When manufacturing a product, its easy to make oversights, and somewhat impossible to predict the future. For this reason, products change over time. This is true for all manufacturers, in all industries. Engineering isn’t easy.
When an oversight is so huge, and the manufacturer accepts responsibility, the product must be deemed faulty, recalled, and discontinued (See Galaxy Note 7). In other cases, the manufacturer may recall the product for rectification of the issue, or release a software update to correct the issue.
This is nothing new. When a team of hardware, software, and structural engineers work on a product, all it takes is the misplacement of a single decimal to cause an issue down the track. Most of these things are picked up prior to the release of the product, but post release issues aren’t going away any time soon.
Engineering the Apple iPhone
Over the past few months, we have gathered a list of the most notable issues in the Apple iPhone, and what was done to correct them in the next model. It’s important to note that correcting the issues wasn’t the only reason for some of these changes. They also had aesthetic and functional benefits. It’s also important to note that Apple never acknowledged many of these issues, however they did rectify them in future models. It’s fair to say based on that, that Apple were well aware of the issues, but didn’t want to admit fault. Admitting fault to an engineering oversight after all, would have negatively impacted sales and marketing.
We have done our best to make this list objective, and include as many sources as we can. This isn’t just a list of our opinion and experience, but almost everything listed can be verified from multiple sources.
iPhone (aka iPhone 2G)
The Original 1st Generation iPhone (released June 29, 2007) featured some of the most cutting edge technology seen at the time. For example, it was the first mobile phone with multi-touch. Unfortunately as a result of being so feature packed, and despite the astounding user reception, it came with its own issues.
Having its own never before seen OS (originally marketed as running OSX, but released with “iPhone OS 1“, now recognised as iOS), the Original iPhone was “buggy as hell“. It’s important to note that this was mostly due to software, and the initial release of any software is always going to have some unforeseen issue. Software engineering on a large scale is after all pretty complex. The instability issue on the original iPhone however was well predicted:
“It often ran out of memory and had to be restarted if made to do more than a handful of tasks at a time… Jobs had a number of demo units onstage with him to manage this problem.”
as reported here by Fred Vogelstein.
It’s not completely accurate to call this an engineering failure, as it was actually intended design. Even so, it was certainly one of the most notable features of the 1st Gen iPhone that left users with a poor experience. Apples decision to recess the 3.5mm headphone jack, left many users forced to use the supplied headphones. It’s not definitively clear what the reasoning behind this decision was, be it structural integrity of the port, or having some level of control over third party accessories dipping into potential profits. Regardless of the reasons that led to the recessed headphone jack, it left a sour taste in the mouths of consumers. Those who already owned an iPod, and planned to replace it with their new iPhone, found their existing high quality headphones, vehicles auxiliary cables etc, did not fit without the use of some type of adaptor.
The iPhone 3G (released July 11, 2008) was the first model we saw here in Australia. A year later it was followed by the slightly faster iPhone 3GS (19 June 2009), which also featured a better camera, and twice as much memory. Like many other smart phones at the time, and unlike its predecessor, it had a plastic housing. Some would argue this reduced the risk of damage to the glass in the event of shock.
The previously recessed headphone jack was made flush due to the level of negative feedback from consumers, resulting in applause during the keynote presentation.
The iPhone 3G came shipped with iPhone OS 2. The new OS not only corrected many of the memory management issues plagued by the original release, but brought with it the launch of the App Store. Finally, third parties could produce apps for the device, such as games, utilities, etc.
The device(s) were capable of connecting to 3G networks, for substantially faster Internet access than the prior model.
Unfortunately, while Apples incorporation of a plastic housing may have reduced shock to the glass front panel of the device, it did little to protect the battery directly attached to it. As a result of this structural engineering oversight, a large number of users saw the battery in their device expand. This resulted in the screen pushing outward, the motherboard to bend under extreme stress, the rear plastic to crack, and in extreme cases fires. There is a number of resources available providing proof of this. This wasn’t the only cause for battery swelling. In addition to the potential to crush the battery due to its flexible plastic housing, heat from the motherboard is directly transferred from the ground plane to the battery itself. This is due to the sandwiched design of the components, and Apples decision to layer the components in these models, rather than situating them side by side with a reasonable air gap between them.
Possibly directly related to the battery swelling issue above, the iPhone 3GS was found to overheat in certain conditions. In many cases, the excessive heat resulted in discoloration to the rear plastic housing. Apple released a support article which has been updated over the years, advising against subjecting the device to excessive temperatures. While many say this was an admission of a fault, others suggest they merely warn that operating devices in hot environments is a bad idea. Regardless, it was an issue that seemed to effect the 3GS specifically, and is widely believed to be due to the battery itself.
“Store the device where the temperature is between -20º and 45º C (-4º to 113º F). Don’t leave the device in your car, because temperatures in parked cars can exceed this range.”
taken from Apples Support Article.
To 3G, or not to 3G
The first of the iPhone Antenna woes, but its arguable what the issue really was, or if there was ever an issue at all. Well, many, many, many, many people say there was. Interestingly enough, you’ll have to see the evidence of this on the Internet Archive, as Apple have deleted all evidence of such from their support forums. All that remains is a very short thread proposing the possibility of a design flaw, with the only responses being arguable rebuttals from a higher level contributor. These comments are fairly unconvincing though, given they show a lack of understanding of how reception is measured. Its safe to assume that the original poster was not imagining things, and that hundreds of people really did experience dropped calls, weak 3G signals, and slower than expected Internet connections. So what was the problem? Was it an issue with an Infineon Chipset? We may never know directly, but there was indeed an issue caused by software that wasn’t picked up until much later down the track. See the iPhone 4 section for more on that.
Upon the release of the iPhone 4 (June 24, 2010), it was pretty clear apple had rectified the issue with the plastic housing from previous models. The new device had an alloy frame around the perimeter of the device, and a full glass back. In comparison to other devices at the time, the iPhone 4 was an engineering marvel, so it’s no surprise it was so popular.
The metal frame extended to the middle of the device between the battery and LCD, preventing battery swelling should it occur, from pushing the display outward. In addition, the motherboard was now a much more dense, compact design, that was no longer at risk of being damaged by an expanding battery. This also made battery replacement far easier, as only 3 screws needed to be removed to perform the replacement. One in the battery terminal, and 2 in the rear panel.
The rear glass panel was held in place by a plastic frame, with small plastic clips. This resulted in a good breakaway point for the panel should battery swelling occur. Apple knew lithium batteries swell, and it’s not possible to avoid every circumstance resulting in that. They did however implement damage control, to isolate the potential issues to the battery itself.
No Microphone in Calls
The metal frame gave the battery far more protection than the plastic frame. Unfortunately, the metal caused vibrations and shock to resonate to other components directly secured to it. The most known effect of this was a fracture at the Audience IC (U62, aka Highland Park). U62 was located very closely to one of the main screws securing the motherboard, in one corner of the device. The IC was responsible for some degree of noise cancellation in phone calls, hence the microphone still functioned fine outside of a call.
A fracture at the Audience IC caused the microphone not to work when in a phone call. This was occurring in a time when microsoldering was a rarity in the phone repair industry, so most attempted to correct the issue by placing high density foam between the EMF shield and the Audience IC, putting pressure on the IC. Because of this, there are still iPhone 4s all around the world, that function perfectly aside from microphone in calls. A large percentage of the ones that do function in calls, have something pressing the Audience IC against the motherboard.
That wasn’t the only issue caused by the metal frame. Within hours of the release, many users reporting a substantial decrease in cellular reception, depending on how they hold the device in their hand. This was caused by the design of the antenna, which was part of the metal frame. Holding the device in a certain way caused attenuation of the cellular antenna, due to connection to the Wifi antenna via the users skin.
Officially, Apple responded to this issue with “don’t hold in that way“, an answer which many new owners did not accept as a sufficient resolution. As a result the “bumper case” became more popular, with Apple stores giving them to consumers who were facing the issue at no cost. The issue did however appear worse than it was, when combined with an apparent bug in software. The bug resulted in an inaccurate signal graph, so the drop in signal seemed more severe than it actually was. With that said, there are unconfirmed reports that Apple knew about potential issues with the iPhone 4 Antenna, and that Steve Jobs swore on stage when confronted with this notion.
“Upon investigation, we were stunned to find that the formula we use to calculate how many bars of signal strength to display is totally wrong… Their big drop in bars is because their high bars were never real in the first place.”
taken from Apples Letter to iPhone 4 Users.
With the change in design, the battery connection was no longer secured by the motherboard.
Instead the battery terminal was now held down with a single screw at one side, and a small dimple on the other that pressed into the EMF sheild. This resulted in users experiencing half disconnection of the battery terminal after a drop on the right angle.
After this, placing the device on a table firmly was enough to cause it to turn off. Repair shops all over the world were flooded with “just needs to be reconnected” repairs. The notable symptom was a device that immediately powered off when disconnecting the charger.
Power Button Failure
Prior models of the iPhone used a Tactile Dome Switch, with switch contacts incorporated into the Flexible Printed Circuit (FPC). As of the iPhone 4, the power button was replaced with a Surface Mount Tactile Switch assembly.
The switch assembly was low profile, but the switches housing prevented actuation of the switch without a spacer. The addition of this spacer which was made of some form of plastic, created a new issue.
As the spacer wore away, the tactile switch would become less responsive, until eventually it failed to “click” completely. As this issue became more common, a class action lawsuit was launched against Apple. The class action was dismissed, due to the issue occurring outside of the warranty period, with Judge Gary Feess stating Apple had “no duty to disclose the alleged power button defect”. In court, the safety concerns raised were found to be “speculative”, and there was little evidence to support misrepresentation.
“Regarding the representations made by the Apple employees regarding the repair ability of the alleged power button defect, Plaintiffs fail to indicate why the representations made to them were actually false.”
taken from Minutes in Chamber.
This issue was never fully rectified, and all future devices have the potential to suffer the same fate. However, the bracket holding the tactile switch would later be replaced with metal, reducing “flex”, and therefore reducing abrasion to the spacer.
The release of the iPhone 4S (October 14, 2011), brought improvements to the previous engineering issues.
The antenna attenuation issue was corrected by separating the metal parts of the frame from 2 to 4 pieces. The battery connector issue was resolved by adding a second screw to secure the connector on both sides. Finally the Audience IC issue was resolved by removing and relocating some of the components on the board.
The boards design was inspired by the CDMA version of the iPhone 4, which had less issues than its GSM counterpart.
Unfortunately as expected, re-engineering of the device introduced some new issues.
Wait, this was fixed wasn’t it?
The battery terminals were now secured together with 2 screws. However, a lack of solder paste, and an incorrect thermal profile during production, resulted in a large number of fractures at the battery terminal.
Having good connection to the battery was important for current consumption, especially when the device was in a high power consumption state, such as when the Portable Wi-Fi Hotspot was in use. Due to this lack of solder at the joint, the previous fault from the iPhone 4 resurfaced, only now it required microsoldering to repair. The few motherboard repair technicians at the time likely remember doing an immeasurable amount of battery connector replacements.
Wi-Fi IC fractures
With the relocation of the Wi-Fi IC to a small tail section of the upper board, all by itself, a new issue occurred. First of all, the IC was placed right next to one of the motherboards main securing screws. This meant shock resonating through the frame and into the board, had a new part to damage.
On top of this, the lack of ground plane in this area didn’t provide as much thermal mass. The lack of heat dissipation, combined with the location of the Wi-Fi IC in relation to the motherboard mounting screw, resulted in a large number of “Greyed out Wi-Fi”. This was made more prominent as Cellular Telcos started providing more data, and Hotspot/Tethering use increased.
If you were using your iPhone 4S as a wireless AP, and happened to drop it, there was a good chance you would have Wi-Fi failure soon after, if not immediately.
There were a number of possible solutions to this issue, including “reflowing” the Wifi IC. The only permanent, and true solution to this issue however, was to reball/replace the Wifi IC. Everything else was temporary, even if it lasted months.
There was never any formal acknowledgement of the issue by Apple, but as you can see here, it was far from an isolated occurrence.
The first ever real case of Touch IC fractures, were seen in the iPhone 4S. Again, this was caused by its proximity to a motherboard mounting screw, and its placement on the narrow most part of the board.
At some point during production, Apple (or Foxconn rather) started adding underfill to the Touch IC. The underfill reduced the cases of Touch IC fracture, but didn’t eliminate it. Now there is an even mix of boards with underfilled IC’s in the wild, as there is without.
The fractures occured at the ball from shock, rather than the pad from flexing. The IC’s with no underfill could usually be repaired by adding flux and reflowing the IC. Underfilled IC’s required a reball or replacement, which was more time consuming, but also less common.
The iPhone 5 (released September 21, 2012) was another major change in the engineering of the Apple iPhone. With a thinner device, redesigned aluminium housing, larger screen, and 4G/LTE, it was a clear winner. The iPhone 5 design was so popular, it barely changed at all when the next model was released. In fact, it was re-used years later.
The battery connector issue was finally solved in multiple ways. First the connector was secured by 2 anchor points per electrical connection, and 3 structural anchors per end. Second, the connector was sandwiched under a metal plate, with screws at either end, and high density foam in between. This model was followed in all future designs, with slight variance.
The Touch IC’s had been relocated to an area of the board above the CPU, away from major mounting screws, and underfilled after proving to be a beneficial step during iPhone 4S production. The Wifi IC had also been relocated with a much better ground plane, closer to the iPhone 4’s design, and not on a separate section of the board.
The iPhone 5 was the first iPhone to introduce the “Lightning” connector/interface. Previously, the 30 Pin Dock Connector allowed the use of a multitude of accessories on iDevices, by providing a range of interfaces. These included USB, Analog Audio Input/Output, Component/Composite Video, Serial/TTL, and Firewire.
Upon the release of the new iPhone, consumers were disappointed to find the full featured 30 Pin Dock, for which they had an abundance of expensive accessories, was obsolete. Apple released a 30 Pin to Lightning Adaptor to ease the pain a little, but at around $50 AUD and with limited structural and electrical support, it did little to soften the blow.
Tristar (U2) Failure
Apples lightning connector was reversible, a first for the mainstream smart-phone market, and boasted an “All-digital, 8 Signal Design”. In reality the connector is 8 pins, only 4 of which are actually used for “signals”. The 4 pins make up two data pairs, which can be assigned to provide up to two simultaneous functions.
This new connector required the ability to switch the appropriate functions to the accessory connected, and as such required a new IC within the device, called the TRISTAR IC (read more here). The iPhone 5 contained one of the first revisions of Tristar IC, the 1608A1 produced by NXP.
This was a huge blow to the third party accessory market, because cables and accessories had to identify themselves in order for the correct data pairs to be assigned. You could no longer buy a cable that was “just wires”, or get line level audio from your device by connecting directly to the relevant pins. In fact, enthusiasts and manufacturers couldn’t even go out and purchase IC’s that were capable of assigning the correct data pairs. Instead, manufacturers had to first apply, and pay thousands of dollars to become “MFi Certified“. Only then could a manufacturer obtain the proprietary IC’s required. This was seen by many as a means of Apple controlling the accessory market, and making sure they got a cut of the profit. While it did help identify counterfeit accessories, the action jacked up the price of accessories, with Apple making profit on every single product.
Due to Tristar being a proprietary IC without any prior consumer testing/use, it should have been expected there would eventually be some issues.
The Tristar IC was susceptible to damage from unauthorised third party accessories, power supplies that produced a high level of noise, and foreign matter finding its way into the lightning port to name a few. Failure was observed in a range of ways, the most common of which being “Accessory not Certified” messages on certified cables, rapid discharge of the battery, and a device that no longer charged once it went flat.
Unfortunately Apples decision to underfill the IC and place it directly below the CPU, made repair of the failure difficult. This wasn’t just the case for third party repair, but also for Apples own service division. Combined with Apples inability to accept the failure as anything short of consumer misuse, it resulted in expensive repair/replacement options. In many cases devices found their way into e-waste, along with all the now discontinued 30-Pin accessories.
Apple never acknowledged an issue with the Tristar IC, even now in 2018 where the IC has been redesigned entirely as “HYDRA”, but the repair community has always been aware of its potential to fail, especially in the iPhone 5. There are countless articles that cover the specifics, not just from us, but from others around the world.
Not long after a bunch of European iPhone chargers were recalled due to overheating, Apple revealed a potential issue with a “very small percentage” of iPhone 5 Batteries, in devices sold between September 2012 and January 2013.
A service program was started for the effected devices, with Apple offering to replace the defective batteries for free. Unfortunately for those who had any other cosmetic issues, such as a cracked screen, they were forced to pay to fix those issues. It’s unclear just how many of those devices actually suffered from Tristar failure, but there are mixed reports of Apple actually performing a battery replacement, vs replacing the device in its entirety. Consumers were instructed to backup and restore the device prior to service, presumably because Apple were aware that many devices would need more than just a battery replacement. Some consumers however were denied any coverage, due to not meeting the criteria provided to the AASP’s, who were largely unaware of the Tristar IC and the symptoms of its failure that were so similar.
Power Button Failure
In addition to the spacer/actuator issue that began with the iPhone 4, the iPhone 5 had a manufacturing defect that resulted in spontaneous failure of the power button, despite still “clicking” fine. This was usually the outcome after some type of degradation, such as worn contacts resulting in a button that wouldn’t always respond, or that responded as if it was pressed more than once.
While there isn’t a lot of information in regards to the specifics of the issue, we can disclose what we found to be the cause from our own personal research.
The two filters/resistors situated closest to the power button, on the narrow most part of the flex, can become detached from their pads due to flexing. This results in an open circuit. Re-soldering these resistors returned the function to normal, with the exception of any degradation that may have been present prior to complete failure.
As such the best fix was to replace the Power Button Flex completely, a service which many consumers paid for themselves, prior to Apple acknowledging the fault. Upon finally issuing a recall for the “small percentage” (sound familiar?) effected, Apple offered to replace a limited amount of the devices based on Serial Number. If you suffered the issue and your Serial was not covered, or if your device had any other form of damage, you were out of luck.
More to Come
This article has been an ongoing project (in our spare time) for a long time. Despite being unfinished however, we have decided to release it early.
This decision was made because in reality, the article will never be finished anyway. We realize that the most notable and well known issues, like iPhone 6/6+ “Touch Disease” , and iPhone 7/7+ “Audio/Loop Disease”, are yet to be covered. We are still working on newer models at this stage, and will add them here incrementally.
Hopefully, we have covered some issues that most are unaware of, and we plan to cover everything eventually.
Watch this space for more!
The Future of the iPhone
It would seem Apple has a track record of making mistakes, learning from them, and correcting them only to have new issues arise. As they do we intend to add them to this article, with as much detailed explanation as we can. If we missed something, feel free to contact us so we can include it.
As for what the future holds, it’s really anyone’s guess, and we can only speculate for now.
Will Apple make us throw away all our accessories again? Well if they do, it’s not clear if the new connector will be a re-designed Lightning connector, or if Apple will finally adapt to USB-C, as they have with their MacBook range. Perhaps, the new connector will be a water tight vacuum sealed USB-C connector hybrid. Regardless, we are preparing for consumer outrage once again.