Apple iBook G4 Design Flaw Proven

empaler writes "Apple has long denied service on iBook G4s whose screens went black after just over one year of use, denying that there was any error. But now, the Danish National Consumer Agency has released a report proving that the error is due to a design flaw. So far, the only news site picking this up is The Register (unless you understand Danish). The Danish Consumer Complaints Board says that Apple needs to get a grip and acknowledge this error in the rest of the world. The NCA also has some photos from the report (explanations in Danish, but easily comprehensible from context)."

It's a dry joint.

[Gordonjcp]

Reflow the solder. Simple.

Incidentally, with the introduction of RoHS-compliant lead-free solder, you will see this more and more. Consumer-grade lead-free is so crappy that it's almost impossible to make a single working board without at least some reflow work. Oddly enough, military- and medical-grade equipment are exempt from needing lead-free solder. Wonder why?

bah

[nomadic]

My G3 ibook did the exact same thing, and it was also a logic board failure. Apple has had lousy QA for several years now, and as someone who actually LIKES Apple products it's extremely frustrating.

Re:photos?

[$RANDOMLUSER]

I got to the photos before the site got borked.borked.borked - it shows solder traces cracking after repeated flexing - huge surprise.

The article doesn't show...

[msauve]

where this component is located. Comments here talk about using a C-Clamp (which is also shown in the Danish photos) as a workaround to the problem. A quick bit of searching produces this site, [coreyarnold.org] which shows that the chip is nowhere near the power button, as you claim. In fact, it appears that the power button mounted to a small, completely separate PC board, in accordance with good design practice.

Re:It's not a bug...

[brkello]

Would you be saying the same thing if this was made by Sony or MS? If Apple does something wrong then they should be called out for it. Just because it is currently trendy doesn't mean it can't have faults.

Powerbooks?

[SanityInAnarchy]

About the same thing happened to my Powerbook, and it still hasn't been fixed. Apple refuses to fix it, because it was dropped about a year ago, and if there is any physical damage at all (so much as a dent), the warranty is void. Since they will only do complete and total repairs, it would cost $1200 to fix.

So, my question: Does this also happen with Powerbooks? And if so, is it something I could easily fix by cracking it open and soldering something? Any step by step instructions on how to do so?

Re:It's a dry joint.

[monkbent]

You don't even need to solder. Just disassemble the iBook (challenging enough!) and put some sort of filler on top of the graphics chip. I used a 3M rubber foot. Close everything up and the bottom of the case will keep the chip in place. I haven't had a problem in the 6 months I did this repair, and have continued to tote my iBook all over on my scooter.

That said, it's clearly a design defect, and should have been fixed by Apple just like the G3 iBook.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:It's not a bug...

[Mylina]

It's not "just a few laptops". There are petitions with over 5,000 signatures of people whose laptops have died and if you google "iBook G4" and "logic board" most of what you will see is boards full of people whose iBooks have died within 13 months. A laptop that you spend over $400 in general should not break after only one year of use. You'd be singing a different tune if it were yours.

Design Flaws are part of Apple's history

[swisswuff]

We love Apple for their design flaws, don't we!!

The SE/30 featured an shielding cardboard that was coated with metal on the bottom; the motherboards pins would at time poke through, and if the machine was sitting vertically for a while, suddenly it could go - blam - and stop working and go into some type of hangup or crash. For two models, I could repeatedly get them going by shaking the machines upside down, but neither Apple, nor Apple dealers, acknowledged the issue. It may be that having parts installed - and having people disassemble and reassemble parts of the Mac - may have played a role.

All of the compact Macs featured floppy disk drives that, over a working period over more than 6 hours, would reliably and predictably cause floppy disk errors. So I'd start using a new OS floppy after 5 hours, and things would be o.k., or not do it and consistently get crashes from 7 hours on upwards. They had built in a bright CRT, and obviously, shielding was some issue there. Nevertheless, this was an obvious design flaw.

Or the iPod mini. The iPod mini featured some weird shielding problem whereas crackling noise would occur. It would disappear as soon as the components that are stuffed together (battery, main board, micro disk) were pulled from each other - then, no crackling noise would occur even when mechanically straining the 3.5mm jack. Another design problem where capacitor- and shielding-related issues determined the outcome.

Or, take the Powerbook G4 Aluminium "Narcolepsy" model Apple built and sold! A design flaw classic. Not admitted by Apple, ever. I guess they switched to Intel partly because there were so MANY of these sold, that switching to Intel may have been the only way to give the Powerbook G4 owners a good reason to buy a new Apple laptop rather than attempting to force Apple to fix their old one. Maybe one day, we will hear the insider story of that botched up piece of hardware?

Also, there were a number of Powermac G5 computers that all had severe logic board problems that I laid hands on - two of them DOA (Dead On Arrival), and on another one I just got it repaired for the cheap sum of around 800 dollars.

So, I think if anything is newsworthy it'd be publishing that Apple actually managed to assemble some parts without design problem. That'd be what Slashdot may want to focus on, not that Apple "yet again" was shown to have screwed up something we all knew they couldn't get right to begin with.

Anyone believe that they can get the iPhone right, at all?

Thank the European Union for this. Seriously.

[SeaSolder]

Disclaimer: Both myself and one of my coworkers are IPC-610D certified inspectors of electronics. For those of you who aren't in the know, IPC-610 is the industry standard that dictates how all electronics assemblies are supposed to be put together. I spend considerable time inspecting fine pitch surface mount devices. I am also the manager of the assembly department at my company, where we make class-2 commercial grade electronics.

To say that I spend a lot of time looking at soldered joints is an understatement.

So, what's the problem here?

The Danish government is just as culpable as Apple is on this one.

The solder used in this joint is obviously (to me at least) a lead-free solder. (Lead free solders are exceptionally grainy and dull in appearance compared to leaded solder that is smooth and shiny.) The switch to lead-free solders was mandated to occur last year on July 1st by the European Union. Any electronics shipping into EU member nations must comply with this new standard, which is called RoHS, or "Reduction of Hazardous Substances". (Look it up in Wikipedia.)

Due to complexities in maintaining parallel manufacturing lines, most companies simply switched to the Lead-free solder for their entire product line. (As did my company). This means that most all new electronics you by are going to have lead-free solder holding them together.

So, why is this a problem? Plasticity.
Leaded solder alloys (SnPb 60/30) are extremely ductile. This means that they will flex a considerable amount before fracture occurs. With electronics that experience heat cycles, or any kind of motion at all, this is an extremely beneficial trait to have. Lead-Free solders on the other hand (like the most common SAC305 SnAu3%Cu0.5%) are incredibly brittle. What is obviously happening here is that the heat cycling from the laptop turning on and off is stressing this solder joint, and causing the joint to fail. Had this been leaded solder, I can almost guarantee that this problem wouldn't exist.

So what does this mean? Exactly what the electronics industry has feared. The EU made a dip$#!t move. Industry experts believe that the average lifespan of an electronic device has been significantly diminished. Down to an average consumer product lifespan of 5 years. There have been early reports of serious quality problems, including SWATCH having over 10,000 watches fail within a month of shipping. You can expect to see a drastic shortening of the lifespan of your electronic goodies. This is just the beginning people. If you see a device that is labeled as RoHS compliant, do not buy it if you expect to use it for more than a few years.

Re:A dry joint on thousands of units is a design f

[SeaSolder]

What do you know about electronics manufacturing? 1. "Higher Quality Solder" There is no such thing as "higher quality Solder". There are two industry recognized solder alloys out there. SnPb 60/40 and SAC305. SAC305 is used on RoHS compliant assemblies, and is mandated by the EU. From the photos, it is obvious to me that this solder was SAC305. (Yes, I can tell. From the side-cuts, you can see that the solder is very grainy and dull. This is the hallmark of lead-free solders.) Again, due to the idiotic RoHS legislation, manufacturers are REQUIRED to use a lead-free alternative, among which the SAC305 is the best. (The proverbial winner of the special-Olympics.) 2. "A far more substantial joint". The solder joint in the photos EXCEEDED IPC610-D Standards for even Class 3 medical devices. This is a SURFACE MOUNT DEVICE, and as such, there is no other way that you can get a more "substantial" joint, unless you went to a TH DIP package, if it was even available, and if you could even fit it into the design. 3. "Right type of solder" See number 1. 4. So no, it is NOT entirely Apple's fault. I'd bet $100 that if this chip were removed, and a new one were installed with SnPb solder, then it would not fail. Period. But since Apple isn't allowed to use the right solder, this happened. Have you ever designed something? Can you say that you tested EVERY POSSIBLE point of failure? No, didn't think so. It's because it's impossible to do. There is a problem, and it's the knee-jerk reaction of the EU to blindly ban the one substance that has been fully responsible for allowing the computer revolution to occur. Way to go!

Re:Consumers should just know what to expect

[v1]

That does look like a manufacturing defect, but it's not common. I have worked on hundreds of macs this year and I have only seen maybe one or two ibook G4s that had dead displays. It's not anywhere near the scale that I would say deserves a recall. If every 1 unit in 200 is bad, the model is not defective. Now if we are talking 1 in 10 maybe we can start saying there is a problem.