Hey everyone,
I'm not sure if anyone is even going to read this but I felt like I had to post what I did and it seemed to have worked. It does involve downloading the software mentioned above, but not only that. Again, it has been working up to now, a week or so of intense ProTools sessions (music recording) and other multitasking.
This is going to be a long post.
First of all, how it started: After realizing that some of the music production software could not run on Snow Leopard I migrated to Lion. A few weeks went by and then the random shutdowns started to happen. I don't recall doing anything different other than upgrading to Lion. On a desperate move I decided to downgrade back do Snow Leopard but it did nothing to the issue.
After watching some of Louis Rossmann's videos I decided to shoot him a message describing the issue, this was his reply: (it might sound harsh but Louis is actually an amazing dude, who's been assisting a bunch of people with Apple products related issues, look him up on Youtube)
+"japasetelagoas" Reballing is constantly being recommended by the dumbest of our industry, with no brain. Make sure to go back to those places where idiots suggest that and credit me with calling them an idiot.Replace C9560 with anything but a tantalum kind. Apple made the same mistake they made with the 820-2610 and 820-2330 where they ran out of space on the motherboard for C7771 so they made it a tantalum one. This is not suitable for a rail that is always on, as C7771 is for an S5 rail - so it dies. Voltage fluctuates, and eventually the machine stops turning on. On the 820-2850, that same 330uf tantalum capacitor re-appears on the framebuffer power supply for the GPU and the VRAM. The reballers without brains are heating this capacitor when they heat the board to remove the chip, so what happens is that they think they fixed it when the capacitor works again post heating! But it dies again when the cap F's up again. The proper repair is to replace C9560 with a non-tantalum one, which won't fit! You will have to scrape the board's ground pad to make this fit. It is doable.
Well, after realizing that I had to do some SMD soldering in order to try to fix it I took the fact that I'm into electronics (have a small lab at home) and decided to buy me a hot air rework station, crucial to swap the super tiny capacitor.
Months went by and I finally came up with enough courage to either ruin my 15 inch mid 2010 MBP or fix it, but before heating up the station I thought about the software mentioned above, the gfxcardstatus, why not give it a try, I'm screwed anyways.
Downloaded exactly as described on the post above, version 1.8.1. Installed it on my MBP running Mountain Lion by the time. After installing the computer failed 2 or 3 times in between both setups, NVIDIA and Intel's GPU. I had lost my faith by the time but after setting it once again to Intel's chipset it sort of stabilized. Three hours went by and no reboots, 5 hours, I then thought to myself, why not downgrade the OS back to Snow Leopard WHILE gfxcardstatus is set to Intel (isolating NVIDIA's chip).
That's exactly what I did, had to format the whole thing, wiped out my HD and after doing that I STILL HAVEN'T HAD ONE SINGLE REBOOT. Almost two weeks went by with constant Protools usage, Adobe Illustrator, Photoshop, not a single hiccup. And before you guys ask, we all know that the gfxcardstatus sort of shuts down NVIDIA's chipset, guess what? My MBP is running exactly on NVIDIA's chipset, the one which was told to be triggering the fault.
The image below points out that the computer is indeed running on it, it even has a Protools session open and the messsage with Louis Rossmann on the background:
This is a message that I've sent to Louis explaining the things that I did:
"Louis, so something weird happened here. I haven't touched the logic board but for some reason my MBP returned to life. A day and a half have passed with extensive Pro Tools 8 sessions and not a single hiccup from the computer. I'm trying to identify any logical reason for it to be working but couldn't think of any, the best thing that I can do is share with you the things that I did prior to realizing that the computer seemed to return to an operational state. It happened on a very short time lapse so I'll just number things out:
1- Finally created the guts to take that cap off after receiving a reply from you 6 months ago.
2- Downloaded the schematic but couldn't find the damn component, sent you a message and received your reply, was confident on doing it.
3- Before actually heating up my hot air station I decided to give the "gfxcardstatus 1.8.1" software a shot, the one that lets you choose between NVIDIA's and Intel's.
4- By the time I had it installed my MBP was still running Mountain Lion (can't recall exactly what version). It did reboot a few times but then after switching back a few times between GPU's I selected the "Intel Only" option and the thing sort of stabilized.
5- A few hours went by and the Mac still hasn't rebooted, which was a relief to say the least. I then decided to downgrade the OS to the one that IMHO is the most stable for music production (Snow Leopard).
6- Did the whole thing, put the Install CD in, erased the hard drive, reformatted the whole thing. Just let me emphasize something, I did all of that whilst the "Intel Only" mode was still selected. I really can't tell you if this has any effect on the final product, if after wiping the HD the command to isolate GPU's simply reset thus ruling out any probability that this is linked to the problem being "solved".
7- Turned the Mac on, now with a fresh Snow Leopard version, installed all of my recording stuff back in, Protools 8, Waves Plugins, a bunch of VST's, still with the uncertainty that it would give me a kernel again.
8- Today 07/06, I decided to give the old Mac a shot on actual recording. Spent 3 hours recording a little project band which I'm in and it worked like a charm. Smooth, 3 different songs with 4 or 5 tracks, nothing major, the bass player didn't show up. But hey, it seems that the thing is back to life.
9- Here comes the intriguing part. I've just checked which processor is being used according to Apple's support website, and apparently, by the time I was writing this email, with a Protools session open, NVIDIA's GPU was running. Should I consider this a fix or should I get some Kleenex to wipe my tears out when this thing fails on me again? Maybe it's just plain luck that it hasn't "decided" to screw me over."
His reply was that he never saw me stating that I've replaced the capacitor, which I didn't.
I apologize for this huge post but I really had to share it, this is something that til now worked for me, I've never seen anyone mentioning that they downgraded their OS while running the gfx software, still I can't guarantee that's going to work considering that I couldn't come up with any logic explanation as to why I still didn't get any reboots, but hey, when I was desperate thinking that I had lost my pricey computer, the first thing that came on Google I was trying it.
Hope this somehow assists people with this particular issue.
Resale value, man. Resale value. REPLACE C9560!!!!
[doublepost=1489642369][/doublepost]
Look at the logs and you will see the kernel panic is related to video. Then ask the rep to facetime you so you can give him the middle finger
It's all a game to prevent you from receiving warranty service on the extended warranty programs - god forbid they accept liability.
For example, on the 2007/2008 A1226, A1260 machines, the issue wasn't a kernel panic. It just wouldn't turn on a good 75% of the time. When the GPU failed, the light would come on but it would not POST. Apple had a test they would run to determine whether the GPU failed. If the machine could not POST, they could not run the test to determine whether the GPU was at fault, and therefore would decline you service under the extended warranty program.
I remember there was this girl in my kindergarten class that knew this one kid would laugh if he heard curse words. So she'd walk up to him when he had to pee, and start cursing in his ear, causing him to laugh and piss himself before he got to the bathroom. People told the teacher but no one believed us - she was cute and got away with murder. Similar to my kindergarten arch-nemesis, Apple manages to skate liability for all wrongdoing. And no one cares...
[doublepost=1489642477][/doublepost]
Glad to hear that worked Groverb
This costs $1 to do yourself if you can borrow someone else's soldering station. I gave away the answer for free, now GO FIX IT!!
[doublepost=1489642679][/doublepost]
You might want to ask yourself why anyone in this thread, or your local computer guy, know the issue is with C9560 in the first place. From 2011-2015 people had no idea and were reheating the GPU!
I think I spoke to you - I had told you we have a flat rate for board repairs as polite as possible. You had said "but it's only one cap, why is it $325? Too much!" an I said "you're not paying for the cap, you're paying for the five years of our life it took to figure it out." Call it a personal failing, I've always replied to politeness with politeness, and attitude with attitude. I couldn't make it to explaining that U9600 or the GPU itself often cause these issues which take hours to replace properly before you rage-quit and hung up. For the best, IMO.
If anyone else is reading this thread and looking to start their own businesses, run like hell from people who arrogantly and combatively try to beat down your prices rather than politely request to work with you. Do realize the dude who gets you to commit to a guaranteed $50 repair will scream bloody murder when it's the mux, the GPU, or VRAM itself causing issues. "BUT YOU PROMISED $50!!!" and there comes the chargebacks and nightmares and DCA claims. This business relies on trusting relationships which never start out with combative conversations: avoid this mess altogether by sending these customers elsewhere. We've given a lot of breaks to a lot of people who called in and weren't dicks, and there's more than enough nightmares(two in this thread!) where people sent in a 7 year old machine, complained it took a week, where they had waaay more at issue than just the capacitor.
We are more than happy to be the ONLY company you'll find that publicly shares these solutions. You can use these solutions to repair your machine on your own without giving us a dime(not giving us a dime is advice I give to every customer who wants to fix a 6+ yr old dinosaur). Or, you can rage that after calling into a licensed business that offers a warranty in the most expensive city in the United States, you got a price that was only half Apple's rather than 1/10th of it.
If you think board repair here is expensive, just look at the apartments.To each their own!
[doublepost=1489643311][/doublepost]
This is a job well done.
[doublepost=1489643449][/doublepost]
Fabio I really appreciate what u did to help me, am writing from my macbook mid 2010 repaired, i want to share to the community my process repairing the macbook.
First of all i used non-tantalum Capacitor extracted from a mid-2012 broken motherboard (C160Z, thanks again) because I did a deep search between tantalum and non-tantalum capacitor and what i found was that tantalum are very unstable and poor reliability, a lot of people have very bad comments with this type of capacitor,I read it's good in high frequency, but also that this type of capacitor is common to fail Bypassing IC Supplies (this is what the capacitors is doing in the mac, That's why it fails in most mid 2010 motherboards? What happened Apple engineers?? well am not very expert in this, i just want to have a powerful capacitor that can get my nvidia never fail again, at least not because the capacitor.
So here is it, i took pictures to show
1. I open up my macbook, then u can see that horrible capacitor making my mac unusable. (yes, it's very dirty i cleaned up later)
2. I extracted non-tantalum 330uF Capacitor from the macbook pro mid 2012 broken motherboard, it's bigger than the tantalum one, but thats no problem
3. I extracted the tantalum problematic capacitor from MacBook Pro mid 2010
4. Okay now here is the non-tantalum capacitor if someone want to see it. (So let's make it and get this machine work)
5. I had to scratch a little to fit in the non-tantalum (Yes this was said before)
5.1 Okay then I weld the scratched part so it can make contact with the bigger one
6. It's ready to battle again for a good while.
Now let's see how thermal paste is (( Okay this is not part of the problem but it can make ur mac live longer,
having lower temperature in the nvidia and intel processor chip))
1. This is a mess, everyone know's that's not the correct way the thermal past has to be. (Fist time I'm changing it) so....Apple???
2.1 Look at this!!! this wasn't doing anything! it's dried
3. cleaning thermal past very very good
4. Please, if ur doing this, don't use cheap past, it's very important to use a good one. My choice was an arctic silver 5
5. This is the correct way that past needs to be
Done!
I have my macbook working properly now. I did a clean install El Capitan OS X, I re-download gfxcardstatus 1.8.1 and am using Nvidia only.
My using macbook normally now, i'm using HDMI with no problems. and i saw 3 movies with my computer now.
am trying to reproduce all types of videos all the time, and again everything is working normally now.
Thank to all that have contributed to the solution. I had a lot time since i did not used this macbook. since my mid 2012 broken this my computer now
This is some true junkyard macgyvering and I am proud of it all
[doublepost=1489644081][/doublepost]
For 2011 it is the GPU itself. No luck with easy fix on that one.
[doublepost=1489644140][/doublepost]
This problem doesn't affect the 17". If you're having this issue on the 17", it is likely something else.
[doublepost=1489644255][/doublepost]
YOU!!! Good time to try this as a project.
[doublepost=1489644877][/doublepost]
It is and I've also ordered non-tantalum capacitors as well, covering both bases in case my repair shop says that they won't bare a track on the logic board. I'd rather it fail over time again knowing that it can be repaired rather than leaving it slightly crippled with the kext fix.
I don't know if it's all tantalum capacitors that are bad, if there are better brands, what's causing it fail etc. It would be great if an electronics expert could shed some light
My preference when I take it to the shop is for them to fit a non-tantalum cap, but if they won't then at least I have one that's the same physical size and whatnot.
Most places hate being told what to do - it isn't from some schoolday anti-authoritatian place. Just that it is VERY rare that customers believe you when you tell them "your fix didn't work" - usually they look at me and think "no, I was right, but you must've F*&#@'d it up" - which is why I steer clear of many DIY diagnosis. Find a place that knows what the issue is from the beginning so you don't have to sell them on the fix of putting a non-tantalum cap & scraping the board, I am sure there is one locally. You're so close to not having kernel panics, don't give up now!!
Last edited:
