View source for Electronics notes/Volts, amps, energy, power

{{electronics notes}}


=Some theory=

{{stub}}


==Nature of energy, charge, voltage, current==

<!--
There are two main practical forms of energy we deal with: electrostatic and electromagnetic.


Electrons are particles with a fixed charge, and when you push them though conductors and other components, you are moving electrostatic energy.

They are a useful medium to push around, so that work can get done elsewhere.

The electrons lose some of that energy you pushed in.
Not their fixed charge, but let's call it their speed, due to let's call it friction.



At the same time, that act of ''moving'' electrostatic charges around induces electromagnetic (EM) effects.

It's a different form of energy.
EM itself is considered photons, but this gets into quantum physics, so in most cases it's a lot easier to think of EM as "becomes a magnetic field, which will probably translate back to electrostatic nearby".



In a pragmatic "I am looking at a circuit board" view, most is electrostatic, with relatively little electromagnetic. 

That said, all circuits have some minor EM induction all over the place,
because current running around any loop (so basically any circuit) acts as an inductor.

In most things we make that inductor is so inefficient that you don't need to think about it, not until...
: you drive serious current (...implicitly through a loop)
: you sense something with very high impedance
:: because higher impedance implies being more sensitive to smaller currents
: you have to meet EM interference certifications (i.e. are making products for sale) 
: have long conductors
: have noise issues (because you or something near you are doing one of the above)
:: (noise will generally be weak, and at relatively low frequencies, kHz to MHzes at most{{verify}})
: you specifically add inductors, transformers, or other add electromagnetic components, for specific EM behaviour.



So you can see static electricity as the simpler part of EM energy
While energy-wise electrostatic and EM is more or less created equal,
tend to only mention EM when things get more complex.

-->

===On analogies===
<!--

: '''On analogies:''' 

There are various analogies with physical kinetics - water, marbles, and whatnot.

These can be useful bootstrappers for some concepts, and some good at pointing out specific behaviour in isolated examples.

For example, [https://www.youtube.com/watch?v=8f3l8KaVcBg this water gate analogy for an NPN transistors] is nice in a functional introduction to transistors, and clearer than long stories about silicon doping.


But '''all analogies break down at some point''' and start adding to the confusion.

Also they don't model EM at all.






But let's try anyway.

The common analogy is water in a hydraulic system, in which you can see...

* '''Charge''' is the thing that lets you do work with something
: 1 Coulomb is just a (large) number of electrons
:: (though mostly because an electron's charge is a fixed, at ~1.6E-19 coulombs)
: in the analogy it could be mass or volume of water
:: though it'd come from mass per molecule, and mass makes potential/kinetic energy analogies simpler, volume is easier to work with. We can consider mass and volume proportional (because incompressible) and ignore this well enough.

: (how it is confusing: relating mass and/or volume directly with energy doesn't really make things clearer)

* Energy is the amount of work put into that charge
: e.g. how much energy is used to move charge. Consider that when electrons flow though a resistor, electrons do not go missing, they are not converted into heat, much of their ''energy'' is, and the electrons carry less energy when they leave that resistor.
: in the analogy, e.g. how much the mass is pushed/moved around
: (how it is confusing: nature of potential/kinetic energy somewhat fuzzy)

* '''Voltage''' is pressure, behind e.g. a pipe 
: e.g. from a balloon filled with water, or more pressure coming from a higher tank of water
: for the same sized everything-else, lower pressure will mean less movement 
: Volts are 1 joule of work per 1 coulomb of charge  (also 1 Watt/Amp)
: in the analogy e.g. in pascals (N / m<sup>2</sup>)
: where it is helpful
:: pressure could be seen as a difference in forces
:: exists before anything flows, and will only ''do'' anything when it is allowed to flow
: how it is confusing: 
:: some sources of pressure are more fitting, others just entangle more concepts (e.g. a higher tank of water works only because of gravity; a balloon only because of air pressure relative to atmosphere
:: saying it's a pump[http://hyperphysics.phy-astr.gsu.edu/hbase/electric/watcir.html] attempts to sidestep the issue but will be extra confusing to those who consider ''that'' part of the analogy in any way

* '''Current''' is how much charge flows past  
: 1 Ampere is 6.241E18 electrons per second
: in the analogy it could be mass or liters per second
: how it's helpful
:: helps illustrate that voltage/pressure can exist, without any current/flow using it
:: might help explain voltage drop moderately well
: how it is confusing: 
:: it's easy to conflate how ''much'' with how ''fast'' (you do not want to get into speed or how much sense diameter then does or doesn't make in regards to e.g. Kirchoff's current law. I've seen serious diagrams using low water-pipe diameter to indicate resistance but this is ''misleading'' at best)
:: doesn't help explain energy loss very well{{verify}}

* Power
: e.g. how fast you can fill a bucket (which depends on both pressure and flow, and how those relate)

: how it's confusing
: pressure and flow seem more entangled than voltage and current - which makes a concept that ''combines'' them extra fragile (and easily confusing)


And there is also some sense of work [https://en.wikipedia.org/wiki/Electric_potential_energy]

* resistance is anything that takes/wastes energy while things glow past
: how it is confusing: 

* capacitance could be like a {{pressure tank cutout view|pressure tank}}, or a balloon stuck on an open pipe
: how it is confusing: actually this is a pretty nice one


Other problems with the water analogy

* does not much help answer questions like 'why and how does a transformer let you exchange volts for amps',
: particularly if you care about the losses.

* does not consider EM at all
: meaning inductors and transformers don't have an analogue at all

* torque and work get confusing

* water systems suggest leakage is another type of loss, but this is misleading at best.

* ground gets even more confusing than it already was

* leaks or open pipes in the physical analogy just don't map to the electrical side



[https://www.slideshare.net/PDiCEOThaneHeins3240/georgia-state-university-hyperphysics-how-newtonian-mechanics-is-incorrectly-being-applied-to-explain-the-laws-of-electricity-and-magnetism]




Say, the 'electrons push each because charge, like marbles because they are solid' is.. iffy. 
I still think it's great as an introductory mental model, but in any conductor, electrons are free and you're still only talking about the net movement of a ''lot'' of electrons.






A similar analogy is a a tube filled with a row of marbles (or water).
They can be used as a medium for energy: force will move it, and once they move, 
they exert it on the other end. 
It's mostly the same as the water analogy, but makes it a bit more visible that, like electron charge (and water), the medium itself are effecively incompressible things, and their effect moves much faster than the individual water/marbles/electrons.

(the speed of electrons is a more interesting question. For more detail, see [http://c2.com/cgi/wiki?SpeedOfElectrons c2: Speed of Electrons] or [http://en.wikipedia.org/wiki/Speed_of_electricity wikipedia: speed of electricity])}}. 



----



'''Conductors''' are things that allow charge to move easily. It's not that they have space for charge to move through easily, it's that they contain charge that can move easily. {{comment|(Consider that vacuum would be insulator. It's not so relevant whether we block charge, it's that it ''has'' no movable charge.)}}

Metals are common as conductors largely because their structure and their bonds mean they do this better than many other materials.

There ''are'' non-metallic conductors, and of course semi-conductors. 

{{comment|(While sort of advanced for an introduction, semi-conductors are interesting to contrast.
Basically, semiconductor means 'not a very good conductor, but not an insulator either'.
If charge in good conductors are analogous to dense fluids, then semiconductors act more like a light gas in that you need little energy to shove the charge out and make it a decent insulator. It is this behaviour that makes semiconductors useful. For example, this is basically all a FET does: it uses a metal plate to make the silicon near it an insulator.)}}




'''RC lowpass filters''' could be understood as a restricting pipe and a balloon.

The resistor working along would only be a current limiter.

The capacitor alone would filter at all, but assuming that the rest is low resistance, 
very little, probably only for very high frequencies, and because you don't know the resistance of the wiring (or its relation to the capacitor's equivalent resistance, also small), not in a very quantified way, and certainly not for specific frequencies you probably wanted to target.

It is the combination of R and C that makes it quantified how fast the capacitor will charge,
which corresponds to a particular frequency.

Which means rounding off fast changes, which is what a lowpass (=high-cut) filter is.






'''Current''' can be seen as flow of charge, as the count of electrons (/marbles) moving past a checkpoint.
{{comment|(In fact, 1 Ampere is equivalent to 6.25*10<sup>18</sup> electrons per second.  Put another way: 1 Ampere is 1 Coulomb per second, and an electron is 1.6*10<sup>-19</sup> Coulombs)}}


Current is only a result of an existing imbalance. It won't exist if stuff doesn't want to move.



'''Voltage''' is an overall property of a (closed) system: an potential for current to flow.

Regardless of whether that flow is happenning, and in fact, when flow happens, it often lower that potential,
because things are trying to balance out.
There are a lot of variations on how this interaction can work out in practice, so those details can come later.



Voltage and current are inseparable concepts.

Consider two adjacent lakes on a mountain, at different heights.
A conductor would be a pipe.
You also need a difference: If they're at the same level, there will not flow to each other.

Consider the extremes.
Maximum flow/current would be like holding one lake above the other and dropping it in at once.
No flow means you're blocking off the pipe with a tap. The pressure's still there.

In practice, you would probably wish to exploit the flow as efficiently and for as long as we can.

* ....no current/flow means no voltage/difference change, and that nothing interesting is happening.


* Current is something of a slave to voltage.
** Voltage is more of a fact, current more of a result.
** There won't be current without voltage (no reason to flow means no flow), but there can be voltage without current (when there is no connection for the flow to happen)



Making the analogy more complex - power and pressure.


You could say that 

* voltage is like the amount of force put behind the marbles/water - more of it means more flow

* ...and more flow usually needs a bigger pipe. 

I=V/R

needing a conductor that 

* a piece of thin pipe would be analogous to a resistor. Less water/current will flow.


Some things you can say:

* the more pressure, the more flow (current is proportional to voltage) -- assuming that pressure/voltage doesn't lower as a result of you using it

* Current can flow through an element without much power loss; this depends on how much energy is used in it. 
** For example, a valve/switch is open/closed and has little friction, so mostly just affects only the application of pressure/voltage
** 

* actual arriving force is the product of both (P=VA). If 




It has a radius, and water flowing through it has speed. 
Take either out and there is no system within something can happen. You can vary them 
Vary one and the 

Current is more of an implication of a setup than a direct property. For example, if the pipe radius suddenly lessens, this is better likened to higher resistance. The ''result'' of this is lower current.





'''Power''' is the rate of transfer.

Consider that the amount of marbles


This is proportional to both the voltage and current. 
This idealizes somewhat; in particular, it assumes an [[#ideal voltage source]]; it assumes voltage will 


 but this is arguably a little misleading - for many real-world power sources, voltage depends a little on current.


In simple cases, such as purely resistive loads, it is exactly voltage times current, but with inductance and capacitance and such, means that power at any one point may not be indicative.


Real power is an average of power that has been transferred - real work done over time, in Watts.

Reactive power is that which moves around because of being stored, and is in units of VAR - Volt-Amps reactive.

Apparent power is what you usually measure. In complex loads, it is a combination of reactive and apparent power.


Joule's various laws

P=I2R

P=V2/R 





The difference between '''energy and power''' is one that should probably be cleared up, 
as keeping it vague keeps some other things vague.

Arguably the most confusing things is that people juggle different units in different cases:
: energy is in joules  (regardless of the kind of work, though what other units get expressed as depends on where the work comes from. For electricity it's )
: power is often in watt, a joule/sec 

: with electricity, power is in watt, energy in watt-hour 


Okay, but wasn't current already an "energy moved" thing?







Some people prefer '''Potential difference''', a (near-)synonym for voltage, because it points out it is only ever a difference/imbalance of applied energy that creates the ''potential'' for flow (current).

Pick two points a circuit, and there likely is a potential difference between them.
Voltage ''at'' a point doesn't make sense, but people usually mean it as 'compare that point to ground', or some other reference that is obvious from context.

In the physical analogy, voltage is the pressure. And more of it means more current/flow if everything else is equal.


{{comment|('''Voltage''' refers to the amount of energy exerted to move 1 coulomb of charge, but pragmatically we usually don't care much about the electrons, only the Volts that come out the end (which is proportional to applied energy, but even that we don't need to care about much with wallplugs and batteries))}}







In electricity it's volts, in the water-and-lake analogy it would be pressure, but in both cases, it's not entirely clear what things you can and can't transfer into and out of the analogy. 
Many of the simpler statements about voltage and current don't actually clarify the two from the other.


A common analogy for this lies with water - involving mountainous lakes, a dam, vats and pipes, a submarine, or whatnot. The pressure would be analogous to the potential for flow, but doesn't ''mean'' there will be any flow until there is a connection/river/pipe/hole. Once there is, then water will flow for as long as there is potential (physical) energy between the two things.

Resistance in this analogy would be whatever limits the amount of flow/current - things like the size of the hole or pipe, how much a tap squeezes down and limits flow/current. 


When you see mention of to positive and negative voltage, this is largely about polarity of application (often simple and indicated). 
You mostly see voltage polarity on dual/multiple power supplies (including computer power supply units). You can also get a basic circuit like this yourself, for example using three 1.5V batteries you can supply 4.5V to ground, or have two different circuits, one on +1.5V and one on -3V -- in that they are both relative to their common ground and don't add up in some way.




http://amasci.com/elect/vwatt1.html


-->

===Best means of transfer?===
<!--
We use transformers a lot, to deliver the same amount of power either at high-voltage (and lower current) or at lower voltage (higher current), this raises the question: which is the better way of moving around power?


The short answer is that it depends on other constraints.

If you care about losses on a wire with some resistance (i.e. real cables), losses are basically proportional to <em>I<sup>2</sup>R</em>.

The longer the distance, the more that moving the same amount of power as high voltage means lower current and so lower losses. This is probably the largest reason power lines are high voltage.


But there are also a few major drawbacks to high voltage. 
* the higher the voltage, the more likely [http://en.wikipedia.org/wiki/Electric_arc arcing] becomes
: The smaller the electronics, the lower the voltage needs to be to avoid this.

* danger 
: to us, anything else alive, devices
:: as V=IR or rather I=V/R, and we can assume our body's resistance more or less as a constant, the current through us mostly depends on the voltage.
:: this is why, though it's current that you feel, we can actually put this in terms of voltage. This is roughly why 
:: a lot of household electronics is in the 5..12V range (safe to touch and you won't even feel it)
:: we use up to 48 in a few cases (e.g. agriculture) which is at 'serious tingle but shouldn't hurt or harm' level
:: over 100 or 200 you are unlikely to come to harm from a touch - but don't ever grab it. 
:: a few hundred volts more and all bets are off

This is why wallplug voltage for the safety-unthinking masses is what it is.

Industrial amounts of power is higher voltage because a fixed set of ''can'' be taught to respect things that can kill - and loads of regulations (that dumbasses like you and me wiring up a lamp plug may forget).

-->

===Speed of electricity (trivia, mostly)===
<!--
DUPLICATE OF ABOVE

Bet you thought electricity moves at the speed of light.

It does and it doesn't. 

The marble analogy also help to show that the ''effect'' of the energy seems to move a lot faster than the actual moving things do: If you push in a new marble into the tube, another will almost immediately fall out of the other end even though the marbles move slowly.


For DC, everyday wires, and usual currents, electrons move at the order of centimeters per second, even though its ''effect'' is pretty much speed of light.


For AC it's actually a vibration, and one with less distance yet (though the wavelength of the ''effect'' is huge).

AC also shows the [http://en.wikipedia.org/wiki/Skin_effect skin effect] which means hat for solid-core wire, electrons flow slower in the center and faster near the surface. At power-line frequencies (50Hz or 60Hz) this is fairly negligible, it's mostly at higher frequencies that this starts making ''effective'' wire resistance higher.



The marbles are a useful but not ideal analogy, actually - it's not that electrons in a conductor are such an incompressible thing,
but these are details don't matter to most use.

-->

==Power==

Power, in general, is the rate of work. 

Electric power is proportional to both the voltage and current. 
That's not a complete story, as those two don't vary freely or independently, but it's a start.


===Real, apparent, and complex power===
<!--

Current generally follows voltage closely, but how closely varies - with whether it is a simple/resistive load, or a complex/reactive load.

[[Simple loads]] are resistive loads, and 
: power is exactly voltage times current
: voltage and current, seen as waveforms, follow each other directly

[[Complex loads]] are reactive loads (due to inductance and/or capacitance), and things become more complex.



'''Real power''' is an average of power that has been transferred - actual work done over time, in Watts.

'''Reactive power''' is that which moves around because of being stored, and is in units of VAR - Volt-Amps reactive.

'''Apparent power''' is what you usually measure. 
In complex loads, it is a combination of reactive + real power.{{verify}}

'''Complex power'''


Things like wires have to consider apparent power, because they have to be sized for how much flows,
not how gets usefully done.




http://en.wikipedia.org/wiki/Apparent_power#Real.2C_reactive.2C_and_apparent_power


Joule's various laws

P=I<sup>2</sup>R

P=V<sup>2</sup>/R

-->


===Power Factor===
<!--

'''For context'''

'''Simple loads''', a.k.a. ohmic loads, are those that purely follow Ohm's law: the current neatly follows the voltage applied. 
I is precisely V/R at all points. If you would measure and plot voltage and current, you would get two lines of the shape shape, 
sinusoids that are in sync.

This describes things like toasters, water boilers, and most other heaters - because most of those are really just chonky resistors.
For simple loads, real and apparent power are the same thing, and the power factor is 1.



'''Inductive loads''' are defined by not wanting to change current flow too quickly.

'''Capacitive loads''' are defined by charging effects. 

'''Complex loads''' {{comment|(sometimes 'Reactive loads', but that is confusing due to how reactive power is defined)}} refer to loads with capacitive and/or inductive components. Any where energy gets stored, meaning those curves are not necessarily
in sync, or
of the same shape.



Interestingly, some of the more power-efficient solutions (like LED lighting) have switch-mode power supplies, 
which may have lower heat losses than their predecessors, but also lower power factor because they are ''by nature''
complex loads.



'''Leading and lagging'''

Inductive's unwillingness to change current will lead to the waveform lagging a little later, up to 90 degrees, i.e. a quarter of its wavelength.

Inductive load often comes from things like motors, lighting ballasts


Capacitive elements can be very willing to charge, and may will make the current waveform seem to ''lead'' the voltage waveform, up to 90 degree early.




'''"What does it do to your power bill?"'''

Most residential are charged for real power, not apparent power. It wouldn't make too much of a difference, because ''most'' of your chunkier devices have high power factor, either because they are resistors (toaster, drier), or sometimes because some PF correction lets it draw more real power over the same wires (presumably true for ACs?{{verify}}),








'''"what does that mean to power draw?"'''


So, in general, power can pass without it doing much.



'''Real power''' (a.k.a. active power, true power) is what is actually consumed by the load in the process of working.

'''Apparent power''' is the power flowing through the circuit at all.

'''Reactive power''' 


'''Power factor''' is effectively the difference.

You could see 
* real power as the instantaneous product of voltage and current, 
: meaning that the more those are out of phase, the lower the power factor is.

* apparent power is the product of RMS current and voltage
: which will not lower due to phase differences

Negative power factor is possible, but indicates that the device is a generator.





'''What are some real power factors?'''

Most simple meters show real power - what's consumed, not what passes. 
So you'll need a meter that actually tries to measure the difference. 
This isn't overly hard or necessarily expensive, but it is ''specific'',
and big brands are willing to sell it to you for more than necessary.


Most of your power-hungry devices have high power factors.
Things that amount to resistive heaters - space heaters, grills, toasters, hair driers, classical electric stoves, are basically 1.0.


There are a lot of things that have inductive effects, like classical fluorescent lamps (in part because they need a [[ballast]]), electric motors, power tools, transformers, fridges and air conditioners. 

Assume these sit around 0.6 (though a fridge's defrost will make it go to 1.0 because that's basically a small toaster in your fridge in the place where the ice builds up).


Switch-mode power supplies vary a little with scale. 
Say, computer power supplies will probably be above 0.8 (and, like their general efficiency, seems to vary with how the actual load relates to their actual range),
whereas a cheap USB charger charging a phone is probably on the order of 0.5, 0.6,
as will a lot of LED bulbs.




Vacuum cleaners are often around 0.9 (presumably due to capacitors?{{verify}})



0.2~0.7. In particular, many water pumps and some of the older fridges which will have a
power factor around 0.2~0.4.



Things like wallwarts also tend to have a low power factor (maybe 0.5) but the draw is often so low that is barely contributes to issues.





'''What does power factor do to the power grid?'''

If power factor is less than 1.0 ('unity'), more current must flow to produce the desired electrical work.

It seems that in many places, your local transformer will be trying to compensate, so the losses and effect on the grid is localized{{verify}}.


What you cannot help even then is that 
: these wires will have some resistive loss 
: the wires between that station and you have to be thicker to not heat as much 

Your utility provider still needs to spend on this, so they still prefer lower power factor.
At the same time, that cost will probably be charged through to you anyway.


This is in part just because most of the power-hungry devices around a home are high power factor.


Regulations calso help. For example, in Europe, [https://en.wikipedia.org/wiki/IEC_61000-3-2 IEC61000-3-2] (and similar in at least Britain, China, Japan) aims to minimize devices (up to 16A) having a harmonic effect on the mains waveform. It turns out that where the mechanical load is reasonably constant (vacuum cleaner, fridge, AC), just one chunky capacitor is a good way of improving improve power factor - and basically spreads the power factor correction around.

However, around commerce and industry, large devices having lower power factor adds up ''much'' more quickly.
If you draw at least dozens of kilowatts at a 0.7 power factor, matters a lot more than any block of houses could ever hope to contribute.
Electricity companies often charge industry for apparent power - seemingly in part so that industry itself is incentivized to become part of the balancing and planning, and presumably to be able to avoid crossing the limits of the cables you already put into the ground.{{verify}}







'''What does power factor do around an inverter (solar, generators, off-grid)?'''

The generator will be asked to draw the real power, so be loaded more than necessary{{verify}}.


Solar outputs real power (inverters adapt to the grid){{verify}}, 
but that means that even if the solar's real power output seems to match the house's real draw,
you'ld think it was balanced and zero, but you might still be drawing 15% of that figure from the grid.








"Does power factor correction save power?"


That is a more complex question than you'ld think.

I've seen lots of statements like "you have a power factor under 100% therefore you are wasting energy and increasing your bills" or "Power factor is a measure of efficiency".

In the middle of which is missing the ''entire'' argument why.

This is because when there is a percentage that people don't understand, marketers's special senses go off, .

It turns out most such statements are mostly false.



"Mostly?"

If nothing did any compensation, it's a fairly clear yes. 

If the grid does compensate, you're mostly left over with questions like "could you do this in the house instead of the local substation, to avoid resistive losses on the wires between the two", and the answer is yes. 

Also, the net draw on the grid may be lower, meaning less coal needs to be burned. Even if you're charged the same, this improves things.

However, the longe form of that answer is "yes, by very little for most residential devices."






So yes, you might be able to put 10% more power drawing devices in your house without changing the wiring, if each of them did power factor correction.
Is that important? Up to you.





'''Examples'''

Electromotors are inductive things, and their current and voltage waveforms will be out of phase.

Intuitively you could think that this is because energy is delivered via the electromagnetic fields.
That field carries power and needs to be there for the thing to work, but you could almost think of it as springy gears or something.
It's not that some power gets used early or late in that cycle, it's that due to how these things work, some of it doesn't get used at all.

Exactly how much out of phase depends on type, load, and some design details. 



You might expect vacuum cleaners to have low power factor, but it seems to be around 0.9. 
(why? is it that with some basic correction they can have more power before running into apparent-power issues?)


Classical transformer power supplies can be a little worse, at around 0.6, 
yet SMPSes aren't necessarily better than 0.7 to 0.8. They ''can'' be designed better, but aren't necessarily.

A few designs (e.g. capactive droppers, some small cheap LED drivers, others) can be even lower.


Note that technically, it is not the power supply that has a . 
It is the whole circuit, with the load playing no small part in where that power factor settles.



In fact, a lot of things with small amounts of draw may be 0.5 down to 0.3. 
It's not much loss because it's not much draw,



Your house may average maybe 0.8?{{verify}}



Infrastructure should be designed with power factor in mind, because the lower the power factor,
the thicker the wires need to be for the same real power - and remember most things will be indicating real power.

Say, if you're 







"So power factor is about efficiency / waste?"

Unaddressed, yes - but not necessarily.



For example
* outdoor HID lighting has a PF of around 0.5 due to its ballast
: there is often a compensating capacitor added
:: It is not necessary for function,
:: it does not lead to more efficient use of power,
:: but it's less load on the wiring



Reactive power is undesirable because
* you need thicker wires for the same amount of eventual work done
* 



'''Is low power factor loss?'''

In the immediate sense, no.

In theory, indirectly so, in doing the same work with an unnecessarily low power factor machine
means more current for that same work, so more resistive waste. But that's a very selective "if".

The real issue is more that power distribution is much harder with power factor involved,
so your power company.

To the point that 




'''Power factor''' is the ratio between real power absorbed by the load, and the apparent power.

For you, you can read that as "I get X amount of real output per the amount of flow I get in"

For your utility company, it's more of a "what part of the energy has to be there but doesn't get used".
Classical metering measures real power (what gets used) rather than apparent power (what gets put in)
which is why utility companies don't like low power factor - it's basically the part of electricity they have to pay to deliver. And they can recover it, but not very efficiently so it's often not worth the hardware.

It's also why people suspect smart meters are one way to 'accidentally' switch to measuring apparent power instead.



For complex loads, the current moved around is lower than the power that was actually consumed. 

But to a simple (slow-integrating) multimeter this is passing current like any other, meaning it will report a power higher than was really consumed by the load.


You could say complex devices will deliver current back to its source, but lagging or leading a little due (because things are charging and discharging) so it won't be in phase.

Also, whether that is just passed pointlessly, or usable again, is another discussion - and actually a complex one. 

tldr: worst case it's all wasted, best case part of it is used, but you need to be somewhat fancy to do this, so this is mainly, and only partly, worth it at power-distribution level.

That's also why power companies won't like complex loads (...that draw noticeable power), because of distribution reasons -- they get current back, some of which will be lost.

(in fact, industrial-scale power users may get charged in relation to power factor. This to some degree is fair, in that cables have to be rated for apparent power. Consumers are typically charged by real power{{verify}}, though in some areas this seems to be changing)




Power factor can also matter to larger-scale services like lighting, because if you're putting in hundreds of kilometers of cables into the ground, if they can be thinner that can save a lot of cost.




=====Considering power factor in power measurements (and power ratings)=====

Cheap power-measuring will assume the current curve has the same shape as the voltage curve. 
And tend to not measure either curve, and instead get away with sine-wave RMS assumptions/calculations.

And for simple loads, this is pretty accurate. {{comment|(Accurate enough for everyday purposes, anyway - it is only as accurate a the AC waveform is sinusoidal, which it isn't ''quite'' true but usually close enough.)}}


To measure real power, RMS-based calculations are off for complex loads,
and the better way is to measure with very short intervals, i.e. do high-detail integration.



There is the following distinction: A device can be be described with either
* apparent power
** V<sub>rms</sub> * A<sub>rms</sub>
** simple multi-meters will measure this

* real power (a.k.a. true power, others)
** V * A (integration of instantanous values)
** your house's electricity meter historically measures this (to reasonable accuracy), so you'll be billed for real usage. (this might be changing with smart meters{{verify}}, roughly because they can)
** good power meters will measure this (to reasonable accuracy)




If you have many complex devices around, you might not like it so much either, because every complex device pollutes your voltage waveform somewhat. 
You won't immediately care, but some (often cheaper) devices may well be sensitive to it -- effectively complex loads interfering with each other.


See also:
* http://www.dansdata.com/gz028.htm
* http://www.lmphotonics.com/pwrfact.htm
-->

====Power Factor Correction====
<!--
Power factor correction takes complex loads and makes them look like simple loads. 

Not perfectly, but usually much more so.


More linear loads, like induction motors or ballasts, may need only some passive components.

Less linear loads, like rectifiers and swich-mode supplies, 



The larger the load, the more it makes sense to put it near or in the load.

This is why for consumers it's often done at the local substation.

For industrial users this may be untenable, and they may simply be charged for apparent power instead,
so that they have the incentive to correct this on site. 
It's partly lazy, but it also makes sense to spread the planning around.



There a few reasons you might care, and ''most'' of them are related to power distribution.



One is to isolate the reactive portion to in or near the device, and there by in your house - the power grid will like you, because it doesn't have to deal with it.
 
It is also ''primarily'' the power grid itself that is interested in doing this,
because it loads the wires less, 
which lowers losses a bit, 
and means they don't have to oversize the wires as much,
lowers the difficulty of getting the same amount of electricity-based work done,
makes the network act a little more predictably.



'''"You can minimize your electricity bill!"'''

Depending on where you are, home installations are usually charged either for real power (~= actual work done) ''or'' apparent power (~= electricity moved back and forth in the process, which).   (Industrial installations differ, but are not the worry of most people who read this)


In many places you were classically charged for real power{{verify}}, 
in part because that turning-disc meter effectively measured real power{{verify}}, 
and only digital meters spend the time to tell the difference.

If you are charged for real power, power factor correction does not change your bill,
and still are. 

This ''may'' change''.



If you are charged for apparent power, then yes - but less than you think{{verify}}{{verify}}

And with modern smart meters, it ''could'' be both. 
You can ask, and check.


If your interest is environment and such, then keep in mind that power factor correction is ''not'' really a change in efficiency, though.

It's basically power companies trying to save money trying to make you spend on this correction inside your house,
rather than them trying to it within the infrastructure. It spreads things around without changing much.



'''It can help reduce voltage drops in an installation.''', '''It can help reduce wiring costs in an installation.''' 

In larger installations, both of these are potential concerns, and both are true to a degree you want to look at.

Most home installations are very predictable,
there are few complex loads large enough for this to matter, 
and the wiring is typically oversized anyway.



'''"Not doing so pollutes the mains-power waveform"'''

Yes, but can be overstated.

Things like larger switch-mode power supplies do this enough that some amount of correction may be ''required'' to even meet certification specs{{verify}}.

...but again, most smaller installations, smaller devices, don't need to worry about this.



'''"Power factor correction is pure savings" / "Power factor correction is pure waste"'''

The dumbest form of power factor correction is ''just'' wasting that power,
in which case the devices are less efficient than without.

In the best cases, it may mean more effective use of the same power.

It seems that in most consumer cases, it's




They will incentivize this, because 



And not more efficient, as some smooth talking salesmen like to say.

Even if it manages to use some of that power rather than just burning it off,
it will not be much, particularly because in consumer contexts this is
often done at the scale of single power supplies. 



Passive PFC and active PFC are different methods,
and seem to practically say more about the amount of correction than of the quality{{verify}}

-->


<!--
====Power saver devices====

tl;dr I haven't yet found one that isn't a clear scam.



So, the claims of what a device with a sticker that says 'power saver' does are as wide as they are vague.
We cannot possibly address them here.

But they will often claim that just by plugging them in, you will save 30% of your power bill.

How?


One thing they claim is that they take out the transients in the waveform.

On face value, that makes some sense - transients are not useful for devices, so if you could smear that energy back into the waveform somehow, great.

There are also several issues with this. One, that requires a complex load to do, e.g. capacitors


Another suggestion they claim amounts to power factor correction. This too makes apparent sense.

And yet this is rarely worth it until you have commercial or industrial devices.

Also, as long as you pay for real power and not apparent power 





-->

==Current capacity==

<!--
Known as '''current capacity''', '''ampacity''' ('''amp'''ere cap'''acity'''), sometimes '''current rating''' or '''current-carrying capacity'''.

Current capacity describes how much current a device or conductor could deal with, before giving up because of protection circuitry, damaging itself, or simply not being able to store or move the change fast enough.


If describing components, PCB traces, it tends to be a design-time specs thing - 'this transistor can move 1A' sort of thing.


It's perhaps more associated with things that provide power.

They'll have a power rating on the side, which is closely correlated with its current capacity.

In this context you'ld also associate low current capacity with the voltage dropping when you load it more than that. 

(Say, I have some wallwarts USB chargers. Using a usb current meter and usb load it's pretty easy to see how some output stable ~5V up to different points, then drop off.)




Note that high voltages do not ''necessarily'' mean high current capacity.

Take an electrostatic generator (a.k.a. Van de Graaff generator[http://en.wikipedia.org/wiki/Van_de_Graaff_generator]) - it has fairly high voltages, but tiny current capacity, which is why these things can sit in classrooms without us having to worry about crispy kids.



Supercapacitors/ultracapacitors have huge current capacity, because their ESR (their internal resistance) is very low compared to other capacitors, to most batteries, and such. This low ESR means they can discharge (and charge) a lot faster.

While batteries can deliver more power over time (for the same volume) the higher internal resistance of most batteries means they can't do it as fast. {{comment|(The limit is their chemical process. Lithium is currently the best of the bunch, but still not comparable to supercaps)}}

That resistance also implies how much the voltage will drop within the battery when current is being drawn, 
and how much heat is generated in it, both of which further the current that a device can comfortably draw.


http://en.wikipedia.org/wiki/Ampacity

-->



<!--
==Other notes==

Which is a reformulation of the perhaps more basic:
 1 ampere  = 1 coulomb / 1 second

An ampere is also 6.25*10<sup>18</sup> electrons/sec (because electrons have a charge of 1.6*10<sup>-19</sup>Coulombs).

Battery capacity is often measured in proportion to charge. An Ampere-second (or amp-hour, or milliamp-hour) refers to the charge moved in time:
 1 coulomb = 1 ampere · 1 second
...which is a reformulation of the above.

-->

==Voltage and current as related properties==
===Series and parallel, and a little resistance; Ohm's law, Kirchoff's laws===
<!--
http://www.kpsec.freeuk.com/voltage.htm


'''Kirchhoff's current law''' (KCL) / '''Kirchhoff's junction rule''' / '''Kirchhoff's first law/rule''':
At any node/junction, current flowing in equals current flowing out.

'''Kirchhoff's voltage law''' (KVL) / '''Kirchhoff's loop/mesh rule''' / '''Kirchhoff's second law/rule''':
Potential difference around a closed circuit must be zero.

Kirchoff's laws can be handy in figuring out basic voltage/current behaviour

Node analysis (KCL, Ohm's Law), most useful when most node voltages are known

Mesh analysis



In series, voltages add up. 
Current is the same though all elements in series.

In parallel, currents add up. 
Voltages are the same for elements in parallel (hence the ability to plug in many things into wall sockets) as long as the power source can deliver the required power.

Say you have a circuit with a bunch of different resistors, and a bunch of reference points:
 (1)----[ R1 ]--(2)--[ R2 ]----(3)
  |             20mA            |
  |                             |
  |                             |
 (4)----[ R3 ]--(5)--[ R4 ]----(6)
  |             80mA            |
  |                             |
  |             9V              |
 (7)----(8)---Battery-----------+


Voltage
* The voltage between (4 and 6) and that between (1 and 3) is the same, namely 9V
* The voltage between (1 and 2) and (2 and 3), and (4 and 5) and (5 to 6) depend on the resistance, and are likely all different (unless R1=R2=R3=R4).

Current:
* The current flowing through (4 and 6) and that between (1 and 4) is different (...as listed)
* The current flowing through (1 to 2) is implicitly the same as that through (2 to 3), is the same as (1 to 3), namely 20mA
* the current drawn from the battery is 100mA (1 to 2 to 3) plus (4 to 5 to 6)

The power is drawn is 100mA * 9V = 0.9W



In the water example, you can think of voltage as the pressure the battery has/applies, of resistors as mostly closed taps, and of current as the speed of flow, and of power as the amount displaced{{verify}}.

Resistors, in the analogy, could be seen as taps or just pipes that are quite thin relative to that of the normal piping/wiring. This explains why current flows slowly though just a single high-ohm resistor (very thin pipe), and why in parallel the lower-ohm resistor sees more current (it would be the wider of the two pipes).


There are similar analogies for elements like capacitors, transistors and more, but the water analogy does stop making sense soon after.
-->


==Real and ideal electrical sources==
<!--
Ideal voltage sources and ideal current sources do not exist, but are useful concepts when describing tendencies of real-world electrical sources. 

For example, lab supplies intentionally tend to act a little more as ideal sources than batteries,
and knowing about battery imperfection is often useful.
-->

===Ideal voltage source===
<!--
An ideal voltage source maintains its voltage regardless of attached resistance.


This is idealized because every real-world electricity source acts like it has [[#internal resistance]]. When a load is attached, the voltage is divided between the load and the internal resistance. 

This tends to matter fairly little to loads on the order of a k&#x2126; or more since internal resistance is typically much lower than that, but it easily matters for heavier loads.

Since the (fairly low) internal resistance is in series with the load, low-resistance loads causes the voltage to drop.

The internal resistance also implies the limit of the power that can be moved from source to load.
-->

===Ideal current source===
<!--
Maintains current regardless of voltage.
-->

===Real voltage sources===
==Some basic volt/amp/resistance circuits and related analysis==
{{stub}}
<!--

...to get you started on some basic reasoning how to deal with things.



Ohm's law is one of the most important in looking at a circuit. May be useful to remember in more than one form. 
Since
* resistance is usually the most controllable property, then voltage, then current.
* voltage and resistance are also a little easier to measure.
...this means that I=V/R is usually the most interesting form. 

The most interesting implications from Ohm's law are probaly:

From I=V/R:
* less resistance is more current, more resistance is less current
* more current with more voltage

R=V/I
* figuring resistors necessary for a voltage drop (at a current){{verify}}

V=IR 
* not so interesting when your voltage is fixed

Each may be useful when you figure out unknowns of V, I and R in a table.
 


It is useful to consider [http://en.wikipedia.org/wiki/Kirchhoff%27s_circuit_laws Kirchhoff's circuit laws] to simplify your mental model of various circuits. These can be seen as an implication of conservation of energy in a 'it has to go somewhere' way. They can be helpful in seeing what is and isn't involved in some calculation you want to do.


Kirchhoff's ''voltage'' law: 
* The sum of the voltage drops around a circuit equals the sum of voltage sources
* Alternatively, and more abstractly: The directed sum of the electrical potential differences around any closed circuit is zero.

Can be useful thing to remember when figuring out relative voltage drops.


Kirchhoff's ''current'' law: 
* at any node, the sum of currents flowing in  equals  the sum of currents flowing out.
* Alternatively: the directed sum of currents is zero.

When looking at things in effective parallel, you can easily figure out the relative currents.


-->

===Voltage divider===

<!--
A voltage divider produces an output voltage (Vout-to-ground) that is a fraction of the input voltage, by partitioning voltage among the components.


Note that in some ways, every load in series with another is a voltage divider, as the resistance of each relates directly to the voltage drop over each such load.

For example, when you have a load for which you have to drop voltage, and you know its resistance (and that it won't vary) and its current requirements, you can calculate the resistor to put in series. For example, say you have a 12V battery and want to drive a 5V load that will need 50mA. You need the resistor in series to drop 7V at 50mA, for which we need R = V/I = 140 Ohm.

Note that larger voltage drops across this resistor as well as larger current needs increase the amount of power wasted in the resistor - as heat. In this case it's 0.35W. In some cases you'll need a sturdy resistor.


Note that 140 Ohm is not a standard value. You can choose something near, but you can also get more exact voltage drops by using two resistors in series, and use the voltage across one of them. Since a particular voltage drop is the ratio between the two resistance values, you can pick whichever pair of standard resistors fits that best, and there are usually several options.


The output voltage being the difference between Ground and the point between the resistors. This is shown in a few different ways:

                                     +---R1--+---o
 Vin ---- R1 ---+-- R2 -- Gnd        |       |   
                |                   ---      R2  
              Vout                   -       |   
                                     +-------+---o 
  +-------+
  |       |   
  |       R1  
  |       |   
  |       +----o
 ---      |     
  -       R2    
  |       |     
  +-------+----o

Basic ciruit analysis will show that:
 Vout = Vin  *  R2 / (R1 + R2)


It is also fairly easy to create an ''adjustable'' voltage divider using a potmeter, either using it for R2 (or R1), or using just the potmeter and using the wiper as Vout. {{comment|(Note that the latter will make R1+R2 a constant, which simplifies the formula that figures out Vout, as it now depends only on R2.)}}




For example, say that you want to drop 12V to 7V. You could pick R1=1.5kOhm and R2=1kOhm, for voltage drops of 4.8V over R1 and 7.2V over R2.



Of course, once you attach a load, you change things, because you make the following (where R3 is the resistance of the load):
  +-------+
  |       |   
  |       R1  
  |       |   
  |       +----o---+
 ---      |        |
  -       R2       R3
  |       |        |
  +-------+----o---+


This will change the the resistor balance - if the load draws more than a little current, that means R3 is lowish, and if it's low compared to R1 and R2 it will noticably change the voltage division.


While all the load's current will pass throuh R1, the lower the R3 resistance is (i.e. the more the load draws current) the more current will go through R3 instead of R2.


Voltage dividers are perhaps most interesting...
* for voltage sensing (ADCs, op amps, LEDs and such draw little current)
* to create voltage that will then be fed through an amplifier


Then we know that 7V will be dropped across 600 Ohm and whatever load you attach, meaning that 

It helps to realize that there is an behaviourally equivalent but conceptually simpler circuit: one with the Vout voltage supplied in series with a resistor equivalent of R1 parallel with R2 - in this case 600 Ohm.

age VOUT with a series resistor whose value is equal to the parallel combination of R1 and R2. The figure to the right shows the equivalent circuit for our example voltage divider.


However, the load's current goes through R1 but since the load is parallel with R2, much of it passes past it - this upsets the resistor balance somewhat. 
Continuing the previous example of our 50mA 5V load - it has a resistance of R=V/I=100 Ohm. This means the R2-and-load combination has a 90.9 Ohm resistance, meaning the voltage divider will suddenly split into 

R2 / (R1 + R2)
90 / 1.590

We can get a smaller difference from using smaller resistors -- but then we start using a lot of current.



, this means that you always draw 0.48 mA




Note that a voltage divider always draws current across the two resistors, as I=Vin/(R1+R2), so voltage dividing always wastes some power.
To minimize this you want R1 and R2 to be high. 
For example, for voltage sensing on an ADC you can easily use high resistances. 
You do however limits the amount of current a load can draw, since that current will be limited by R1.

This is a tradeoff that you make according to the application.






http://en.wikipedia.org/wiki/Voltage_divider

http://www.play-hookey.com/dc_theory/voltage_divider.html

http://www.tpub.com/neets/book1/chapter3/1-35.htm

http://www.play-hookey.com/dc_theory/voltage_divider.html

http://www.facstaff.bucknell.edu/mastascu/eLessonsHtml/Resist/Resist3.html

http://hyperphysics.phy-astr.gsu.edu/hbase/electric/voldiv.html

-->

===Current divider===

<!--
A current divider is mostly just what happens to loads in parallel things on the same voltage, which is proportional to 1/R {{comment|(think V=IR for I with V crossed away as a constant)}}.


  +-------+----+
  |       |    |  
 ---     R1    R2
  -       |    |
  +-------+----+


Current is proportional to
 R1/(R1+R2) of the current moves through R1,
 R2/(R1+R2) of the current moves through R2,



Often more of an implication than intent, though current limiting and short circuit protection are extensions of the idea{{verify}}.


http://en.wikipedia.org/wiki/Current_divider

-->




<!--
==semi-sorted==



With all possible reformulations, there are a lot of seemingly entangled formulae:
 1 volt = 1 joule/coulomb

 1 Watt = 1 joule/second


 1 ohm = 1 volt/ampere (you can recognize Ohm's law in that)

 
 1 farad = 1 coulomb/volt (so 1 coulomb = 1 farad / 1 volt)    (Capacitance)
 
 1 joule = 1 newton · meter

 1 newton = 1 kg · meter/second2
 


Voltage refers to voltage difference 

http://en.wikipedia.org/wiki/Voltage#Historic_terminology_-_.22potential_difference_.28P.D..29.22


We associate '''shorting out''' with Very Bad Things&#x2122; because most power sources can deliver a lot more power than you want going through anything, and because wires may not be able to take the current. In the case of batteries (and a lack of fuses) we can cause chemical damage (a battery can only supply so much current at once).

But shorting things out isn't a bad thing per definition. Consider:
* protection diodes across a coil short out the coil -- taking care of pretty little kinetic/electrostatic energy, and the short only changes the way the magnetic field collapses (mostly time taken)
* shorting a solar panel means it supplies just as much current as it would with a load. Short-circuit current measurement is also fine on a solar panel.
-->

=Some practice=
==Is AC the peak voltage or RMS figure?==

RMS, conventionally.

Deviations from that should and typically will be mentioned.

<!--
{{skipme|
So technically
: 230V is approximagely 325V mid-to-peak, and approximagely 650V peak-to-peak
: 120V is approximagely170V mid-to-peak, and approximagely 340V peak-to-peak
...but these are not voltages you ever really care about. 
}}
-->




Multimeters on AC setting try show RMS value. 
Or try to - cheap ones effectively ''assume'' the waveform is sinusoidal.

Only the fairly expensive actually sample values quickly enough to be able to integrate.
Since the waveform is broadly sinusoidal but never perfectly (and use and even wiring can influence this),
this can matter a little.




One reason for the 'AC is RMS' convention seems to be that the power delivered through the same voltage VDC and VAC<sub>rms</sub> is roughly the same, which is pretty convenient.

Also comes with footnotes - there is a difference between the root mean square and the average (of the absolute) voltage - see [https://en.wikipedia.org/wiki/Form_factor_(electronics) form factor], which for a sine wave is approximately factor 1.1 {{comment|(and approx factor 1.5 for half-wave rectified output)}}, which also seems to be the source of figures like 110V, 120V as they typically come from AC generators{{verify}}.

==Powering your projects==

===From-the-wall power===
{{stub}}

====Adapters====
<!--
Most [https://www.google.com/images?svnum=10&q=wallwarts wallwarts] and black blocks like them - are little more than a transformer, and if they're for DC, a rectifier and capacitor to get a vaguely constant voltage.


The cheaper kind isn't regulated. Regulated versus non-regulated refers to whether there is also a voltage regulator to try to keep the output voltage constant. 


In '''unregulated adapters''', output voltage depends mostly on the transformer, 
and varies with the current it is supplying - more current means less voltage. Under no load they'll often show more volts than is on the label, under a lot of load it'll show less. It should be the right voltage for the device it's made for. 


Regulated adapters have a voltage regulator added, which basically means the voltage will stay stable 
while there is enough energy to do so. 

For more on how they do so, see [[regulators]], but you may wish to know the cheapest way to do so is burning off an internal voltage difference as heat.

Adapters doing multiple voltages are typically regulated. {{verify}}



See e.g.:
* http://www.apogeekits.com/power_adapter_selection.htm

-->


'''Unregulated adapters'''
* cheap
* transformer and rectifier
:: where voltage lowers noticeably with more draw (in the cheapest, most basic designs)
:: are fine for smaller draws (e.g. microcontrollers without much load, and since boards often have regulators, regulating twice is pointless)
* Price for approx ~200mA-2A: EUR/USD 5-10? {{verify}}

'''Regulated adapters'''
* usually meaning the above plus a linear regulator (linear regulator generate heat, but voltage is stable)
* slightly more expensive than unregulated
* Price for approx ~200mA-2A: EUR/USD 8-15? {{verify}}
* Price for combinations up to ~100W: on the order of EUR/USD 100? {{verify}}

'''Switch-mode power supplies'''
* more power-efficient, and smaller for the same power, but also more expensive
* In a few cases the fast switching may interfere with high frequency (E.g. RF) components, produce a hum in audio, and such. 
* Price for approx ~200mA-2A: EUR/USD 10-40? {{verify}}
* Price for combinations up to ~100W: on the order of EUR/USD ~100? {{verify}}

<!--
Wallwart-type supplies 
* price correlates with complexity (generally, switching &gt; regulated &gt; unregulated -- but the difference can be pretty small) 
* vary between brands
-->

'''Variable / lab supplies'''
* can be exactly controlled (voltage limit and current limit), which can be very useful to figure out behaviour
* bigger, expensive.
* The simplest start at maybe EUR/USD 70
* serious versions easily 200-300, more depending on power, precision and [[ripple]] reduction, etc.

====Repurposing an ATX power supply====
See e.g.:
* http://www.wikihow.com/Convert-a-Computer-ATX-Power-Supply-to-a-Lab-Power-Supply


{{stub}}
<!--
Having a previously internal fan screwed on the outside may be handy space-wise, since you may be using more space inside for nice plugs.
If it's going to sit on a desk anyway it's fine.


The green wire would usually be the power button - power-up request by shorting to ground, which in a computer would be the power button (plus debouncer, as per ATX specs, to avoid possible damage. Usually the motherboard does this).

Soldering green to ground is easier than making a debouncer, and when there is a switch on the supply already you don't really need a soft power button.

(TODO: wouldn't that mean the 4-second poweroff?)


Newer ATX PSUs (approx what year?) have a sense wire, recongnizable as a wire connected {{comment|(regularly thinner than the rest since it need not carry current)}} to one of the 3.3V wires on the many-lead ATX plug. If you have one, the PSU is measuring what that voltage is doing, and may refuse to power up if it is not connected to anything. 

If you have cut off the ATX plug and it was wired to one of the 3.3V pins on that side (it often is), you may need to wire it to one of them again.


The power wires are often 18 gauge (AWG) or so, so can comfortably carry at least a few amps each.
This is more than you'll use to power most things, so you can cut away most wires. 

If you're cutting away a lot of things, it may be smart to leave at least:
* two red (5V), probably the thing you are most interested in)
* one or two yellow (12V), depending on how much you think you'll use it
* one orange (3.3V)
* three black (ground)
** plus, for wiring ease rather than current, separate blacks to status led and/or power switch.
-->

==Power and cables==

===Wire gauge, current limits===
{{stub}}

Wire gauge markings is often seen as one or more of (all convertible, but tables exist mainly for the common ones):
* AWG,                American Wire gauge (also sometimes Brown & Sharpe, B&S)
* diameter           (inch or metric)
* cross-section area (inch or metric)
* SWG,                Standard Wire gauge. Imperial as in British. Seen less often. Same order of magnitude as AWG, but a little different - see e.g. http://www.clag.org.uk/swg.html) 

There's some others, like Birmingham Wire Gauge, but these tend to be more specific.



<span style="color:red">'''The current ratings are approximate'''</span> because beyond '''cross-sectional area''', the main and obvious variable, '''it also varies with''' things like
* '''Material'''
: the below is for copper, because it's by far the most common for wiring

* '''Solid''' versus '''stranded''' wire
: braided has maybe 15% less material for the same outer diameter, so current rating will be accordingly lower
: AWG (and probably others) is the total cross sectional area, though, so braided wire of the samw AWG will just be a slightly thicker{{verify}}

* '''Enclosed''', or '''free in air'''
: these figures are for in-air cables
: enclosed cables warm up more with the same heat input, so ''should'' carry less if it's going to be sustained current
: If you like large safety margins, you could say a third.

* '''insulation temperature rating'''
: you can safely get away with running wires one or two dozen degrees hotter if it's in better insulation, which can be a good factor 1.5 in terms of amps

* '''duty cycle'''
: power transmission assumes constant use
: things that are always intermittent can get away with factors more current - at least in heat terms (it's still lossy)

* '''frequency''', also because braided has more [[skin effect]] than solid, and surface matters more than diameter
: assume these figures are for DC, mains AC or other low-frequency AC
: you can find some tables with "solid will be fully used for frequenies up to X Hz" figures

* '''installation'''
: coiled wire heats more. Don't do that on anything that carries a bunch of power. Also a thing on spooled extension cords.

* '''safety margin''' 
: case in point: the below current figures are moderately conservative, people reading these charts may not about any of the above. In well-considered cases you can safely get away with more.


These and other reasons mean the figures vary between different resources

'''Not thoroughly checked''' - Don't stake your life or home on chances there's no typo in here. 


{|- border="1" cellpadding="4" cellspacing="0" class="wikitable" style="text-align: center"
|- style="background:#eee"
! AWG   
! Diameter <br/>(Inch)
! Diameter <br/>(mm)
! SWG (approx)
! Cross section area (mm&#xB2;)
! Ohm/km or mOhm/m (for copper)
! Approximate&nbsp;current&nbsp;limit<br/>(Amp, for solid&nbsp;copper)<br/><span style="color:red">Big footnotes, see above</span>
! Notes

|-
| 40	
| 0.0031
| 0.08
| ~44
| 0.005
| 3440
| 0.05

|-
| 38
| 0.0063
| 0.10
| ~42
| 0.008
| 2160
| 0.07
| approximately a single strand from (mid-20 AWG range, by size) stranded wire 

|-
| 36
| 0.0050
| 0.13
| ~39
| 0.013
| 1360
| 0.10
| Probably the thinnest you can easily find/buy

|-
| 34
| 0.0040
| 0.16
| ~38
| 0.020
| 850
| 0.13

|-
| 32 	
| 0.008
| 0.20 	
| ~36
| 0.03
| 530 	 
| 0.32

|-
| 30 	
| 0.010 	
| 0.26  	
| ~33
| 0.05
| 330 	
| 0.52

|-
| 28 	
| 0.013 	
| 0.32 	
| ~30
| 0.08
| 210 	
| 0.83
| computer ribbon cables for signaling (PATA, SCSI, Floppy, USB port connectors, and such) seem to often be 26, 28, or 30AWG {{verify}}

|-
| 26 	
| 0.016 	
| 0.40	
| ~27
| 0.13
| 130 	
| 1.3
| 

|-
| 24 	
| 0.020 	
| 0.51 	
| ~25
| 0.20
| 84 	
| 2.1
| Seen e.g. in Cat5 network cable

|-
| 23
| 0.023 	
| 0.57
| ~24
| 0.26
|  	
| 
| thickness of various office [[staples]]

|-
| 22 	
| 0.025 	
| 0.64 	
| ~23
| 0.33
| 53 	
| 5.0
| Seen e.g. in computer power distribution

|-
| 20 	
| 0.032 	
| 0.81 	
| 21
| 0.50
| 33 	
| 7.5

|-
| 18 	
| 0.040 	
| 1.02 	
| 19
| 0.82
| 21 		
| 10
| Seen e.g. in computer power distribution

|- style="background:#eee"
! AWG   
! Diameter <br/>(Inch)
! Diameter <br/>(mm)
! SGW (approx)
! Cross section area (mm&#xB2;)
! Ohm/km or mOhm/m (for copper)
! Approximate&nbsp;current&nbsp;limit<br/>(Amp, for solid&nbsp;copper)<br/><span style="color:red">Big footnotes, see above</span>
! Notes

|-
| 17 	
| 0.045
| 1.15
| ~19
| 1.04
| 17 		
| 11
| roughly equivalent to 1.0mm&#xB2; copper


|-
| 16 	
| 0.051 	
| 1.29 	
| ~18
| 1.3
| 13 		
| 13

|-
| 15
| 0.058		
| 1.45
| 17
| 1.5
| 10.5
| 15
| roughly equivalent to 1.5mm&#xB2; copper

|-
| 14 	
| 0.064 	
| 1.63 	
| 16
| 2.0
| 8.3 	
| 17

|-
| 13	
| 0.072
| 1.83
| 15
| 2.6
| 6.6
| 20
| roughly equivalent to 2.5mm&#xB2; copper

|-
| 12 	
| 0.081 	
| 2.05 	
| 14
| 3.3
| 5.2 	
| 23

|-
| 11 	
| 0.091
| 2.31
| 13
| 4.2
| 4.1
| 27
| roughly equivalent to 4mm&#xB2; copper

|-
| 10 	
| 0.102 	
| 2.6	
| ~12
| 5.2
| 4 	
| 32
| 

|-
| 9 	
| 0.114 
| 2.91
| 11
| 6.6
| 2.6
| 38
| roughly equivalent to 6mm&#xB2; copper

|-
| 8 	
| 0.128 	
| 3.26 	
| 10
| 8.2
| 2.1 	
| 45
|

|-
| 7
| 0.144	
| 3.67
| ~9
| 10.6
| 1.6
| 52
| roughly equivalent to 10mm&#xB2; copper


|- style="background:#eee"
! AWG   
! Diameter <br/>(Inch)
! Diameter <br/>(mm)
! SWG (approx)
! Cross section area (mm&#xB2;)
! Ohm/km or mOhm/m (for copper)
! Approximate&nbsp;current&nbsp;limit<br/>(Amp, for solid&nbsp;copper)<br/><span style="color:red">Big footnotes, see above</span>
! Notes

|-
| 6 	
| 0.162 	
| 4.12 	
| ~8
| 13.3
| 1.3 	
| 60
|

|-
| 4 	
| 0.204 	
| 5.19 	
| ~5
| 21
| 0.81 	
| 80
|

|-
| 2 	
| 0.258 	
| 6.54 	
| ~3
| 33
| 0.51
| 100
|

|-
| 1 	
| 0.289 	
| 7.35 	
| ~1
| 42
| 0.40 	
| 125
|

|-
| 0 (also&nbsp;1/0) 	
| 0.325 	
| 8.25 	
| 1/0
| 53
| 0.32 	
| 150
|

|-
| 00 (also&nbsp;2/0)
| 0.365 	
| 9.27 	
| 
| 67
| 0.26 	
| 175
|

|-
| 000 (also&nbsp;3/0)
| 0.410 	
| 10.4 	
| 
| 85
| 0.20 	
| 200
|

|-
| 0000 (also&nbsp;4/0)
| 0.460 	
| 11.68 	
| ~6/0
| 107
| 0.16 	
| 230
|

|- style="background:#eee"
! AWG   
! Diameter <br/>(Inch)
! Diameter <br/>(mm)
! SWG (approx)
! Cross section area (mm&#xB2;)
! Ohm/km or mOhm/m (for copper)
! Approximate&nbsp;current&nbsp;limit<br/>(Amp, for solid&nbsp;copper)<br/><span style="color:red">Big footnotes, see above</span>
! Notes
|}


<!--
Notes:
* Cat5, Cat5e is often 24 AWG, and can be a handy source of such wire in electronics projects (Cheaper cables can have 26 AWG, DIY plugs may require 22 AWG or 24 AWG).




The wiring for your house is likely due to regulations. 

Something like:
* light switches: 1,5mm<sup>2</sup> is enough
* regular wall plugs: 2,5mm<sup>2</sup> diameter is often required
* mixes: 2,5 mm<sup>2</sup> is often required
* plugs for ovens, washing machines, and other heavier use may be required to be 4 mm<sup>2</sup> or 6 mm<sup>2</sup>
-->

===Labeling===
{{stub}}

Loose-strand cable tends to list voltage and gauge.


Ready-made power cords are often labeled with 'Designation code for harmonized cables' -- see HD 361, &nbsp; DIN VDE 0281 / DIN VDE 0282 / DIN VDE 0292 {{verify}}

Example labeling:
* H07RN-F
* H05VV-F
* H05VV-F 2x1.00
* H07RN-F
* H05VV-F3G1,5
* H05VV-F5G2.5


This conists of...

* '''character''' indicating
** H: Harmonized standards (usually)
** A: Nationally authorized/recognized standards (meaning? different interpretation of the following?)


* '''Two digits''' indicating nominal voltage
** 01: 100V
** 03: 300V / 300V
** 05: 300V / 500V
** 07: 450V / 750V


* '''character''': Isolation material (see merged material list below)

* (optional) additional structural elements

* '''character''': mantle material (see merged material list below)


* '''Dash-character''', indicating the wire core type:
** -F: fine wired, for flexible cords (common for extension cords)
** -K: fine wired, for fixed installations
** -H: very fine wired (for flexible cords)
** -U: single rigid round core
** -R: standed rigid round cores
** -Y: Tinsel
** -D and E: Flexible conductor for use in arc welding cables


Optionally, there is a trailing:
* Number of (independent) cores
* earthing details{{verify}}, one of:
** G - a green/yellow core{{verify}}
** X - no green/yellow{{verify}}
** Y - tinsel conductor with no cross-section specified{{verify}}
* all conductor's (nominal) cross section in mm<sup>2</sup>, often something like 1.5, 2.5, 0.75 or so



The material list used for isolation and mantle materials - which can be the same, of course {{comment|(a merged list; either type can only pick a subset of these)}}:
* V: PVC
* V2: PVC, heat-resistant
* V3: PVC, low-temperature
* V4: PVC, cross-linked
* Q: Polyurethane
* R: Rubber (natural or synthetic)
* B: ethylene-propylene rubber (Synthetic rubber)
* G: Ethylene-vinyl-(co)acetate
* S: Silicone rubber
* N: Chloroprene
* N2: Chloroprene
* N4: Chloroprene
* J: glass figre braid
* T: Textile braid
* T: Textile braid, heat resistant


So, for example, a H05VV-F3G1,5 is a 300/500V cable with three fine-wired 1.5mm<sup>2</sup>leads that uses PVC for both mantle and isolation.


For the implication of the wire cross-section, see e.g. [[AWG]]



http://www.eland.co.uk/cable-genius/1/an-introduction-to-harmonised-cables.html

===Wire colors===

May vary per country. May also have seen a standard change so that houses may be wired old-style.
<!--

Netherlands:
* Blue: Neutral 
* Brown: Phase, a.k.a. Mass
* Black: Switched mass
* Yellow/green: Earth

Note that there may always be country-specific and/or historical variations, and those introduced by semi-knowing DIYers.
-->




===Outdoor power===

In terms of requirements, this comes in roughly three types:
* '''Movable cables''' (on lamps, power points and such that are made to be basically weather resistant)
* '''Fixed underground cabling''' - e.g. to support serious lighting
* '''Fixed overground cabling''' - e.g. to wire up a shed

<!--


Movable cables that go outside should opt for neoprene insulation (and not vinyl or rubber) as it is more weather resistant.


'''Fixed overground/underground cabling'''
(note some of this is Dutch requirements/suggestions):

It is recommended that all connections conform to IP44 or better.

Grounding is important, and a [[ground-fault breaker]] may be required in some cases/areas.

Outdoor cabling may be suited for direct burial, or to be put in protective conduit (may be preferable).

Rules about minimum depths may apply, e.g. 50cm in the Netherlands.

Overground cabling may be suggested/required to be fastened, e.g. every 30-40cm, and <10cm next to connector boxes.


XMvK 
* used for washing machines, overground outside applications
* up to 450/750V
* a little thinner than YMvK
* should not be bundled; preferably use YMVK i

YMvK
* less flammable
* up to 1kV
* more isolation than XMvK, can be bundled


YMvK'''-as''' and XMvK'''-as''' (as == aardscherm) refer to armoring that can be used as an earthing line for protection. (it seems YMvK-as is a lot more commonly sold than XMvK-as).



http://www.acehardware.com/sm-installing-outdoor-wiring-lighting--bg-1284747.html
-->

===Electricity and protection===
{{stub}}


* IP rating[http://en.wikipedia.org/wiki/IP_Code]
* IEC/ISO 60529 (previously IEC/ISO 529)
* DIN 40050
* NEMA rating [http://en.wikipedia.org/wiki/IP_Code#NEMA_Rating]


====IP rating (objects and water)====
[http://en.wikipedia.org/wiki/IP_Code IP code] (International Protection Marking, IEC 60529) is two separate numbers.
'''Roughly,'''


The '''first digit''' describes protection against '''solids''', such as fingers, screwdrivers, and dust
* 0: none
* 1: Protection against objects thicker than 50mm
* 2: Protection against objects thicker than 12mm
* 3: Protection against objects thicker than 2.5mm (fingers, many tools) 
* 4: Protection against objects thicker than 1mm (most tools)
* 5: Protection against large amounts of dust
* 6: Protection against all dust
* X: no information


We often care more about the '''second digit''', the way it deals with '''liquids''' (water and other inert stuff)
* 0: No (added) protection
* 1: Protection against vertically falling drips
* 2: Protection against falling drips up to 15 degrees (tipped device / slanted falling)
* 3: Protection against falling drips up to 60 degrees (most rain), and atomized water (condensation?)
* 4: Protection against drips and sprays from all directions (e.g. good idea around pools - to protect this added object, and if it is fixed to something)
* 5: Protection against a stream from all directions (you can e.g. clean it with a garden hose)
* 6: Protection against strong jets, situations on sea (like 5, but can deal with more force)
* 6K: jets of very high pressure. Rarely used.
* 7: Protection against submersion (...under given conditions of time and pressure/depth)
* 8: Protection against constant submersion (...under given conditions of pressure/depth)
* 9, also 9K{{verify}}: introduced later[https://en.wikipedia.org/wiki/IP_Code#IP69K_and_IPx9], basically for high-pressure, high-temperature washing, and implies 69. Separate standard[https://en.wikipedia.org/wiki/IP_Code#IP69K_and_IPx9].
* X: no information


X is often deliberate, often because it doesn't quite make sense to test or specify in the first place.

You can often assume or estimate what rating something might have. 

For example, 
* wallsockets would typically rate IP2X
* Even cheap bathroom lights are probably IP3X, and probably approximately IP33 in that you'ld have to intentionally spray it before it has trouble.

* many things rated IPX7 are waterproof enough that you can often assume it's IP57 or IP67 (...there's just no tested guarantee).

...assuming that checks out with common sense, e.g. that there are no big moving parts, access ports, or whatnot.


"Can I leave this outdoor" depends on use. For example:

* IP67 usually means 'entirely closed', but you don't always need that

* "Takes a bit of splashing" does not necessarily mean "good to leave where a deep puddle may form"

* Outdoor speakers that you place in the garden would be nice if they were X4, 
: ...yet X1 may be enough if you put them under a patio, since the worst they'll deal with is moist air.



See also:
* NEMA enclosure ratings[https://en.wikipedia.org/wiki/IP_Code#United_States_(NEMA_rating)] (tests different things, so no direct conversions to IP ratings)

==Cable connector types==
See [[Common_plugs_and_connectors#Power_-_wire-to-wire_connectors]]



==Measuring your house's power use==
See [[Electronics project notes/DIY home resource metering]]

==Solar panels==
ElectroVoltaic (EV) panels, that is  -- solar panels can also refer to water heaters.


===Yield and planning===
{{stub}}

You'll want to look at:
* panel peak rating 
* how much sun you'll get (per year, in the darkest part of year, etc.)
* (if necessary) how you're going to bridge the times you have no direct sun, or only diffuse overcast sunlight




A solar panel will only give anywhere near its given current in direct sunlight, so the figure for a panel is often in Wp - Watt-peak[http://en.wikipedia.org/wiki/Watt-peak], the highest Watt output the panel gives, with the panel at 25°C and the sun delivering around 1000W/m2 (its approximate peak, which varies with place on earth, pollution, weather, and more).

Of that rating, you will often see less: 
* up to the rating for ideal situations
* perhaps 70% for a decent but not ideal day and/or setup
* perhaps 30-50% in good sun but when badly oriented, on overcast-yet-still-brightish days, behind reflective glass (e.g. phone charger in the windowsill), etc.
* perhaps 15-30% in dreary overcast weather, or shadows
{{verify}}


When you look at [https://www.google.com/search?q=insolation+maps&tbm=isch insolation maps] and such, you'll often see figures in peak sun hours per day. It looks like most places can count on at least two hours for a yearly average, but relatively few more than five.

For example, in the Netherlands we get around 1000 peak sun hours per year, which is ~2.7 per day, and that's an average - a long-term graph will look roughly like a sine wave with a period of a year, with perhaps 1 in the darkest months (...and that's still an average; seeing a week with less isn't crazy).


If you're powering some electronics on a solar cell and a battery, you want to consider some of the worst cases:
* if you're going to power your device 24 hours per day with perhaps 1 to 2 hours of peak power (and a little more current than that from diffuse or indirect light), that means the panel current should be perhaps eight to twenty times the average current your device uses.

* using the power is lossy. For example:
** inverters (perhaps 80-90% efficient{{verify}})
** regulators (loss depends on voltage drop and regulator type, but again, ~80-90% with some consideration{{verify}}).

* consider losses in battery charging (often involves regulation), although matching panel and battery voltages means this doesn't have to be serious.

* consider short-term variation - you may want a large enough rechargeable battery so that you can deal with perhaps a week of nasty overcast weather and little charging.



'''Orientation''' usually means pointing the thing south on the northern hemisphere, and north on the southern hemisphere -- basically, orientation and inclination should point to the place where the sun is while the peak sun hours happen.


'''Inclination''' matters mostly to direct sun, and you can choose a good fixed value based on your latitude.

For diffuse light, lying the panel flat seems to work best (since that means you're getting light from all of the sky) - in practice meaning ~15-20 so that rain and wind may wash off fallen leaves.



'''Tracking''' (primarily for orientation) is only worth it if you get a lot of sun hours, because only then will a fixed position mean noticeable yield variation within in those hours.
Of course, the occasional manual seasonal adjustment can't hurt, particularly if it's easy enough to walk up to them and move them a bit.



'''Reflectors''' are ''not'' useful for peak sun hours, because solar panels are built for peak sun (~1000W/m2) and more than that will heat the panel, ''lessening'' its efficiency (will noticeably lower its voltage).
Getting it particularly hot may even shorten its lifespan, and for this reason, using reflectors ''may'' void your warranty.

However, reflectors can give you better yield during overcast hours, which can be worth it if you get more overcast hours than peak hours where you live.

You would need to be clever about it, because you want to reflect only diffuse light, and specifically ''avoid'' reflecting direct sunlight onto an already directly lit panel.

Note that a reflector setup may take a lot of area - and that filling that same space with more panels ''may'' give you more power (assuming of course that you can afford that many panels).



<!--

For solar + battery powered projects:
* consider average charge and overage load. For example, if you get 2 peak sun hours where you live, and a constant (24-hour) load, you need to charge the battery at a factor 12 of the load current to balance that. For example, assuming 80mA draw, you would need a panel capable of charging your battery at ~1A (and you may want to multiply that by 1.2 or so to calculate in 20% for various inefficiencies).

* The batter is a buffer. At the ''very'' least it should last it a night of power consumption, but in practice you'll also want it to comfortably last over any expectable period of not-so-good sun. 

* Many batteries last longer if you never discharge them completely. You could perhaps double the capacity you need.
-->



See also:
* http://en.wikipedia.org/wiki/Insolation
* http://www.solar4power.com/solar-power-global-maps.html
* http://www.solar4power.com/solar-power-basics.html

===Panel types===
{{stub}}

* '''polycrystalline''' (currently the most common)
** slightly cheaper than the other options
** ~14-17% conversion rate (efficiency of conversion of sunlight to electricity)
** ~20% efficiency drop in hot panels
<!--** ~120W/m<sup>2</sup> {{verify}}-->

* '''monocrystalline'''
** ~12-15% conversion
** ~15% efficiency drop in hot panels
** a little less W/m<sup>2</sup> than polycrystaline <!-- ~100W/m<sup>2</sup> {{verify}} -->

* '''amorphous''', a.k.a. '''thin film'''
** ~5-7% conversion
** Noticably less W/m<sup>2</sup> of poly/monocrystalline <!--** ~60W/m<sup>2</sup> {{verify}}-->
** little efficiency drop in hot panels
** easier to apply in many surfaces (so easier to spend the space you need)
** combined with a laminate at production, often glass (often plate glass and not tempered{{verify}})


Watt-Peak per cost (yield efficiency) and cost per area are usually the most interesting figures, but it does depend on your situation.

<!--
Price per Watt(-peak) is slowly dropping, with better technology and more production. (Price also tends to be better for larger versions of individual panels) It has come done from ~10EUR/USD not so long ago to hover around USD/EUR 3-7 (variation between types).

The highest efficiency to date is ~24% (at high cost), the lowest cost ~$1USD per Watt (at cost of space efficiency)
-->


<!--
(the sun's irradiation sun is ~1.3kW/m<sup>2</sup>, not all of which is easily used.)
-->





===Wiring===

Wiring panels in series adds voltages,
wiring panels in parallel adds currents.

In both cases, having some panels shaded is a potential problem:
<!--
'''When a panel within a series is shaded''', this limits the current for the whole series (meaning that the output isn't proportional to the shaded/unshaded section - it is primarily limited by the shaded section). This also leads the shaded cell to be reverse biased.

A '''bypass diode''' will provide an alternative path -- shunting out the cell/group the diode has been put across (and only in the reverse-biased direction).
Theoretically you could add a diode to each cell, but this is fairly costly. There is a tradeoff between the diode amount/cost and the amount of power loss.


'''When a panel within a set of parallel panels is shaded''', the other panels will reverse bias in the panels that are delivering less. 

A blocking diode, placed in series with the panel, will stop that from happening, and is the reason you'll often see diodes in arrays of solar panels.

Schottky diodes have very low voltage drop, which makes them attractive to solar panel constructions.

Note that redundant diodes will just waste energy because they drop your voltage unnecessarily. You may want to check whether your panels have the built in.



Note that reverse biasing can also happen in very simple charger circuits, when there is more voltage in the battery than is currently provided by the panel - which will easily happen at night. Many solar charger modules will prevent this. 


-->

===Glass===
{{stub}}

Preferably tempered glass, as it's stronger than regular plate glass, and not so sharp if you manage to break it. 
Plexiglass is even less likely to break (but a little more likely to cloud).


Anything that reflects considerably, or clouds over time, will reduce your efficiency.

=Voltage-related labels=

===Circuit labels related to voltage===

In circuits and specifically on components, you'll see power pins labeled as:

* '''V<sub>CC</sub>''' {{comment|(positive)}} and '''V<sub>EE</sub>''' {{comment|(negative/ground)}} are used with BJT circuits (and also more generally).
: the name seems to come from "the voltage '''c'''ommon to all transistors' '''c'''ollectors" {{verify}}

* '''V<sub>DD</sub>''' {{comment|(positive)}} and '''V<sub>SS</sub>''' {{comment|(negative/ground)}} are used with FET/CMOS circuits

The distinction between these two is blurred in practice. 
You regularly see V<sub>CC</sub> in the general meaning of 'positive supply voltage', regardless of component type, 
and probably see 'Gnd' more than V<sub>EE</sub>.

V<sub>SS</sub> and V<sub>DD</sub>, being newer, have a more marked meaning when they are used.



V+ and V- are also sometimes used, 
as are VS+ and VS−. 
These are even more vague.

<!--
V<sub>CE</sub>, 
V<sub>BE</sub>, 
V<sub>BC</sub> indicate potential difference between the pairs of the relevant ports (B, C, and E) on BJT transistor.
-->

See also:
* http://en.wikipedia.org/wiki/IC_power_supply_pin

=Some notes on delivery of electricity=


Generation is often on the rough order of 10kV{{verify}}.

Transmission over any distance is done at at higher voltages (increasing over time, long distances currently around 1MV).
There are a handful of reasons this, some intertwined, and not all about losses.

City-scale distribution steps it down, often in a few stages. 
Near your house it's kilovolts again.

In your house it's a few hundred volts tops, whether split-phase or three-phase.
<!--
typically either 3-phase 440/400V or around or 3-phase 220/200V.
-->




==Polyphase power==
{{stub}}

<!--
For background, phase-wise the options are:
* One-phase power is a single sine wave

* Split-phase power, technically "single phase, three wire" system
: means single-phase transformer, which is center-tapped.
: e.g. In the US this would be a 240V transformer giving two 120V legs, in Europe it might be a 460 transformer in two 230V legs.
: '''Is not two-phase''' in that phases are meant to refer to separate transformations{{verify}}, and these two come from the same transformation, if center-tapped.
:: But you could call it the same phase and opposite polarity, or 180 degrees out of phase compared to the center tap)
:: and some will think legs=phases and call it 2-phase.
:: People will understand what you mean.
:: Actual two-phase did exist once, but for specific purposes (90 degrees out of phase, which avoided the need of starter motors on industrial electrical motors) and wasn't efficient for general use

* Three-phase power 
: Three transformations, 120 degrees out of phase


<!--
What's more confusing is that which variations you have varies by location.

Where I live, you typically get three-phase power, each 240V to neutral, or sometimes just a single one if your house is small.

You'd be forgiven for thinking the US is all split-phase power (giving 120 and 240V) except for industry, but it turns out that apartments may actually use three-phase for efficiency reasons (and give you two - which amounts to 120V and 208V, which can get confusing, and in a few edge cases actually dangerous)

-->

<!--
'''Some background'''

Polyphase (which are quite often specifically three-phase) systems are often used for high-power transmission, for medium-voltage distribution to higher-desity residential areas, and for (areas with) commercial and industrial buildings.


The main reason to choose three-phase over single-phase ''transport'' is that the cost of materials(/maintenance) and cost of use:

One-phase puts one usable leg on two wires {{comment|([https://en.wikipedia.org/wiki/Single-wire_earth_return single-wire, earth-return] exists but does not scale due to safety issues)}}, while a three-phase system puts three phases on three wires, and because they are out of phase in a fixed way, every ''pair'' can be used.

Additionally, consider that in single-phase there are times of low and of zero transfer (also true of split-phase), whereas in three-phase systems the total current (when the legs are loaded equally) is better spread over time.

These details combine to mean that in comparison to single-phase systems, three-phase can carry the same power with 25% less conductive material, so the material, installation, and maintenance costs are lower.
This is one of the larger reasons for its use in transmission and most of the way in distribution.


The last few kilometers are more interesting, 
because there are various choices you can make between local power station and the building you're wiring.

And, if you're larger than a house, ''in'' the building you're wiring.


Industrial and commercial areas have higher power needs,
and three-phase is the more efficient and flexible way to deliver it.

{{comment|(The three-phase motor typically comes up as an example, partly because it's a nice example of something that loads the phases equally and more constantly, and these motors end up running more smoothly. They are not a highly influential reason by themselves, though - there's not ''that'' many of them around.)}}
(There are other high-powered that are likely to use multiple of the available phases. For example, a sauna heater may have three completely separate heaters)

In industry this is also about wiring. For example, in datacenters it makes more sense to pull through the three phases right to the racks (as thick buses), to make it easier to balance the load with less transformation, bother, and losses.


Something similar goes or residential areas dense enough
to collectively want a bunch of power from a single power station.

High-powered devices that some houses may have, such as a sauna heater or instant water boiler{{verify}} come in variants that use three-phase (or split-phase), though in these cases may use them as independent legs, to neutral.



Sparse residential areas may be 

For sparse rural areas you may even see single-wire ([[https://en.wikipedia.org/wiki/Single-wire_earth_return single-wire/phase-earth-return earth return]]) because the wiring is cheaper, but it can't carry much power due to safety issues. Single-phase is simpler to design for low-power uses and potentially more flexible - and uses less conductor when you go a long distance to just one or a few houses. 
This is one reason single phase (regularly specifically split-phase) is more common in the US.

On the other hand, within cities the difference is smaller, so in planning there often seems to be a "you might as well" bias towards three-phase.

Dense areas have higher power requirements (per house, even)
and you can locally spread the smaller houses between the phases.


Existing infrastructure matters. In the US there are more above-ground power poles than people, and changing what they do would be a very pricy undertaking, so final legs may not be changed until it's necessary.

-->

===Various voltages; wye and delta===
<!--
: '''Various voltages'''

There are a lot of voltages out there.
It's not just (approx) 120, (approx) 230, or sometimes both.

For starters, three-phase means you can get two different voltages depending on which wires you use.
Three voltages, if it's a configuration with a tap.


And then there's larger-scale high-voltage lighting, used in car parks, industry and commerce, and the latter ''on top of'' more regular voltages for some offices.

The higher voltages are more efficient for the longer distances this lighting often involves (consider voltage drop), and safe enough when you can easily keep it away from people.

Mostly you don't see these at your house because there is no reason for it. There is little demand for the higher voltages, and even if there is ''some'',
the typical transofmer is made for large setups, so too pricy to install for just one house.

The US apparently has:
* 120/240 single phase (split-phase), typical in consumer installations
* 120/240 three phase (unusual/rare, does not allow 240V appliances{{verify}}, sometimes seen in apartments?{{verify}})
* 120/208 three-phase wye
* 277/480 three-phase (e.g. in machine shops; 277V is the lowest voltage you can easily get
(one wye leg), so e.g. 277V lighting exists - it lets you sometimes avoid also wiring in 120V)
* 347/600 (industrial, but less common){{verify}}
* 600V delta


208V is the voltage between phases of a 3-phase wye circuit that is 120V from neutral to any hot



: '''Wye and delta'''

Once you care about details, there are a lot of details and variants -- that most of us never have to care about.

...even if you have to deal  with three-phase, because residential three-phase is very typically wye on your end.


The basis is a transformer with three coils, which can be wired in two different ways (Y/Wye/Star and Delta).
Actually, the primary and secondary both can, so there's four basic combinations (Delta-Y being the most common in residental and commercial areas, Y-Y being unstable and rare, Y-Delta useful in high voltage trasmission, Delta-Delta useful in some industrial settings) -- but even more practical variants variants in practice due to the fact that using tapping the secondary coils in different ways makes for further  useful variants.

Since Delta-Y is the most common for residential areas, the output from a Y secondary is probably the only one you may sometimes meet{{verify}}.
If you deal with the other ones you're an electrical engineer and probaby laugh at this page :)


Y means you get three live wires and one neutral.
: If you don't need higher voltage, this is wired into three completely distinct legs in your house (largely because that Y is more efficient when the legs are loaded roughly equally)
: If your needs are smaller, you may just be using one (in theory you could use two of the phases, but the preference is to use three, apparently for the efficiency reason{{verify}})

: some devices will use three phases, but completely separately
:: E.g. a sauna heater can easily switch three heaters. Again, better balanced, plus you avoid putting a much larger load on a single wire/breaker
: You can also get ~400V by using two 230 pairs together (or ~208V from a 120V basis).
:: It's not twice because these these are 120 degrees out of phase (It's sqrt(3), ~1.73)


For example, our house has a power box that says
"Three-phase four-wire, 230V/400V"
Three-phase four-wire implies Y (Three-phase three-wire means delta)

...yet this is not the only way houses are wired. 


In the US three-phase Y means 120V L-N, 208V L-L   (or 110Y190V)

In Europe three-phase Y means 230V L-N to neutral, 400V L-L

The US also sees Y with 277V L-N, 480V L-L, 
e.g. in commercial lighting, large parking lots


-->


===See also===
* http://www.epanorama.net/links/wire_mains.html#usacanada
* http://en.wikipedia.org/wiki/Split-phase_electric_power
* http://en.wikipedia.org/wiki/Three-phase_electric_power
<!--
http://aegispower.com/index.php/blog/179-what-s-the-difference-between-single-phase-and-three-phase-ac-power-supplies

http://www.electrical-contractor.net/forums/ubbthreads.php/topics/140687/Russian_220V_single_phase.html
-->


===Unsorted===


<!--
https://en.wikipedia.org/wiki/Electric_power_distribution




A fourth, neutral wire may be introduced at the local transformer before distribution into houses, {{verify}}
because it allows for more constant voltage when you cannot assume all phases are balanced. {{verify}}

These days, most houses are wired to use one or all three of the locally available phases.
In almost all cases, wallplugs will have the neutral wire and one phase's live wire.


Note that if you try to study this, inconsistent naming gets you very confused for a while.


You can also combine two or three phases for higher voltage -- though this comes with various extra footnotes.

* e.g. in 110V countries to power 220V dryers, etc, or ({{verify}} -- some people sugges this is actually a center-tapped two-phase setup, suggesting that the US is not three-phase)
* in 220V countries to power 380/400V instant water heaters, etc.
: There are two (main) different ways of doing this, each with its own upsides.
: using more than one live phase for the same voltage is is not the same thing, because they are both live, and out of phase. Some devices are designed for it, some don't care, some do.

* hand a device three phases
: see e.g. https://en.wikipedia.org/wiki/IEC_60309
: sometimes used separately, e.g. in sauna heaters, just to spread the load
: sometimes using the phases, e.g. in industrial three-phase motors, which vibrate less (because of the more even power transfer{{verify}})



Implications:
* In general, very few. You plug things in, they work :)

* current-clamp power monitoring:
** Use of a single leg makes monitoring simple (you only need one clamp)
** In the US you may find two different legs in a house, in which case you would monitor both (US: red and black are those two 120V lines, white is neutral)

* Power-line communication
** Often assumes to all work on the same single phase
** Multiple-leg cases are problematic. Could be worked around [http://www.hansotten.com/attersjo/powerDistrubX10.html]

* [[RCD]] wiring - ?


http://www.selfhelpforums.com/showthread.php?t=5646
http://forum.ih8mud.com/chit-chat-section/410188-208v-vs-230v-3-phase.html
-->


<!--
https://www.reddit.com/r/explainlikeimfive/comments/12asui/in_the_us_why_do_we_keep_our_power_lines_above/

http://physics.stackexchange.com/questions/280611/why-do-we-use-ac-for-long-distance-transmission

-->

==Some notes on house wiring==

===Open hot, open neutral, open ground===
<!--
Means said cable is not attached in a particular socket.


'''In cases where all phases are separated''', this is pretty straightforward:

Open hot - the hot wire is not connected
: you'll be getting no electricity on it, so you can't even zap yourself.

Open neutral - the neutral wire is not connected
: meaning that devices won't work (because they sit between hot and neutral)
: you ''could'' zap yourself
: and shorting hot and ground will still flip the breaker

Open ground - the ground wire is not connected.
: most devices will work (except e.g. those that test for ground)
: you're just not protected against
: devices detect this by testing whether neutral and ground conduct (because these are bonded)
: this is a code violation -- a socket with an ground prong ''must'' be ground, as to not mislead you about being protected.
:: and non-grounded sockets are getting rare, so this now mostly applies not extension cords



'''When you use sockets with two live wires''', things are a little different.

In most countries, this is rare in residential installations.

In others it is not uncommon - for example, the US is split-phase 240V,
so each leg is 120V to neutral, but wiring between both legs is 240V.

...while in much of europe, 230V is either 230V to neutral (wye), or 230V between phases (delta).


There is also a distinction between
* two-phase center-tapped transformers, which seems common in e.g. the US
* three-phase transformers



 (e.g. the US's 220V outlets), things are a little different.
This situation is actually fairly rare in many places.


If one of the hots is off (breaker off, or outside your house),
then all devices on that hot will notice, as will 220V devices.

You may find the 220V device influences what the dead leg does.


If the neutral is broken but the two hots are fine, then 220V devices won't notice,
but all others won't work.


One of the hots being off means there is still power

There is no neutral

-->

===Putting your safety ground in the dirt===

<!--
Unlike the earth that's on your devices,
where the point is that it ''can'' carry significant current when something is broken,

The connection between your house's earth and the dirt is
''not'' intended to carry much current.

It's just meant to make sure that your house's earth wire
is about the same potential as the puddle you'll be standing in.


This particular wire
isn't guaranteed to carry much current, and
isn't part of the blow-a-fuse.


It mainly avoids floating grounds - case where different

-->

===Floating neutral versus bonded neutral===
<!--
'''Bonded neutral''' :
* Neutral line and ground line is either at essentially zero resistance because they are tied somewhere closeby
* A '''ground fault''' indicates that this bond isn't found. 

* Bonding neutral to ground should be done at most once in an installation.
: This is typically done at the main power box.
: Doing this more than once means XXX
: your power company has a few options, because XXX


'''Floating neutral''', a.k.a. '''unbonded neutral''' , '''open neutral''':
* will, to devices, look the same as an open ground
* an RCD won't work{{verify}}


Perhaps easiest to expa
It's relevant in everyday life e.g. when you use a generator,
because many generators do not make that bond
(in case they're wired into a house?{{verify}})
and ground will be
devices that check for a ground, or want a ground, will not get one


Larger generators tend to be neutral-ground bonded,
smaller inverter generators are not.
(why?)


Consider a distribution transformer, or generator, going to your hourse.
If the neutral line breaks, it is no longer connected to ground,


Three phase systems may or may not have a neutral wire.
* in houses you typically want one, using each phase for a separate leg in your house ()

* Having one allows use of higher voltages via multiple phases, ''while'' still supporting single phase appliances.

* If you ''only'' care about high voltages, you don't need the neutral
: it also helps if you can ensire the phases are loaded more or less equally (which at larger scale is easier)




-->


==On zapping people==
See [[Electronics notes/Electricity and humans]]

=Power supply considerations=

<!--

==Switch mode or not?==

Pretty much all larger power supplies are switch mode, 
because it's more efficient than the transformer-rectification


The main exception is probably audio equipment.


There are some odd cases, like the [[capacitive dropper]]



-->



==Power conditioning==

<!--

First, some terms:
* surge - a 2+ second of voltage higher than the specced range
* sag - similar length of too little voltage
* spikes - ''significantly'' more voltage for a short time (often just a fraction of a second)


Computer PSUs and other switch-mode power supplies
can handle surges and sags to a degree, and that degree is larger when it is ''not'' working near its capacity.
This comes from 

So yes, overspeccing your PSU somewhat is implicitly surge protection.



Spikes are nastier.

They happen fast enough (microseconds) that your circuit breaker (tens of milliseconds) will not catch them.

Surge protectors that also do some spike protection usually have [[MOV]]s (Metal-Oxide Varistors) which work fine but wear out.

Cheap surge protectors



Of course, what people think of is lightning.
Which falls very much under spike.



(in a matter of microseconds, against the tens of milliseconds a circuit breaker takes).



One thing to realize is that if it's close enough, that spike will ''not'' be handled.

If it's a few blocks down and dissipates in part via the power net,



http://www.dansdata.com/gz039.htm






http://www.dansdata.com/gz039.htm


http://www.sto-p.com/pfp/pfp-transients.htm

http://www.eetimes.com/document.asp?doc_id=1273228


-->



=On mains voltage=

====Why all the different mains voltages?====
<!--
Early days, US service voltage was once around 90V or 100V, and was raised over time to 110V 

100V, 115V, 117V and 120V are restandardizations, and can be considered to be the same thing. 
: when someone in the US says 110V or 115V, they mean "the standard that has said nominal 120V for most of my life"

Similar for 220V, 230V, 240V, and other closeby values.
: e.g. europe is nominally 230V. ... -6% to +10%, which includes 220 and 240V. 


The US is nominally 120V, and has been since the 50s (1954, see ANSI C84.1 {{verify}}) 
: that is, 120 VAC volts &#xB1;10%, which means 108 to 132 V is in spec
: and places seem to aim for around 117V delivered {{verify}}


Japan is roughly the only country with 100V{{verify}}, and also a mix of 50 and 60Hz.
This seems to originate in having various localized grids, different areas initially importing different systems, 
doing so from different places, and never standardizing probably because it doesn't ''quite'' matter enough for the amount of bother and cost.

You'll find switches include 100V on some older audio audio equipment, because the Japanese are pretty good at designing and making those.



Note that what ends up at your socket will be a little lower than what is supplied by your local transformer,
and it is likely to vary a few volts over the terms of minutes or longer.

Not by much, but this and more is part of what makes all of these figures 






There are more voltages if you combine different phases.
...and that varies by country too. E.g. the US sees a lot more split-phase than three-phase power, while three-phase power is common in a lot of other places.

Values that come from phase combinations have similar variation, e.g. 206 and 208 from the various voltages arond 120V, 




[[Switch-mode power supplies]] can adapt to input voltage, sometimes intentionally including both voltage levels (see e.g. many laptop adapters being labeled "100-250V").

Older and simpler devices run on different voltages will generally just run the device a few percent hotter and, particularly if it was basically a heater to start with (including incandescent lightbulbs) may well burn out a little faster.



--> 


====Is the US 110 or 120V?====
<!--

The US has nationally been at a nominal '''120V''' &#xB1;5% for a long time now (1954, see ANSI C84.1).
: (...actually it provides 240V, but safety-wise the split-phase setup means touching any ''single'' wire would never expose you to more than half that, which is actually fairly clever design)


Before then, things were more interesting.

Early days, the the power grid was used for street lighting, where exact voltage didn't matter so much.


Once domestic electricity was a thing, apparently they started aiming for 110V, so that 100V would make it to most houses {{verify}}.
That was later raised to 115 (1930s?) - not sure why or how quickly.

Neither of which seemed to be national standards, just widely understood conventions{{verify}},
and both were common long enough that people still refer to it that way,
and that rating on various things (like larger motors, or general load calculatings) might lag.



More technically, most service is 120/240, split-phase.

There is also some 120/208, 
based on three-phase transformers, and houses served this way will typically get two legs of that (balanced among more houses).

Devices expecting 240V (AC, driers, etc.) will now only get 208V - but most are actually designed to take either, so that's fine.

-->

====Is europe 220, 230, or 240V?====

<!--
Yes.



That's a smart-ass answer, but less than you'd think.


Historically, mainly the UK was 240V, and many of the mainland European countries were 220V.

So now it's unified, all having moved to 230V?

Not exactly. It was 'harmonised' to 230V around 2003, meaning it was ''specified'' to be 230V -6% to +10%.

Doing the math, that's 216 to 253 V, and you'll note that both 220V and 240V fall in that range. 

Which means that no infrastructure had to change in the short term.

You ''could'' choose to change it to 230V at the next scheduled replacement, but you don't even need to.
The country I'm in was previously 220V, and right now there's 232V



'''So is this all a numbers-on-paper game?'''

Partly. 
It is expected that we end up with a more 230V over time, and various 220V countries have gone to 230V. 
The UK is for a large part (but not fully) still 240V.


One real difference is that devices aimed at Europe now have a slight better-defined range to design for (230V AC -6% to +10% = 216 to 253) - before that, it was maybe 220 &#xB1; something or 240 &#xB1; something, meaning the possible range used to be larger and less known, and it was a bit more "design for the country you're in, and point at the footnotes if it's sold internationally" {{verify}}

The other implication is that consumers no longer have to worry 
about the difference when buying devices.


It's the common sense long term fix.

And the reason why, say, it's perfectly fine that a bunch of the UK is still at 240V.


Also, any of these voltages were never precise for practical reasons,
like that conductors aren't perfect,
that it fluctuates with load,
and seem to err on the few-volts-over side,
presumably to account for unpredictad dips.

Say, sticking a multimeter in my socket usually tells me 234V in my 230V country, while 240V in the UK it's often around 245V. 



For many things, this doesn't matter anyway.

For example, switch-mode power supplies are often made to deal with 100-240V (labeled, practically will work even a little outside that), in part because you can make it that way, and having fewer designs makes it easier to certify.

For things like lightbulbs, it can matter.
For example, an old 220V incandescent bulb run on 240V will burn a little brighter and not last quite as long (and that curve isn't linear).



https://www.leadsdirect.co.uk/knowledge-base/what-is-the-difference-between-uk-voltage-and-european-voltage/
-->

====Isn't 230V twice as dangerous as 120V?====
<!--

Since current is the real danger, 
and twice the voltage on the same human means twice the current,
it is certainly more dangerous.


Yet 'twice as dangerous' suggests a lot of assumptions that you probably don't want to be making,
for some actually quite practical reasons, see [[Electronics_notes/Electricity_and_humans#Is_120V_safer_than_230V.3F]].


If you really cared about safe-to-touch electricity, we need to go below 50V or so.

But the problem with lower voltages is that they are inefficient. 
Half the reason Europe tried and stuck with 200+ volts was more efficient use of wires.


-->

====How stable is that voltage?====

<!--
It will wander a few volts, but rarely more.


The following plot is a once-per-second sampling for three hours, from a random socket (not the board), in a country that aims for 230V:

[[File:230V.png]]

(a few hours of samples)

Note that there can be more variation between houses,
because 


Most devices don't care much. 
Everything that does switch-mode power (which is a lot of things these days) are designed to accept a wider range in the first place.

Incandescent bulbs might burn out a little faster when it's 245 rather than 230V, 
but we don't really use incandescent anymore.


-->





====Why don't we use DC?====
<!--



'''Transmission'''

For various reasons, you want AC for longer distances. 
AC is more flexible, and generally more efficient, than DC (...certainly the low-voltage DC Edison was using - it could only reasonably do a few km)


'''Isn't AC more dangerous?'''

Sort of. 

But once you adhere to electrical code, not so much.

Sure, Edison tried to associate his rival's AC with death (elephants, electric chairs).
What he mumbled in the fine print is that the equivalent power of DC would do roughly the same. 
Not exactly, but 



'''But wait, what about in the house?'''

Initially, AC or DC it barely mattered. 
Motors can be made for both AC and DC.  

These days, most devices in the house run on DC,
so have to rectify AC first, a somewhat lossy process.

And we have inverters to make AC when necessary, so wouldn't it make sense for local power distribution to do DC?

...sort of, yes. There are purposes where this would be convenient, but it doesn't scale so well, and needs thicker wires for the same power{{verify}}.



-->