Apple iPhone charger teardown (2012) | Hacker News Reader

Apple iPhone charger teardown (2012)

righto.com

107 pointsofrzeta71 comments

Power engineer here. Something that is not apparent is that many small power supplies are limited in power handling not by the components, but by maximum safe touch temperatures for meeting regulations (probably around the 10W level). Therefore making a more efficient power supply means that you can package it smaller since the heat needs lower surface area to dissipate safely. Right now, gallium nitride MOSFETs are gaining popularity allowing for higher efficiencies in power conversion, and some wildly high power densities for USB chargers. I'm a fan of ravpower's 30W USBc supply which is about one cubic inch with folding prongs.

For context, the "safe touch temperature" in some countries is 50 celsius, in a place where air temperatures can reach 40C. That means the charger can't rise more than 10C above ambient temperature, which is tricky to achieve without a cooling fan.

Yep, you are completely right.

It is a triangle of size, efficiency and cost.

Simplest flybacks, and half bridges can't compare to topologies with extra magnetics on efficiency, but you need to find space for extra xformers.

The appeal of high end 3-5 switches is that you can get high efficiencies even in simplest topologies, and employ no extra magnetics.

My engineers too hit icebergs few times when we tried making power supplies "the smart way."

PZT xformers are smaller than iron, but their output may not be completely sinusoidal.

Fancy vitroperm, finemet, and other similar magnetics can have rather low tolerances for magnetic properties, and some even drift over time as their metallurgy changes at relatively low temperatures.

GaN switches can be easily damaged without carefully made gate driver.

Capacitive isolation is not foolproof, and substandard safety caps are everywhere.

SiC switches are said to be as reliable as galvanic iso, are not so reliable in practice.

> It is a triangle of size, efficiency and cost.

It's interesting to consider that the reason there are tradeoffs is because uniformly bad technologies get discarded, so the remaining choices have tradeoffs.

For chargers, consider the 1990s-style line transformer and linear regulator (big wall warts). These are larger, less efficient, and cost more than a modern switching power supply, so essentially nobody uses them.

So it may seem like you're stuck in a triangle of choices that are bad in some way, but in the historical picture, these choices are all very good.

Yeah, you’re sort of describing engineering in a nutshell. It’s all about tradeoffs, and there’s so much min-maxing going on. Then eventually you discover some property or design where the tradeoffs aren’t as bad while still mostly maxing the desired property and it becomes a common solution.

What’s interesting, is that periodically you can go back to what looked like a bad solution, and because of advancements in related designs, this solution can actually end up being superior in some ways.

Yet there seem to be no designs with any ventilation. Why is that? Are heat sinks not allowed because the case is (usually) not grounded?

Useful ventilation means holes through which someone could insert a thin metallic part (wire, needle...)

Never underestimate the ingenuity of fools. The safest design is one without holes.

You don’t even need people inserting anything.

Dust is a problem on its own it can ignite from ESD and the charing causes carbon built ups that eventually cause a short.

Even when it doesn’t ignite from ESD it can change the capacitance of circuits which can cause a flash.

This is compounded even more by the fact that many dust particles in cities can be conductive carbon particles form combustion, platinum and other metals form catalytic converters and fine metal particles from things like car brakes, roads and wheels are all over the place.

A lot of that black/dark grey dust that can settle around windows and other leaky parts of your house in a city isn’t dead skin and dirt.

The general rule is that on the mains side one needs either grounding and single insulation, or no grounding and double insulation. In Europe (except UK and Ireland) there us the Europlug standard plug with two prongs, that works everywhere, but no single standard three-pronged plug with grounding. For economy if scale and simplicity of logistics, double insulation is the best option, but as you note it makes a heatsink less practical.

I’m constantly amazed by these things in daily life. The charger is just barely large enough safely house a USB jack, and that safety margin is large enough to fit a high-efficiency converter.

If you are curious how it works, the Horowitz and Hill book, 2nd ed, has an extended section dedicated to describing how the power supply for the Apple II works. The iPhone uses the same technology, broadly speaking. Switch-mode power supplies were fairly new at the time, and older computers will generally use bulky, inefficient linear power supplies.

Every once in a while I come across something from before the era of small SMPS, like a constant-voltage transformer that was bundled with my color photographic enlarger. The idea is that you regulate the voltage for the lamp to make the light color more consistent, but since the technology was old, it consisted of a massive, heavy ferroresonant transformer. I ended up selling it for scrap.

The general way that you make power supplies smaller is by increasing the frequency they run at. For SMPS, this means increasing the switching frequency, but there are also some AC standards that run at higher frequencies in places where space and weight is at a premium. For example, AC power within an aircraft might run at 400Hz rather than 60Hz. The transformers are correspondingly smaller.

But higher switching frequency means more losses and thus higher temperature.

The world is a better place when companies like Apple do stuff like this properly. Besides, I’d guess that well over 95% of the chargers Apple ships are bundled with a device. Possibly 99%. So the retail cost of the charger on its own is beside the point.

If you want a cheap aftermarket charger, consider products sold by IKEA. While I can’t say it’s an immutable truth, IKEA electronics tend to be well engineered considering their low cost.

Here’s a bigclive tear down of an IKEA KOPPLA USB power supply, it seemed well-built https://youtu.be/uRe9w5PKmsE

Another advantage is that IKEA has its own online shop. No risk of getting a fake.

IKEA LADDA AA and AAA batteries are some of the best I've ever bought and for the price are very hard to beat.

There are some reports and writeups that suggest that LADDA 2450 AA/900 AAA rechargeable batteries are the same as Panasonic Eneloop due to extremely similar characteristics and the fact that they very likely come from the same factory in japan because only one makes Ni-MH low discharge. For $6.99 per 4 batteries I have yet to find any that can beat these and I have been using some for years without any noticeable drop in capacity yet.

https://www.slrlounge.com/panasonic-eneloop-vs-ikea-ladda-ar...

Irrelevant anecdote - "Ladda" is Swedish for "(to) charge".

They are basically OEM ( at least it was in 2018 / 2019 ) from Panasonic. But not all Eneloop and in fact LADDA are from Japan. Panasonic do make Eneloop from China and they are of slightly lower quality than the one produced in Japan.

I believe there are a few variant / types / size of LADDA, and there are one or two make in Japan.

Unless you are perfectionist you would want the best one from Japan. Otherwise most LADDA are good enough for 99% of its intended usage.

Initially I thought Tradfri IoT stuff would end up being a glorious mess but it’s well designed, secure, and privacy respecting.

The hub has a very reasonable API, it reaches to WAN only to get software updates and blocking that has no ill effect, and the UI of the app is both very simple and for some cases makes more sense than Home.app (Default Moods are nice, and setting the light levels/Mood across all lights of a whole room can be done without reaching to voice or setting up a dedicated scene per room)

https://learn.pimoroni.com/tutorial/sandyj/controlling-ikea-...

Yes, except for IKEA LED bulbs, which are cheap but terrible (low CRI, huge latency to switch one)

In the more distant past I found this to be true, but recent LED bulb purchases from Ikea have convinced me they've solved the earlier problems. Their LED bulbs are now great performers with fair prices.

Hard disagree. IKEA LED lights are excellent. Maybe some early models were bad, but so were CFL back then too. Current IKEA LEDs are reasonably high CRI and are available with low lumens options. It’s often difficult to find a quality LED bulb that isn’t super bright.

(By the way IKEA was almost single handedly responsible for the rapid plummeting of the cost of LED lights. They invested early and drove up volume before anyone else.)

>The world is a better place when companies like Apple do stuff like this properly.

I don't see what's special about Apple's charging solution when compared to the competition.

Sure, it's better than fake knock-offs but chargers and batteries from OnePlus, Xiaomi and Huawei have way more engineering in them allowing for faster charging that's equally safe.

2 points:

1) The blog touches on it, but can you imagine the temptation of shady (and non-shady) companies out there to try to engineer a cheaper version or to cut corners to gain some of the margins charged for such a seemingly-simple piece of hardware?

2) Do you remember 20 years ago, 5-10VDC power supplies for computers used to be this gigantic ugly black brick with cooling fins that weighed 2 pounds and had embarrassing Apollo-style connector plugs? Amazing what miniaturization has achieved.

there's plenty of breakdowns out there of cheap chargers. some with shoddy soldering work, or faulty chips that does not stop overcharging, etc -and a safety hazard if you leave it plugged in overnight.

Author here. I've written a couple of articles about counterfeit chargers. You are correct; people cut every corner possible to make a counterfeit charger that looks exactly like an Apple charger. (If you buy a $3 Apple charger on eBay, this is what you're getting.) The power quality of the knockoffs is awful. They also ignore safety, so the fake chargers occasionally kill people.

For point 2, I also wrote an article in the IEEE Spectrum about the history of switching power supplies and how we ended up with such tiny chargers.

Links: http://www.righto.com/2012/03/inside-cheap-phone-charger-and... http://www.righto.com/2016/03/counterfeit-macbook-charger-te... https://spectrum.ieee.org/computing/hardware/a-half-century-...

Even though Australian houses have full RCD protection for the entire house which would often help in these cases -- I still am terrified of these and pretty much just use official apple chargers from an Apple store or otherwise very reputable brands. The level of cheap chinese crazy is, well, crazy.

Thanks for those informative articles!

How can I know whether I have a real of fake charger?

Directly order from Apple, Samsung, etc. Definitely do not trust anything that comes off Amazon or eBay.

Yes, but what if I already have a charger and I don't know where it came from?

It is probably hard to say, from the outside they often look legit. I have seen one once, but it became more apparent when we compared it to a genuine charger (usually the print is off in some way). Since counterfeit chargers typically have fewer components, apparently weighting the charger can also help finding counterfeits:

https://www.electricalsafetyfirst.org.uk/blog/how-to-spot-a-...

Of course, it's probably not beyond them to add some weight to make you believe it is real.

I guess if you have the right tools, you could also measure output quality.

They do add weight (and go through the trouble of implementing ring structures to fake writes)

https://youtu.be/g8ovFkd8myE

Anecdata: My A1265 charger weighs 23g and my A1385 weighs 25g.

This article shows how to tell between an authentic A1265 and at least one fake one: https://www.righto.com/2012/10/a-dozen-usb-chargers-in-lab-a...

In my limited experience, counterfeit iPhone chargers do not have a serial number printed inside the USB connector. Real ones do have it. I've found this picture that shows they've started printing a "serial number" on the fake ones too https://apple.stackexchange.com/questions/243541/how-to-diff...

If you order something that is sold by Amazon.com (not merely fulfilled) you're extremely unlikely to receive a counterfeit.

Yes I know about comingling, but Amazon does not comingle their own interventory with 3rd-party sellers. (Apparently they may have in the past but do not anymore.)

So realistically the only way you'll get a counterfeit is if someone bought a genuine one, swapped it for a counterfeit, and returned the counterfeit, which was then re-sold to you.

However, that can happen with literally any store that sells items that were previously returned, which is a common practice.

1. This is not true. Amazon still commingles inventory with 3p sellers.

2. For Apple specifically, Amazon blocked everyone but an Apple provided list of authorized sellers, so you're pretty much guaranteed to get an authentic product.

>Yes I know about comingling, but Amazon does not comingle their own interventory with 3rd-party sellers. (Apparently they may have in the past but do not anymore.)

Source?

Edit: This page from Amazon does not state shipped and sold by Amazon is excluded from commingling.

https://sellercentral.amazon.com/gp/help/external/200141480?...

From the page you linked to:

> For inventory tracked with the manufacturer barcode, each seller’s sourced inventory of the same ASIN is stored separately in our fulfillment centers. We can also track the original seller of each unit.

That doesn't say items sold by Amazon.com are not subject to the following:

>We may fulfill that order using another seller’s unit of the same product if it’s closer to the customer

Another piece of evidence that Amazon doesn't really want to be a retailer is that they removed the option to filter items to only shipped and sold by Amazon.com

I threw out all my cheap chargers after reading your post years ago and have shared it many times, thank you.

Are Anker chargers engineered well enough to match?

Can't say anything about Anker chargers. But I have been somewhat disappointed by their engineering. It is well known that USB 3 can cause interference in the spectrum used by WiFi and BlueTooth (2.4-5GHz), but their USB-C hubs seem to be particularly bad causing a lot of interference. I had one that would reliably disturb wireless keyboards, etc. I did not have the same problem with Apple or Aukey USB-C hubs/dongles.

The Amazon reviews are also full with people who encountered interference:

https://www.amazon.com/product-reviews/B01D0WE99C?reviewerTy...

If it interferes that badly shouldn't the FCC or similar be having a word?

I have no idea about the interference claim, but incidental radiation within an ISM band is allowed to a great degree, that's the idea of ISM. The device may have been tuned to switch in that band to keep it affordable.

edit- I think I mis remembered this, I can't find an easy to interpret rule or graph that gives an incidental radiation pass unless you're a specific category like wireless charger. it could also be explained by conducted emissions into the laptop that other devices do better with, or just a legal level of interference that happens to be at the high end

I have more of a problem with iphone charging cables - the official ones disintegrate, and both official and non-official seem to have problems with the fourth pin getting corroded over time and the cable failing.

We have this problem with both apple cables and micro usb. The kids yank on them and the cables break. Charging cables just seem to be a consumable item. The difference is that the ports in the micro usb devices also get damaged so pretty much all the android stuff we have that the kids are allowed near only charges if you wedge it in a special angle with a something bending the cable in the port. The lightning port seems to be much more robust than micro usb.

one useful advice I heard in this context is to clean your iPhone (Lightning) port regularly with a soft brush. There may be dust etc. inside interfering with the charging.

A (dry) toothbrush works great in my experience.

I'm always curious - what are people doing with their lightning cables? I've had apple devices for over 10 years now, my house is littered with lightning cables everywhere, and they are all in perfect shape. No fraying, no problems with the plug, nothing. Can't say the same about some USB-C cables I bought, pretty much anything that isn't Anker just stops making a contact after few months.

Same here. I think the issue is the charger connector is on the bottom and they sit there with it rammed in their stomach at one end and tugging the other end. I don’t do that and mine have always been fine.

Not just charging cables, but thunderbolt cables, most of their dongles, and so on. Apple's strain relief seems like it's optimized for aesthetics rather than for, you know, strain relief. I'd say half my Apple products with cords have tears and exposed shielding in the same spot: 1-10mm away from the end of the strain relief.

I'm used to Lightning cables breaking, but it's a pity that the thin, charging-block-to-MagSafe section of my MacBook charger has corroding insulation that Apple refuses to take responsibility for.

Some of their Geniuses have even tried to blame me for not storing it properly.

Do you have access to heat shrink and a heat gun? That fixed my insulation issue.

I bought heat shrink in a number of sizes, but haven't managed to stretch any to fit through the MagSafe 2 connector.

I bet you also held your iPhone 4 wrong /s

I always wondered what the specific differences between expensive original and alternative cheaper chargers were and whether the price differences were justified... "The Apple charger is higher quality and I estimate has about a dollar's worth of additional components inside. But it sells for $20 more."

And yet all the USB chargers I use are Apple chargers bought directly in an Apple store. Peace of mind that the house won't burn down is definitely worth $20 per charger.

Previously on HN: https://news.ycombinator.com/item?id=3996171

I‘m old. I remember reading that article eight years ago.

> Two other R-C snubbers filter the diode bridge, which I've only seen elsewhere in audio power supplies to prevent 60Hz hum; perhaps this enhances the iTunes listening experience.

Is this really the reason? That when engineers are designing chargers, they are considering the music listening experience?

Clean power does have lots of benefits, cleaner audio is one of them

In 2012? It's not a HI-FI

You can't make clean audio without clean DC power. The year doesn't matter.

Old Apple used to care about things like this. One of their primary customers were audio engineers and sound designers.

And their customers still are audio engineers, it’s just that they also have more general consumers.

The new Mac Pro rack mount version basically makes no sense unless you do professional audio. Why else would anyone need a quiet rack mount server with really bad component density per U that has no lights out management?

It’s because it’s a studio rack mount workstation intended to coexist with other rack mount audio hardware.

Apple’s hardware seems overpriced until you consider the license cost of Avid ProTools compared to paying for Logic Pro once. Apple is still quite aware that audio professionals are an important niche customer.

I think that's a joke.

FYI only chargers that one one day just stopped working was made by Apple. Good they are well enginereed, bad that they just broke.

They didn't even test it under different loading conditions and with transients on the input?

They did, just in a different article: https://www.righto.com/2012/10/a-dozen-usb-chargers-in-lab-a...

Looking forward to seeing the Samsung teardown for comparison.

I wrote about a Samsung cube charger here: http://www.righto.com/2012/10/a-dozen-usb-chargers-in-lab-ap...

The short answer is it is slightly less complex than the Apple charger, with a completely different internal design, and it performs about the same.

Reply on news.ycombinator.com