risk and likelihood of a particular event happening, it's not possible to decide if you are at a higher risk than average.
Go talk to an _honest_ insurance broker (not some State Farm or other company you see on TV commercials) or better yet a financial adviser. You should reserve an insurance claim for catastrophic issues only. You should also have max deductibles. Forking over 500 for a lost iPhone is a MUCH cheaper deal than paying insurance premiums for that coverage and then the INEVITABLE rate hike because this claim appears on your CLUE report.
I made the same naive mistake as your assumptions when I was young. Lost a 1k watch, made a claim, went about my jolly day with my nice check for $750 (250 deductible). 1 year later when I go to buy a house I am surprised by how much my insurance rates are. FOUR years later after that claim fell off my CLUE report my premiums went way down. I paid the insurance company more than the $750 I got from them.
I was lucky in that my neighbor was an insurance broker and told me low deductibles are taking advantage of suckers and the lesser fortunate who are scared into these policies. What he advises most of his clients to do is take the highest deductible they can, then put aside that amount of deductible in some low earning liquid account. Now when you have an accident you "pay your self" and keep your insurance rates low.
Don't be a sucker.
I'm in the UK, so I can't help but assume there are differences in the insurance market. I have a comprehensive insurance policy that covers buildings, contents and accidental damage, and that covers things like smartphones too. Despite a couple of previous claims for stolen and damaged electronics, the rates are pretty good. The marginal cost for accidental damage is minimal.
But yes, low deductibles are a sucker's game.
Don't be a sucker.
As a counterexample, for all but the richest people having catastrophic health insurance is a good idea.
If you can afford to replace the item yourself you can instead spread that risk over time (and multiple items) by essentially self-insuring.
I absolutely agree you should buy insurance for "large, uncertain costs", but a smartphone or laptop is not that for a lot of people.
Healthcare and liability (auto, homeowners/renters, etc) are two examples of actually "large, uncertain costs" for most people.
Consider also the overall effect if you invested the premiums instead of paying them.
I know when you're young 45 years seems like it'll never happen, but if you're lucky it does.
Have tons of cash? You should still probably look at a cheaper plan with a super high deductible.
In fact, my laptop is a $500 one I bought 2 years ago for travel use and didn't want to be too upset if it was lost/broken/stolen.
My previous one was 10 years old and was a bit heavy :-)
Next to me a woman had brought in her laptop because of problems (did not overhear what), and they replaced it for free despite it being 3 years old.
I definitely agree that the first step with Apple stuff is to see what they'll do for you. Hitting the Apple store is the best way to do that, but they're not everywhere. I wonder if the mail-in folks have as much latitude to please the customer. (And of course if you mail it in, you're missing your computer for at least a week or two.)
I've had 1 day turnaround from my local Apple store on a repair. On another repair (replace failing SSD) they fixed it in a few hours.
They also replaced an iPhone 5C screen in a few hours.
Well yeah, that's a problem as thermal paste is only meant to fill in the microscopic irregularities of the contact surfaces and isn't some magical thermal bridge. Thermal paste has thermal properties on par with most condiments. Thermal pads are particularly bad. I would use a metal shim to bridge such a gap, not just more thermal paste.
As a side benefit, Apple could begin referring to the machines as "laptops" again in their marketing from time to time (which they have abstained from for at least the last 5+ years).
The liability issues are just too great. We're talking class action levels of expenditure.
[0]http://www.webmd.com/infertility-and-reproduction/news/20041...
I don't understand this claim. They are still calling hot devices laptops today. Do you mean that at some point in the future they will stop, and then they will never again call these devices laptops?
It was funny watching Apple a while back market its products to consumers as a quality item, but tell the government that it was 'just another computer', because they didn't want to honour their warranties past 1 year.
[1] https://www.apple.com/legal/warranty/
[2] https://www.apple.com/legal/warranty/statutoryrights.html
[3] https://www.apple.com/legal/warranty/products/uk-ireland-uni...
Laser drilling is an option too:
https://www.ncms.org/wp-content/NCMS_files/CTMA/Symposium200...
It might be also I don't game on laptops...I save that for a desktop with a GTX670 card and plenty of fans..
The reason this matters is because the process by which those solder balls are made does not check for the presence of voids in them. At least 10% of these balls (regardless of composition) x-ray as having voids.
GPUs go through the most radical thermal shifts of any surface mounted component. Lead has a sense of humor about that kind of temperature change; tin alloys do not and begin to crack after a sustained number of shifts. This happened on my 2011 MBP which was not used for gaming.
It's worth noting that the iMacs made the same year, where the discrete GPU was a replaceable daughtercard, had a voluntary recall for exactly the same reported video problems. The irony here is that the average cost of having a GPU reballed is about $150US, with a high rate of reliability compared to reflowing. Part of that reliability comes from reballers refusing to use anything but lead.
The EU is directly responsible for this problem but Apple is responsible to its users and with a cash reserve outstripping the US Treasury's, could have absorbed the cost of repairing these MBPs easily.
I've changed enough capacitors on power supplies to know the damn things have solder that's practically indestructible, our old Pace solder station iron cranked to something like 850F and still I'm sitting there 5 or 10 minutes cursing it all the time, and nine more caps to go x two leads each.
Plus that heat would degrade the life of components too so who knows if it's ever worth it. But I guess cheaper than spending thousands on a new Macbook Pro.
This is the solder that comes from the factory we just use the lead-free regular stuff when it goes back in.
Intel had long been pushing for cooler chips and longer battery life. Beginning with its Centrino line and the Pentium M processor.
Apple hopped in right after the Core chips landed (dual core), and these chips were amazing for Apple. They could continue with the macbook pro design with a new and faster chip.
Then Intel went on a speed binge again and tried to yield the most bang for your buck, sacrificing thermals and battery life.
Now finally with the Haswell chips they took a step back and tried to improve battery and thermals. And hopefully Broadwell will be even better.
Putting vents at the hinge actually makes the most sense. It guarantees ventilation ports won't be fully obstructed. The vents face away from the user, reducing fan noise. Air ducts are short, since that area is close to the CPU/GPU.
It's hard to cool a laptop that can generate 85 watts of heat. Moreso if one cares about it looking nice and being thin. You may not like the trade-offs Apple chooses, but they put a lot of thought into their industrial design.
I have an rMBP 15" and a 15" Lenovo Y500. The Y500 cost less than half of the rMBP for equivalent performance but the experience, battery, screen, trackpad, etc are of course much better on the Mac. But for intensive tasks, even with nearly identical internals and the same OS, the Y500 is far more pleasant to use. Say what you want about Lenovo and the terrible things they've done to their laptops in the last few years, but the Y500 is designed to expel heat in the least obtrusive manner possible and it does an amazing job.
At full bore, all cores, maybe a game, the Y500 happily pushes a lot of air out the side vent, and the keyboard, trackpad, bezels, etc., stay very cool.
With the same workload, the rMBP gets very, very hot and seems to blow a small amount of air out of the keyboard (?), occasionally burning my fingers, It's loud and very uncomfortable to use.
In a sense it definitely makes me feel like I overpaid for the extra speed, since I can't use it comfortably anyway.
So, when it's on my lap it doesn't get enough airflow and burns my leg.
So much for 'sensible' design.
Also, Apple could place their CPU/GPU wherever they wanted.
Critics hold Apple to a higher standard than other companies, and the media reports Apple's failings more than other companies. There are tons of examples: iPhone 6 bending (it's just as sturdy as other phones), solder separation issues (rare; limited to a few first-gen models), overheating (again, rare and usually first-gen models). Other manufacturers have similar issues, but you never hear about them because they don't get eyeballs. Few care if some model of HP laptop overheats.
In general, Apple makes quality hardware. If it were otherwise, they wouldn't be able to charge so much.
Edit: I have no idea why this comment has been downvoted. Anyone want to clue me in?
I have three macbooks, two airs and a 13" mpbr, and none have had any thermal issues at all, even in the heat of summer.
If you are using a laptop in a way that generates that much heat, you probably get about 1 hour of battery life, which sort of defeats the purpose of having a laptop in the first place. A larger fan or heavier case material would also make the laptop less user-friendly for the vast majority of users.
I'm sure Apple has done some pretty exhaustive (no pun
intended) thermodynamic analysis of the case design.
All they do is "cooling" the aluminum bottom case, which has no contact (or though holes) with the inside parts of the laptop generating heath.
The fact that you can only get a discrete GPU'd MBP now if you buy the most expensive model while the Iris Pro is capable of doing <i>everything</i> a discrete GPU did indicates that Apple's cut their losses with this technology. The motherboard's fine. The decision to accede to EU environmental mandates was the point of failure.
No one in the industry sufficiently tested leadfree solder in GPU BGA mounts before going into production and to be fair, it was a year before people started noticing failures. Apple's dismal head-burying response only made it worse.
2012 Retina MBP are generally good but have some known issues (eg LG display losing pixels). Mine's currently getting a bunch of things fixed which will probably end up being a new keyboard, logic board, and another LCD.
2013-2014 rMBP seem to have the 2012 issues fixed.
Airs are ok so long as you don't intend on flogging them. They can have overheating issues too.
EDIT: I must also state that it has a plethora of ports on it that are lacking on the retina models. Gigabit ethernet (with AVB for audio, hurray), Thunderbolt (or Mini DisplayPort for driving a screen/TV with audio over it for HDMI too) FireWire (for audio interfaces), two USB3 ports, SD slot, mic input, audio out (headphone or optical S/PDIF), button to indicate amount of battery remaining (so you don't have to turn it on to find out) and a DVDR/CD drive. Can't beat them in my opinion.
I chuckled observing the MacBook Air user with his bag of appendages for ethernet and powering a display. Sort of defeated having the super slim device!
I've been running on MBPs since 2008 and never had one give up. (My oldest one still is running as a test device).
I personally own a late 2011 MBP (same as the article, I believe) with 8GB RAM, also no problems, running a pretty standard development set up.
I have another 2012 MBP through work which is a bit beefed up, though still nothing fancy. Also no problems. Also manage a couple of 2010 Mac servers there which have been running steady since late 2010.
Oh, and none of these have SSDs. So, say what you will about non-SSD, they work well enough for development.
All in all, have to agree -- whatever fits your budget, at least in my personal experience, will work pretty well. I know that people do certainly have issues with them occasionally, but when I compare my luck with them to my luck with various other brands (Compaq, Dell, etc), they've lasted significantly longer with few or no problems.
It amazes me that Apple still ships computers with spinning hard drives in 2014, let alone a laptop.
Its a 2011. Trying to sell that in, lets say, 2019, it'll sound a lot like when I sold my 1997 Saturn in 2013 with a hole for a ham radio antenna, in other words it'll be so old no one will care. Also they were more interested in the cracked heater core, broken sunroof opener, leaking sunroof, worn out brakes, and high oil consumption, than being worried about a little hole.
It's not that hard to understand.
The first thing I would have tried is to replace the fans. I had to do this with other laptops I have had in the past (HP mostly) but never with the two models of MBP (and I run pretty hard too), so I guess it has to do with the model as he mentioned.
While I sympathise with the idea of improving the design of a product massively consumed, I first try to think that the engineers did a well design and try to look for the malfunctioning component to replace.
I try and make sure it's ventilated properly (when I use it on my lap I often hold it in the air when the CPUs are busy and the temp is rising), but I'm not convinced the design is good for cooling in general.
On a serious note, I'm not sure what it means for a cable to be corrupted.. wouldn't this just mean a bad cable? Could you elaborate?
Also, at what point is he willing to acknowledge that maybe this thing is not all that well designed, especially going by how widespread this problem seems to be?
Shouldn't be much in the way of fumes. The circuit board goes through a similar process at manufacturing time. After that, the board is washed to remove excess flux and other gunk. Assuming the OP didn't leave any stickers or such on the circuit board, it should be fine.
As a note to anyone thinking to do something like this... you should only try to reflow the board itself. Take everything apart. Do not, under any circumstances, heat up any kind of battery, or else bad things will happen.
Perhaps the bigger question is at what point will Apple acknowledge that it is a real problem?
I don't really know why it gets hotter over the year. I guess fine dust end up clogging it, and since laptops are very slim, everything is packed, so it might be pretty hard to remove that dust. Some dust can stick to some surfaces, but I'm not sure.
One thing for sure is that I wont use it for gaming or anything that might get it hot.
Instead of putting it an oven, I turned it upside down, put a heating pad on it and placed the whole thing under a blanket for an hour. Turned on just fine after that.
Heat/GPU issue killed my first 17" Apple Macbook Pro. Just as 3 year AppleCare ended.
Heat seems to be the reason ethernet in my mac mini randomly dies every few days and the fix is rebooting the switch it's connected to. https://discussions.apple.com/thread/1341108?start=45&tstart... I believe a heat sensor either correctly or wrongly detects high heat and disconnects ethernet. It's never shutdown completely though.
My last iMac screen was burned in just as 3 year warranty ended due to high cpu heat over the 3 years. The burn in happened right where the cpu was located.
And current imac is even thinner, meaning less space for air to flow, which is why I'm forced to buy Mac Pro for my video editing. iMac would be more than enough for my need but I know the monitor will get burn in after 3 years and so i will just go for mac pro. At least it has good cooling system or so I hear.
Well played apple. Well played.
This work might also be useful to keep 2011 MBPs with a damaged dedicated GPU working with just the Intel card. If you have a 2011 MBP and are willing to spend some time developing a software workaround that at least lets you use it with Intel only, check out https://github.com/ah-/gmux-scripts/issues/1#issuecomment-68... and https://github.com/codykrieger/gfxCardStatus/issues/150
Granted, it took a lot of "let me speak to your manager" and I believe needs to still be under Applecare in most instances, but it's always an option to pursue if you have the time and patience.
Thankfully the 2010 I received lasted until this year with no issues, and I finally replaced it after beating it to hell with a 2014 retina MBP that's been running smoothly so far.
They must have applied the heat transfer compound better than before, because with SMCFanControl running a bit higher than normal I haven't seen it go above 65 degrees C much now (admittedly I haven't done much heavy lifting with it yet) and feels cooler to the touch up the top.
If doing anything slightly intensive, my 15" MBPR is the same; I've had it hover at 100-105C for periods. Why not scale down the CPU at crazy temperatures? Intel CPUs have a cutout, but the entire machine needs a more gradual solution. Running at 90C+ is not viable long term.
Now it's over 2 years old, I might just buy a entry level MBP instead rather than maxing out as I usually do. They seem fast enough now and what's the point if I can't even use the full power?
I have reapplied thermal paste (which improved things a little bit), used a cooling laptop stand (useless), fiddled with smcfancontrol and other applications, to no avail.
But I'm not drilling holes on the case, as someone said it would be hard to explain when reselling it :).
Are you kidding me? What's the point of even buying the faster CPU then? You could just, you know, design the computer properly. You hive-minded Apple fans are so unaccustomed to critical thinking that you don't see Tim Cooks writing on the walls.
For temperatures near 100C the conversion is easy.
Ovens report temperatures in F, computers in C, shrug.
In general you use F for human temperatures (air temperature, fever, and cooking), and C for everything else. F has a better range for air and body temperature anyway.
Maybe Europeans will switch to F, and Americans will switch to Km :)
(In case you wonder why F is better: The coldest temperature normally experienced by people is 0F and the typical temperature is near 100F. Unlike C where you go negative, and the typical is 20.)
It is really a matter of what you are used to.
Of course, for science and cooking we need more dynamic range, which F and C work well for, but just for our daily experience, we need something new.
That's exactly what Celsius is, except it pegs 0 and 100 at non-arbitrary temperatures that are both scientifically precise and exceptionally relevant to everyday use.
Only 0C is relatable.
On the other hand, I grew up in Miami. In college in northern Florida, those of us from the south stayed up late so we could actually see what it was like when it was freezing outside. We left water outside to see that it actually froze! Back then I couldn't relate to temperatures below 32F/0C.
In other words, "we" is a rather nebulous definition.
Beyond a certain age everyone has had the experience of suffering thru having total weirdo coworkers (weird even by my standards) who seemingly randomly oscillate the office thermostat setting from 60 to 80 thru the day because they "feel cold" "feel hot" so they turned the thermostat all the way up or all the way down. Nutcases.
I'm just saying if you create a hot and cold human scale, it'll have to be personalized for each individual and seemingly randomly vary over time. May as well use a magic 8-ball.
Of course, the people on this board are probably running generally more punishing tasks on their computers than the average, so if anyone was going to find this problem, it would be us.
Gaming and BitTorrent (e.g. Battle.net launcher) are the only things that "upset" my machine heat/utilisation wise. Maybe full virus scans but that is more IO again.
i also noticed my mbp also runs very hot even when i am not doing anything fancy. the most i have is 15 chrome tabs open. i know chrome is a hog but hey ... its a powerful system, should be able to handle it without heating my like an oven ready to bake chicken.
compressed air many times .. no use i had to run smcfan but still its pretty hot.
Instead of the oven, I used a heatgun and a stack of nickles on top of the GPU chip. I almost fell over in surprise when the laptop booted right up afterwords.
Get a desktop already. At least then, when your fancy GPU has a meltdown, you can salvage things from the wreckage.
Laptops that were built with thermals in mind actually do quite well at full load. I used to play a decent amount of games on my Lenovo Y500 when it was new and it could do full load for hours and hours without getting hot. It was still less noisy than my Macbook. The secret is to build a high diameter fan and a wide fan vent. Unfortunately Apple seems to think that being thin is more important. It's certainly easier to market.
What I trade for portability is performance and I replace the machine more. I just keep everything important stored online and when my laptop dies I grab another and keep going.
In almost every instance of someone using a laptop for heavy CPU work, there is certainly a better, more practical reason to use a desktop. There is no need to sit at a coffee shop and make music, unless you do not have access to a studio or even your parent's garage. Video editing will not be done out in the field. It will, if done professionally, be in a quiet room with the proper equipment. Most of the time, someone that insists that doing either of these is just fine on a laptop, is someone that is young, broke, and doesn't have access to the proper tools for the job. Of course, the person pointing this out will get many downvotes...
Compiling Importing and Exporting from a database Photoshop load tests.
Also, I run a web server and database on my machine. It's just a personal preference for me. I do have a mac mini at home that I use also for stuff like video editing.
1. They just spent $2000+ on a laptop and don't want to / can't afford to spend another $1500+ (desktop + decent display).
2. They only game occasionally.
3. They don't want to own more stuff.
4. Their laptop has a warranty which covers any GPU issues.
In my case all of these apply.
But it sounds like his warranty has expired, hence the oven.
Lead free solder melts somewhere in the 200C-300C range. The solder we use melts at 230 degrees, and we reflow our PCBs in a Vapor Phase Soldering oven, which precisely limits the PCB temperature to 230C. Reflowing a PCB multiple times risks solder paste flux exhaustion, and also risks parts on the bottom side of the PCB falling off. Also, some parts are rated only for a very limited time at these temperatures, even when they are not running. Cooking these parts multiple times results in a dramatic reduction in lifespan.
It wouldn't surprise me if Apple did an x-ray inspection of every BGA part on every device they produce. This isn't common practice in the manufacturing lines that I know about, but I know that it's done in some cases. This would help catch cracks or other defects that would result in a reduced lifespan of the device, as well as detect show stopper issues.
There are a lot of other things that can fail on a PCB due to heat long before the solder will melt on a BGA pin. My guess that this was a mechanical failure inside the PCB. Either a through hole failure (such as a via disconnecting from a PCB trace, since the PCB is undoubtedly a very high layer count producing many fragile connections to long copper vias, which would expand vertically and in radius during temperature cycles) or an inner layer connection defect due to PCB or copper expansion. In either case, cycling the temperature by a large amount could temporarily fix the problem by creating a good enough connection for operation.
The bottom line is higher operating temperature always hurts mean time between failure. This is well studied, and many manufacturers will include plots of the operating temperature vs MTBF. Even though the laptop is working now, all of the components have been exposed to additional extreme additional thermal stress. Everything in the laptop is now much closer to failure. I don't expect the laptop will last much longer.
I'm not familiar with the guts of a MacBook Pro, but if there was a larger than normal gap between the chips and their heat sinks, then this would definitely explain the higher than normal system temperature. Even if the heat sink compound was able to bridge the gap with minimal air bubbles, it's still a pretty poor thermal conductor. This may have been a manufacturing defect that Apple could check for in the future. (Although I would hope they already run the devices for some length of time in their final enclosures while monitoring component temperatures)
Apple normally 'does not mind' if you open the case. Obviously you have to know what you are doing, hence the special screws. If you know what you are doing, you'll find where to spend $1 to get the correct screw driver.
Alternate ending sentence: "So that's how I ruined my MBP with duct-tape-level thermal reengineering! Thermal resistance, wazzat?"
As someone with over 30 years in electronics manufacturing the first thought I had when reading this article was: This is an incredibly stupid idea.
Do not try to play manufacturing engineer with your kitchen oven. You will, without a doubt, cause more harm than good. Not to mention coating the oven where you cook your food with stuff you don't want in your food.
Modern high density RoHS electronics boards require higher temperatures in precisely timed and controlled vapor phase reflow ovens. Components, solder joints and the board itself will almost certainly suffer damage from uncontrolled non-vapor profiles in a kitchen oven.
It is far more likely that components will be exposed to far greater temperatures than rated as well as all kinds of imaginable differential heating and thermal slew rates. You can destroy the vias and pads connecting multiple layers of a multi-layer PCB simply by having differential heating or high temperatures. You can also damage dielectrics in precise impedance controlled boards. You can cause problems with solder that might very much affect signal integrity in high speed signals (for example, around the memory subsystem).
In general terms, you are very much opening Pandora's Box if you think you can use a kitchen oven to reflow a modern mid to high complexity circuit board.
Beyond that, you don't have the necessary optical, x-ray and high-speed testing equipment required to verify operation. You don't even have access to design data needed to understand where critical paths might lie.
Sorry. This is a really dumb idea.
Then we move on to drilling holes for the fans. OK, great, yes, more cool air molecules per unit time is what cooling is about. However, this design does not have any air filtration mechanisms. So, by drilling a bunch of holes you are far more likely to create a situation where you will coat the entire inside of the computer with a layer of dust and dirt with the potential to cause major issues.
If I had no choice but to cut holes on the bottom on one of these machines I'd take the time to engineer a bolt-on filter plate with washable filters so that the internals might be protected from contamination (within reason).
A far better idea might be (more on that later) to engineer the attachment of a surface extension feature on the bottom of the machine rather than drilling ventilation holes. "Surface extension" is what you commonly see on heatsinks, in other words, fins. Anyone who's done thermal FEA quickly understands that designing thermal management by eye only works if you have a lot of experience. These machines probably have decent thermal design. Perhaps all that is needed to make them a bit better is to add 10 mm of metal to the bottom with a nice thick set of fins widely spaced in order to extend the surface and promote thermal exchange with ambient air.
Final point on TIM's (Thermal Interface Materials) like thermal paste. Their main purpose in life is to take two non-perfect surfaces and bridge the little imperfections so that more molecules make a thermal connection between surface A and surface B. If you could polish both surfaces to absolute perfection and mate them to absolute perfection you would not need TIM's. When it comes to thermal paste, you want as little as possible and you do not want any air gaps. To some degree there's an art to applying these kinds of TIMs.
The author talked about buying some copper sheet in an attempt to fix the problem. Again, horrible idea. You have have two layers of TIM, on one each side of the copper sheet.
In the hands of those without the requisite knowledge or experience I feel that ceramic sheet-based TIMs are far safer (http://goo.gl/Q6e0Es).
But I digress. The best idea would have been to walk the machine to Apple and deal with it through their channel. That's what they do best.
The good news is that this kitchen-reflow machine is unlikely to make it to the used market due to the drilling on the bottom. I would hope that the author will be honest enough to disclose the amateur reflowing of the main board if and when the machine is ever sold in the used market.
I have a BookArc[1] laptop stand, which is about the worst possible way to keep a running MBP (lid closed, fan slots pointing down) and even with lots of gaming, I have never experienced any problems.
Yes, I do clean the fans once a year, but all laptops need that. It goes to show that thermal wise the MBP is actually very well engineered.
[1] https://www.twelvesouth.com/product/bookarc-for-macbook-pro-...
Your study size of one doesn't 'go to show' anything at all regarding the MacBooks thermal performance. It shows that you yourself haven't experienced any issues but that's where it ends.
I for instance have a 2010 i7 MacBook Pro and it will spin up and get ridiculously hot (too hot to touch) simply watching a 1080p flash video on YouTube. Not the greatest thermal design there with my study size of one :-)
What I will say though is that Apple seems to use the Aluminium case as a heat transfer device, so it purposefully gets hot to disapate heat. That's great until there is a fault with the solder being too thin to withstand the planned temperatures like is happening to the 2011 MacBooks with the discrete GPU.
Also, I didn't notice the MBP in the article is a 2011 model, mine is mid-2012, so I guess I can't compare those.
Before I bought my first macbook air I wrote a program on the console that slammed the CPU and left it running at the apple store. When I came back a few hours later the program was still running (100% CPU on all cores) and the machine was still cool to the touch. I then purchased the laptop.
Yes, but have you ever specifically had a 2011 MBP like in the article? It's a very specific set of models that have the heat issue.
Mine could easily hit 90C under load (it just bricked itself last week). Which mean it was quite hot to the touch. After I changed the thermal compound, it would hit the low 80s (C) at peak, but even then, it was quite hot.
http://appleinsider.com/articles/14/12/30/apples-latest-mac-...
Engineer: "We can get the computer to half heat if we add a few more intake holes..."
ArtsySalesDouche: "Nein! Nein! Nein! I don't want that! I know nobody sees the bottom, but I don't want it, I don't want it, I don't want it, I don't want it!"