Hacker News new | past | comments | ask | show | jobs | submit login
USB power supply puts 220 Mains power out on USB jack (youtube.com)
369 points by software_radio on April 11, 2016 | hide | past | favorite | 156 comments



Not sure if HN knows about Mehdi Sadaghdar's (aka ElectroBOOM's) YouTube channel, which is somewhat similar - but takes a more lighthearted view on electronics.

E.g. Compare Big Clive's mention of a full bridge rectifier in the video vs Mehdi's YouTube video of the same thing :)

https://youtu.be/sI5Ftm1-jik


I love this guy. His skids are like slap stick for the makers.

However, on the serious side, I wish that he would publish a complementary video showing the real story of what he is doing.

My kids would watch his videos, and then run off the find the bits to do exactly what he did. Hmm..Luckily I was there to let them know of the real consequences.


Both channels are pretty funny in their own way. While BigClive churns out a lot more videos and is more eclectic, Mehdi's ElectroBOOM is clearly more comedic.


Man, I know I shouldn't have taken his videos seriously, but I got a little careless playing with my Nixie Tube Clock project with 170Vdc on it, thinking that it wouldn't hurt me.

It hurt A LOT. Granted, my hand was wet a little, but man it fucking hurt.


Ah, yes - the FULL BRIDGE rectifier-man :D


I much prefer Clive's joke-to-useful info ratio.


If you are wondering "How could an isolation transformer ever get past QC with abraded or nicked insulation?", you might find this video interesting:

https://www.youtube.com/watch?v=HEJn74Z3wHk


I've had a portable Bluetooth speaker blow up in my bedroom a couple of years ago while charging. Something malfunctioned in the battery section.

The only reason that didn't set the building on fire was that I was right next to the speaker when it burst into flames. My immediate reaction was to assume my SO playing a really unfortunate firecracker prank on me, but half a second later it was obvious it was the speaker's fault. Turns out batteries can make pretty flames and a lot of fairly unpleasant smoke. To this day, still can't believe how lucky I was to be right next to it with a pillow as it blew up.

The company was kind enough to send us their top of the line product for the incident, in exchange for the faulty unit (for them to study) and a written agreement to not take em to court.

Still pretty uncomfortable around charging devices with batteries to this day, but the reality is that it's unavoidable and there's no point in developing a phobia about it.


>for them to study

For them to study, or for them to destroy the evidence so that you absolutely could not take them to court.


> so that you absolutely could not take them to court.

Did he miss an opportunity or an opportunity cost?


He's the poster child for a potential class action lawsuit. He witnessed their product explode with his own eyes.

And yes I would imagine that there could have been some cost associated with cleaning up the smoke smell left in the room and possibly any burned areas.


I wish their was a class action every time a developer leaves an unpatched bug in their software...


The cost of us wanting everything cheaper, and yesterday...


I don't get it. They made transformers like that for about 80 years in America, but if a Chinese person does it, it's crazy and dangerous? It looked like she was doing a pretty good job.


Nobody said anything about America or China. The listed uploader on the video even suggests it's from India. So quit yer trolling.

The transformers made 80 years ago were indeed dangerous. Generally they used paper insulation soaked in fire retardant chemicals.

There are a couple of observations I make about this video:

1. She's working amazingly fast and clearly with great skill. However, there's very very little time being spent inspecting the result at each stage.

2. Observe that she is holding sharp wire cutting scissors in her hand as she performs many operations on the insulating tape and wire. It would be extremely easy to accidentally nick the insulation when working at that speed. This insulation is the only barrier between your charging iPhone and the 120/240 mains voltage (or higher in a switching supply).

3. This transformer design is not the safer more modern "split bobbin" type as suggested in the original video. Instead of having a plastic barrier between two separate windings, it relies entirely on the enamel insulation of the transformer wire and a few layers of insulating tape between concentric windings. The precise number of insulating tape layers is essential to achieve the necessary safety insulation. The white tape wound along the sides is important too. Watch the video again closely. How careful is she being to count the number of layers of insulating tape? It's possible her winding machine is counting them, but this step flies by so quickly it's a bit worrying. Edit: This may be extrapolating a bit because it may be we're seeing the outer, rather than an interposing, layer of tape being wrapped in the video.

4. The transformers are piled haphazardly as they are being constructed. They are heavy and the bottom pins can have get a pretty sharp point on them. I bet some of those transformers at the bottom of those piles are getting their insulation punctured.


(She speaks Mandarin, briefly, so this is very likely China.)

It's amazing though! Working that fast.

It's also just what the client ordered, if they wanted better quality they'd have to pay for it.


But would a machine do better? Machines can miss too.


You can still run those through a bed-of-nails tester to verify insulation, continuity, and pinout.


Yeah, that will catch a lot of direct shorts and opens but some problems just don't show up until the device has been thermally cycled and baked in for a few thousand hours of usage. It's also possible that there was another problem in the circuit which caused overheating in the transformer.


I think that video has been sped up somewhat....


Why do you think it's sped up?

There's surprising amounts of skill in these unskilled / semi-skilled jobs.

If you're sitting at a machine winding transformers for 8 hours a day, and paid for each item you make (and fined for each item that fails test) you quickly get faster at winding transformers.


  >Why do you think it's sped up?
Primarily because of the extreme precision when moving the arm over a long distance


That seems sped up, like a 1920s movie.


A true isolation transformer has an earthed copper shield wound in between and physically separating the primary from secondary windings, anything less is unsafe.

One of the many reasons I detest "double insulated" equipment, as it is impossible to determine if it has been designed for safety or not without ripping it apart and destroying it.

This is a real issue in Australia as our regulators do no regulating and most equipment I have investigated does not comply with Australian Standards even though manufacturers lie and put standards compliance numbers on them.

Also the Australian Standards should be considered an absolute Minimum not "super safe".


Does the Australian Standards body require that devices are sent to them for testing? Then there could be a database you could check as to whether a device has passed.


technically yes.. but it is not enforced


I should also add that if you are an Australian manufacturer, you have to submit your product for testing for design compliance and you won't last long if you dont, but importers can import anything with none, yet another impediment to Australian manufacturing.


> even though manufacturers lie and put standards compliance numbers on them

The CE marking in Europe is the same (in most cases). It just means that the manufacturer declares it is compliant, not that it has been approved and tested by anyone.

https://en.wikipedia.org/wiki/CE_marking


I found that at least in the past, the S markings from Sweden where quite good. Now, I have not been back for 20 years so no idea if it still valid...


What about ordering things that are also UL/CSA-approved? Lots of electronics sold in the EU have their CE mark, but also list their UL/cUL/CSA approval.



Hehe, I don't know whether CSA/UL/cUL are faked as often though -- I think you can even look up the certification record to check it matches the product.


If you're worried you can order 2 at the same time and then rip one of them apart. It's an expensive solution but it should give peace of mind.


But would it give peace of mind really? Right in the linked video. It would seem the widget is properly designed, it has a thick separation gap between the two sides on the pcb. It would seem, that you could test 10 of these and they are all fine, and the 11th kills you. Or is that transformer design obviously unsafe?


How much would you charge me to do teardowns / tests of a few chargers?


Big Clive, who makes these videos does some brilliant tear-downs. I recommend watching his other videos.


Big Clive is _the_ channel for teardowns of everyday electronic stuff. I am actually not that interested in this kind of stuff but somehow I still watch most of his videos.

If you are into more... entertaining teardowns you should have a look at AvE. Even if you don't like it you will still learn some useful canadian terms ;) (edit: just to make it clear, AvE is actually very knowledgable. His "boltr" teardown series of electronic power tools is most likely the most thorough you will find on the internet)

My third goto channel for electronic stuff is EEVBlog who also does amazing stuff. If you are going to watch one teardown today then please watch his rather famous video about the worlds worst tablet computer: https://www.youtube.com/watch?v=2o8MDCIlOEk


> If you are going to watch one teardown today then please watch his rather famous video about the worlds worst tablet computer: https://www.youtube.com/watch?v=2o8MDCIlOEk*

I'm crying out of laughter right now. This is absolutely brilliant. That is, the teardown - the tablet itself is an abomination. Thanks for posting this. This channel is great and seems to have a lot of educational content besides teardowns too.


> the tablet itself is an abomination

Yes, but I find myself strangely impressed. I like to imagine someone walked up to an engineer and said "So, I just told client $foo that we can provide a tablet version of our PC software. They need it by end of week. That's not a problem, right? Because the CEO is fully on board with this..."


And then the engineer went home, got drunk, and talked his niece/nephew through building this on a bad skype call.


> My third goto channel for electronic stuff is EEVBlog who also does amazing stuff. If you are going to watch one teardown today then please watch his rather famous video about the worlds worst tablet computer: https://www.youtube.com/watch?v=2o8MDCIlOEk

Oh boy, that thing is a MESS. When he first opened it, I thought it looks like the sort of thing I'd knock together - not pretty, probably mildly unstable, but gets the job done. Then he took a closer look, and it's the sort of thing that wouldn't even be acceptable as a secondary school project, never mind a device to be sold.


Yikes... That tablet makes the cheapest, crappiest Chinese Android tablet I've bought look like an amazing feat of engineer in comparison (though that's probably mostly because almost all of them are slight tweaks on Allwinner or MTK reference designs - most of them, regardless of manufacturer look almost identical).


Just this weekend I watched entirely too many AvE videos and now I am fluent in Canadian.

His style is... unique. I think it's great but I'm sure it's not for everyone.


Mikes electronic stuff is also quite good.


See also scanlime http://youtu.be/UTUYl-2g-r4




Tesla500 as well


AvE is definitely one of my favorites. The only youtube channel I contribute to. I'll probably starting contributing to Big Clive as well.


Thanks, I'd not seen that tablet computer before. Incredible!


I like that the company that produced this apparently still exists.


Actually found that incredibly annoying. These guys managed to get something working, probably with very limited knowledge and funds, that deserves an award.

It just seems this guy expects everything to look like an Apple ipad when you open it up.


He definitely doesn't expect it to look like an iPad; he points out that he's cool with them cobbling together off-the-shelf parts like the motherboard and the touchscreen controller. His issue is with how badly things are botched, like soldering the batteries directly, and then selling it as a fully fledged industrial product. If you have "very limited knowledge" as your comment says, then you shouldn't be doing business. That's irresponsible. At worst, designing products without experience could lead to electrical fire.


Including sexual lubricants

https://www.youtube.com/watch?v=5i7_-NtJeB4

Someone has to review them...


From Poundland, which - for any non-Brits who haven't come across it - is our more upmarket equivalent of a dollar store and sells all kinds of interesting items that he's done videos about.


Poundland? Sex lubricant? The marketing just writes itself...


The vibrating cock ring is the next related video, and is much more interesting. There's a tiny motor, possibly the kind used in mobile phones, which he takes apart.


Here, Clive looks at an Ikea USB power supply: https://www.youtube.com/watch?v=uRe9w5PKmsE&nohtml5=False

Much better made.


The basic issue is that if that thing hurts or kills someone, Ikea's rep is on the line for selling crap products.

Over on Amazon, anyone can offer something for sale. And the only real indication you have is the "sold by XYZ, fulfilled by Amazon" line somewhere near the top.

The site is amazon.com, the logo and design is all Amazon, but Amazon never had to place an order for the item, nor is on the hook if something blows up. They just act as a combo of UPS and Paypal for the transaction.


> Over on Amazon, anyone can offer something for sale.

The practice is called white labeling: http://www.investopedia.com/terms/w/white-label-product.asp

Basically an Amazon seller sources product from China and gets the manufacturer to label the product with their brand name. Quality can be either really good or dodgy.

However, I have seen a lot of this stuff with the CE mark but not the UL mark.

Do the CE underwriters just let anyone stamp their mark on any product with minimal oversight?


There are no CE underwriters and you put the mark on your own products. It's self-certification.

The manufacturer/importer is supposed to be able to produce evidence of compliance on demand, but if it's posted from China, good luck with that. This is why the products are so cheap.


I recently discovered AvE and now bigclive. I'm not sure how I missed this particular internet rabbit hole but I love it.


If you want another rabbit hole to start down, I can also recommend Techmoan. He doesn't do teardowns, but instead likes to buy old HiFi gear and reviews them.

Some of them are incredibly obscure such as the Tefifon: https://www.youtube.com/watch?v=nBNTAmLRmUg


This is great, thank you!


I just copy/paste a comment I made below:

> My third goto channel for electronic stuff is EEVBlog who also does amazing stuff. If you are going to watch one teardown today then please watch his rather famous video about the worlds worst tablet computer: https://www.youtube.com/watch?v=2o8MDCIlOEk


If you like AvE, check out This Old Tony. He is more machining focused, but he did do a clever multimeter 4/1 video.


If you are into machining then check out Clickspring, oxtoolco, Abom79, Keith Rucker, Keith Fenner, Shadow HKW, ...

If this is not enough then go and visit this list:

https://www.reddit.com/r/ArtisanVideos/comments/3v264a/meta_...

(But be careful, this is the ultimate rabbit hole)


Keith Fenner is astounding. Years ago my girlfriend said "enough Keith Fenner videos, You have to take a welding class", I did and it was amazing.

As a side note, I'm starting a blog where I interview heavy equipment operators and write about heavy equipment. Anyone here interested in that?


Heavy equipment operator and computing science student here. Let me know if you'd like to talk.


Fuck yeah.

Um, I mean, yes. Yes, I am interested.


Yes


It's sad in a way that I only find out about these other channels on a HN thread. Youtube really should work on making channel discovery much more effective.


How do I know what I'm buying is safe? I've sworn by the Nexus chargers in their various incarnations and http://www.amazon.com/Charge-Charging-Station-Micro-USB-Sams... (I have a Tronsmart and an Aukey vertsion)


I play Ingress a lot (a GPS based mobile game where you run around a lot in the real world so most players use external batteries) and at least 60-70% of the whole community uses Anker products. I personally know at a dozen people who use Anker wall plugs and all of them (including me, 1 wallplug and 2 battery packs) are very happy.


Anker and recently Aukey. And iOrange for cables. The problem is that smaller brands hit the Amazon lottery and get a lot of early 5-star reviews, then swap out their product with one made of cheaper components.

I think for Anker, their brand has become a bit bigger than is gamble-able on a bait and switch, so I'm continuing to use their products.


Anker is ex-Google employees.


Interesting! Source?


Just like Ingress used to be Google!


Do you know of a reliable 4 in 1 USB cable manufacturer?


I do not, sorry :/


Just want to add: With Anker make sure you're buying through an authorised channel. Fake Anker products DO exist.


Unfortunately there doesn't seem to be an authorised reseller in Australia


Yes there is. macfixit.com.au and mushtato.com.au both stock anker. Unfortunately, because of this Amazon has stopped shipping the brand to AU and the prices here are twice or more that overseas.


Bugger. Off to priceusa.com.au


When talking about Ingress, I suppose you primarily talk about USB power packs. These are not as dangerous as power supplies that are connected to 110V or 220V (although they can catch fire).

Instead of trusting a particular brand, I think it makes more sense to trust the various security certifications that were created for this purpose.

The ones I am aware of are:

* TÜV Nord, TÜV Süd, TÜV Rheinland (TUV, Germany)

* GS (Geprüfte Sicherheit, Germany)

* UL (Underwriter Laboratories, US)

* CCC (China Compulsory Certification, China)

There are probably more that are trustworthy and that actually check the safety of the devices.


> I suppose you primarily talk about USB power packs

No, somehow we have to charge all those batteries ;) A lot of people use Anker products for this as well.


I pretty much only get Anker products now and recommend them to anyone, super high quality and very affordable. Pretty good customer service as well.


Beware: Anker became so successful that now Anker knock-offs are a thing.


Does anyone know if the RAVPower products are any good? I have a few (4-port USB charger, 6-port USB charger, and a battery pack) and have been impressed with their performance (they are the only chargers I have found that my Vita will recognise, for example), but can't find any teardowns etc. discussing build quality.


I have this funny feeling that Tronsmart, RAVPower, Aukey and Anker are using the same factory for these - very similar specs and features for all of these chargers. If I had the cash to get 5x ~$50 devices tested by someone like Clive, I'd definitely do it. I do worry about the internals of the device being swapped out silently by the manufacturers, though.


I'm happy with the one product of theirs I have. They have multiple products on Amazon with several thousand positive reviews each. I don't know if that means they will be reliable in the long term, but they seem to have one of the best names on Amazon to me.


Anker is fantastic. With one weird exception - when I use my Anker USB to Ethernet adapter, my (Microsoft brand) wireless mouse stops working unless it put it right next to the USB receiver. I bought a much crappier-looking Ethernet adapter and had no issues.


I believe many USB3 devices emit noise in the 2.4 GHz spectrum, jamming nearby WiFi and Bluetooth.


Yep. This is a known issue, and there are ways hardware manufacturers can mitigate it:

http://www.usb.org/developers/whitepapers/327216.pdf


I can answer from an Australian perspective. Here's how it's supposed to work:

Suppliers register and prove that their products comply with Australian standards, and are awarded a compliance number. That number is the consumer's indicator that a product is safe. Any interested party can check that the certification is genuine by looking up the compliance number in an online database.

Here's how it works in practise: you're screwed. The relevant government department took the database off line years ago, and there still isn't a replacement. When you go to do a search, you arrive at a page that talks about the database in the future tense [1]: "The national database will list all the equipment registered by the responsible supplier". A supplier can put a fake mark on a product, and there is no easy way for a consumer to know if is genuine, making the marks virtually useless. The result is electrocution whilst listening to a plugged in music player with headphones [2].

[1] http://www.erac.gov.au/index.php?option=com_content&view=art...

[2] http://www.smh.com.au/nsw/warning-over-usb-chargers-after-wo...

---

Edit: spelling



I wonder what the hell's gone wrong in that department to not have a replacement after all that time.


You should read up on Myki


Public transport payment system != product safety system.


Government IT projects are government IT projects, no matter what the end goal.


Nothing is 100%, but the chance should be sharply reduced via certain tests.

Supposedly the CE marking in Europe should be a good indicator, but sadly there are various ways that a company can self-certify products.

I think the biggest problem is that Amazon allows anyone and anything to use them as an outlet by default. Thus many fly by night resellers can used Amazon's brand as a cover.


All CE certification is self-certification.

The manufacturer/importer is supposed to keep and present on demand a "technical file" detailing compliance to various standards, which is supposed to include particular tests, but there's no real guarantee that they've been done.

On the other hand, this is a major barrier to entry to small-run products, as is WEEE.


> How do I know what I'm buying is safe?

I think I answered my own comment a few sections above. But for safety you should look for the UL mark:

http://www.ul.com/code-authorities/resources/faqs-for-code-a...

The CE mark is the manufacturer's self declaration of compliance with EU law, sometimes including safety.

Whereas the UL mark can only be put on products that Underwriters Laboratories have authorized themselves.


Although counterfeiters aren't above putting it on unsafe products - e.g. http://www.righto.com/2016/03/counterfeit-macbook-charger-te...

"The photo below shows the safety certifications that the charger claims to have. Again, it looks genuine, with no typos or ugly fonts."

Or http://www.righto.com/2014/05/a-look-inside-ipad-chargers-pr...

"The counterfeit charger displays a bunch of certifications (such as UL) that it doesn't actually have. As you will see below, there is no way it could pass safety testing."


That's scary. Not enough isolation between the line side and the output side. Classic cause of electrocution and fires. The standard for this is UL 1310, Class 2 Power Supplies. Not sure if this design complies.

There's a lot to be said for split-bobbin transformers, where there's a solid plastic barrier between the two windings. This just has wire insulation and some tape.


The really scary thing is that as he points out, it actually seems to be designed with an apparently reasonable amount of isolation between the primary and secondary - it's just that something managed to completely abrade the insulation. Not sure if it was UL compliant as designed because their requirements for transformer construction are quite complicated, but it certainly seems plausible.

He's reviewed numerous really questionable power supplies[1] with a gnat's whisker of separation between the mains and low voltage sides and much more dubious transformers, but somehow none of those actually managed to put 240 volts on the output...

[1] Such as the cheap shitty pink USB charger from China: https://www.youtube.com/watch?v=ioAq7PI1Uwg


The really scary thing is that as he points out, it actually seems to be designed with an apparently reasonable amount of isolation between the primary and secondary - it's just that something managed to completely abrade the insulation.

That's the argument for a split bobbin. With a hard plastic wall between the two coils, in addition to the wire insulation, it's hard to get a short between both side of the transformer.

Here are some split-bobbin transformers for switching power supplies on Alibaba.[1] Split-bobbin seems to add about $0.10 to the price, raising it from $0.89 to $0.99.

[1] http://wholesaler.alibaba.com/product-detail/EE22-split-bobb...


Split-bobbin seems to add about $0.10 to the price, raising it from $0.89 to $0.99.

So, there it is.

Good luck finding a miscreant to hold accountable if this sort of cost reduction kills someone. And even if you could hold the culprit accountable, that still can't undo the loss of life anyway.

It's all just collateral damage from the race to the bottom.


Price aside, split bobbin construction also leads to lower efficiency since there is less effective coupling between the two sides.

It's always about tradeoffs, sort of like how you can ride a motorcycle or drive an armoured vehicle. What is safer won't necessarily be better in the other dimensions.


> Not enough isolation between the line side and the > output side. Classic cause of electrocution and fires.

I'd doubt that any significant amount of fires are caused by loss of isolation between line and output side. While it might be very unpleasant to touch any device that came into contact with the live wires, there's not really much current you can have flowing. And if any significant current flows, a GFI should catch this, and shut down the power.

Regarding the amont of deaths by electrocution: How much people are actually electrocuted by badly designed house appliances? As far as I see there's only very little injury related to electrical shock outside of actual electrical or construction work, where people might naturally be more easily exposed to live wiring carrying higher currents, possibly higher voltages and, most importantly much lighter standards in terms of circuit breakers.


Finally a possible explanation [1] as to why Apple chargers (USB and laptop) give you tingles when touching the metal casing of the device they're plugged into.

Would love to know if the mentioned capacitor is indeed the reason and whether it leads to enough efficiency to justify the scare it causes with first time customers.

[1]: https://youtu.be/3Hdn0MuCK_0?t=2m14s


As a reminder, if you're buying a USB charger or battery, or MicroUSB, Type C, or Lightning cable please buy Anker products.

They are not expensive, but they are the best products you can get your hands on. I own multiple Anker products, and have never discovered anything better than these.

The only non-Anker charger I currently use is the Google Nexus branded Qi charger (plugged into an Anker multiport brick), and the original Chromecast brick plugged into my Chromecast.


The one thing that I can't find from Anker or any other reputable, reasonably-priced supplier (last time I checked, anyway) is a single-port charger that is small enough, horizontally and vertically, to not sprawl out onto adjacent outlets in a power strip.

I don't understand why this is so hard. My first smartphone, a Samsung Galaxy Nexus, came with one, but it failed within a year and Samsung wanted like $50 for a branded replacement.


If you look at some of bigclive's other videos you'll probably understand why it's so hard. Small, single-port chargers skimping on isolation requirements to meet the size restrictions is kind of a long-running theme, most recently https://www.youtube.com/watch?v=tQK2Z0wRND0 A lot of engineering goes into safely fitting power supplies into compact packages and companies want to charge a premium for it.


Anker power pins are oriented to only block one socket in a typical up/down arrangement in the older version of the product. So, my 4 port charger from them? Only uses 1 socket, and doesn't block the one above or below it.

Also, their 5 and up port new and old ones don't hang off the plug, they have a short normal power cable.

https://www.anker.com/products/71AN3654WS-WA

This is the one I have. Their new version of the product doesn't seem to be oriented that way anymore. So, yeah. Not quite what you were looking for, but they do exist.


Unfortunately Anker / Amazon don't want my Australian dollars. I've got a similar Tronsmart charger I got from Geekbuying (similar enough that I want to send one to Clive for testing). It's even more difficult to get decent USB battery packs here as only DHL will ship them


I buy those little 3-inch extension cables to get around this problem.


Like an Apple one?


The Apple ones do not speak standard USB charging resistor ladders, thus max out at 500mA when charging non-Apple devices.

On the flip side, Anker chargers speak both, and charge newer iPhones and iPads faster than that huge iPad charger does.


I bought an Anker 5 port USB charger and one of the port died after just 1 month of infrequent use. Their products are very well polished but their QC is bad as you can find so many similar reviews in Amazon. They refused to replace my unit as it was sold by: Amazon Export Sales LLC, according to them which is not official reseller.


I bought five anker USB cables after reading about them on HN, and a wall charger, and I was actually really disappointed. All the cables stopped working within three weeks. Monoprice USB cables didn't fare well either, but the OEM cables that came with my Icons, kindle, and S3 have lasted years.

Wall charger still works great, but the cables are what I was really interested in because people are always running off with them. Still haven't found anything satisfactory.


I had a similar problem with some Anker cables - they'd die with light use. I found at least one culprit in that my wife would rest her thumb on the cable connection while the phone charged. This bent something enough the cable would stop working.

But some of them have simply stopped working after use on my desk or being carried around in my backpack for a while. I've switched to aukey for cables for that reason. Whether they're any more reliable is TBD.


What did Anker say when you contacted them? Anker has a warranty.


I'm very lazy and didn't bother. :) Given Anker's reputation, I would be surprised if they wouldn't have replaced them. (In fact I haven't bought new cables either - just been making do with the few cables I have which has been a bit of a pain.)

I just want cables without the hassle of going through warranty and RMA processes. It's possible I just got a bad batch - I can't believe people would be as positive about Anker if my experience was that common - but it did sour me on them. I just want to find a manufacturer where in my experience, the cables I buy are good 100% of the time and last for at least a year so I can buy them in bulk and recommend them to others with confidence. And I haven't found that.


I've looked at Anker products in the past but it looked as if they didn't have any certification (TÜV, UL, CCC, GS). I'd rather trust a non-profit safety consulting company than one particular brand. That said, UL LLC is no longer a non-profit since 2012.


I've always dismissed Anker out of hand as they are practitioners of the comically verbose product name school of Amazon listing, which I've come to associate with poor quality. Add to that the usually accompanying infeasibly high review count, and I'm genuinely surprised to hear they're any good at all.

I probably will give them a go now, but they're in bad company.


I love the simplicity of and the humor in the Big Clive videos. The methylated spirits video was quite... hilarious. DO NOT try this! https://www.youtube.com/watch?v=ep2I3Gf3Sec


Hmm, I've always bought Anker and occasionally bought other brands like Mpow, Oldshark (haha) and Aukey which seem like yet another rebrand from the same batch of Shenzhen factories. I know Anker has considerably more Customer Service backend and better packaging but I wonder if their products are engineered any better?


Google engineer Benson Leung tests a bunch of USB 3.0 cables/chargers and has generally found Anker to be good. Spreadsheet of results: https://docs.google.com/spreadsheets/d/1wJwqv3rTNmORXz-XJsQa...


He's like the "Bob Ross" of electronics.


Apple chargers, USB-C... When will Amazon offer a hardware fault bounty?


Seeing the wire damage reminded me of this Boeing 787 fire...

http://asndata.aviation-safety.net/reports/2013/20130712-0_B...

...caused by a battery short resulting from the two wires being trapped against the case and eventually the insulation abrading through when subject to vibration. Even the highly regulated aviation industry isn't immune to simple defects like these.


Can someone explain why the transformer is on the DC side of the circuit in the diagram he draws at the beginning of the video?


This is a switch mode power supply. In order to use a smaller transformer it uses very high frequency AC rather than 50/60Hz mains frequency.

In order to do this it rectifies the mains input to high voltage DC (330/220Vish, depending on the mains voltage) then using a power transistor rapidly switches this on and off at around 1-10kHz. Then the transformer outputs low voltage AC which is re-rectified and filtered to produce low voltage DC.

The switch mode chip monitors the output voltage and modulates the switching frequency to maintain a constant output voltage.

If you look closely at the board (and he talks about this in the video) there's a separation between the high and low voltage sides. Only the transformer and certain specialized components should cross this (Y-class capacitor and optoisolators). Cheap power supplies often miss this out which could allow a small bit of loose solder being able to send mains voltage to the output. Look at other videos by Big Clive to see the horrors.

The high voltage side is a bit of a misnomer. It's really just the non-isolated side - the chip runs on 5v or similar, but it's not completely separated from the high voltage DC so isn't safe for human contact.


> the chip runs on 5v or similar

How does the chip get 5V, out of interest? Given that it's not allowed to be connected to the non-isolated side?


Normally though an additional winding on the transformer.

Switch mode transformers have three or four sets of windings. The primary (high voltage), secondary (low voltage) a bootstrap winding (for the chip) and sometimes a flyback winding.

The bootstrap winding takes off a variable voltage depending on the local supply and uses a linear regulator to produce the required voltage.

For a better explanation than I can manage in text the EEVBlog channel goes in to all sorts of regulators and how they work: https://www.youtube.com/watch?v=cM7t1Mpu7s4&list=PLn-ZIFhWxh...


This happened to me with a cheap reflow heat gun. [1]

Got a nasty bite as soon as I touched it.

Got what I paid for I guess. Made me rethink super-cheap AC equipment from the internet, that's for sure.

[1]: www.amazon.com/Kohree-Digital-Rework-Station-Solder/dp/B00ITMPQS2


Aren't there legal tests that any appliance which plugs into the mains must pass? I'm surprised this kind of thing is possible, but I guess not performing such tests cuts costs.


Sometimes even large well trusted companies get fake products shipped to them.

I think EEVBlog highlighted this with the FTDI case. People were buying what they thought were authentic FTDI chips (and paying the proper price) but being delivered fakes that then went on to be bricked by the driver update.

There is so much margin in swapping good stuff with dreadfull stuff there are several points in the procurement chain where people can try their luck passing off crap.


There's so much forgery of UL and other certification marks and so little enforcement or punishment for doing so that you can't rely on them.


oh man i don't even need to click, i know its big clive! i spent all of saturday watching his videos holy shit there are some dangerous products on the market!


Never seen a video from this guy but I love the massive beard moving at the bottom of the video when he speaks!


Why does he handle the bare wires of the test lamp near the mains? Is there no risk the way he hooks it up?


The wires are insulated. Yes, there is a risk but this is what he does.

Check out this video for a story of a risky thing he did with a heating element and a bath (4:30 is where it really gets going): https://www.youtube.com/watch?v=uUn17ccR2a4


The scary part is most of us would plug our phones into one of these chargers without thinking there'd be any more than 5 volts at the output. Best case - destroyed phone. Worst case - fire or electric shock.


Heh, ground fault interrupters should probably be required on every circuit.


That only works if the device is grounded, and the fault is to ground. Neither is the case here. :(


The most likely danger with this fault is that someone would touch both the live device and ground, causing a path to ground. This would cause any RCD to trip. It's obvious that whoever owned this didn't have one installed because even the cable shorting to a grounded radiator didn't cause a trip.


Question for the knowledgeable: why did one of the insulation layers have some metal foil in it? He commented on it as he was unwinding it, as if it was a little odd but not completely bizarre.


I get that tingly feeling when I hold my ipod after plugging it into the wall through a usb adapter. It's given me a minor shock once I think. That's very interesting.


great video. I don't know a lot about electronics, so I am left wondering two things:

why did the fault not stop after i pulled out the first two components? wouldn't that break off the power circuit inside?

why did the fault stop after he removed the transformer?


The first two components he removed (the filter Y-capacitor and the opto-isolator) are supposed to be open-circuit to direct current. And since they were not at fault (ie, they were not shorted), removing them from the circuit didn't change anything.

The fault probably stopped after he removed the transformer because in the process of desoldering and yanking it from the board, he caused enough strain/vibration that whichever wire was shorting the primary and secondary windings moved and was no longer causing a short. That sort of intermittent fault is common in hardware, and quite annoying to debug.


I have always buy Amazon Basics chargers.


This takes some horrific failure to accomplish, I imagine.


The problem appears to be the transformer: https://youtu.be/3Hdn0MuCK_0?t=20m49s (and specifically a bit of shredded insulation https://youtu.be/3Hdn0MuCK_0?t=25m35s)


Got your Power Delivery right here.




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: