What caused the teenager to die while charging her phone in the bath?

[Guineafowl]

http://www.telegraph.co.uk/news/2017/07/11/teenager-dies-charging-phone-bath/

The poor girl got electrocuted, I assume, from the USB case being live at mains voltage.

I don't understand - between you and the mains, there should be a transformer, class Y capacitor, and an optocoupler. No direct reference.

Now, you could argue it was a clone charger, with poor isolation. However, I can measure 135Vac (our supply is 240V) between my Apple charger USB plug's metal case and Earth ground. This, however, drops to 13V when I touch the case. I assume an Apple product is well isolated electrically.

The scope shows a 50Hz (our frequency) sine wave 400V p-p. Is this just coupling through the class Y capacitor? Do you think the girl's charger had an underrated cap that broke down when given a good reference to Earth ground? Is it OK for the manufacturer to supply lethal equipment with impugnity? She shouldn't have been using it in the bath, but anyone can get a good ground reference by leaning on a radiator or kitchen sink.

There's a frustrating lack of technical detail in the article, so I'd love to hear your take on the tragedy.

 

[anorlunda]

There's a frustrating lack of technical detail in the article, so I'd love to hear your take on the tragedy.

I've seen the articles and there is a lack of detail. Because of that, we can't rule out the possibility that she brought an extension cord into the bath to plug the charger into. Speculation is fruitless until we have the facts.

 

[Guineafowl]

I've seen the articles and there is a lack of detail. Because of that, we can't rule out the possibility that she brought an extension cord into the bath to plug the charger into. Speculation is fruitless until we have the facts.

Absolutely right. But from what I can gather, I'm pretty sure the phone dropped into the bath - its metal case being connected directly to the USB chassis.

Since this happened in the US, I wonder anyone there can find out more detail. My working hypothesis is that the charger lacked a class Y capacitor. This does happen - BigClive (youtube) tore down a switching supply with a simple ceramic disc cap where the Y should have been. No prizes for guessing where the thing was manufactured, or the auction site it was bought from.

 

[anorlunda]

Absolutely right. But from what I can gather, I'm pretty sure the phone dropped into the bath - its metal case being connected directly to the USB chassis.

You can't depend on a statement like that being reliable. The phone and extension could have fallen. I would consider popular press accounts as having zero evidentiary value.

 

[berkeman]

Because of that, we can't rule out the possibility that she brought an extension cord into the bath to plug the charger into.

Yes, I've read that she used an extension cord, but I'm not finding that news article at the moment. Looks like the parents have changed their story to claim that the phone was plugged into a bathroom outlet (which of course would have had GFCI protection).

The phone and extension could have fallen.

That seems more likely, or even just water running down the USB cord to the extension cord outlet.

 

[russ_watters]

Thanks for posting this - I saw the same article and had the same doubts.

 

[russ_watters]

I've seen the articles and there is a lack of detail. Because of that, we can't rule out the possibility that she brought an extension cord into the bath to plug the charger into. Speculation is fruitless until we have the facts.

Agreed. My problem with the scenario presented isn't the electronics, it's the wires; I don't see how such small wires could carry enough current to electrocute someone.

This may be another misundertanding tech issue, like last week's false Alexa called 911 story.

 

[Guineafowl]

Agreed. My problem with the scenario presented isn't the electronics, it's the wires; I don't see how such small wires could carry enough current to electrocute someone.

This may be another misundertanding tech issue, like last week's false Alexa called 911 story.

Well, a typical USB phone charger can carry 1A at 5Vdc. Just 30mA of AC can reportedly cause fibrillation of the heart, and of course therefore loss of consciousness. If you're in the bath, drowning becomes a real risk at this point. Also, in many cases, it's not the shock that directly kills you, but what the shock causes you to do - fall off a ladder etc.

Further testing of my Apple charger reveals that the 135Vac between USB chassis and Earth ground will only drive 37.5 uA of current, and drops to 10 mV p-p when doing so. Whether this is a 'ghost' voltage and reactive current/power I'm not sure, but it seems pretty safe.

Anyway, as you say above, the point is moot until we know the exact circumstances of the accident. But we really should know, because if it turns out the charger coupled the fatal current to the phone case, a tragic accident like this could just be the first of many.

 

[anorlunda]

Well, a typical USB phone charger can carry 1A at 5Vdc. Just 30mA of AC can reportedly cause fibrillation of the heart, and of course therefore loss of consciousness.

Use common sense. Ordinary AAA, AA, and 9V batteries can also supply much more short circuit current than that. How many deaths have you heard of about people handling those? No doubt if you dig you can find reports of fatal falls from a height of 30 cm, but that doesn't make us worry about standing erect.

 

[berkeman]

Ordinary AAA, AA, and 9V batteries can also supply much more short circuit current than that. How many deaths have you heard of about people handling those?

I think the question came up when Russ mentioned low-current-carrying capability of the USB cable. The voltage conducted along the cable would have to be high to cause a lethal shock, so somehow there had to be a conductive path back to the (non-GFCI-protected) AC Mains supply...

 

[Guineafowl]

Use common sense. Ordinary AAA, AA, and 9V batteries can also supply much more short circuit current than that. How many deaths have you heard of about people handling those? No doubt if you dig you can find reports of fatal falls from a height of 30 cm, but that doesn't make us worry about standing erect.

Ouch! I was really just brainstorming about whether a USB cable is beefy enough to deliver a fatal current (30mA, for argument) at 240V or 120V, as per Russ's question in post #7. I would guess it probably is, and wouldn't trust a USB cable to fail before I did.

 

[anorlunda]

Ouch! I was really just brainstorming about whether a USB cable is beefy enough to deliver a fatal current (30mA, for argument) at 240V or 120V, as per Russ's question in post #7. I would guess it probably is, and wouldn't trust a USB cable to fail before I did.

Sorry if I was harsh. But you were trying to equate the danger of 30 ma @ 120V to 1 amp @ 5V. That's invalid. 1A @ 5V is akin to AA battery range of numbers and we know by common sense that those aren't dangerous.

This article discusses the lethality of currents. https://en.wikipedia.org/wiki/Electric_shock
But the currents discussed are defined as passing from hand to foot. It takes substantial voltage to drive 30 ma from your hand to your foot. That's why 5V is not dangerous. A USB can deliver 1A to a load, but it won't drive 30 ma through your whole body. If USB ports were lethal shock hazards, they would have much more safety in the way they are packaged and handled.

 

[Guineafowl]

Agreed. My problem with the scenario presented isn't the electronics, it's the wires; I don't see how such small wires could carry enough current to electrocute someone.

This may be another misundertanding tech issue, like last week's false Alexa called 911 story.

Further to this, I rigged up a 2ft/60cm length of USB cable between a mains socket and two GU10 LED lamps in parallel. The lamps consumed 55mA of current at 240V for a good 20 seconds. I concluded the USB cable is more than capable of delivering a lethal shock. It didn't even get warm.

 

[Guineafowl]

Sorry if I was harsh. But you were trying to equate the danger of 30 ma @ 120V to 1 amp @ 5V. That's invalid. 1A @ 5V is akin to AA battery range of numbers and we know by common sense that those aren't dangerous.

This article discusses the lethality of currents. https://en.wikipedia.org/wiki/Electric_shock
But the currents discussed are defined as passing from hand to foot. It takes substantial voltage to drive 30 ma from your hand to your foot. That's why 5V is not dangerous. A USB can deliver 1A to a load, but it won't drive 30 ma through your whole body. If USB ports were lethal shock hazards, they would have much more safety in the way they are packaged and handled.

Ah, yes, OK. I meant that if the cable can carry 1A at 5V, it can surely carry 30mA at 240V. But see post above too.

 

[CWatters]

If it can carry 1A then it can carry 1A at voltages high enough to kill you.

Not really any more info but..

http://www.dailymail.co.uk/news/article-4690766/Tearful-father-stepmom-warn-electronics-bathroom.html#ixzz4mpQORVo5

'She had her phone plugged into the extension cord and it was by the bathtub and I did it, she did it, we all had sat there in the bathtub with our phones plugged in and played our games,' she explained tearfully. The phone plunged into the water while it was still being charged, electrocuting the girl instantly.

If the only thing that fell in was the phone then there must be some problem with the charger.

Most countries have regulations that require anything electrical you can tough to be safe. Think crawling baby putting the USB cable in their mouth. I would hope all countries have regulation in place that make it unlikely such a child would come to any harm.

Sadly this isn't the first time it's happened...
http://www.mirror.co.uk/news/uk-news/man-killed-after-electrocuting-himself-10044838

A man was electrocuted while charging his iPhone in the bath as he got ready for a family Christmas party.Richard Bull, 32, is said to have died when he ran an extension wire from his hallway into the bathroom. He then rested the cable on his chest while charging his phone, an inquest into his death heard

 

[Guineafowl]

If it can carry 1A then it can carry 1A at voltages high enough to kill you.

Not really any more info but..

http://www.dailymail.co.uk/news/article-4690766/Tearful-father-stepmom-warn-electronics-bathroom.html#ixzz4mpQORVo5
If the only thing that fell in was the phone then there must be some problem with the charger.

Most countries have regulations that require anything electrical you can tough to be safe. Think crawling baby putting the USB cable in their mouth. I would hope all countries have regulation in place that make it unlikely such a child would come to any harm.

Sadly this isn't the first time it's happened...
http://www.mirror.co.uk/news/uk-news/man-killed-after-electrocuting-himself-10044838

Indeed. I even started a thread a while back about the last story you mention. The consensus at the time was that the mains extension cord must have fallen in too.

Are fatal shocks are being coupled across poorly isolated USB supplies? You may not be in the bath, but the USB chassis is just waiting for you to ground yourself with a wet hand on a radiator.

I've taken apart a few USB supplies that I had lying around. The transformers are wound such that the primary is wound in with, or sometimes sandwiching, the secondary. You are relying on very thin insulation as you hold your metal phone chassis, or your young child sucks on the cable end.

One of the cheaper ones had a ceramic disc, not class Y, bridging the USB chassis to the mains supply. I commoned the live to neutral, then attached the +ve of my insulation tester. The -ve went to the USB chassis. The resistance at 500V was less than 1 Mohm. If I gripped the -ve lead of the tester, and with the same hand touched the USB chassis, I got a moderate shock. (Intra-hand, not hand-hand - I'm not that stupid!). This did not happen when I lifted a leg of the cap.

(500V is the standard for testing UK appliances - At the normal 240V rms, the peak will be around 340V.)

All the opto-isolators were tested right up to 1000V and showed no sign of leakage.

Was my testing reasonable?

 

[CWatters]

Plenty of suspect chargers around... 72,000 in this case alone..

https://www.netmums.com/life/poundworld-fined-for-selling-dangerous-phone-chargers

"During the investigation it was found the adaptor did not have adequate insulation..."

 

[jim hardy]

I've taken apart a few USB supplies that I had lying around. The transformers are wound such that the primary is wound in with, or sometimes sandwiching, the secondary. You are relying on very thin insulation

What ? That's why we have "Split Bobbin" power transformers...

Some manufacturer is stumbling up the learning curve. They need more gray hair in design department.

old jim

 

[OmCheeto]

...the phone was plugged into a bathroom outlet (which of course would have had GFCI protection).
...

Not necessarily. I just did a Zillow scan of the houses in the area of the incident (800 block of West Avenue, Lovington, NM), and they were built in 1965. The GFCI requirement for bathrooms didn't come into effect until 1975. [seemingly reliable ref]

Also, I've been mowing my neighbors lawn for the last 2 years, and they don't have a GFCI outlet on the outside of their house. Those have been required since 1973. They just moved in a few years ago, so there seems to be some loophole about GFCI rules. Not sure if they are renting or own the house.

When I bought my house in 1989, it still had its original 1945, two prong outlets, in every room.

 

[berkeman]

Holy crap. That's scary, Om. Time for a PF educational campaign?

 

[OmCheeto]

Holy crap. That's scary, Om. Time for a PF educational campaign?

Maybe. I replaced all my wiring in my house within the first 5 years, as I couldn't figure out how GFCI outlets would work without a ground wire.

Funny thing is, I never replaced the wiring to two outlets, which are in my kitchen.
About a month ago, whilst doing some "load testing", I discovered that my refrigerator has a ground fault when I open the freezer door.
Quite minor compared to the "if you touch the stove and refrigerator at the same time, you'll get a mild 'electrocution'" when I was 40 years younger, at my mothers house.

Anyways... Yes, education is good.

And we should probably notify @Greg Bernhardt , as he just bought a slightly used home.
Curious if his wiring is up to code.
I'd hate to see either him, or his lovely wife, turned into toast.

Just sayin'...

 

[Greg Bernhardt]

Curious if his wiring is up to code.

The bathrooms do have GFCI outlets :)

 

[dlgoff]

Quite minor compared to the "if you touch the stove and refrigerator at the same time, you'll get a mild 'electrocution'" when I was 40 years younger, at my mothers house.

[offtopic]Reminds me of my first experience with electricity at age 5. On a hot summer day, barefooted on the wet ground, turned on the outside faucet to get a drink and was promptly put on my butt. I leaned then that water pipes didn't make the most reliable grounds. [/endofftopic]

 

[Janus]

Not necessarily. I just did a Zillow scan of the houses in the area of the incident (800 block of West Avenue, Lovington, NM), and they were built in 1965. The GFCI requirement for bathrooms didn't come into effect until 1975. [seemingly reliable ref]

Also, I've been mowing my neighbors lawn for the last 2 years, and they don't have a GFCI outlet on the outside of their house. Those have been required since 1973. They just moved in a few years ago, so there seems to be some loophole about GFCI rules. Not sure if they are renting or own the house.

When I bought my house in 1989, it still had its original 1945, two prong outlets, in every room.

Here's something I've seen quite often: three prong outlets installed "upside down". It may not seem like that great a deal, but the ground plug should be at the bottom. The reason for this is that if the plug is somehow wiggled so that it starts to the come out of the outlet, you want the weight of the plug/cord to cause it to tilt so that the "live" prongs pull out before the ground plug does.

 

[Tom.G]

I've seen quite often: three prong outlets installed "upside down". It may not seem like that great a deal, but the ground plug should be at the bottom.

Probably because many (all?) of the packaging I've seen shows the outlet with the Gnd on top.

...tilt so that the "live" prongs pull out before the ground plug does.

I always 'assumed' the plugs were designed so 'upside down' didn't matter. But I decided to check before inserting foot in mouth. You are right, it is possible to make the Hot prong connect before the Gnd pin makes contact.

Learn something every day.

 

[Guineafowl]

For interest, I wired my rural UK house, having only an Earth rod in the ground, with ALL circuits protected by an RCD (GFCI). This is required by regulations. Also, quite sensibly, you're meant to have the downstairs sockets (ring main) and upstairs lights on one RCD, and vice versa on another. Thus if someone gets a shock from a socket, the lights don't go out.

The reason is, the impedance of the Earth loop is such (100 ohm or so) that not enough current would flow to trip an MCB in the event of a fault. The largest MCB is 45A, for our 10 kW electric shower.

 

[jim hardy]

Also, that's why the ground pin is longer than the other two.Interestingly NEMA draws them ground pin up

There's an old electricians's saying "White on Right" which for the receptacle would be ground pin up. But i don't know on what it's based.

There's one school of thought that ground pin up will deflect something that's falling and keep it away from the current carrying blades.

But so far as i know there's no standard dictating any orientation.

Looking at LOWE's site
Here's a BOSCH right angle extension cord with ground pin UP

and a Morris with ground pin DOWN

Wikipedia:

Being an OCD type i mount them so the writing on the metal frame is right side up.
Note this Hubbell GFCI has buttons printed for both orientations.

and ground pin up gives "white on right" .

This Leviton metal is stamped the other way

I just looked - my house is about 50/50 up or down.
At least Fair Anne's kitchen is all the same. Doubtless because I bought them all the same day.

old jim

 

[NTL2009]

Probably because many (all?) of the packaging I've seen shows the outlet with the Gnd on top.

I always 'assumed' the plugs were designed so 'upside down' didn't matter. But I decided to check before inserting foot in mouth. You are right, it is possible to make the Hot prong connect before the Gnd pin makes contact.

Learn something every day.

But I've also seen some threads where codes were being discussed, and some argued for the ground pin on top (talking NA 3-prong 115V here), or if oriented with a line drawn through the hot and neutral being vertical, to put the neutral on the top. The reason was, if something metal fell down the wall ( a metal ruler, a loose metal plate cover, etc), it would most likely hit the ground pin or neutral blade, and not create a big problem.

If I could go back in time, I'd like to convince Edison to re-think what a power outlet should look like. I would think something more like a co-axial connector would be far safer, and still simple to construct. The outer ground cylinder would completely protect the power pin, and the power socket would be recessed as it is now. Something similar to a larger RCA/'phono' jack connector, with the power pin recessed so ground is made first. I bet a lot of lives and property damage would have been saved.

And I would have allowed enough space for two pins inside the outer cylinder, in anticipation of three prong sockets to replace the old two prong. In that regard, hindsight is wonderful, but I do think the protective outer ring is almost obvious. I often go around the house pushing plugs in tight to the socket, it bothers me to be able to see those hot and neutral blades exposed an 1/8 of an inch or so. It seems common for less attentive people to not seat them all the way, or for them to work loose a bit. I know other countries have some better arrangements, but they seem more complex and bulky than a co-ax like this.

 

[OmCheeto]

[offtopic]Reminds me of my first experience with electricity at age 5. On a hot summer day, barefooted on the wet ground, turned on the outside faucet to get a drink and was promptly put on my butt. I leaned then that water pipes didn't make the most reliable grounds. [/endofftopic]

Actually, I don't consider this off topic.
AFTER I bought my house, I discovered that much of the iron plumbing under the house had been destroyed by going through a couple of unoccupied winters. 99% of the damage was ruptured elbows, but I had one 10 foot straight section that had a 2 inch rupture. Rather than replace it, I hacksawed out the bad section and replaced it with a plastic slide on compression repair piece. Had I not also installed a copper jumper across that piece, it might have been disastrous.

ps. Though, the house also has the obligatory grounding rod, so maybe not. But as a former naval nuclear power plant technician, I'm a firm believer in redundant safety devices. I was also an electrical safety inspector on my first vessel (6 months, aircraft carrier), and the sole electrical safety inspector on my second (4 years, submarine). Hence my "better safe than sorry" attitude towards just about everything.

 

[dlgoff]

Hence my "better safe than sorry" attitude towards just about everything.

You are a very wise man. I wish all could have this attitude.

 

[dlgoff]

... so far as i know there's no standard dictating any orientation

From my experience, with the ground pin on top and it being longer, cord weight has caused less pull-out. I think it has more to do with torque; rotation force captures long pin better when on top.

 

[OmCheeto]

You are a very wise man. I wish all could have this attitude.

My younger brother told me once; "You drive like a paranoid schizophrenic".
After my last drive with him, I will never ride in his vehicle again. (He trusts red lights to stop people. I look both ways before driving through an intersection. He didn't.)

Anyways, back to the topic:

I wrote up a REALLY long chastisement post for Greg, because I still didn't understand how GFCI's could work without a ground wire, but decided that my post was in error, as they apparently do work.

In my defense, it has been ≈30 years since I last studied GFCI, and I don't think the internets, as we know it now, existed.
Though, after I thought about it some more, I think I may have understood how they worked, but have forgotten, in the meanwhile.

 

[berkeman]

I wrote up a REALLY long chastisement post for Greg, because I still didn't understand how GFCI's could work without a ground wire, but decided that my post was in error, as they apparently do work.

I believe they sense the current in the Hot and Neutral wires, and if there is a difference of more than the threshold, they trip. It seems like I learned that a while back when I had the same question as you. I haven't looked up the circuit, though.

 

[dlgoff]

In my defense, it has been ≈30 years since I last studied GFCI, and I don't think the internets, as we know it now, existed.
Though, after I thought about it some more, I think I may have understood how they worked, but have forgotten, in the meanwhile.

A picture may be helpful in remembering.
image compliments of http://www.ecmweb.com/basics/how-gfcis-work

 

[OmCheeto]

I believe they sense the current in the Hot and Neutral wires, and if there is a difference of more than the threshold, they trip. It seems like I learned that a while back when I had the same question as you. I haven't looked up the circuit, though.

That's what I read yesterday, which is why I preemptively deleted my post.
But the page seemed to be filled with a lot of "ifs, ands, or buts", so I didn't share it, as most non-old, non-electrically literate people might get confused.

Replacing 2-Wire Ungrounded Receptacles

When in doubt, call Om.

Though, after I thought about it some more, I think I may have understood how they worked, but have forgotten, in the meanwhile.

My vintage 1945 bathroom vanity fixture, upgraded in about 1992.
I'm pretty sure I would not have done that, had I known it wouldn't work.