Recently, I've noticed a trend in power supply manufacturers touting their PSUs (generator or battery inverter etc.) as having a pure sine wave output.

I've also seen people saying that warranties will be invalidated if devices/motor homes etc. are connected to anything other that a power source with a pure sine wave output.

I wonder what the world did before such power sources existed.

Is there science behind this? Surely a standard petrol generator with a good automatic voltage regulator (AVR) or an old-fashioned coil regulator will be enough to stabilize the output to run sensitive electronics like LCD televisions or computers?

  • \$\begingroup\$ Since all electrical items have a start up surge, wouldn't this suggest that this equipment is able to handle voltage spikes, most likely with inbuilt regulators suited to the said equipment? \$\endgroup\$
    – user46041
    Jun 22 '14 at 3:46
  • 2
    \$\begingroup\$ There's a great answer on SU with oscillograms of several UPSes, detailed discussion and examples of devices which perform poorly when connected to a cheap UPS. \$\endgroup\$ Oct 2 '15 at 22:35

Historically, inverters (electronic circuits that take DC power and convert it to AC to simulate the power line) were pretty awful in the waveshapes they produced.

Early inverters produced little better than square waves. This means they included significant power at frequencies that devices were not designed to handle. Most devices that are intended to plug into wall power take the sine shape of the voltage for granted. Some might count on the peaks of the sine being a particular voltage, while others count on the RMS. For a sine wave, the peaks are at \$\sqrt{2}\$ times the RMS, whereas for a square wave the peak and RMS are the same. This presents a problem in deciding what voltage square wave to produce. If you match the power line in RMS, then lightbulbs, toasters, and other "dumb" devices will largely work. However, electronic devices that full wave rectify the line will see a significantly lower voltage. If you raise the square wave voltage, then you might overdrive and damage devices that use the RMS.

The extra harmonics in the square wave can also cause problems on their own. Transformers designed for the power line frequency, like 60 Hz, might not deal well with the higher frequencies. Or these frequencies might cause extra current and heating without them being harnessed for more power. The sharp transitions can also overload electronics that is expecting a maximum slope from the power voltage. For example, just a simple capacitor accross the AC line would in theory conduct infinite current if the voltage changed infinitely quickly.

The next step in inverters was "modified sine", which had a extra ground "step" in the square wave. The point here is that this reduces the power in the harmonics relative to a full square wave. However, many of the problems with square waves were still present, although generally reduced.

Modern electronics that can efficiently switch at many times the power line frequency can produce a output voltage that is pretty close to a sine, meaning it has little harmonic content. This eliminates the issues with square wave and modified sine outputs, since the power line itself is ideally a sine. It is still a bit more expensive to produce inverters with sine wave outputs, but the extra cost is no longer that much and is getting steadily lower. Today, sine wave output inverters are common.

Note that inverters intended to drive the power line backwards, called grid-tie inverters, are all sine wave output. This is due to a lot of regulations covering what you are allowed to do with the power line, especially when you feed power backwards.

  • \$\begingroup\$ Note that the harmonic content of inverters that switch faster is not necessarily less from the offset, but rather shifted to higher frequencies according to the modulation index. Your typical analog filter is very good at attenuating these frequencies, not so good at the low frequencies characteristic of square and slow modified sine wave PWM inverters. \$\endgroup\$
    – RYS
    Oct 2 '15 at 23:25

Inverters that have a "modified sine wave" output can put additional stress on some devices.

enter image description here

(the diagram shows 50Hz waveforms, for 60Hz, the cycle would be 16\$ 2\over 3\$ms rather than 20ms.

I don't think that caveat is meant to apply to non-electronic sources of power such as conventional (non-inverter type) generators.

The blue waveform in the above diagram is what is called (in marketing-speak) a "modified sine wave" (as marked) and is what most inexpensive inverters produce. It has the desirable (or even essential) characteristics that the RMS value and the peak values are the same as a sine wave, so a peak-sensitive device like a switching power supply for a CFL sees the same voltage as if it was a sine wave, and an RMS-sensitive device such as an incandescent light bulb or a heater sees the same voltage as if it was a sine wave.

The down side is that things that are sensitive to rate of change of voltage (capacitors, for sure, and maybe others) see a much higher dv/dt than with a sine wave. That can cause additional stress.

In my (limited) experience, it's more likely to manifest itself as a requirement to derate the inverter (iow you might have to use an inverter rated for much more wattage than the load requirements or it would shut down) than to cause actual damage to the load.

  • 1
    \$\begingroup\$ Did you mean "modified square wave" in your first sentence (to match the figure)? \$\endgroup\$
    – The Photon
    Mar 27 '14 at 16:05
  • 2
    \$\begingroup\$ Hey, Photon, I don't make up the marketing-speak. \$\endgroup\$ Mar 27 '14 at 16:09
  • \$\begingroup\$ OK, then what's the difference between "modified sine wave" that you talk about in text, and "modified square wave" that's shown in the figure? Can you add a figure that shows what's meant by "modified sine wave"? \$\endgroup\$
    – The Photon
    Mar 27 '14 at 16:10
  • 2
    \$\begingroup\$ "Marketing speak" calls the three-step wave (blue in Spehro's drawing) "modified sine wave", but it looks much more like a modified square wave to me. \$\endgroup\$ Mar 27 '14 at 16:15
  • 1
    \$\begingroup\$ What @PeterBennett said. I'll edit to make it clear. \$\endgroup\$ Mar 27 '14 at 17:43

Petrol generators usually produce pretty good sine waveforms, it's just the frequency and amplitude that can be off.

What's of more concern are the square wave and stepped-sine UPS designs. Some older (and even some new) power factor corrector front ends could not track these waveforms and would not operate properly. They contain lots of harmonic content that can cause currents to flow that were not designed for in the original design, so it can be a problem for even non-PFC designs.

Most quality manufacturers today test with these types of waveforms, but some may still specify sine only.

  • 2
    \$\begingroup\$ The decent sine wave from a gas generator derives from the fact that it is actually spinning an armature coil in a magnetic field at an appropriate(ish) frequency, not unlike the spinning generators at the coal, oil, hydro or nuclear power plant down the road, right? \$\endgroup\$ Mar 27 '14 at 17:29
  • 1
    \$\begingroup\$ Could you give more details -- perhaps a link -- to these PFCs that allegedly can't handle square or stepped waveforms? \$\endgroup\$
    – davidcary
    Mar 27 '14 at 17:51
  • \$\begingroup\$ @Craig That's a good point, the analogue mechanism leads to a 'natural' sine-wave. This stack question came about because a friend was told to purchase a very expensive pure sine-wave inverter generator because his gas/petrol one would damage electrical equipment (that converts straight to DC anyhow). \$\endgroup\$ Mar 27 '14 at 20:11
  • \$\begingroup\$ @davidcary Since a PFC is a boost converter, the boost choke is often bypassed by a diode (in the inrush direction) so that the inductor doesn't saturate. Depending on how they do inrush management this can cause an overload on some UPS devices: apc.com/site/support/index.cfm/faq/index.cfm This condition is more due to the transfer to the modified sine rather than an always on UPS. A PFC is also trying to track the input voltage and force the input current to match and be in phase with the voltage. With a low frequency sine wave that's relatively easy. For sharp edges it's not. \$\endgroup\$
    – John D
    Mar 27 '14 at 21:40
  • 1
    \$\begingroup\$ @Craig: I would expect that with today's electronics, letting the generator turn at variable speed and generating the frequency electronically would be more efficient than using a mechanical constant-speed drive; a mechanical system might be more robust in the presence of an electrically-charged atmosphere, but I wouldn't expect consumer-level generators to use a constant-speed gearbox. \$\endgroup\$
    – supercat
    Mar 28 '14 at 18:12

See the attached picture of my WattsUP Pro ES meter when connected to the output of an old APC BackUPS Pro 650 unit for about 30 seconds.

WattsUp Pro Es burnt by connecting it to a UPS with stepped sine-wave output

I had changed the battery and was trying to figure out if the battery runtime reported by the software was consistent with what I would get. The load was about 20-25 Watts. The meter was reading 179 Volts RMS (nominal voltage is 230V here) and I was wondering why. Then smoke came out and I immediately unplugged everything.

The meter is still functional and the 10 Ohm SMD resistors still read 10 Ohms but the plastic on four of the resistors have melted, along with some of the closed-up holes (probably these were included because through-the-hole parts were used in one revision of the product).

I have made two mistakes and learned two things in the process:

  1. I incorrectly remembered the APC models with "Pro" in the name were units with true sine-wave output. Apparently that was not the case (and I verified this by finding the manual afterwards).
  2. The meter is not true RMS. Read this note by APC regarding details. If I had read this earlier, I would probably disconnect the meter when I saw 179V RMS readout. This would not be because I would expect a meltdown but because I would understand that the meter would not be correctly measuring the voltage and current (i.e. it would be useless for what I was trying to do).

I wrote this answer in response to comments asking for examples of "catastrophic failure". BTW, the unit still powers an Ethernet switch, a media player and a 32-inch LED TV fine.


Historically and generally, it is a bad idea to use a cheap genset for expensive electronic equipment. This is because cheap gensets have poor voltage regulation. Apart from sparking, that wasn't really about the shape of the waveform.

High voltages can destroy your expensive electronic equipment, low voltages will destroy your refrigerator, and rapid voltage changes can give you random results on your expensive electronic equipment.

But modern electronic equipment is often much less sensitive to this problem than old electronic equipment: if your plug pack can take 70 to 250V AC input, it's not going to be bothered by the supply voltage going up from 110 to 135.

The alternative to using a cheap genset is to use a properly regulated AC power supply. In modern terms, any properly regulated PSU will have pure sinewave output, and any PSU that doesn't have pure sine wave output will be so ancient, or so cheap and nasty, that it won't be properly regulated.

In the old days, you started with the expectation that in the medium term you would destroy your electrical or electronic equipment by hanging it off a typical poorly regulated generator. Also, your light bulbs wouldn't last very long. But at least they were cheap to replace.

And by the way, those big coal or nuclear power stations output pure sine waves because they are carefully wound to do that. If they weren't so careful, they'd get distorted waveforms too. The distorted waveforms wouldn't bother you, but the power supply companies would loose substations.


I also have heard people saying that anything other than a "pure sine wave" will allegedly cause "noise" or "unwanted harmonics" or "stress" or unspecified "damage". Like you, I am skeptical about the alleged benefits of pure sine waves.

With only one exception so far (the problems with the "CR dropper" technique), every device I've seen so far will run just as well and in some cases better with a simple square-wave inverter than with a pure sine wave inverter.

When I look at the local mains power with an o'scope, the actual waveform is pretty far from a perfect sine. And so these hypothetical devices that allegedly won't work on anything other than a near-perfect sine won't work when plugged into any real wall outlet I've ever seen.

  • 5
    \$\begingroup\$ The problems with square waves are not all theoretical. Consider a synchronous spinning motor. Of all the frequency content you drive it with, only the fundamental will make the motor spin. The harmonics still cause current, and therefore heat due to the resistance of the coils, but do nothing to move the motor. Only 81% of the power of a square wave is in the fundamental. To make a square wave spin the motor with the same torque, you have to put 23% more power in, with all that extra going to heat. That can be quite significant. \$\endgroup\$ Mar 27 '14 at 18:15
  • \$\begingroup\$ The tl;dr is that sensitive digital equipment, old digital equipment, or anything analogue will experience anything varying from a minor decrease in efficiency to catastrophic failure from a poorly shaped wave--though usually it is the former; actual damage isn't very common with household or commodity stuff. \$\endgroup\$
    – user39062
    Mar 27 '14 at 20:20
  • \$\begingroup\$ @OlinLathrop: my understanding is that at high frequencies, including high harmonic frequencies, most of the energy going into a coil such as a coil of a synchronous motor is later returned to the AC power grid -- it's not all dissipated as heat. I agree that harmonic currents will make the motor warmer than if it had hypothetically been fed a perfect sine wave. It is not clear to me if the ugly harmonics I see on the actual waveform at my wall outlets are significantly better or worse for motors, etc. than a square wave. \$\endgroup\$
    – davidcary
    Mar 28 '14 at 4:14
  • \$\begingroup\$ @user39062: I would be fascinated to find out more details of this vague "equipment" that allegedly experiences "decreased efficiency" or "catastrophic failure". Since I design electronic devices, I would like to know the details of how things could go wrong, so I can avoid mundane repeats of those failures and instead experience new and more exciting kinds of failure :-). Please tell us (perhaps in a separate answer to the original question) any details you have. \$\endgroup\$
    – davidcary
    Mar 28 '14 at 4:23
  • \$\begingroup\$ The harmonics into the synchronous motor cause less current at higher frequencies due to the inductance, but any current caused by the harmonics times the DC resistance of the coils is wasted power. Equipment that starts with a full wave bridge will see considerably less voltage from a square wave since that square wave is usually adjusted to the same RMS as the sine to keep from blowing out light bulbs and the like. \$\endgroup\$ Mar 28 '14 at 12:39

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