LFC#95 - MacBook Pro GPU Fault: C9560
Вставка
- Опубліковано 22 лип 2024
- Credit to senpai Louis Rossmann, I've spent the past few months grinding through his videos and today we see the fruits of my learning!
So we've got a 2011 (thought it was a bit older in the video) MacBook Pro that's failing GPU tests in diagnostics, and we're going to fix it by replacing the faulty capacitor on it's power rail.
Louis Rossmann: / rossmanngroup
Music by;
Synthis - / samsynthis
Etrii - / ertrii - Розваги
Oh God ! I could not continue my full length documentary film editing process because of this exact failure. Did reflow couple of times , then just gave up. Now I downloaded this precious video to my phone, and going smiling tomorrow to a repair shop with detailed explanation how to fix it. Priceless, since I can't afford new Mac purchase. You are the MAN !
Updates?????
What temperature C/F did you put the Hot Air Gun?
In circuit did you check the cap for continuity or anything other multi-meter tests outside of ESR meter which you said was good?
Hi, instead of trying to fit the bigger cap in the same location (not easy) would it be a problem to find a ground via to solder the ground side of the capacitor and wire the + side of the cap to the original location ? Would it work ? Would it change the capacitance of the cap ?
Great video thanks for the info, just a quick question if I may. Would the logic board of a mid 2010 MBP 15" i7 2.66 be situated in the same place as this? Or is there a way of finding it within the logic board?
Asking as I,ve taken mine to get this repair done and the guy says that the capacitor you show on here isn't in the same place in mine...?
Thanks!
where can I buy the capacitor in the u.s. because my Mac has done this twice and it will probably do it again
Hello teacher thank you very much for the great videos that you share, I'm doing the test that shows in this video and in the capacitor I have 1.3v on with the battery so is it correct to have 1.3v or should I only have 1.5v? My problem is that when turning on the MacBook pro 2011 it restarts just midway of the boot. Please ask for help thanks
i have a 17 inch early 2011, my monitor has any light and the back logo too, could be the same problem?
someone can share the schematic for this model?, I can't find it on the internet
Hi I have early 2011 mbpro 15 inch, after say 10 minute the screen goes blue with lines. Would you say it's gpu or cap issue?
Just wondering here, when you moved the coil wasn´t it too much against the heatsink? Maybe the thermal expansion of the sink when it gets hot will press against the coil and induce structural stress on it and maybe knock it off the board with enough cycles. Nothing a Dremel couldn´t solve though
Hi, just a short question:
When I power my locigboard, the old C9560 goes to 1.503, later to 1.37. Is it the failing capacitor for sure then what makes my MBP crashing while booting or could it be the graphics chip entirely? Cause you said something about "low-power-mode" for 1.37 in the comments below. I have an 820-2914-A logicboard from 2011. Thanks :)
i am having hard situation with my macbook pro, its freeze when login in user mode, entering recovery mode and safe mode still freezing, i am taking out the ssd and trying to boot with another hard drive the result is still freezing, so i try to find out the answer through googling and ask some folks, they told me the problem is in the GPU, so what i'm gonna ask here is, is it possible to fix the GPU chips without change the whole logic board that Apple told me earlier to replacing the logic board with the new one which is cost me a lot of money that i cant afford..
HI there. how do I go about sending in my logic board? have a tech remove it and then ship it to you? (I'm based in the Caribbean )
hello bro, I have a MBP 15 late 2011 i7 that does not start after the apple logo and a little later shows blue lines on the screen, do you think it is also because of the tantalum capacitor? Greetings from mexico from a mpb 15 2010 nvidia 330m repaired thanks to people like you and louis rossmann
Would this work for a1286 (emc 2353-1
A little off topic, but I saw you had to remove the heatsink assembly. Would using Thermal Grizzly or any other after market thermal paste give you better temps?
I'm thinking in buying a 330uF 2V cap. With ESR of 9mOhms, should I get a lower ESR cap instead ? Does it even matter?
Please friend, how do I enter this GPU test mode?
Hi Adamant It
I got a macbook pro mid 2010 with the gpu issue. I can't got the capacitor 330 uf 2.0v. Can I use a capacitor 330uf 2.5v?, if will work ?
hello, i thought this fix only applies to boards that uses 330uf tantalum capacitor?
Hi! I've a similar problem, but the error message is "GPUPanic.cpp:170"; the solution is the same?
When I plugged the logic board to power I get intermittent fan on off
I’m getting 1.36 volts at those capacitors, which is what your voltage dropped to. Is this common in the failed condition?
Hi Graham, I have a late 2011 MBP which I assumed has a GPU fault.
When booting, some 3/4 way through there appears a large multicolour band through the top third of the screen and it flickers off then completes the boot and works for quite some time usually but then reboots occasionally. Sometimes there is a ghost on the boot screen which consists of lots of horizontal red lines which tends to cause multiple boot loops. Is this a GPU fault or could this CAP repair fix things? The thing is is appears to be a lot more stable than it was 12 months ago when I retired it for a new one.
Just last weekend I sat in front of it for maybe 8 hours transferring all my old pics which I thought I had lost to the apple cloud photo share thing with minimal issues.
What if there is not AMD graphics but just the Intel graphics ... Is that possible.. to route the power to the Intel graphics because it seems to be dependent on the AMD graphics? Just curious
I don't understand why the gpu can be replaced but not upgraded?
Good job
Hey man great video, quick question for ya though that I can't seem to find answers for... I removed the current cap thats faulty and replaced it with one market 330 from another dead logic board, but I still can't seem to get past a white screen during boot up , recovery or re-install , did I use the wrong cap or is there something else wrong with my MBP?
Are you sure you installed the new cap with the correct polarity?
Hi, what is the capacitor replacement value that you used? Thanks for the video!
Sorry it's taken so long, I don't get replies sent to me for some reason. The Apple number is EMC 2353. It's a MacBook Pro 6.2 (i7 2.8, 15", 2010).
Can you fix mine? I am based in Poland! How much to fix mine? 2010 MBP
This doesn't apply to 820-2915 motherboards! It is the GPU on those.
I tried mightily with my 820-2915. I may have used an incorrect value, though; used a 330uF from digikey. My MacBook Pro has the lines-onscreen-during-bootup-on-the-Apple-screen/ white screen-crash-reboot cycle problem.
My repair was unsuccessful and I've been wondering if this repair video (and the Rossman video you reference which shows this identical procedure) actually applies to my 820-2915 board. I s'pose it does not after all. Really, a 330uF vs a 220uF on a power supply rail shouldn't make a helluva lotta difference so I guess I'm stuck buying a new logic board from eBay or getting my GPU replaced.
(Edit: I visited the Rossman repair website and have now seen that they aren't no longer offering 2011 MacBook Pro GPU repairs. So it's down to buying a new or used logic board from Amazon or eBay!)
Adamant IT : thanks for the reply; I certainly wish the news were better, but at least I still have the option of buying a new (well, it'll be a used) logic board, the prices for which seem to be around $430-550, still much cheaper than a new MacBook Pro. I have no reason to upgrade this computer as yet, since it was running what I needed it to run just dandy before it decided to fail.
you can use a heat gun to remove the gpu and replace it with a new one they are like 30 bucks on ebay
type the model number of your chip on the bottom of the shiny part rhat has some writing numbers star like this in example 816-568029
new one on ebay, next you'll tell me all the clean std-free women are on backpage...
Adamant IT: you know, that sounds like a pretty good option right now. Thanks!
Found c9565 and replace it but still no luck , I think is the gpu .
It measure 1.35v from c9560 but it fail to boot to Mac OS now
Hi, i have question - what do you think about replacing the bad cap with another tantalum cap? Isn't there a possibility that it will fail again in near future?
Adamant IT, hi.. Thats really nice info.. Can you please give me what type for my MBP 15" early 2011 .. Thankyou very much 🙏😥
Do the numbers on the capacitor mean anything? mine says "J8 eW"?
J8 = 220uF e = 2.5V W = Lot No. B2 in the schematic is the chip size code. Look it up at the Panasonic site. It's a TPE POSCAP capacitors.
Do you have a video that shows how to remove the logic board please as you have in this video
Adamant IT thanks
Hello great video , I follow your video and change the c9560, but have the same issue , I double check and find out there is a short on the board where c9560 was , I think the c9565 was short , can you point me out where is the c9565 10uf is ? I have a mid 2010 MacBook Pro with Nivida gpu. Thanks
Thanks for the reply , the MacBook still up and running but when it switch to the gpu , it will do the same panic , i think I will try your suggestion to remove the c9565 cap but I do not know where it located ? Is it next to the L9560 or at the back side of board , thanks
When the GPU chip fails what other logic board components might it take with it?
Thank you for replying. Yes I know about the "replacement" or "new and improved" 216-0810005 chips situation. However, once all other options have been exhausted and the GPU chip is so far gone that the nvram command hack will no longer work (because the mbp won't even boot in recovery/single_user mode), I may have to spin that roulette...
I got a 330uf 2.5v low ESR capacitor, but the casing is yellow? is that okay? did I miss something? can I replace a 220uf for a 330uf or would that cause issues?
Totally new at this, thank you!
I thought going up in capacitance was fine, but I guess that doesnt apply to super sensitive laptop components! I better keep it the same. According to See Sharp J8 = 220uF but the number I got from the board (printed on the charger circuit) is "820-2565-A" Its a late 2011 15".
I heated the gfx chip once and it worked for a month, then I found your video and heated only the capacitor to test it and its been working since! I have even been gaming on the thing!! Apple wanted to replace the whole logic board its ridiculous ! THANK YOU !
Please could you help me identify what capacitance and voltage I need to get. I think its 330uF 2v but that is just guesswork really. My motherboard number is 820-2850-A.
And a 2nd request. I live in SA. We have this great local website for capacitors but there are so many to pick from even when just looking at Aluminium Polymer Capacitors (I read somewhere that these are much more stable than the tantulum ones) 330uF 2v. I've posted a link to which will hopefully show you the dilemma.
za.rs-online.com/web/c/passive-components/capacitors/aluminium-capacitors/?applied-dimensions=4292123263,4294450449
Could you help me pick out one of this website? (of course this will need you having verified that the capacitor in question is indeed of the specifications I think ti is).
Thank you so much. Both you and Louis have potentially given my laptop new life!
Thanks for a great video. What are the specs of that small component (capacitor) sitting next to the plus side of the one you changed?
I changed the big one successfully but as a noob without the proper tools and experience I managed to lose the small one and will need to order some.
Very grateful for any help.
Thanks a lot for responding.
It looks exactly as the board in your video as far s I can see.
EDIT: put on a strong light and found the number on the board: 820-2915-B
Thanks a bunch!
I'll order the smaller ones and try to make it work.
In case it doesn't, do you think there is any way to completely disable the the discrete graphics and simply have the computer not detect it?
Thanks, I managed to change the two capacitors even if it took some time and a couple of tries as this is the first time ever I tried something like this.
Now I get 1.5V over that capacitor but it still doesn't work.
If I try to get ahold of a new board, which are the ones that will fit as a replacement for the 820-2915-B (or where can I find out?
I hear you. I'll see what I can find and what I'll do. :)
Thanks a lot for your help and the video!
I have macbook pro i7 early 2011 with AMD graphics card (logic board number 820-2915-A) what is the right capacitor for me ? Thanks in advance 🙏😊
Adamant IT, Oh noo 😳 thankyou anyways 😬🙏
its ok i will start digging
hi can you let me know why pcs do not use the full memory if 4gb then it should use 4 not 3,49 gb
thank you very much for your reply
hi, can this problem also related to the problem if the fans spin on off for one second ?
@@Adamant_IT thanks for your reply
Hi! If I wanted to do a DIY of this, can you list out the tools I would need? Thanks! :)
Adamant IT (newbie question) ..after replacing the cap and put back everything do I need a bootable disk or any software? Or just turn on the MOB? Thanks
Thanks for the info so far. (Also: nice beard!) I'm attempting to replace the capacitor on a late 2011 macbook pro 15". I've got her opened up and the cap' is a, "J8 ec". So, 200uf 2.5V. I've used info from other postings (thanks all) to get to the panasonic page where all is good except I'm being asked for some specifics: ESR's ranging from 13 to 35 mOhms. Any suggestions on that? Or should I just throw in the 330 as listed? Thanks again.
Notice in the schematic in video that it has an output frequency of 500kHz. I used the 2R5TPE220MFGB because it has 15mOhms/300kHz and its the best I can find available. ESR is important at high frequencies. Adamant IT probably used a T591*series from KEMET which have 15mOhms/100kHz judging from the video.
I also replaced the KOcap 68uF 16V (Yellow color) and the Panasonic 68uF 16V (below the KOcap) with T521V686M016A(1)E040 KOCap by KEMET. In my case these capacitors near the heatsink are bad due to the heat.
Sorry to answer a question with a question but how do I identify the logic board? I've typed my serial number in on a number of sites but I garner no info starting with "820" or "logic..".
Is it physically listed on the board? (Serial number is: c02gm49ldv7m)
Thanks again. (And thanks See Sharp for a note on the schematic; afraid I'm guilty of not spotting the details). I found 820-2565-A on the small circuit board where the power supply mag' attaches. I think that means I'm looking for a 220uf. What's interesting (and tragic) is that I rolled the dice and ordered a 220uf capacitor. Upon arrival the paperwork shows a 220uf but the label on the static bag shows a 300uf (and the caps have no writing on them at all). The SKU (below) shows that I was sent a 300uf (if that's what they actually put in the bag). So... Can you (please) verify I need a 200uf; or, if I'm lucky, tell me I can put this 330uf in. Thanks.
PS: is there a way to 'test' the uf on a capacitor with a multimeter?
SKU: A700X337M2R5ATE010
820-2915-B. List your PayPal and I'll buy you a long, dark beer. Seriously.
Hey man. Great series of videos.
What’s the software you used to test the computer? The one that came up with the green “Passed” text.
Thanks again. Your videos are invaluable
That software is called "Apple Service Diagnostics" or ASD for short. There is a separate version for all of the older MacBooks.
This should be enough to get you going - you need to do some googling, to find out where you can get it and how to install it. You might even come up empty, since this software is supposed to be for Apple internal use only, so it might not be so easy to find. That's all the information, I can give you on that.
Hi, I just want to ask how's the laptop doing so far, did it show any sign of GPU panic after the repair? I think it's been more than a month now and I hope it doing fine. I really want to try this method to my laptop...
Great, thank you for your reply, I'm currently running with forced integrated gpu using gfxcardstatus because if I use the discrete GPU my laptop will reboot it self from time to time, and, one other thing that I notice is that my GPU die temperature has gone high 129C or 264F according to HWMonitor I've never notice this before but the HWMonitor didn't red-flaged this, and my laptop is usually quite warm just like other MBP but it's been a bit cooler ever since I replace the thermal paste. Hopefully this is not a sign that my GPU is already shot, but, I will give this method a try when my work load is not high, thanks, have a good day...
Both the fans spinning fine at around 2000 rpm, also I put the laptop on a stand that has a cooling fan too, the other temperature reading for the CPU cores etc also normal around 55-65C, and here's the odd thing the GPU Die temp is constant at 129C while the rest is fluctuating +- 5-7C but never reach above 70C unless on high load which usually the fans will start kick in too...
Hi, finally I did the replacement of the faulty capacitor like you did and everything worked except that now my laptop won't charge, I don't know if this is because I left my laptop uncharged for some time or maybe I did something wrong on the procedure, however, the graphic card now passed the ASD test, twice...thank you for your video, I just keep my laptop pluged-in for now...
How do you power up the logic board with the power switch disconnected? Critical info.
Aha! Good to know. I wouldn't have disconnected the keyboard. One more thing: When testing the voltage, it looks like you're not touching the capacitor in question, but rather the one next to it. What is the exact point you're touching with the probe? Thanks for your reply.
Does it matter which end of one of the neighboring caps I use?
I have a 2011 macbook pro 15" with the same grey kernal panic.
I don't have a 330uF 2.0v that i'm aware of, however i have some capacitors that are on my old macbook '09 unibody. are they the same as the 330uF 2.0v and will they work?
I have these one's:
33/C93
330e8D
330d92f
Please let me know.
If it's a 2011, is it an AMD GPU? If it's AMD, then it's game over.
Adamant IT thanks for replying!
It has an intel HD build in GPU
And a Nvidia geforce GT 330M
What’s your take on it ?
my problem is i have not the test! not easy to find!
perfekt jobb
!!
With 2011 models 15/17 inch (Radeon GPU) this is mostly not going to be a total fix, only Nvidia models like 2010 A1286 820-2850 will truly work with this method. You will have the problem back in a little while. With these kind of boards you can also just heat area near GPU or on GPU and that is what is causing the stripes to dissapear for a short period of time.
I did replace the capacitator but still getting black screen 2 or 3 seconds after turn on and some time randomly restart my is MacBook pro 2010
Adamant IT I replaced capacitator twice and switch ram too ,I'm soo frustrated on this computer
Adamant IT can you tell me please how to set voltmeter to check if capacitator or something is ground on board
it worked perfect but i tested the old fujitsu capaseter it show 1.4 v i think its bekouse my kaise multimeter oly charge this little. im a windows repairer dell and hp. sorry for my bad english am swedish swe!
00:23 yes Louis and others like yourself, Adamant IT are great at showing us how to do this repair. However credit should also be given to Duke, from the Netherlands as he is the one who first identified this capacitor as being the culprit. No, not Apple 's multi-million dollar engineering dept, but a guy working out of his garage using a paint stripper as a hot air source, and no microscope. Louis acknowledges this in his own video in regard to this repair ua-cam.com/video/DzcgT_fiVTA/v-deo.html
Greetings. 820-2915-B board here. Both quick and extended Apple Hardware Tests (option+D) show no hardware issues. Intermittent non wake from sleep and when it happens 4-5 reboots needed to get desktop. Currently using nvram hack to prevent 6750m from being used. Schem for 820-2915 shows C9560 220uF 2.5V, could it be different for the -B board version? Could C8942 & C8943 (330uF 2.0V) also be faulty?
Appreciate your response. Will check and report back. If the sodimm socket is at fault, is it replaceable?
What about MacBook Pro (A1260) early 2008
I haven't looked in detail at anything that old, but I believe those were GPU faults, rather than the frame-buffer capacitor we see here.
@@Adamant_IT - I've looked at many UA-cam clips on this issue and it seems more logical to me that it would be the capacitor (C7771)! I guess I'll give it a try as I have nothing to lose. The re-balling of the Chip does not make that much sense - if the cause was originally overheating, then why would heating it a gain fix it - seems illogical? I guess I'll find out sooner or later?
I have an early 2013 MBPr which seems to suffer from this. I have not been able to find any GPU diagnostics on mine, though. Nothing like your flashy ASD screen. General diagnostics (Apple Hardware Test 3A244) shows nothing other than one SATA error.
Thanks for your reply, unfortunately my macbook is not covered anymore. I do not experience artifacts, only shutdowns when switching to external displays, and when the nvidia card is utilised. I will have to get a pentalobe screwdriver before I can test the offending capacitor.
Ok, then I'll just continue troubleshooting. Thanks for you helful replies, and also for taking the time to make this great video. Lots of useful tips on this channel as well, will recommend them to the apprentices who do the repairs here :)
what would happen if the cap was revesred
Adamant IT thank you
i soldered it in backward and now my mac is dead iv replaced the cape as it went pop and i now have 12v between the cap and ground anyideas
👍👍👍
Hi I tried everything down to the letter. except I used an none polarity capacitor. Tayio yunden multilayer ceramic. got it from mousser electronics.
www.mouser.co.uk/Search/m_ProductDetail.aspx?Taiyo-Yuden%2fPMK325AC6337MM-T%2f&qs=8jfIC9vqKDtSgjVVM80Syg==
my reading still 1.37 and I never get the 1.508 volt
see my board
www.dropbox.com/s/unmtucmgh7kwrlz/20171101_212936.jpg?dl=0
can anyone help. is there something else need to changed.
is 1.37 volt is correct on the mid 2010 board
Adamant IT I guess this one should work than.
uk.farnell.com/rubycon/2-5swz330mr15/cap-alu-elec-330uf-2-5v-smd/dp/2346118?CMP=i-ddd7-00001003
Also its interesting that the one I put in reads low. and it does fail on the asd test. however it runs perfectly the latest os. before every time was an external display were used and kept toggle in on them it triggered the go I panic screen and restarted. now it does not do that. Do you think I still need to change it to the one I showed above. or not.
Adamant IT I thought but I could be wrong the voltage can be higher just not lower than the spec
Adamant IT guess what I think the reason it's never read more than 1.37v is because I am using 2.5v not 2v . I have tried all my caps 330uF with the 2.5 v. and all of them even the polarity one too is reading 1.37v . so I just ordered the one you told me I'll get it by max week . let you know if it worked. thanks you were a sport.
Adamant IT well this what happened I ordered for farnell .uk the one u told me . Because the guy at the ebay messed me up.
but then instead what I see on their site they sent me this
With a 3 contacts. is this ok to use. or they made a mistake see screen shots for reference
www.dropbox.com/s/1jkrm70rk8h42c3/Screenshot_20171111-152044.png?dl=0
this should be the one I ordered.
www.dropbox.com/s/0k9wh3b2bz3y7zy/Screenshot_20171111-132829.png?dl=0
Adamant IT I am clueless. I just measured on the board where the caps should go and without anything been replaced it reads 1.34 v so I am sure doing something wrong or my board has other issues. is the usual if you remove the caps would read anything like that?
hi where do you get the information from mate the one you show on your laptop. Louis Rossmann, videos are good but language is not so good to hear.
I am not sure I fully understand the root cause of this failure. I understand that Apple used a commercial digital voltage down-converter regulator with a 330µF tantalum cap on the output as a charge accumulator. I believe there are many of these caps on the board, but none of the others fail. Unless there is some research showing that a particular signal may cause a cap to age faster, which I seriously doubt, then why does this particular cap fail? I would expect it to age faster if it were in a higher temperature environment than the others, but I do not believe this is the case. I have not checked it with a FLIR camera, but it is mounted right next to the fan, so I do not believe it is mounted in a particularly hotter environment. I am aware that the fundamental physics of Ta caps causes them to lose capacitance slowly with age. I am also aware that Ta caps (counterintuitively) decrease slightly in capacitance at cooler operating temperatures. I am also aware that some old Ta capacitors can often be made to temporarily regain some of their original manufactured capacitance by re-annealing the part. Observations have been made WRT to each of these influences on Ta capacitor capacitance which correlate with a decrease in the capacitance of the C9560 capacitor. I now understand all of these things. What I do not understand is this:
A significant number of this precise model of 2010 Macbook Pro reportedly experienced this failure when they were new, within weeks of being purchased. Additional units have failed with some consistency pretty much continuously with time as these approximately 10 million computers age. Yet Louis Rossman regained function of one of these failed computers when he accidentally replaced the old cap with a polarized replacement cap which he installed in the wrong polarity. (I done seen it on TV with my own eyes.) At these low voltages, the caps have a fairly high effective capacitance in reverse polarity. From past experience, I would guesstimate about 80% of the rated value.
There is no question that the GPU generates this fault. You can read it in the first line of the dump. And it is almost certainly true that it happens because the voltage across the GPU supply lines instantaneously sinks too low. For example, my 2010 Macbook Pro only crashes when I am changing "spaces", and then only infrequently. But the frequency is slowly creeping upward on me as the C9560 capacitor ages. The largest slug of current the GT330M GPU gulps from the power bus is when the largest number of bits per clock cycle in the frame buffer are changed. This, in turn, occurs when large areas of the display in high definition (large numbers of pixels) change color. I do not (yet) know enough about the internal architecture of the GT330M to know what color1 to color2 transition will cause it to flip the largest number of bits in a high-def full-screen write. But that is just a matter of reading the GPU datasheets in enough detail. This is the condition to avoid if you are on the hairy edge of losing your machine to frequent crashes until you can repair it. It is also the condition you should want someone to include in a test routine to verify that the post-repair machine will continue to function properly under the worst possible condition.
But here's my real question: Why should we believe that whomever designed the chopper voltage down-converter circuit which powers the GPU did a proper design? There is apparently not much to the design beyond selecting the right charge accumulation capacitor to hang on the output of the chopper. BUT IN ANY CASE MY ASSERTION IS THAT THIS POWER SUPPLY DESIGN IS FAULTY. I have not (yet) dug deep enough into the chopper regulator datasheets and appnotes to do the design with the same digital voltage down-converter myself. But my suspicion is that using a 330µF cap on for the charge accumulator/filter capacitor on the output of the chopper voltage regulator is going to be on the hairy edge of being theoretically large enough. Perhaps, in unusual circumstances, it may even be found to be a bit too small. If such negligence could be proved, it might still have significant legal ramifications.
Conversely, the spec for the nVidia GT330M GPU may be in error. Perhaps the GPU draws more power in certain transient conditions than is specified in nVidia's chip spec. Again, if such fraud could be proved, it might have significant legal ramifications.
But it seems to come down to this: Either the cap is too small, or the load placed on the supply by the GPU in certain transient conditions is too large. Which is it? Does the fickle finger of fate point at Apple, or nVidia?
In any case, it seems fairly well established that the GPU supply voltage slips down below the acceptable limit in transient conditions of high GPU load. This correlates with the failure being due to the capacitance of C9560 being too low to handle the instantaneous in-rush current of the GPU under transient high load conditions. A proven hardware repair is to replace the slowly diminishing capacitor with a new one which will usually be of slightly greater capacitance. My last question is this: Why wouldn't it make more sense to replace the failing cap with a new one of a value slightly greater than 330µF - or perhaps even one of twice the capacitance? Perhaps a larger filter capacitor would cause some kind of potential transient start-up overload problem with the chopper voltage down-converter, but that seems unlikely. From the POV of the GPU, there is NEVER a reason to introduce more, rather than less, noise into the power rails. So increasing the capacitance on the circuit may, in fact, turn out to be the proper design that some nameless, faceless engineering group at Apple was incapable of originally doing correctly. Or maybe they didn't test and correctly measure the actual maximum instantaneous load generated by the GPU. In either case, I WOULD FIRE THEM.
Considering the mismatch between filter capacitor capacity, and maximum GPU load, wouldn't an easier fix be to simply jumper a new capacitor across the old one? Couldn't that be done with a simple soldering iron instead of a SMT soldering & desoldering station?
Am I missing something? Why is everyone going to so much trouble to remove the old open-circuiting filter cap?
interesting idea!
might actually try this. I mean it sounds a bit daunting to not use the tried method, but not having to learn smd soldering sounds tempting :)
So you would suggest that using any capacitor with values between something small like 20µF up to 330µF in parallel with the existing on should work?
What a mess !!!
it is a 2012 win computer is my donnor
i did watch L.R. video to but it started to act like a super star and started to not give a .... about quality so... i learn from another place.
get a proper iron man! its 50% of the job if not get a wet sponge and wipe before apply solder.
would love a contact!
tjhagroo@gmail.com
330u and ok its ok
loooool