How does power spikes work?
November 3, 2011 11:45 PM Subscribe
Calling all electricians: please explain to me how power surges work.
All right, I sort of understand the basic principles. Sudden boost of electrical charge somewhere across the power lines causes an increase in the potential, increasing current to the wall sockets.
For something like a lightening strike, this makes sense to me. Sudden boost in energy (current/voltage), transmit across power lines (blowing through all failsafes), blows up everything in their way. Okay.
But the more mundane examples escape me. E.g. power outages. The electricity reaching my house during a power outage is zero, so yes, there is a difference in electrical potential compared to electricity from the source/plant...but surely within the complex power grid there are transformers and capacitors and other -er's that changes the tons of energy coming from the power plant into the usual household voltage and current. Is it simply those failsafes are overloaded by a power outage, and if so, why? If the power plant had stopped and then started (making a humongous difference in potential), or as the lightning strike example above, I can see that sheer amount of energy being problematic. A short duration of power outage should be within the electrical grid's capabilities, right? Why would it cause an issue with my electronics?
An example from today: an electrician from the power company was fiddling with my meter and shut off power to my house for 10-20 seconds. My phone was plugged into the wall. Does 0 energy reaching my phone, and then the sudden availability of electricity again, create a power differential as to damage it? (My phone actually has had spotty internet but okay voice service for about 20 minutes in the morning after the 10s shutoff, was fine in the afternoon, and now is erratic again, but I think that's my provider). I'd assume the power grid can handle this, but power outages are frequently warned for causing spikes. And my brain is trying to relate this to on/off of electronics in general somehow. So, if I turn my computer off, 0 energy is reaching my computer...so the "surge" of electricity when I turn it (or any electronic) on might be enough to damage it? That makes no sense, but it sounds kind of (faultily) logical from the above train of thought...
Also, internet googling tells me that just heavy use of the electrical circuit (heavy machinary, appliances that use a lot of energy, etc) causes dips and recoveries in the electricity available to the house, acting like another form of surges, and can slowly damage the electric wiring in the house. But if my hairdryer works with a typical socket at 110V, and less energy is reaching my hairdryer because the dryer (or air con, or elevator, or whatever) is on, then once the dryer is off the voltage flow returns to its 110V maximum which should be what it's designed for, so how these 'mini surges" induce (gradual) damage in the system is also a mystery to me.
I'm actually embarrassed to ask this because I somehow managed to get decent to excellent grades in physics classes all my life, but obviously the wisdom imparted was not wholly understood. Google gave me the general principles listed in the first paragraph, but doesn't further my understanding. Enlighten me, Mefites!
All right, I sort of understand the basic principles. Sudden boost of electrical charge somewhere across the power lines causes an increase in the potential, increasing current to the wall sockets.
For something like a lightening strike, this makes sense to me. Sudden boost in energy (current/voltage), transmit across power lines (blowing through all failsafes), blows up everything in their way. Okay.
But the more mundane examples escape me. E.g. power outages. The electricity reaching my house during a power outage is zero, so yes, there is a difference in electrical potential compared to electricity from the source/plant...but surely within the complex power grid there are transformers and capacitors and other -er's that changes the tons of energy coming from the power plant into the usual household voltage and current. Is it simply those failsafes are overloaded by a power outage, and if so, why? If the power plant had stopped and then started (making a humongous difference in potential), or as the lightning strike example above, I can see that sheer amount of energy being problematic. A short duration of power outage should be within the electrical grid's capabilities, right? Why would it cause an issue with my electronics?
An example from today: an electrician from the power company was fiddling with my meter and shut off power to my house for 10-20 seconds. My phone was plugged into the wall. Does 0 energy reaching my phone, and then the sudden availability of electricity again, create a power differential as to damage it? (My phone actually has had spotty internet but okay voice service for about 20 minutes in the morning after the 10s shutoff, was fine in the afternoon, and now is erratic again, but I think that's my provider). I'd assume the power grid can handle this, but power outages are frequently warned for causing spikes. And my brain is trying to relate this to on/off of electronics in general somehow. So, if I turn my computer off, 0 energy is reaching my computer...so the "surge" of electricity when I turn it (or any electronic) on might be enough to damage it? That makes no sense, but it sounds kind of (faultily) logical from the above train of thought...
Also, internet googling tells me that just heavy use of the electrical circuit (heavy machinary, appliances that use a lot of energy, etc) causes dips and recoveries in the electricity available to the house, acting like another form of surges, and can slowly damage the electric wiring in the house. But if my hairdryer works with a typical socket at 110V, and less energy is reaching my hairdryer because the dryer (or air con, or elevator, or whatever) is on, then once the dryer is off the voltage flow returns to its 110V maximum which should be what it's designed for, so how these 'mini surges" induce (gradual) damage in the system is also a mystery to me.
I'm actually embarrassed to ask this because I somehow managed to get decent to excellent grades in physics classes all my life, but obviously the wisdom imparted was not wholly understood. Google gave me the general principles listed in the first paragraph, but doesn't further my understanding. Enlighten me, Mefites!
Oh, and the reason it's worth protecting expensive electronics with a proper power conditioner (which will often contain a big fat inductor in series with the supply - you can spot these because all that copper winding and iron core is heavy) is that the simple surge suppression offered by the MOVs built into the typical switching power supply or surge suppressor power strip becomes less effective over time; every time a MOV soaks up a spike it suffers cumulative internal damage.
posted by flabdablet at 12:42 AM on November 4, 2011
posted by flabdablet at 12:42 AM on November 4, 2011
The problem with outages (and the damage that comes from powering back up after an outage) happens because of the various inductances and capacitances of the components of the grid, the inrush of current in the system can develop voltage spikes. It is sort of like water hammer. For example, if you have ever been in a building where the lights get brighter when the air conditioner starts, you are seeing that happen. When the entire neighborhood comes back online, those thousands of amps getting drawn through the system can cause trouble.
Unlike simple DC circuits, AC power is more like radio and sound waves. Spikes are like pops on records and the crackle you get on AM radio when there is lightning nearby. The ones that damage your equipment are like that, but stronger.
posted by gjc at 6:49 AM on November 4, 2011
Unlike simple DC circuits, AC power is more like radio and sound waves. Spikes are like pops on records and the crackle you get on AM radio when there is lightning nearby. The ones that damage your equipment are like that, but stronger.
posted by gjc at 6:49 AM on November 4, 2011
Another problem with brief outages is known a 'brown out'. Devices that run off DC usually store some charge in capacitors, and at power off, instead of the voltage dropping to zero instantly, it will slowly drop to zero over several seconds. You can see this in some devices when the power LED stays lit for a few seconds after you've powered it off.
If you reapply power before it has reached zero, the internal state of some devices can get messed up - into states that would be impossible in normal operation, unless the manufacturer has added special circuitry to prevent this. Powering the device off completely should cure this.
posted by HiroProtagonist at 8:00 PM on November 6, 2011
If you reapply power before it has reached zero, the internal state of some devices can get messed up - into states that would be impossible in normal operation, unless the manufacturer has added special circuitry to prevent this. Powering the device off completely should cure this.
posted by HiroProtagonist at 8:00 PM on November 6, 2011
« Older I made out with a girl at a party, but I really... | And way to include in PHP across public_htmls? Newer »
This thread is closed to new comments.
No. The grid has no storage at all. If the piece of it that supplies you gets disconnected from the generators that feed it, your house gets an outage. The only way your electronics will keep running through this is if they themselves contain enough energy storage to keep themselves alive until the power comes back on (which will often happen fairly quickly as substations automatically switch over to an alternate feed). For example, a laptop with a battery in it will survive an hours-long power outage, while a typical PC power supply contains smoothing capacitors that might keep it running through an outage that lasts half a second.
Does 0 energy reaching my phone, and then the sudden availability of electricity again, create a power differential as to damage it?
No more than switching it off and on at the wall outlet would do. That said, the most common time for modern switching power supplies to fail is at switch-on, because of the fairly high inrush currents that occur as the aforementioned smoothing capacitors try to go from no charge to full charge instantaneously.
power outages are frequently warned for causing spikes
Inductors store energy in the form of a magnetic field, and want to keep the current flowing through them constant. If you try to stop current flowing in an inductor by suddenly disconnecting it, the current will in fact continue to flow through the inductor until the stored magnetic energy is all used up. There is no theoretical upper limit to the voltage that a suddenly-disconnected inductor can generate across its terminals in order to maintain this current flow. In practice, the height of an inductive voltage spike is limited by the fact that the current will be soaked up by other things connected to the inductor or ultimately by its own self-capacitance.
If there are inductive devices such as motors connected to a piece of the grid that suffers disconnection at the instant when current happens to be flowing in those devices, the resultant dumping of stored magnetic energy will manifest as a voltage spike on that piece of the grid. If there are lots of motors, that spike can be quite tall and fat.
The fact that the very wiring that connects your house to everything else has its own inductance can act to damp down the worst effects of these; inductively-generated spikes generally cause more damage close to the inductors concerned than far from it. The inductance of the transformer inside a typical old-style wall-wart power supply will also reduce the potency of spikes arriving via the grid. But switching power supplies, such as you will find inside most new electronic appliance, have semiconductor devices connected straight to the grid and spikes can literally punch holes in these, which is why they need protection from surge suppressor devices such as MOVs.
another form of surges, and can slowly damage the electric wiring in the house
Damage to electric wiring can happen two ways: damage to the conductors (especially at points of locally higher resistance such as terminal blocks and wire nuts) due to overload and consequent heating, or damage to the insulation caused by heating or by pinhole punctures generated by very tall voltage spikes. Neither of these should happen in properly sized and rated house wiring; your home's fuses or circuit breakers should disconnect an overloaded circuit well before the point where the wiring starts to heat to anything even approaching dangerous levels, and there should be a lightning arrestor (basically just a spark gap) that clips the top off any incoming voltage spike well below your insulation's maximum voltage rating. You'd need to be doing some pretty savage local switching of some pretty heavy inductive loads to induce spike damage in the wiring around those loads.
posted by flabdablet at 12:36 AM on November 4, 2011 [4 favorites]