Protection from cheap power supplies.

Many of the cheap power supplies rely on a capacitor to limit current from mains voltage, yesterday someone on died after the phone charger failing and sending 120/240vac into her phone...

What's the easiest/best way of stopping this? I want to sit a circuit between a cheap power supply (5v out) and your phone/ device, if it goes over 5v kill it.

But will a dose of 120/240v also kill the protection circuit.

TVS diodes?
MOV?

I'm not sure... if it's possible without some kind of isolation? (Transformer) but if so that defeats the purpose of cheap power supplies!

Use better power supplies. Good 5V switching supplies can be had for <$5
http://www.dipmicro.com/store/DCA-0510
A input/output box with a fuse, resettable or otherwise, that blows with overcurrent would do it.

Used residual-current device (RCD), or residual-current circuit breaker (RCCB) or Earth Leakage Circuit Breaker (ELCB) at your house

BTW. This question looks like its related to a news story here in Australia, where a woman appears to have been killed while holding her phone while it was charging, while at the same time using a laptop with headphones attached.

See

In which case the exact problem has not been determined i.e the article reads

It is believed a dodgy $4.95 phone charger sent a high-voltage electrical pulse into her phone, which transferred to the earphones she had connected to a laptop.

Note the word believed

I don't think they know for sure.

And I'm not personally convinced that the diagram in the article is correct, as it implies that somehow the laptop is the return path to earth.

DO NOT USE CHEAP switching power supplies. There is no transformer in them. That is very dangerous.

Looks like a carefully covered up murder to me.

steinie44:
DO NOT USE SWITCHING power supplies. There is no transformer in them. That is very dangerous.

That is not true, most SWITCHING power supplies used transformer in them.

Apple iPhone charger teardown: quality in a tiny expensive package

Only some of the cheap power supplies rely on a capacitor to limit current from mains voltage, if you buy one of these cheap power supplies then it will be dangerous.
You can always open up these cheap power supplies to find out yourself, how it wire up.

Conclusions
Stay away from super-cheap AC adapters built by mystery manufacturers. Spend the extra few dollars to get a brand-name AC adapter. It will be safer, produce less interference, and your device's touchscreen will perform better.

steinie44:
DO NOT USE SWITCHING power supplies. There is no transformer in them. That is very dangerous.

That is of course, complete nonsense. If anything, the opposite as the transformer in a switchmode supply being smaller, has a smaller area of insulation and could be argued to be on that account, less likely to fail.

If of course, it is properly designed and constructed. The strict requirement of these two-pin appliances is "double insulated", so there must be two separate and complete layers of insulation. One fascinating problem is whether the feedback opto-isolator is technically compliant with this.

michinyon:
Looks like a carefully covered up murder to me.

Hmm, I suspect you have been watching too much television!

Anyway the article was on TV last evening, my wife as a matter of course said to me "Do you have any of those?" to which I replied along the lines of "only about half a dozen". From eBay of course, costing little over a dollar each. At $4.95, those local sellers are making a good profit.

Not sure what to do with them. Use an isolation transformer I suppose!

cjdelphi:
Many of the cheap power supplies rely on a capacitor to limit current from mains voltage, yesterday someone on died after the phone charger failing and sending 120/240vac into her phone...

You seem not to understand what a SMPS is. Perhaps you should study up on it. It does not "rely on a capacitor to limit current". What is however interesting, is the popular type of LED light globe which I had (innocently) presumed to contain a swithchmode device like the CF globes. I have now realised that they must use a capacitor as a "ballast", with a bridge rectifier and as many LEDs as possible all in a series chain to use the minimum current at the maximum voltage.

cjdelphi:
What's the easiest/best way of stopping this? I want to sit a circuit between a cheap power supply (5v out) and your phone/ device, if it goes over 5v kill it.

That is of no use, since the problem is not the differential voltage between the output wires, but the possibility of the mains voltage being imposed as a common mode on both due to insulation failure.

cjdelphi:
But will a dose of 120/240v also kill the protection circuit.

It's not a protection circuit. What is required is insulation.

cjdelphi:
TVS diodes?
MOV?

cjdelphi:
I'm not sure... if it's possible without some kind of isolation? (Transformer) but if so that defeats the purpose of cheap power supplies!

And indeed, that is the only answer to a faulty manufactured device - in this case.

rogerClark:
In which case the exact problem has not been determined i.e the article reads

It is believed a dodgy $4.95 phone charger sent a high-voltage electrical pulse into her phone, which transferred to the earphones she had connected to a laptop.

Note the word believed
I don't think they know for sure.

If indeed she had burns on her ears and on her chest, that would certainly suggest something.

rogerClark:
And I'm not personally convinced that the diagram in the article is correct, as it implies that somehow the laptop is the return path to earth.

If the laptop has a three pin plug (as most seem to) you will find that the earth pin is common to the negative output. Why? Because certification on the power supply is easier if it does not have to be "double insulated". This arrangement is safer, albeit a complete menace if you are doing any sort of audio work.


Added:

BillHo:
Only some of the cheap power supplies rely on a capacitor to limit current from mains voltage, if you buy one of these cheap power supplies then it will be dangerous.

That does not make sense. Any such power supply which was not isolated with a transformer would electrocute everyone using it - at least 50% of the time.

Reminiscent of the old "transformerless" radios in the days of valves.

BillHo:
You can always open up these cheap power supplies to find out yourself, how it wire up.

Looking into that ...

Hi Paul,

Have you seen this Teardown by Dave at Eeblog

Basically, he starts the video thinking he has one fake and one authentic Apple charger, however things aren't quite what he expected.

I think it's going to be very difficult for you to replace all of your USB chargers with ones that are definitely authentic.

Well if you buy direct from Apple, I think you should be getting Apple chargers, buy buying from anywhere else could be clever fakes.

I have lots of chargers on various devices I've bought from shops like Target, for various bits of cheap electronics.
I suspect that the chargers that come with a lot of these devices can't be particularly good quality,because the prices are too low of the whole package.

I'm not sure that the charger linked to by CrossRoads would be any better than the one that caused this death, it seems to cheap.

Re circuit path for electrocution

Unless the user was dripping wet, I'm not sure how the circuit was completed.
A lot of phones, albeit not an iPhone, are plastic, so not likely to conduct mains.
Most laptops are completely plastic, albeit again, not mac books.

I will check my mac book, to see if the metal case is connected to earth, but to be honest, I didn't think it would be.

Update..

Just checked a couple of laptops

Very old HP laptop. Earth is connected to GND on the laptop.
But the whole laptop is plastic.

Not so old Mac book.
Metal case is not connected to earth, at least not when the device is not plugged into the mains.

The GND of the USB connector has a path to earth of around 1k ohm. But to connect to this, I had to plug in a USB extension cable and put my meter into the internal pins.

I'd be interested to know if your devices have a good path to earth.

I've used a lot of the dipmicro walwarts - mostly 5V and 12V, couple each 7.5V and 9V.
Here's shots of the "top" and "bottom" of the internal card for the 12V/1A supply I was using today to power a TPIC6B595/328 card for controlling LED strips.
The big chip near the output wires is marked "NEG 2501 LF118". Can't find any info on it.
Can't read the other multi-pin parts. There are two diodes, can see the cathode markings on those.
Unloaded voltage is 12.39V, and 12.14V driving a '328P, 4 x TPIC6B595, and 28 3-LED segments with 15-20mA/segment, so ~ 500mA total.
Seems pretty well regulated. Haven't looked at it with a scope. 5V signals on other boards where I was checking out 8 MHz SPI operations looked solid, no ripples at all, powering a Bobuino from 5Vdirectly (no onboard regulator) and 27 TPIC6B595s.

I've not had any issues and regularly power designs from the 5V regulators directly. I'l vouch for them.

@CrossRoads

That PSU looks well made, but the diode in the top right corner looks a bit dodgy.

It's soldered in well tho. Last minute design change I guess.

After carefully analyzing all the relevant data in this article I have come to the conclusion that 240V is dangerous and Australia should convert immediately to 120V like the U.S. and hire me as their electrical safety czar to oversee the awarding of the relevant contracts.

JoeN:
I have come to the conclusion that 240V is dangerous and Australia should convert immediately to 120V like the U.S.

Why? So that everything needs twice as many amps and all the cables need to be thicker?

Do they need to convert to 60 Hz too, instead of 50Hz? 8)

I see an awful lot of power supplies are rated 100-250V, 50/60 Hz. So the legit folks seem to have it worked out.

The 3 small units at my desk are here all 120V, one is 50/60Hz and two are 60Hz input.
The larger PS for my Dell Laptop is 100-250V, 50/60 Hz.