Turning the Belkin WeMo into a deathtrap

The Belkin WeMo is a small, WiFi connected outlet controlled by a mobile device that adds Internet control to a desk lamp, coffee maker, or, if you’re feeling daring, your home server. It’s an interesting device, but of course there are a few security implications of having your electric kettle connected to the Internet. [Daniel] was able to get root on his Belkin WeMo and with full control of his Internet-connected outlet was able to turn it into a deathtrap.

[Daniel] says his exploit could be developed into a virus that will scan for WeMo devices. Once these Internet-connected devices are found, it’s easy to turn these devices on and off really fast; something not too dangerous for a desk lamp, but potentially lethal if it’s plugged into a space heater.

In the video after the break, you can see [Daniel] exploiting the WeMo with a flaw in its UPnP implementation. There’s footage of his terminal hacking and of his desk lamp being turned on and off really fast, something that could be very dangerous for higher current devices.

90 thoughts on “Turning the Belkin WeMo into a deathtrap

  1. Nice!

    Supposedly Belkin showed a WeMo lightswitch replacement at CES. Although the WeMo stuff is pricey (early adopter fee), it looks promising. Of course, ANYTHING is better than X10, so I’m hoping this drops in price and becomes more viable.

      1. Those boards speak SWAP, not wifi, and they don’t appear to be UL listed. Apples and oranges are different, even if you can eat either of them.

        1. I think the point of Rasz comment is you can do a DIY remote switch easy. No need to do anything in the outlet you could make a box or use a power strip and build in the TL and relay. You don’t need to be an electrician you just need to be somewhat smart and not work on anything will its on.

  2. I wonder if one could modulate the switching of a WeMo-controlled high-current device in more interesting ways; maybe to control an associated X10 system, or transmit morse. :D

    1. I think they mean if you remotely hack a desklamp, ok, there is light where none should be. But if you remotely turn on a heater and no one notices, a fire may break out.

    2. When you switch a device on the current spikes (I believe due to inductive effects?) so switching on a heater, it will initially draw far more than the cable is rated for. This isn’t a problem for short periods and the cable will hardly warm up at all, but prolonged use would create a lot of heat. Although surely a fuse or an RCD would help out a lot here?

      1. The current spike you get in the case of a space heater etc is due to inrush because the cold heater coil still has quite a low resistance until it heats up. If you turn it on and off quickly, it would still heat up, and the current spikes would get quite small.

        You could probably damage some power supplies, and other active circuitry though that isn’t designed to cope with rapid power cycling, and if switching a heavily inductive load, you might put a lot of stress on the relays in the wemo and get them to burn out or melt or something.

        1. Heaters amongst other devices do have the wiring in coils often, those will induce a backcurrent and if it’s a device without or with insufficient compensation by a capacitor then switching it rapidly can cause a failure and a fire.

          Of course having a heater on such a switch doesn’t seem a smart idea in general to me.

          I wonder what the rating is on its relay anyway, some such devices can’t even do too much wattage, and some use cheap relays that themselves might break down I bet.

    3. Quickly turning on and off High amperage appliances could cause damage to the circuitry. By flicking the switch over and over would cause strain to the relays and/or transformer.

      1. Are you high? Don’t speak on things you obviously don’t know. I have been designing products that comply with UL for 26 years. There are intermittent power interruptions that change in both frequency and duration, line spike, line surge, brownout and overvoltage. About the least likely thing to suffer from these is anything inductive. Good God man, take a basic science class.

        1. In the civilized world we don’t have constant voltage fluctuations and brownouts and spikes, are you in some third world country or california?

    4. Arcing of the relay. Listen to the audio in the video (before the music starts) and you can clearly hear the clicking of a relay every time the device is turned off or on. Additionally, some devices will cause problems if switched on/off rapidly. The balasts in the older florescent fixtures, as well as a few DC power supplies were burnt up by an automatic generator switch that went bad and was rapidly switching between mains and generator at the fire station near me a few years ago.

      1. We call it PPPD which means pizz poor product design. If you have a relay rated for 15 amps, protect it. There are plenty of relay snubber circuits that can do this. If you know what the load will be (a motor at startup) a simple 10 cent circuit consisting of a cap and falmeproof resistor will reduce the arcing. It ain’t rocket science, but seeing what qualifies for engineers nowadays, I guess it is.

    5. I believe it is not directly related to the high current, but rather the inductance of the load. Switching of inductive loads can cause arcing within the switch, normally this isn’t much of an issue with AC as the zero voltage crossing of the AC power will rapidly extinguish the arc, but if you switch rapidly then you will likely create repeated arcs, heating up the contacts and causing a potential fire hazard.

      Another may this may be dangerous could be the fact that boost voltage converters work by switching inductive loads very rapidly, potentially producing very high voltages. However, for that you would have to get the frequency just right.

    6. Inductive loads, such as motors will put extra stress on the device. CLF bulbs also have a high current draw when switching on, the non-dimmable ones at least. That’s most dimmer modules say for non-inductive loads and not for use with fluorescent lighting.

      Heaters are generally resistive loads so this flickering is not as much of a problem.

    7. I think it’s because large AC voltages can be generated, causing arcing. This effect becomes more pronounced, the greater the current draw and the faster the device is switched.

      From my electronics textbook, regarding inductors:
      v = L*di/dt

      This means that the voltage at the terminals of an inductor (like heater coils, motor coils, etc.) increases with the rate at which the current changes (di/dt) and, of course, the inductance. The peak di/dt for a typical 15A, 60Hz sinusoidal mains is max(dsin(t)/dt from 0 to 2pi)*60Hz*15A = 900A/s. In a typical heating element of maybe 1mH, this amounts to just one volt of inductively-induced voltage. An (ideal) switch like a relay, however, will instantly and completely isolate the inductor by opening the wires leading to it. This means that di/dt shoots to +/-infinity, and likewise the voltage.

      This effect of boosting voltage is used in a controlled fashion in circuits like the MintyBoost often featured on HoD.

      1. The danger is heightened the longer the relay continues to oscillate, because any arc induced ionizes the air through which it travels, turning the air itself into a conductor. From then on, it takes lower voltages (potentially with deadly currents) to continue the arc from then on. (This is why with tesla coils, you have to bring the ground probe close to the anode before you can draw the spark out to a longer length.) As it continues to arc, heat would be concentrated, which could turn into a fire.

        Even if there is no fire, it could be deadly to touch any surface electrified by the arc. (This is what the third prong is for; it redirects arcs into the earth. But, many people will use three-to-two prong adapters without connecting the grounding tab, and maybe the outlet was mis-wired and isn’t really grounded to begin with….)

    1. You know like when you’re taking a bath with a hair dryer submersed under water, plugged into a WeMo, and a non GFCI outlet.

        1. Still if the only thing keeping your heater from burning your house down… A heater should be safe to run all day. Just unplug it if you’re gonna store fireworks on top of it.

      1. Imagine this device plugged into a space heater. Then an attacker could remotely execute the sequence causing the heater to potentially catch on fire. This is the common problem in SCADA/ICS environments. This is the first exploit in the consumer space though.

        1. Reference please? I can’t imagine a scenario in which a space heater would be safe if on continuously but would be dangerous switched on and off in any pattern whatsoever. Having said that, I think that space heaters should not be left on unattended for any reason– excluding the oil filled ones which you can touch, leave a shirt on, etc even when left on indefinitely.

          1. +1. Anyone that plugs a space heater (especially one without a thermostat, tilt sensor, etc) into a remotely switchable outlet deserves everything that happens to them.

      1. Wouldn’t turning a heater on and leaving it on be more dangerous than turning it on and off really fast…? Also don’t heaters normally have a thermostat and a main power switch that is downstream of the remote power outlet? Heaters dont really strike me as something someone would want to turn on remotely in the first place…

        If anything Id be slightly concerned about sensitive electronics

        1. doesnt have to be a heater, could be something with an induction motor. those things draw up to 15x their rated current on startup. easy to burn out by power cycling.

        2. It’s easy to imagine “I’m leaving for home, so I’ll turn on the heating to warm my house up, or turn on the A/C to cool it down.” It’s harder to understand why I’d want remote power-on on my home electronics – most home electronics is only useful if I’m actually there. About the only useful remote I can think of would be remote programming for a DVR.

          1. Home automation isn’t necessarily only for remote control, but also sometimes for timed control when you’re nearby. For example, I wake up every morning to a radio that is switched by X10, controlled by a program I’ve got running on a beagleboard (single board unix computer). Sure, I could buy a clock radio, but I like the plain old radio I have, because its sound quality is very good. And my network connected unix box has its clock disciplined by NTP, so my radio has always switched on at precisely the top of the hour, for years now, without any manual clock resetting ever needed (self-adjusts for DST, too). Finally, by putting this under the control of software I write, I can easily make it so that I get extra sleep on the weekends, without having to manually adjust things or remember to reset anything before I go to bed on Sunday night.

            Of course, I don’t think fast switching of my radio is a potentially serious hazard, though.

  3. seeing there is no pictures of a teardown i am unable to know how the safety could be an issue.

    if the WeMo uses a relay to do the switching then you are not going to damage the device .

    if it uses a solid state relay, triac or diac or mosfet or an scr then it is possible to switch it thousands of times a second (pwm) and cause the solid state device to heat up or even burnout short and send power over the control wires that would be the gate pin.

    a relay can only be switched about 10 times a second.

    i can see the possibility of limiting the current so much that the heater tries to draw too much current like if you try to run a heater from a lamp cord and could start a fire

    1. Surely running a simple resistive heating element PWM would just make it less warm. I realise they have a low cold resistance, but that’s gone within a fraction of a second. If you could really cause enough of a current surge to damage something, wouldn’t the fuse in the plug go first?

      This really isn’t as obvious as people are claiming. I’d like to see some proof somewhere.

      1. The low cold resistance takes about 0.5s to go away on a large nichrome element. It’s possible to conceive of a situation where you pick a frequency and duty cycle to keep the resistance low, but switch the contacts enough to damage them and heat up.

        Regardless, even when heated up, a heater can present a 3kW load. This is not small.

        You need to bear in mind that a lot of the relays used in these switches are not great – cheap Chinese ones only just large enough. The RF sockets in the UK sometimes have 10A relays – not the 16A ones required to switch 13A (the packaging does warn you), but they will work for a while with a 3kW electric heater.

        The symptoms I have seen for failing relays are contacts welding shut, contacts not closing, and most commonly overheating caused by pitting and soot. This melts the outside of the relay.

        A fuse will do nothing. There is no excess current being drawn. Even at 26A (2x rated current for a 13A fuse), the fuse will take minutes to blow.

        1. >A fuse will do nothing. There is no excess current being drawn. Even at 26A (2x rated current for a 13A fuse), the fuse will take minutes to blow.

          *cough* bollocks *cough*

          What world do you live in where fuses allow 2x rated over current for minutes?

          Regular fuses blow in a second at 2x current.
          Slow blow might be tens of seconds, not minutes.
          Fast blow fuses, which are probably most common will blow in a tenth of a second.
          You may be thinking of fuses that are time delay. Certainly not standard.

          So… Fuses are still likely to blow well before wires catch fire.

          1. I have no idea where your misconception comes from, it’s one I am seeing a lot at the moment – there seems to be a magical world where fuses blow instantly at 1.000001 times rated current. This isn’t how it works. I’m also really not sure why you are being so assertive about something that is trivial to disprove.

            Here is the datasheet for a common, 13A plug-top fuse from a respected manufacturer.

            http://www.farnell.com/datasheets/10867.pdf

            Look at the envelope diagram – and notice how the 2x rated current line can take more than 100s to blow. They call that a fast/medium.

            Look at this datasheet, and see that most of the fuses take more than 60s to blow at 2x rated current:

            http://ep.mersen.com/en/pdf/Blue_Dot.pdf

          2. I never said anything about 1.0000000000001 times rated current, you put that in there.

            In any case. Where do you get the idea of hundreds of seconds? The data sheet only goes up to 100 seconds. The worst case is basically a minute and a half for 2x rated current.

            This is still a bunch of crap, this hack is worthy of listing if only because it adds a better way of switching the appliances (as in another way of switching this currently only iPhone controllable device),

            The idea that an Internet controlled space heater or kettle is somehow a security risk is laughable!

    2. This is an issue with inductive loads, (possibly one with resistive that have an air-gapped component on it like my oil filled radiator space heater which visibly arcs upon start up). I repair commercial kitchen equipment, refrigeration and hvac for a living and have seen plenty of relays, contactors and smoke from cycling these on and off too rapidly. Cause could be a loose wire, failed capacitor filter, or idiot operators doing something the unit isn’t meant to do. In the real world plugs get replaced (there goes any fuse, if it was present), high limits get jumped out by lazy or cheap people, contacts get carbon build up (or bugs crawl in them and get vaporized), terminals get loose due to vibration and cycling. Relays sometimes fail closed too, due to arcing and welding of contacts together. I have seen bi-metals fail closed or fail to open upon fault, heating elements fail rapidly and catastrophically shooting sparks a yard across a kitchen due to bad wiring and a flooded floor chase (3 phase flat top). Fortunately I have only seen two fires caused by such failures and neither caused any real damage (but this was mainly because of isolation from combustible materials found in a commercial setting). I have also seen loads of burnt wiring from voltage spikes and drops due to storms, where loads were starting up at a bad time and blammo!

      That being said I can’t understand why this would be a useful device for much besides lamps. A timer or controller is better for grow lights or fish tanks. I wouldn’t mind turning my AC or heat on remotely but this isn’t a way to do that. I wouldn’t use one of these primarily b/c it is a solution looking for a problem, but also after seeing that video for safety.

  4. *yawn* Toasters were PWM’d first. I was quite ammused to make all the lights in half the apartment building flicker :P The cord however did not heat up.

  5. I’m ignorant, but not satisfied with the above answers. The article calls out the rapid switching: “…it’s easy to turn these devices on and off really fast; something not too dangerous for a desk lamp, but potentially lethal if it’s plugged into a space heater.”

    Why is rapid switching of high current dangerous? I suspect that if there’s a short and the pulses are short enough, the breaker might not have enough time to trip. But, if that’s the case, would the RMS current be high enough to cause any damage? What’s going on? Brian, can you provide an explanation?

  6. Interesting… A casual internet search [including the Belkin web site] did not turn up any specifications on the device – just a bunch of marketing hype.
    Has anyone actually seen any useful specs on the max ratings for AC devices??
    6lz

  7. “but of course there are a few security implications of having your electric kettle connected to the Internet.”

    Like what?

    I leave my kettle on at the wall all the time, I press the manual button on top when I want hot water, -the kettle turns itself off when it’s boiled.

    What implications are there of connecting a kettle to the internet? Some little scrote will turn your kettle off whilst it’s boiling? massive security problems if I don’t get my morning coffee!!
    And what about space heaters? Oh no, not turning off my heating too, without heat or coffee I’m guaranteed to be super bastard grumpy on that day!

    I’m curious as to what exactly you’re believing the issue is with rapidly turning off an on a space heater?
    I’d have thought that the biggest danger with operating a really high current space heater with this is that the wemo only rated to handle ~3KW.

    In which case the problems is much more user stupidity rather than exploit problem.

    I suppose that the very worst case scenario regarding the kettle connected to the internet is that I turn it on.
    then someone kills the mains,, then re-applies power -and I forget about it.
    then the person keeps letting my kettle nearly boil (but never reach the point where the bi-metal switch will kill the power, so that my kettle boils dry.

    But then usually when a kettle boils dry, and there is no longer water cooling the element, the element just burns out.

    Clearly the person who developed this hack does though, if you google wemo shell access, you see that they’ve linked their video on lots of sites for tracking exploits.

  8. Heaters are resistive devices, they can be switched at THOUSANDS of times per second and not matter one iota. Also, they are PTC, which means they are self current limiting.

    Second, if it uses a relay, then damaging the relay is the most common outcome, unless they use zero cross switching on the relay.

    IF they use a Triac, I don’t expect any faults because Dimmers work for years and they chop the output every wave.

    I just don’t see any REAL threats here, except to the device plugged into it. Due to rapid on/off cycling, such as computers, televisions, anything with power supply capacitors.

    1. Inrush current on a heater can be large. I suspect the biggest risk is the relay contacts potentially burning out (a lot of remote sockets use cheap relays, only just over the max current rating of the socket), potentially sustaining an arc or overheating.

      1. If the heater is already warmed up, the PTC of the element will reduce the inrush current. Electric heaters that are on their own (not ones with a fan to circulate the air as well) are just a big (low value) PTC resistor. Not really any different than an incandescent light bulb.

        1. If the heater is already warmed up, yes. But it isn’t. Switch it with a 5% duty cycle at as high a frequency as allowed.

          1. And it’ll heat up, and not be a problem. the temperature coefficient is not at all linear, and the higher the frequency you drive it at, the less cooling and heating time per cycle.

            Regardless you’ll get the typical light diming spike the first time, 1/4 the amount of dimming the next time. 1/16 after that, etc until you reach steady state at a level where it really doesn’t matter in the slightest.

            The higher the inrush current is, the more quickly the element will heat, and the more quickly it will get warm enough to bring the current down to a reasonable level. if this weren’t true, you would have massive problems with dimming light bulbs, which are basically lower power higher temperature space heaters.

          2. Where do you get the 1/4, 1/16 etc? I don’t understand why steady state has to be achieved. I’m not in the slightest bothered about the heater getting hot – I don’t want it to get hot. I want to drive the relay as rapidly as possible, trying to take advantage of the inrush current to damage the relay. If this means switching on for 50ms and off for 950ms, so be it. There’s no way a heater at those kind of duty cycles is going to reach steady state.

          3. But you’re missing the point:

            Your duty cycle is also applied to your mains wiring. That also has a chance to cool, just as the heater element does.

            If the inrush current can be high enough and sustained enough to heat the wires then the fuse should still blow, and if you believe that you can limit the current enough to not blow the fuse then the wire should be fine.
            If you switch with such a low duty cycle that the element never comes to temperature and increases resistance then you’re driving with such a low duty cycle that you’ll never head your wires and cause a fire anyway…

          4. No, Dan, you are missing the point. I don’t care about the wiring heating up – I have never mentioned this as an issue. You’ve brought that to the table for some reason.

          5. ok, now i’m lost.
            it was brought up by others earlier, I’ve always thought that the argument that heating the wires was weak. Even if they are old wires that are non code etc…

            If we’re not talking about turning on a heater to start a fire. By the heater being accidentally covered when you thought it’d be off.
            Or rapidly power cycling to increase inrush to make current spikes to hopefully heat wires melt insulation and cause fires in the wall.

            What is exactly is the point in power cycling the heater. How is this a security risk??

          6. I have seen these switches fail before – the relay contacts overheat, sometimes causing a thermal cutout to operate, sometimes melting, and sometimes burning.

            The contacts tend to fail for a few reasons:
            * Use with heavy and/or inductive loads – some of the switches are only 10A and struggle with 3kW heaters.
            * Cheap relays that fail or don’t work up to spec
            * Rapid switching (caused by failure of the controllogic it seems)

            My point is that you can engineer the switching of some loads to make this failure happen quickly.

  9. Or if you don’t want to (potentially) burn a strangers place down you could just have it flash dirty messages in Morse. eg. “I wanna stick my golden spike where your Union meets your Pacific!”)

  10. I agree that this attack could damage the outlet, or the device plugged into it, but it’s unlikely to cause any damage beyond that. Unless the device is already so inherently unsafe that it shouldn’t be used at all, much less plugged into ANY remote outlet.

    “Deathtrap”? I was surprised and disappointed to see it was Brian that posted this headline. Sensationalism is more Mike’s thing.

  11. Smells like FUD to me.

    The real deathtrap is the fatass syndrome that people will get by setting up all their devices in their house to remote switching so that they NEVER have to get up off the sofa.

  12. don’t buy the space heater danger scenario, they have fuses, and as was noted they are resistive loads. You can probably find something dangerous to do this to, I just don’t think space heaters are it.

  13. Uh, spaceheaters can be very dangerous and there are plenty of old ones still in service, i live in Wisconsin where there’s pretty much at least 1 space heater in every home. I’ve seen plenty still in service that had to be 30 years old that had no tilt sensor or thermostat. It’s also fairly common to see these things automated with light timers and remote switches like this. While a heating coil heats because of it’s resistance that doesn’t mean it’s not an inductive load, it’s both. When an inductive load is switched on there’s an inrush of current until the coil is saturated, then when you switch it off there’s a back emf voltage spike, which is why you put snubber diodes on inductive loads. By toggling this thing fast enough, continuously saturating and collapsing the magnetic field you will be pulling a higher current for a longer time, yes this could fry your heater or the relay, but the scary part is the house wiring; any circuit intended for an inductive load will have a delayed action fuse or circuit breaker which may not trigger with this type of behavior. Having worked as an electrician i’ve seen a lot of wiring jobs that were not up to code, it’s not uncommon to see wiring that hasn’t been updated since the ’30s or ’40s. So while it might not do anything with a new heater or a new house, this could easily cause an electrical fire in a real world situation.

    1. People have touched on this briefly but here goes:

      When a high load is switched often with a relay the relay will arc at every switching, this will burn out the relay contacts very fast. The thing is that you have no way of telling what happens to the contacts at this point since the relay was never meant to be used this way. The contacts may gain resistance creating even more heat.

      Also, if this thing uses an AC relay instead of a DC one, the coil itself could catch fire since an AC relay pulls more current during turn-on than in a steady on state. We actually achieved this on a couple of larger AC industrial 3 phase 25A relays when a controller failed and started switching at 50Hz (mains frequency), the coils heated and carbonized all the insulation, thankfully the relay coil supply was fused so the fuses just blew.

    2. From what I read on the Internet, American house wiring is really in a state. Don’t you have regulations and stuff? I’m pretty sure in most countries you can’t sell a house with not-up-to-scratch wiring. Certainly you can’t rent one.

      1. You can’t sell a new house or do major electrical renovations without a permit which would mean bringing it up to code, but as long as nothing is radially changed original wiring is still legal (as long as it was code at the time it was installed) The other thing is just because there is code doesn’t mean it was followed or inspected, there are unscrupulous renovation contractors that will not bother taking out a permit do dangerous work and cover it up. And for smaller things like replacing fixtures and outlets the homeowner (even if they are not a quallified electrician) is allowed to do it. True when you buy a house you typically have an inspection done before you buy it, but that doesn’t mean they’re going to notice that an outlet is spliced onto the wires for a floor heater hidden behind a wall, or that one outlet has no ground connection.

  14. i think the biggest danger is when you purchase a CERTIFIED and UL APPROUVED space heater and open it up to find a 6A on/off switch when the dam thing says it takes over 10A !!!!!!!!

    IF I DESIGNED AND SOLD THIS I WOULD GO TO JAIL FOR ATTEMPTED MANSLAUGHTER, YES, ***_MURDER!_***

    no names, no brands, no dates, no locations
    why no comment?
    because i fixed (under duress) the switch by opening it and sanding it.
    i have no idea if the owner is still alive…
    if i could meet him, i’d tell him to sue the company that designed and sold that ILLEGALly certified appliance

    PS: the faceplate where switch was attached was white-plastic-turned-black-and-burnt-plastic.

    1. “CERTIFIED and UL APPROVED”

      Or so the sticker says… I’m sure you can buy ‘em on big rolls for next to nothing in Shenzhen. Doesn’t mean it really is certified. It’s about inspection and enforcement. Report the shop that sells such stuff. That’s where the responsibility is on your side of the ocean, since the Chinese don’t appear to have any sort of safety / reliability / non-stealing requirements for their industry.

  15. I am not justifying the headline, but darn this is a great discussion! Best reading I’ve done all night. I just wish I felt more confident that the maker of the device had been through all of the scenarios debated here.

  16. I’m gonna call bullshit on this, its another HAD sensational headline.

    If the thing was connected to a space heater and switched on and off rapidly, the space heater would be mostly on, or mostly off depending on the mark space ratio, either way, its not full on or more on than on.

    The device uses a relay which is safer than using a triac, because rapid switching of the relay will cause mechanical failure, triacs have a nasty failure mode of short circuit.

    uPnP hack, not a surprise, device discovery for an item like this is moronic, but then consider that the target market won’t know about or care about IP addresses, they just want to plug in and go.

    @hsiboy

    1. The mechanical failure of a relay can be nasty too you know, the contacts can fuse together in a basic failure or so much ‘dirt’ can form on the contacts that it creates a resistance that then makes it overheat by the arcing.

    1. There was a time when TV’s in standby had an issue which caused them to catch fire, ever since those incidents people have been clinging to it as a universal scare for many decades. It’s a bit like a hindenburg effect. And although silly now it did have an original basis in real life.

  17. 3000 watts at 120 volts is going to be 25 amps

    http://www.the12volt.com/ohm/ohmslawcalculators.asp

    that is dc so i think there is a 20% difference factor for ac but still it would exceed a 15 amp fuse or breaker and a heater that pulls that much would trip the breakers or fuses of the house.

    an electric kettle should have 2 thermal protections.

    1. a thermal switch that looks like

    1a. http://atechfabrication.com/images/htpc/products/Temperature_Switch_057_320.jpg

    most common style used in microwave ovens, furnaces

    1b. http://images-en.busytrade.com/106244400/Sell-Thermal-Breaker-For-Hair-Dryer-Or-Heater-heating-Cord.jpg

    mostly used in hair dryers but could be found in some toasters

    self resets after it cools usually used as thermostat

    2. thermal fuse looks like 1 of the following

    2a. http://media.digikey.com/photos/Cantherm%20Photos/SDJ1%20DF%20SERIES.jpg

    used in hair dryers, overhead projectors, copiers and laser printers in the fixer roller

    2b. http://i01.i.aliimg.com/photo/v0/200326709/Normal_open_type_thermal_switch_w_UL.jpg

    mostly used in electric motors but can be found in some electronics like high power amplifiers and power supplies.

    2c. http://i00.i.aliimg.com/photo/v0/253785036/17AMJ_Series_Thermal_Protector_thermostat_THERMAL_SWITCH.jpg

    these too are found in electric motors

    and are non resetting so they have to be replaced

    any maker that does not include one in their product is a tight wad because they are fairly cheap in the large bulk that is needed.

    also if the ul and fcc can approve a product with under rated parts means that they are rushing the process

  18. another idea maybe.

    if you are not up to cracking open the wemo to modify it and you intend to power anything that contains a switching power supply you can build (IN THEORY) you could build a box that takes the ac and converts to dc then using a bunch of capacitors in parallel to act as a battery to power the device for a few seconds after bring unplugged.

    that way if some black hat hacker or script kiddy decides to write a virus that can cause the unit to cycle on and off the external ups like box would keep the device powered so it would not be flickered.

    if you are up top opening up the wemo you could add a large capacitor across the relay to keep the relay closed for a couple seconds after the off command is sent.

    the warnings:

    1. this involves potential exposure to high voltage electric and should be done with care and make sure all capacitors are fully discharged before handling the board.

    if you are not up to the risk of exposure to electric you can just buy a ups at and computer shop and it will provide a few minutes of ac power at maybe 2 amps for small loads like a lamp or fan or an older tv that pre dates the remote.

    2. i dont know what the unit looks like inside so it may not have enough room internally so it may have to be gutted and a new enclosure and wiring may be needed.

  19. Once current through an inductive load starts flowing (anything with a coil such as a motor), it doesn’t want to stop flowing. If you break the circuit at a switch (such as the relay), the current is still being “pushed” by the inductor, and the voltage potential across the switch can jump 100x. This is how an automobile ignition works – when 12v current through a coil is interrupted, a voltage of over 20,000 volts is generated sufficient to create a spark (or shock you pretty good, if you’ve ever pulled off a spark plug wire when a car is running). In this device, the spark would occur within the relay, creating an arc between the contacts, or it would find any easier path, such as between traces on the circuit board or between components in the plugged-in device. By switching at 120hz, you could turn the relay into a rectifier, a mini-welding torch; you could melt the relay contacts, the relay housing, and set something on fire.

  20. If any of you were paying attention in the beginning, the whole purpose for this hack was to “cause” damage by rapidly cycling the switching relay within the WEMO device therefore rapidly cycling the object device itself The WEMO is a very well constructed switching device with a high 15A rating and I believe a 1500watt capacity. Given these facts, this switch would easily serve to power and high resistance unit toaster, compressor, etc, and switch it on and off at hundreds of times per second thus causing a severe overload situation on the structural wiring, NOT the switch. Scroll back all the way and learn..

  21. “Death trap”? Where is the fire then? And where are the victims? Why make a video that does nothing except proving your point wrong?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s