iBook G3 and G4 logic board problems technical aspect

Why not? Do you have any particular reason? And if not... why is their name your handle?

Superior Reball sold me on the repair of the logic board on my 14" G4 iBook 1.33 ghz because I could squeeze it while booting and it would boot all the way to the desktop. If I didn't squeeze it near the left side of the track pad it would not boot to the desk top. I didn't want to shim it or clamp it since a $75 reball sounded like the right fix. Well....since I have shipped it to them I have never received an e-mail or phone call from them. I have always initiated the conversation. I made my first phone call 1 week after I confirmed they had received my iBook through Fedex tracking. " We are swamped and it should ship back to you today." ...Great... another week later I called "Failed Tests. Reworked. Replacing GPU" and "Should ship today if not I will let you know." That was on the 27th of September 2007. Today is October 18th 2007. I have yet to get a reply e-mail or get someone to answer the phone. I would like to add the I am a very calm professional person. To be treated this way by a company I trusted to repair my logic board is very unprofessional. Even if they could not repair it at least contact me and explain and send me back my still broken logic board. But no. Superior Reball in Gilberts Illinois doesn't do business that way if they can't fix it they will keep it and do not contact you and expect you to just go away. I caution anyone considering using Superior Reball to fix their logic board. Good luck in your endeavors to keep you iBooks running. Mine is almost complete 120gb HDD, 1gb Ram, SuperDrive etc.
Anyone got a logic board I can squeeze?

I think he decided to use "Superior Reball" as his handle, to catch people's attention about the problems he's had with them.

So it look like the other company mentioned in this thread, www.firstphasetech.com, might be the better choice.

If you are going to risk getting any reball, rework, or reflow work done on a G4 iBook do not use superior reball. Squeeze, shim or clamp for as long as it takes to buy a working logic board that you can swap into the laptop in one sitting. And then do you best to keep the inevitable logic board flexing to a minimum and heat dissipation to a maximum.

Don't forget to sign the iBook petition. http://www.petitiononline.com/ibook/petition.html
Maybe we can get some satisfaction from Apple like those fortunate Danes are.
http://www.engadget.com/2007/09/21/apple-forced-to-pay-up-over-ibook-g4-flaw/

Anyone tried and succeeded at putting a 1.4ghz logic board where a 1.33 ghz board once was in a 14" G4 iBook?

The iBook G4 14" 1.33 GHz logic board (Apple part 661-3417) and 1.42 GHz logic board aren't interchangeable--there are several connector differences, preventing them from fitting into each other's housing.

I think using a reliable reflow or reball service might still be the better solution, since you don't know if a replacement logic board will develop the same problem, and reflow/reball is substantially cheaper. If the reflow/reball doesn't hold up, you won't have spent much money (though there's extra downtime), and then you can get a replacement logic board.

But whether one gets their existing logic board reflowed/reballed, or replaces the board, it would be wise to install a shim on the graphics chip anyway to reduce the chances that the chip's solder joints will break again.

It should be said here that there is also a chance in any of these reheating schemes that the chip will fail during the process. I estimate that chance at about 15%. So you're gambling but it's $50/$79 vs $300 ($190 at DT & T Computer Services)

Yes Dale at Superior was down for quite a while with cancer and since it's just him the whole place went to pot. So slow and nice. On the other hand that guy Tom at First Phase is "difficult" so take your pick.

I'm developing a new theory about these iBooks. It has to do with the hinge. I have found many with hinges that are growing stiff over time. Often they break or break at the mounting where they attach to the frame.
I feel that the stiffening hinge action causes a lot of stress to be transmitted to the frame and thence the logic board and BGA chips.
This 700 G3 I'm working on right now has been reballed and flashes occasionally when the display is opened and closed. It doesn't seem to do it when I "pin" the area above the keyboard to the table. I've lubed other hinges with Molybdenum Disulfide grease and they loosen up nicely. But it requires a take apart.
So I guess I'm going to do that here. I use Hob-E-Lube from the hobby shop since it's "plastic compatible". When you get down to where the hinge is exposed you can see the axle inside the hinge body. The hinge body is formed in such a way as to limit the travel. I use a toothpick to apply a small bit of grease on the axle itself being careful not to get it anywhere else.

Richard

hey.

I have a faulty G3 700MHz (16vRam) which a got of a friend that just had it lying around as a welldesignet paperweight.

the problem that I have is that nothing happens what so ever when i press the ower botton. It does charge the battery and there 25v on the 6 middle wires from the dc-board to the logic board.. thats what I know so far.

Anyone has had This version of the logic board failiure?

Regards.
Contact on Dzarster@gmail.com

ohh well.. I got the beast totally diasembled using the brilliant guide at www.ifixit.com.

It seems that only obvious faulty I can see it the onboard Memory Cap 0.33F 5.5V (close to this one: http://parts.digikey.com/1/parts-cats/super-ultra-aerogel-electric-double-layer-capacitors ) has some brown substans on the top edge.

when i plug in the power-cord (or mount the battery ont the Logicboard) I can messure around 5V from the + to - pole on the power Cap. and as soo as I pull the cord theres only 0.5V. Im not sure yet whether it should ceep the power for longer and therefor it faulty, or if it should act this way???

Im still only a Eletrinictechnician Apprentice so im still learning.. which is also part of the purpose of the project..

Regards.

ps. Ill be uploading a few pictures later

spudnuty: You're sure your iBook's flashing when you move the display, isn't a bad backlight cable, that runs through the right side of the hinge? Or a bad data display cable, that runs through the left side of the hinge?

yes im sure.. it WAS the memory cap the was the fault.. since its acid had destoyed som leads.. I fexed them and put a 5v5 0.047F super CAP in its place and now it works..

I got loads of picture that I will add to this post ASAP. (DONE. LOOK AT THE LINK)

ohh.. bad news is that I proberly also have the BGA solder fault.. it made strange stuff on the screen right after it was booted up and I moved it a little. just like everybody describes it. so I did the shim fix.. it works.. but now my trackpad is a bit odd.. might be because of the bow that the shim fix is making in the laptop..

anyways.. gotta work on uploading my pics..

Regards

EDIT: Heres the Pics: http://www.putfile.com/d_zarster/images/158452
they are taken with my 2Mpix SE W810i mobile via. a Leica microscope.

Fantastic job reviving the iBook!

Charles Moore over at Applelinks says the 700MHz iBooks are the most prone to break - perhaps in addition to the GPU BGA issue, the 700MHz boards also have this memory cap issue?

M

Good job tracking down the leaky capacitor issue, and documenting it. Leaky electrolytic capacitors destroying logic boards have been a problem ever since the SE/30, IIcx, and IIci, starting in the early 1990's--too bad some capacitor manufacturers still stick their buyers with junky caps.

In my experience fixing dozens of iBooks, it's the 500 MHz model that breaks the most often, but all the 12" iBook G3 models experience these problems--cracked solder joints under the graphics chip, or/and a broken backlight cable.

D_Zarster:

I was responding to spudnuty's post about a flashing display when he moved it back and forth, even after the graphics chip had been reballed. I can see how your iBook wouldn't power up with a leaking capacitor and broken traces, but spudnuty's iBook was powering up.

But again, good work on tracking down the leaky cap problem. I'll add that to the list of things I check when I work on iBooks.

sure thing mate.. I did also figure it out later on.. but didnt feel like editing my post.. good luck in the future.. I might come back when I try to reflow the ATI chipboard..

Will the pics that you have help on the G4 or does anyone have more details instructions/pics to help with the G4? Many thanks.

Hey everyone,

I've been developing repair solutions for manufacture defects for well over 7 years now. Sony has been dealing with problems like this for many years. A few pages back, someone posted a great article regarding the design problems and board flex that are usually to blame with these types of problems. Very accurate info there.

I did notice a few shops offering their services and also noticed that they were relatively inexpensive. The first thing I asked myself is: "Why kind of service are they offering for $50 and $75?" To price a BGA rework / repair at that level is unheard of. Then, a few posts later I see why.

BGA repair is not uncommon but it's NOT easy work. Secondly, what are they doing to avoid this problem in the future after the repair? (Nothing). Thirdly, why in the world would ANY shop ask for the end user to dis-assemble the laptop and only send them the logic board? Are they crazy? I don't mean to discredit the end users abilities but that is preposterous.

There is a right way to perform these repairs. It certainly doesn't cost anything near $75. The correct way to repair these GPU issues will run more like $200 per board / unit.

..and you ask: "why wouldn't I simply buy a replacement board from Apple?" Because Apple is simply replacing your logic board with the same thing, that will have the same problem down the road.

-Joe K

Shims

Are never a good thing. They are a temporary solution as a last ditch effort to possibly get limited run time from your system. This is at the cost of creating more damage to the problem area. When you add shims to ANY chip losing contact with it's corresponding pads, you're actually driving the chip down onto the pads and driving the pads down further into the surface of the PCB (board).

This action not only creates more damage but it also creates more severe damage of "torn pads" under these BGA chips. Once torn pads exists most repair facilities will have very limited success with a reliable repair approach.

Even with torn pads present, it's still repairable but like I said, you won't see anything under the $200 mark for this type of time consuming repair work.

X-rays, BGA machines etc.. etc.. don't mean a thing when you don't have the proper outlook and the ability to think "outside the box" which is what it takes when it comes to these types of manufacture defects.

If you're going to use shims, use a shim made from some kind of plastic. Don't use any type of metal, if it works itself loose, you're screwed if it's conductive.

can someone please help me to add the shim to my pc i really would like to try it out and see if it works but i cant take off my whole bottom case because i dont have the right tool for that screw thanx

>BGA repair is not uncommon but it's NOT easy work. Secondly, what are they doing to avoid this problem in the future after the repair? (Nothing).

Nothing. How would a repair shop prevent the problem ? It is most likely due to flexing of the motherboard. The user would have to prevent this by not picking th board up with one hand.

>Thirdly, why in the world would ANY shop ask for the end user to dis-assemble the laptop and only send them the logic board?

Because tearing out the motherboard in an ibook is a complete pain in the ass ?

> I don't mean to discredit the end users abilities but that is preposterous.

$50 shipped back to you is a good deal, as far as I am concerned. OTehrwise, you pay $200 to DTT for a replacement motherboard that is just as likely to fail if you stress the motherboard.

Mad Dog

What I want to know is if somebody has a complete listing of all the models affected. I know my 500mhz icebook is safe, but I have been looking in to getting something a bit faster, and want to know what machines are prone to failure. I assume the last iBooks, the 1.33ghz 12" and 1.42ghz 14" are alright, but I would like to know for sure.

-digital

I fix Macs, so I've had a chance to get some idea about which of the iBooks fail with the loose graphics chip problem. The iBook G3 500 MHz isn't immune to this problem--I'm not sure where you got that information, since it's one of the most affected by it. Not all of the 500 MHz models will have the problem, maybe not even the majority, so if yours doesn't, it's not that much of a statistical fluke. Many of the iBook models are affected by the loose graphics chip problem, though of the iBook G3 series, I see fewer of the 800 MHz and faster models with this problem. That might be mainly because fewer people bought these, having already bought the earlier models. On these faster iBook G3s, the graphics chip is on the top of the logic board, but on most of the slower iBook G3s, it's on the bottom. Of the iBook G4 series, I haven't seen enough of these with the problem to get a statistical idea about them, but so far, like the slower iBook G3s, it seems the slower iBook G4s are more affected by it.

Like you all I've suffered through this problem too. One very unusual thing I've discovered while using my iBook as a firewire drive to my iMac (to retreave files from the iBook) is that the screen worked everytime I powereed it up this way. Is thie just random and has nothing to do with the logic board? I'm hoping this information might help with a solution.

Bentshape, that's interesting that your iBook's display would work while the iBook was in Firewire Target Mode. I think I've seen this before too. Instead of cracked solder joints under the graphics chip, it or the video RAM (which is outside the graphics chip) might have some internal damage, that might not affect its output when it's displaying the Target Mode's simpler, mostly black screen, with just a Firewire icon bouncing around. Or, the graphics chip might still have some cracked solder joints, but ones which somehow aren't used when running the display during Target Mode. That sounds far-fetched, but when I've pressed on the graphics chip in some iBooks while they were running, I got a few surprising results on the display that I wouldn't have expected.

It's also possible, as you're wondering, that the logic board might not be at fault. If this were the case, the next suspect is the inverter board, inside the display housing, which delivers the high voltage to the backlight bulb in the display. But this would be possible only if, when your iBook's backlight sometimes isn't lit, you can see a dim image onscreen (you might have to shine a flashlight on it to see the image). I've seen a few inverter boards light up a laptop's display, even when the inverter board had a problem, when the logic board was driving the inverter board in some way that the inverter board was "happy" with, mainly when the logic board was telling the inverter board to run in some reduced brightness/lower voltage mode, or when coming out of sleep (briefly lighting the display, but then going black again quickly, etc.). Maybe with these inverter boards, they stop putting out voltage when the logic board tells them to put out full voltage.

As you can see, there are several things that can cause identical, or nearly identical symptoms, like a black screen. Everything in the chain of electronic components that runs the backlight, has to work properly, or else you can get a black screen--either one that never lights up, or lights up intermittently, etc.

I see this forum has become rather quiet. So I'm hoping it's still being watched some.

I recently had Firstphase technologies reflow my G3 900mhz iBook GPU. When I received it back, the video was totally gone while the rest of the laptop appeared to work (judging by sounds and lights). They offered a complimentary hotter reflow in hopes it would fix the video. Unfortunately I still have the same symptoms. At this point they state the problem is something else and they can't fix my board. So I'm out everything with them. I don't fault their customer service. They just didn't fix it is all.

I thought about Superior reballing the video chip, but given the horrid reports I'm reading here, I shudder to think of releasing my board to them. Even though the process sounds like just what I need.

How else does one find a board or who else has been found that can rework these boards? I have found new boards priced from $500-900.00! I would rather get a new MacBook then spend that on my G3. But this computer works just fine, if it only had video. I just hate to throw it away.

>BGA repair is not uncommon but it's NOT easy work. Secondly, what are they doing to avoid this problem in the future after the repair? (Nothing).

Nothing. How would a repair shop prevent the problem ? It is most likely due to flexing of the motherboard. The user would have to prevent this by not picking th board up with one hand.

-----> No, case / board flexing has been happening for a long time with various manufacturers / models. Once again, this is not the first time this type of problem has occured. This can be eliminated by doing research and developing a real solution. Period.

>Thirdly, why in the world would ANY shop ask for the end user to dis-assemble the laptop and only send them the logic board?

Because tearing out the motherboard in an ibook is a complete pain in the ass ?

-----> Thank you, even more reason for the end user to not attempt to disassemble this laptop. This blows my mind eveytime I think ANY repair service would even remotely suggest the end user tamper with and increase the potential for additional damgage to the circuit board. Unbelievable.

> I don't mean to discredit the end users abilities but that is preposterous.

$50 shipped back to you is a good deal, as far as I am concerned. OTehrwise, you pay $200 to DTT for a replacement motherboard that is just as likely to fail if you stress the motherboard.

-----> You do realize that you're paying these people $50 to place your motherboard in a low grade, early release BGA (can't believe I'm calling it that) machine for 3 minutes right? You think thats a good deal?

The chip should be removed
The area under the chip needs to be inspected for trace damage
If damage under the chip exists (usually it does) this damage should be micro welded (no jumpers, wires or Bull Shi* like that, not soldered either) welded! Yes, yet another $10,000 machine used during this process.
The chip should be reballed
The chip should be re-mounted to the board
proper post repair steps (usually company specific)should be taken to avoid this problem in the future.

Guess what, you're getting none of this for $50. I was beside myself when I seen the repair charge.

Hi there

My dual usb 12" G3's gpu went poof, and I'll be getting it back from superior any day now(after 4 months).
Now, I was thinking of shimming when I put it back together. Since this gpu failure is caused by the chip getting too hot, and then separating, would substituting the shim for a heatsink be a good idea?

I know there isn't enough space, but I don't mind cutting holes, and putting raised feet on. I've got a little heatsink that's 40mm square and 12mm high, and that'll cover the chip+ram. I'd mount the 'sink to the bottom casing and use its clamping force to maintain contact withe the chip.

I also saw a heatsink+fan on ebay, that has the same dimensions as my 'sink. And that got me thinking, could I find a power take-off somewhere on the board to power the fan? Would it be worthwhile, or will I be OK with just a regular heatsink?
I can picture the fan blowing hot air into the casing, with nowhere for it to exit. Unless I made some ducts, and drilled vent holes around the heatsink.

This is the fan+heatsink I was considering:

http://cgi.ebay.co.uk/VGA-Video-Card-Chipset-Socket-Cool-Cooler-Cooling-Fan_W0QQitemZ180300695512QQcmdZViewItem?hash=item180300695512&_trkparms=72%3A1300|39%3A1|66%3A4|65%3A12|240%3A1318&_trksid=p3286.c0.m14

Since this gpu failure is caused by the chip getting too hot, and then separating, would substituting the shim for a heatsink be a good idea?

I thought it was from the flexing of the board and a not so great solder job on the GPU solder balls. The GPU shouldn't be putting out enough heat to have an effect on the solder.

From what I've gathered, it wasn't that the ball grid array was badly done, it was just the wrong system for the job.
In the same way that a steam locomotive can pull a load, but a diesel engine does it better.
And yes, flexing played on the BGA systems' achilles heel, and combined with the heat generated by the chip caused the solder balls to fail(crack).
If you take a strip of metal, fold it in half, then flatten it out, then fold it in half again. Repeat this a few times and you get heat(compare this to the heat generated by the GPU). Keep going, and the plate snaps in half(just like the solder balls).

So, Heat+Stress=Fatigue(weakening). With the right amount of both, you get failure.

Correct if I'm wrong(probably am), I'm not an expert

I get what you are thinking of, but I doubt that the heat generated in the solder balls and by the board flexing is in any way damaging much beyond the flexing itself. Repeated flexing of a metal, even without the addition of heat, will cause failure if the flexing is beyond the ductile strength of the metal.

So anyways. I've gone ahead and fitted the heatsink. Looks cool(black heatsink+white ibook, very stormtooper-ish).
I still have to get some extended feet, as the 8mm ones I put on before wont do now. I'm also going to fit a fan grill around the 'sink area, just to be safe.

>BGA repair is not uncommon but it's NOT easy work. Secondly, what are they doing to avoid this problem in the future after the repair? (Nothing).

Nothing. How would a repair shop prevent the problem ? It is most likely due to flexing of the motherboard. The user would have to prevent this by not picking th board up with one hand.

-----> No, case / board flexing has been happening for a long time with various manufacturers / models. Once again, this is not the first time this type of problem has occured. This can be eliminated by doing research and developing a real solution. Period.

*** *[comments by cayoopr]

*** * What a self-serving, disgusting comment ! "Doing Research . . . for a REAL SOLUTION" ? Who in their right mind would do such for an obsolete machine with a market value of _maybe_ $200 ? What situation would justify this effort and money ?
Nothing in the real world, for certain. And, most certainly, not Apple.
________________

>Thirdly, why in the world would ANY shop ask for the end user to dis-assemble the laptop and only send them the logic board?

Because tearing out the motherboard in an ibook is a complete pain in the ass ?

-----> Thank you, even more reason for the end user to not attempt to disassemble this laptop. This blows my mind eveytime I think ANY repair service would even remotely suggest the end user tamper with and increase the potential for additional damgage to the circuit board. Unbelievable.

*** * Another totally bogus comment, typical of ripoff mechanics of all stripes. If they can do it, you can do it. Their "experience" simply means that they've already broken it a couple of times, and, most likely, charged it to the sap who owned it.
________________________

> I don't mean to discredit the end users abilities but that is preposterous.

$50 shipped back to you is a good deal, as far as I am concerned. OTehrwise, you pay $200 to DTT for a replacement motherboard that is just as likely to fail if you stress the motherboard.

-----> You do realize that you're paying these people $50 to place your motherboard in a low grade, early release BGA (can't believe I'm calling it that) machine for 3 minutes right? You think thats a good deal?

The chip should be removed
The area under the chip needs to be inspected for trace damage
If damage under the chip exists (usually it does) this damage should be micro welded (no jumpers, wires or Bull Shi* like that, not soldered either) welded! Yes, yet another $10,000 machine used during this process.
The chip should be reballed
The chip should be re-mounted to the board
proper post repair steps (usually company specific)should be taken to avoid this problem in the future.

Guess what, you're getting none of this for $50. I was beside myself when I seen the repair charge.

*** * More bogus intimidation.
What do you think Apple did under their replacement scheme ? They gave you another bad board that was set for failure, for the same fundamental reason. And, if it hasn't failed yet, it will.
$50 bucks will get you at least as good a fix as Apple gave you. And, Apple didn't consider for one nanosecond all the baloney fix procedures listed above.

If you want to see how simply it is to get the board back and your machine into working order, just go to this little page:
http://geektechnique.org/projectlab/726/diy-obsolete-ibook-logic-board-repair

Now this little routine is a good way to incinerate yourself and your neighborhood, but it is an irrefutable example of what the problem is, how fundamental the 'fix' is, and how very fundamentally it is taken care of, with whatever equipment is at hand.

But, if you want the "$600 toilet-seat-solution" as our master repairman suggests, you are certainly welcome to it.

Argh, I've got a similar problem to this on my iBook G4. It will run okay sitting on a counter with my wireless keyboard attached for hours. I even played a session of darwinia to stress the chip a little. no issues. as soon as I start using it as a normal laptop, it will randomly crash... be it within a few minutes or under an hour. the 'freeze' is where the cursor still moves but I can't interact with the desktop at all and the time in the upper right corner is frozen. I can also cause an instant freeze by applying a little pressure on the left track pad. I think it sounds like that power chip that CWArnold found on the logic board. what other major chips would be in the middle of the left track pad?

Hello,

So, I've been experiencing this same problem as well. I loved my iBook G3 so much. It went through three logic boards during the "not a recall" recall. I was able to get the third one after the expiration of the recall through a friend of a friend situation Genius bar encounter, but not more than a year later, I was unable to. My argument was that I've had it done after the recall expiration date free of charge before so why not again?

Anyway, I've tried the heat gun method twice with limited success.

A good friend of mine is an engineer at Harman Becker and is willing to help me reflow or reball the solder.

He had two questions I don't understand but was hoping someone would know the answer to.

First, what is the pitch?

Second, he asked if the board and solder balls are clean, how do I determine this?

Thanks!

I also had a similar problem on our 500Mhz ibook with the corroding PRAM capacitor. Her ibook one day booted up and no video, tried zapping PRAM, reset switch, no joy. Then the ibook decided it didn't want to power on at all. Took it apart and found the bad capacitor. Luckily I had a broken ibook logic board from when I repaired my sisters ibook. The capacitor on that board looked and measured ok. Soldered it in and the machine still doesn't want to power on. When I removed the bad cap there didn't appear to be any trace damage so I'm a little stumped. My wife says to let it go and she wants me to buy a Macbook for her birthday in April. I've had it with trying to repair ibooks as I have already repaired this ibook as well as my sisters and unless I have any moment of genius, its going into my pile of broken computer parts.

Currently I'm typing away on an old parts queen Titanium G4 that I resurrected from the trash-pile at work, but I don't know for how much longer. The screens got a flickering video issue with a green line on the far left side, the DVD drive is intermittent and makes squealing noises, the airport connection is flakey and it has the usual TI broken latch issue. Now if it only makes it to april...

As a follow up to my G3 with the GPU solder connections issues. Rather then throwing my laptop away, I decided to just send Superior Reball and Rework my G3 board. This was a worry after reading the reports (old reports) about Superior's lack of customer service if not their ability to repair these GPU problems. I gave Firstphase Technologies 2 tries at my expense already. Both times were absolute failures.

I was very pleased with Superior's service. Not only did they actually reball the GPU chip as evidenced by the shinny new solder connections, but the unit works flawlessly. And the turn around time, given that I live in Alaska, was very acceptable.

The only negative I could possibly submit concerning Superior Reball is I wish they would answer their phones or return calls so I could have paid for my repair faster! I was very anxious to see if I had a repaired board or not.

In closing, I wouldn't try these self fixes again. It's a lot of work for shotty results at best. I say this as now I have a working G3. I have all the confidence that for years to come it shall stay that way.

had an electrical engineer check the SOIC chip on my ibook g4 1 ghz. one solder trace looked bad so he used a loop and soldered it. now works 100%. what an odd component to fail.

Me TOO!!!

This MacGyver type fix really worked. I am a novice and opening up my laptop was a little worrisome, but alas, eight business (1.5MM) cards taped to the underside at the square indentation, slap back together and presto! It works! Just thr right amount of pressure. Thanks John Sawyer!

Just get the right tools and a dish for your screws and springs. Frankly, it was more difficult and time consumming to get this post up.

Happy New Year!

Ok, I know this thread is quite old and we all know about the GPU problems that plagued the G3 ibooks, but I may have a solution that's a little on the crazy side but it works in some cases.

We all know the fault lies in the GPU processor having poor solder connections on the BGA, and the solution being having the BGA reflowed or "reballed". Reflowing or reballing a BGA requires special equipment like a hot work station. However, unless you know someone who has this equipment, it can be expensive to have it reflowed and for a 10 year old computer, may not be cost effective.

Now I noticed on the internet how some macbook owners with similar faulty logicboard problems were taking their logicboards out of the macbooks, stripping them, and then sticking them in a conventional oven to reflow all the solider joints. In many cases, their logicboards were returned to fully functional condition. Their formula varied on time and temp, but the majority of them used 7 minutes @ 385 degrees (Fahrenheit), with a gradual cool down (oven off with the door slightly open)

Now I have several old ibook G3 800Mhz logicboards with the dreaded GPU problem. I bought 3 of them off ebay last year for 10 bucks, knowing I was going to dink around with them to try to build up one working machine for my 4 year old daughter. At this point I really didn't have too much to loose, 10 bucks in and three boards to try.

Prep work is also key, I prepped the board by removing everything that could easily be taken off, removed all the stickers including serial number sticker (it isn't going to hurt at this point, the boards are soo far out of warranty) because paper is flammable at 451 degrees and we close to that, not taking any chances. I removed the modem port as it is just clipped in, and I desoldered the PRAM battery/capacitor, again being safe. I also removed the plastic trim square around the GPU and clean off all that yellow foam heatsink crap off the GPU. Also after removing the plastic trim, pick off the glue that held it in place around the four corners of the GPU with an Xacto knive.

To make a long story short, I was able to resurrect one of the boards using this process (cooking the board GPU side up) and am now typing this post on the resurrected ibook. I going to try to experiment a little more with the other boards and see if I can the "recipe" down. While I did get the board working, my oven may be off a little as 3 large surface mount components desoldered (fell off) from the other side, so I had to resolder them on, but the board works flawlessly so far.

So, if you are daring and have nothing to loose give this a shot, here is one of the websites, otherwise google "macbook logicboard in oven"

http://phaq.phunsites.net/2011/03/08/baking-recipe-to-fix-broken-macbook-logicboard/

bon appetite