18

I don't mean damage caused to the socket and prongs, but only to the appliance itself.

  • 6
    You should ask this question on the electrical engineering stack exchange ( electronics.stackexchange.com) to get more than anecdotes. Unplugging a device that is drawing current through an inductor can cause transient voltage surges. Devices are usually protected against this. The protection is often not designed to be relied on. Fun story: caeai.com/blog/what-was-not-lurking-heart-my-washing-machine – trognanders Mar 20 at 18:55
  • 1
    Without knowing the specific appliance, the answer to your question is all of the following: yes, no, maybe, only in extreme cases – Mast Mar 21 at 11:51
  • @Mast Exactly! Someone should at least post a link to this question in electronics.stackexchange.com chat. Perhaps it should even be migrated there. – Mike Waters Mar 21 at 19:13
  • 1
    @MikeWaters With the current lack of detail, it's even more off-topic there than it is here. Especially since they outlawed general advice on consumer electronics. – Mast Mar 21 at 20:52
  • In general you need to be careful about unplugging and then immediately replugging (into a live circuit). Wait 10 seconds or so (or several seconds after all lights have gone out and motors stopped turning) before replugging. You also need to observe similar delays when turning circuit breakers off and then back on. – Hot Licks Mar 22 at 0:08
21

Another example of needing to be careful is in situations where heat removal is involved (such as internally driven forced-air cooling of a video projector bulb). In these cases, to prevent short-term (or longer-term) damage, you need to wait until the device has had a chance to properly cool itself down internally.

  • For which appliances does it apply? I guess TVs don't have fans, and desktop computers, on the other hand, are usually well aired. So maybe mobile computers are a good example? – OMGsh Mar 20 at 23:00
  • 12
    3D printers. They have a fan running on the top part of the hot end because if the heat transfers up the print head it will cause the plastic to melt and cause a blockage. – Qwertie Mar 21 at 2:40
21

Typically, an appliance will not be damaged if it is currently ON and then you unplug its power cord. If you plug it back in the device will simply resume operation as if it were ON.

I say typically because there are some cases where this may not be the case. There are many devices such as computers and some other electronic devices that almost always recommend that you interact with the device to gracefully shut it down before you think of yanking the power cord out of the wall. This graceful shutdown process ensures that the device has finished all critical internal operations before losing power. This could include operations such as closing open files or saving important system state to a storage memory.

With many of today's appliances having more and more smart electronics, it can be more of a concern that you not arbitrarily pull power cords. A well-designed appliance from a reputable manufacturer will have made provisions in the electronics design and software so that the device does not get bricked when the power is suddenly lost. On the other hand, as more designs are done by less experienced people, this possibility of damaging operation of an appliance by the sudden loss of power is more likely than ever.

To make my case, let me comment that the ever more popular Raspberry Pi (rPi) microcontroller board is leading many people to think of embedding devices like that into appliances and other gadgets. It is a known fact that removing power from an rPi board can brick the boot media on its micro SD card and keeping it from being able to reboot when power is restored.

  • 3
    I'd also be careful about unplugging something with a motor that is running and then plugging it back in before the motor comes to a stop. – JimmyJames Mar 20 at 15:15
  • @JimmyJames Why do you think that? – Graham Mar 20 at 15:24
  • @Graham Honestly: because my father told me not to do that to my fridge. I should note that he spent over 4 decades engineering refrigeration equipment. I can ask for the details if you want. – JimmyJames Mar 20 at 15:34
  • @Graham If I had to guess, it might be related to the motor having the most torque at 0 RPMs. – JimmyJames Mar 20 at 16:13
  • @JimmyJames Possibly an issue on a fridge compressor, where if you've moved it around then you want to leave the coolant to settle so that any bubbles disappear. It's not something which applies to other electric motors though. – Graham Mar 20 at 17:01
4

tl,dr; it is probably OK for very low power devices. For high power devices you can accelerate wearing on both the plug and the wall socket.

In most devices the power switch does the same thing you do when you pull the plug from the wall: it interrupts the circuit. There are other devices that are always "on", and the power switch just puts it into hibernation mode.

For the second kind there is a bit of electric noise when you unplug it but modern, well designed electronic circuits can withstand this (at least within their nominal lifetime). As some people have mentioned, some sophisticated devices have computers inside, and some of them are plugged into the Internet. Although the physical device itself is not likely to be damaged, if the device is receiving a software update through the Internet and you unplug it before the upgrade is completed you can "brick" it (it will not start because the software will be incomplete). If your device does not receive "OVA" (over the air) updates, you should be safe - of course do not unplug it if the screen says something like "Please do not unplug until the upgrade is completed".

Simpler devices like a heater or a vacuum cleaner will not be damaged, but depending on the currents involved doing this often can accelerate wearing of both the plug and the wall socket. For higher currents the switches are designed using "arc suppression" techniques in order to withstand the stress of cutting the circuit. Look at this picture of pristine contacts from a brand new switch and the same contacts after 100,000 on/off cycles:

enter image description here

Where the damage comes from? When you cut the circuit, the electrons will try to keep flowing and will fly through the air causing an "electric arc" (if the room is dark enough you will see the spark). The temperature of the resulting electric arc is very high (tens of thousands of degrees), causing the metal on the contact surfaces to melt, pool and migrate with the current. The high temperature of the arc cracks the surrounding gas molecules creating ozone, carbon monoxide, and other compounds. The arc energy slowly destroys the contact metal, causing some material to escape into the air as fine particulate matter. This very activity causes the material in the contacts to degrade quickly, resulting in device failure. Often the switch contacts are coated with some noble metal like gold or silver.

Now look at your wall plug: it is not designed to suppress the electric arc, and it is not coated in a special material. It will fail way sooner than a well designed switch. Because of course wall plugs are not designed to be used as an electric circuit switch!

How dangerous is that? A good quality wall plug can withstand a lot of abuse but once the contacts lose the outer protective layer they can decay very fast causing the plug to heat making plastic parts slowly burn and become conductive eventually causing a short circuit. The circuit breaker should be triggered by the excess current and shut down before the house is on fire so under ideal conditions it is not that dangerous. That said, my advice is to avoid pulling the plug with the appliance on for anything above 1200 w (10 ampere at 120 v).

There are outdated (or plainly illegal) electric wiring, old/defective circuit breakers, cheap low quality plugs and so on. If your plug is humming, flickering, heating or smelling smoke replace it immediately.

Source: I'm a certified industrial electrician and electronic circuit designer.

  • Actually, most wall sockets and plugs that are made to a decent standard are rated to be used as a disconnecting means under load ("horsepower rated") – ThreePhaseEel Mar 20 at 23:37
  • 4
    I would not recommend disconnecting anything above 1200 watt (10 ampere at 120 volts) by pulling the power chord. Yeah, the power chord can survive this hundreds of times but you are still shortening the MTBF by 2 or 3 orders of magnitude. – Paulo Scardine Mar 21 at 1:12
4

Some electronic devices with on-off switches can be damaged when being plugged in. The arcing can damage static sensitive devices. Many electronically controlled devices have debounce circuits because even pressing a switch causes spikes, but if plugged in with the switch on the spikes are many times higher as the electronics power up so at least when plugging things in, it will reduce the possibility of damage if the power switch is turned off.

2

Practically none will be damaged by an unexpected power failure. Customers get cranky about that sort of thing.

Anything software driven with non-volatile storage may have difficulties if it was either in the middle of some critical operation (rare) or you frob the power on and off multiple times without giving it a chance to repair its file system.

A bunch may require some judiciousness or care when plugging them back in. For example: Dishwashers and washing machines will still have water in them, and may require some poking at the controls to convince them to drain rather than attempt to fill a second time and overflow. Also I happen to have a freezer on which the compressor motor is not strong enough to start if there's already pressure in the system and it will blow a fuse if you try, so if it's interrupted mid-cycle you have to let it sit for ten minutes or so to allow the refrigerant pressure to bleed down before you restart it.

  • To a very good approximation, pulling the plug is equivalent to a mains power failure, which happens occasionally (or more than occasionally, depending on your location). Appliances are okay with that. For example, a PC won't usually power up by itself once power is restored, but rather must be manually restarted. – Phil Freedenberg Mar 21 at 20:41
  • If your dishwasher fills by running the water for a set time rather than by using a float valve, it's already an overflow risk, whether you unplug it mid-fill or not. – Mark Mar 21 at 22:15
  • @PhilFreedenberg Most modern PCs are configurable about whether they should start back up after a power failure (default is often off to prevent data corruption if the power comes and goes several times.) Older PCs had manual power switches that stayed where you left them. – Perkins Mar 22 at 6:57
  • @Mark Indeed. And yet such designs are surprisingly common because they're cheaper to design and build. Fortunately I've only ever had trouble with them underfilling due to low water pressure rather than overfilling due to power interruption or anything like that. It is one reason why I prefer appliances with mechanical controls though, they're easier to adjust to compensate for non-standard circumstances than electronic ones (and they don't keel over dead in the first big thunderstorm.) – Perkins Mar 22 at 7:02
1

It depends on the device.

To take Michael's example of the Raspberry Pi, Flash memory (which works like RAM) is destructive on reads, so to read data you have read it out and then write it back. If you pull the power when it's in use you might be mid-cycle and lose data. Many SSD(flash) drives for more expensive computers have a capacitor that keeps the power on just long enough to complete the cycle if the power goes out.

Appliances don't generally have this issue because they tend to use ROM chips instead. A power failure doesn't matter because the data on the chips isn't volatile. If they are using Flash memory anywhere, it's probably not for mission-critical parts (an appliance is something you're likely to repair and warranty repairs are expensive to companies).

  • 8
    Are you thinking of FRAM instead of flash memory? Standard flash memory doesn't have destructive reads. – Segfault Mar 20 at 14:05
  • Practically nobody uses ROM or PROM any more. The days when the cost of an EEPROM were higher than the cost of technician time to physically swap chips for firmware updates are long past. – Perkins Mar 20 at 23:48
  • 1
    @Perkins Does this make any difference to the end user? Typically once a device is shipped to the customer it may as well be ROM because it won't be written to again. – Qwertie Mar 21 at 2:44
  • @Qwertie as devices get smarter and are more commonly connected to the Internet the importance of upgrading their software to patch security vulnerabilities and malfunctions increases. That's why most appliances use EEPROMs instead of just PROMs or ROMs. The fact that the typical consumer is too lazy to take care of their devices is a different issue. – Perkins Mar 22 at 7:07
1

Aside from some wear on the plug and socket that you specifically mentioned that you aren't asking about, there is no harm in pulling the plug on something that is running unless it's a Windows 7 or later computer that is running Windows Update.

Everything is made to handle an unexpected power outage unless it's from Microsoft. One rare exception might be something like a machine tool that has a small light bulb and a large motor that is controlled by a switch. Pulling the plug while the switch is still on and the motor is still running under load may cause an inductive spike created by the motor to get absorbed by the light bulb and burn it out. This situation also applies to any large inductive load that is connected in parallel with something more fragile. When a circuit breaker or switch disconnects everything together the inductive spike created by the inductive load will be exposed to the more fragile device, without the original power source connected to absorb the current. This is a common situation and pretty much everything is designed to handle these types of Voltage spikes, as long as the size of the inductive load is within reason. You'll often find MOVs (surge suppressors) built in to power supplies.

A more interesting version of your question might be: What happens if I unplug a power strip with a variety of appliances connected to it, such as a large operating non inverter type microwave oven and a cheap imported USB power supply?

  • Really interesting! Do you have an example of an appliance of the first type? And concerning your question - in case all the appliances on the power strip are turned off (but still drawing some power as e.g. TVs usually do) the load is so small that no damage is possible? – OMGsh Mar 21 at 11:44
  • Hard disks have a cache. Writing to a hard disk will write to that cache first. If the power shuts down, that cache is lost. It's nothing to do with microsoft - all OS's are exposed to data corruption from this. – UKMonkey Mar 21 at 14:47
  • @UKMonkey fun fact - you can disable the cache on USB sticks. This is now built into Windows as an option, and lets you pull out USB sticks without "safely removing" it first. In fact, it might now be default behaviour. With that said, I would still use the safe removal option - all it takes is to forget one time when using another machine... – Baldrickk Mar 21 at 16:50
  • @Baldrickk You can indeed! Hence my comment about "hard disk" rather than "USB device"... infact you'll find that the cache for a USB device is more often than not (but not always) in the OS anyway; and not part of the hardware. – UKMonkey Mar 21 at 17:07
  • @UKMonkey yep, and I think that Windows 10 defaults to having the cache off, which is safer, but slower. I just thought that it was an interesting addition to the conversation. I definitely know the difference between the devices. ;) – Baldrickk Mar 21 at 17:11
1

Another issue not yet mentioned is that some devices have series capacitors without bleeder resistors. In some cases, if such a device is unplugged without switching it off first, an internal capacitor connected to the AC line might remain charged for awhile, creating two related hazards. First of all, touching the exposed prongs of the AC plug when the capacitor is charged may deliver a rather nasty shock. Secondly, if the line phase when the device is next plugged in doesn't match the line phase when it was disconnected, it may cause a nasty current spike the next time the device is plugged in.

Properly-designed devices which use series capacitors of any significant size should include bleeder resistors to ensure that any stored energy is dissipated within a few seconds at most. Bleeder resistors would waste power continuously while a device is plugged in, and a bleeder resistor which discharges 63% of the stored energy in one second would waste twice as much power as one that would take two seconds, so there are advantages to having things discharge somewhat slowly. Thus, even if a device includes a bleeder resistor, unplugging the device and immediately touching the plug could yield a nasty shock. Generally, however, such a shock hazard would only exist while the switch is in the "on" position. Turning off the device before unplugging it would make the plug safe to touch, at least as long as the switch remains off (turning on the switch sufficiently soon after unplugging the device could cause the plug to yield a nasty shock).

Note also that while any safely-designed device that stores significant energy should include a bleeder resistor, some cheap import devices that aren't subjected to any kind of regulatory inspection or testing might save a fraction of a penny by omitting the bleed resistor. Touching the exposed prongs of such a device could yield a nasty shock even if it's been unplugged for hours.

  • How nasty? Could you estimate it by numbers for an appliance of up to 2000W? Only once did I feet such a shock from prongs - in that case of a vacuum cleaner, and after I've turned it off (but maybe with a still slowing motor). It didn't feel more then a bit annoying. – OMGsh Mar 21 at 21:25
  • @OMGsh: The kinds of devices which use capacitive series droppers tend to be relatively low power devices which need to cheaply convert AC120 down to a low voltage for internal use. The stored energy from a few microfarads charged to 100+ volts isn't terribly likely to be directly injurious, but would be severely unpleasant and may cause a startle reaction which could cause one to injure oneself indirectly. – supercat Mar 21 at 22:49
0

The answers so far state computers as an example; but they're very poor examples. The damage done to the computer hardware is nothing if you rip the power cord out. They don't run at high currents, and the PSU tends to be very well designed to ensure that the output voltage is sane.

However data caches will not be flushed, and data consistency may not be promised. You could end up with a perfectly good computer - which needs a re-installed OS; and again with the pi example mentioned in an answer; the hardware itself is fine.

However, if you have something that has 100A+ running through it, and you kill the power - the hardware itself can be destroyed due to the very high amount of stored energy in anything with inductance that will suddenly be going in the wrong direction. It's not that you can't protect against this; but that the cost of doing so can be prohibitive.

The things that will have this much current though them however will tend to be something you won't have in your house; but in an industrial setting.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.