Limitations of Recommended UPS Usage

Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View

Those who insist that Consumer UPSs produce sine wave output, please check
your egos at the following doors:


All others would be well advised to listen to the manufacturer's warning and
NOT use UPSs for anything but computer equipment.

Re: Limitations of Recommended UPS Usage

You are correct and the warning is appropriate, but .....

It is still perfectly fine to use UPS' with many (many, not all)
non-computer loads.

Generally, anything with a switching power supply is fine.  Any load
that is resistive is fine.

After that, things are more gray.

Billsey wrote:
Quoted text here. Click to load it

Re: Limitations of Recommended UPS Usage

On 3/25/2010 2:45 AM, Billsey wrote:
Quoted text here. Click to load it

What incident triggered your decision to post this advice? I am yet to
have any difficulty feeding APC UPSs output to my video equipment,
ranging from Reel Tape recorder, Cassette records, Amplifiers, VCRs and
TVs (I live in an area where the power is frequently restored less than
a second after a cut; very effective at destroying equipment). Of
course, this also includes computers, ranging for full size desktops to

I am not as qualified as Barry, but I believe that any equipment that
incorporates a regular stepdown transformer should be able to handle
this kind of output. Square wave would be a different story, of course.

Of course, the yellow flag "Use caution, Wet floor" left everywhere in
hotels to protect somes might be adequate, but until you specify
in which circumstances an equipment fed with UPS output failed, this
kind of post does not appear very helpful to me.

John Doue

Re: Limitations of Recommended UPS Usage

John Doue typed on Thu, 25 Mar 2010 20:39:34 +0200:
Quoted text here. Click to load it

Hi John... there was old advice to be very careful with UPS which puts
out square waves. As theoretically anything with a coil or transformer
can kick out an inductive kick during such a cycle. This creates a
voltage in the opposite direction and can be very harmful if the circuit
isn't protected. There are things as voltage and current phases which I
won't get into here. But the truth is laid out below.

That is theory though and nothing happens at two states at a time. As a
square wave implies that at one moment in time, the voltage toggles
completely differently and changes polarity. Although in reality, two
things can't happen at the same time. As if you slice the time to
smaller chunks in time, there is a sharp increase in the other
direction, but there isn't really anything as a true square wave. It
maybe far quicker than the duty cycle and all (assuming 50 to 60 hertz
here), but there still is a ramp up or down regardless (regardless of
the hertz).

Quoted text here. Click to load it

I too ran many things off of UPS and I haven't had a problem at all. But
there are two kinds of UPS and I run the kind that doesn't change
anything unless the power cuts out. Then the UPS kicks in. Some others
run the UPS all of the time and get its power from the battery alone
(although the AC charges the battery). Mine allows the AC through and
doesn't do anything except charge the battery and if the AC cuts out,
then the UPS kicks in.

The kind I have isn't the most reliable. So don't look towards one. As
it switches modes when the power drops off. If it doesn't happen very
fast, everything is fine. But if the power comes and goes within a cycle
or two, the UPS gets confused and can drop power to the computer long
enough to loose everything. So it isn't the best, but it depends a lot
on your AC source. I think I have been burned twice in 5 years. So that
is acceptable to me. Anything higher and it probably wouldn't be.

Quoted text here. Click to load it

Speaking as an electronic engineer. anything with a transformer almost
always has rectifiers and filters. And they should be able to handle any
inductive kick unless the tolerance is so close. But the tolerance is
supposed to be 10% over norm anyway. And it might be pushing close to
the limits, but most everything should be okay anyway. Only poorly
designed supplies might be in trouble.

In practice though, I have been watching the Internet for any signs that
this is a real problem and I can't find anything very serious. There are
some reports that the transformer gets about 10% warmer on square waves
than it does with sine waves, but that is all. And I don't doubt this at
all and I would expect it.

Quoted text here. Click to load it

It is true that transformer supplies would take the most beating from
square wave sources. In practice though, only very tiny number of them
ever fail. And those that do, I bet they would have failed later anyway
under normal AC operation. As the tolerance was so close anyway.

So the scare of square wave UPS, is founded on a tiny bit of truth. But
the real truth is that it pushes a transformer a little bit more and
that is all (everything else should be okay). And I have seen very few
cases where this would be a problem. But they were poorly designed from
the start and are truly rare.

Gateway MX6124 ('06 era) 1 of 3 - Windows XP SP2

Re: Limitations of Recommended UPS Usage

John Doue wrote:

Quoted text here. Click to load it

Contrary to what some self-proclaimed experts will tell you, there is a
reason for it when a company that WANTS YOUR MONEY tells you NOT to use
their equipment in a certain application. Usually it means that the damage
to your equipment is going to cost them more money than what they would gain
by selling you the equipment. If you care so little about your equipment
that the manufacturer's own warning doesn't bother you, then I can't help
you anyway.

In addition, if you don't mind the effect that a consumer-level UPS will
have on the output of your audio- or video-equipment, then you really don't
care about audio or video. The photos shown were taken of the backs of UPSs
that pass my way at work. The harmonic distortions shown are at least 100
times what should be used with audio or video equipment. Now, of course, if
you bought your equipment at someplace like K-Mart, or Walmart, or BestBuy,
then don't worry about it because that level of equipment isn't worth the
bother; on the other hand, if you only purchase audio- or video-phile
equipment, then you're wasting your money on a UPS because that level of
equipment is already built right in the first place.

For my own part, I have UPSs at home, that I use with my computer equipment;
with my audio equipment, though, I use an ISOLATION TRANSFORMER—not a UPS.

Re: Limitations of Recommended UPS Usage

On 3/26/2010 12:57 PM, Billsey wrote:
Quoted text here. Click to load it

I do not believe that the way you expose your view post is conducive to
a useful and interesting discussion. Bill's post on the other hand
brings technical facts. I would not demean equipment based on where it
is bought, although I would not expect top quality from some of the
outlets you cite ...

FWIW, I have a TEAC reel tape-recorder and 2 Nakamichi cassette
recorders, among other things.

Then, I am confused when you state;  "you're wasting your money on a UPS
because that level of equipment is already built right in the first
place". UPS are not built-in in any consumer product I know. You seem to
forget that, unless I am way off, a UPS is essentially a product that
protects equipment, to some extent, against power losses.

If you lived where I live, it would not take long before you decide to
use UPSs for most low power equipments. I am talking about almost dayly
cuts, ranging for milliseconds I do not notice to a few seconds.

The consumer APC UPS I use do not kick in unless the power is way out of
its acceptable range, or is cut.
So, the point you make about sound or image quality - which is probably
valid - is moot AFAIAC: I prefer degraded quality for a few seconds than
no sound at all. And I prefer continuous usage than off-on power to my
equipment, which has always been professed to be a no-no.

Protection against this on-off thing might be built-in some equipment
(is this what you meant?) but how do I tell until it is too late?

At the end of the day, if you have very expensive equipment, I agree
that isolation transformer is probably the way to go, continuous usage
taking a back side to protection.

John Doue

Re: Limitations of Recommended UPS Usage

Billsey typed on Fri, 26 Mar 2010 05:57:07 -0500:
Quoted text here. Click to load it

And what do you think you are gaining by using an isolation transformer?
As an isolation transformer will be more than happy to transfer any
frequency say above 45 hertz or so. And the biggest danger from
equipment running off of AC isn't from 45 hertz or lower, but from those
high spikes on the power line. Sometimes hundreds and even thousands of
volts peak. And an isolation transformer won't stop any of them and will
let them right on through.

The only good thing about an isolation transformer for audio equipment
would be any ground loop would be gone (if the isolation transformer
isn't grounded on the output and/or the equipment isn't either).
Otherwise you are not gaining anything really.

Gateway MX6124 ('06 era) 1 of 3 - Windows XP SP2

Re: Limitations of Recommended UPS Usage

The advice to at least be cautious is good.

I teach PC Repair, Upgrading & Maint. at a local college and I am an
engineer who has worked for power supply companies (which also made UPS').

The output of MOST of the CONSUMER grade UPS's is a discontinuous square
wave.  For one of my classes, I connected an oscilloscope to a UPS (APC)
output and I was kind of shocked at what I saw (I knew it wasn't a sine
wave, but I was expecting a better approximation).

I can't give good guidance on what you can safely use it with, beyond
saying that switching power supplies and resistive loads should be ok.

To your comment "Square wave would be a different story, of course" ...
THAT IS WHAT IT IS, except that between the positive and negative halves
of the square wave there is a period of NO output.  In other words, it's
a Positive square wave pulse, then nothing then a negative square wave
pulse, then nothing .... repeated.

Not what I was expecting, but that's what it is.  There are only 3
levels; zero, positive and negative.

John Doue wrote:
Quoted text here. Click to load it

Re: Limitations of Recommended UPS Usage

On 3/27/2010 12:24 AM, Barry Watzman wrote:
Quoted text here. Click to load it
Hi Barry,

Could you post somewhere a copy of the output you saw from an APC UPS?
Have you tested this way several models? If as you say - and I do not
doubt you - it is squarely (forgive the pun, too tempting) a square wave
with a no output in between, I am shocked too.

As most people here I guess, I did not wait for Billsey post to read the
warnings of UPS manufacturers: they have been there for as long as I can
remember. But one point bothers me:

What is the fundamental difference between a computer (laptops excluded
for obvious reasons) and most consumer electronic equipments which would
make UPSs acceptable for the former and not for the latter? Computers do
have motors (hard disk, motorized trays) as VCR, tape recorders and CD
players do , and LCD displays, as TVs do. That point escapes me, but
admittedly, I do not have or claim to have any in-depth knowledge in
electronics beyond what I learnt, almost centuries ago! I would very
much appreciate if you took the time to explain this, since this would
go a long way complementing common sense and avoiding blinding
compliance to self-protecting manufacturers mentions.

I personally never have had any problem with any of my various audio and
video equipment that could be even remotely related to a UPS (I only use
APC UPSs), and I have never had people I know report any such problem.
Since usual consumer grade UPSs only provide power in case of sudden
loss, and then for a very limited time, there would have to be a severe
incompatibility to cause damage.

One other way to look at the issue might be that, UPS manufacturers do
not bother to improve the quality of their outputs ... knowing perfectly
well that it is *acceptable* for most equipments. The improvement of
output quality would not be cost effective considering they would still
not be able to claim perfect compatibility with any equipment, which
would be foolish in any case.
John Doue

Re: Limitations of Recommended UPS Usage

John Doue typed on Sat, 27 Mar 2010 12:07:41 +0200:
Quoted text here. Click to load it

Hi John! No this is a very common output from inverters (UPS uses
inverters) to convert DC into AC. As it is easy a cheap to build. I know
they were using this back in the 70's when I was getting my EE degree.
And that waveform that Barry speaks of has a name and I can't remember
what it is called right now.

And without getting too technical, the reason why it rest at 0 volts 120
times a second (assuming 60 hertz here). Is because 120V@60 hertz also
passes through zero 120 times a second. And you might say so what? Well
transformers and motors, etc. gets a big break around 70% of the time.
So what does this mean? Well current causes heat. Feed it a true square
wave and you have current flow virtually 100% of the time. Thus it
generates heat at nearly 100% of the time.

But transformers, motors, etc. are not designed to handle current
without any breaks at all. Well you can, but that doesn't help all of
the transformers, motors, etc. already out there in use. Thus creating a
wave like what Barry saw, keeps the temperatures pretty close to what
they would be if used on standard AC.

And this works fairly well instead of a pure sine wave. Works for
transformers and motors. Works for rectifiers and filters, switching
power supplies, lamps, etc. Frankly, I can't think of anything that runs
on AC that wouldn't work okay with this.

And while what Barry had seen is what is commonly called a stepped
square wave (probably incorrectly). And I know of no inverter (or UPS)
that actually generates a true square wave output. If they do exist,
they would be especially bad for transformers and motors. As nearly 100%
current flow would cause a lot more heat and their life would be much
shorter. Although there is a way around this too. That is if the square
wave output voltage was lowered by 30% or more. This would keep the heat
down, but it wouldn't work well for devices that needs the full amount
of voltage to operate. Motors for example might not even start up.

Quoted text here. Click to load it

I believe I explain this above without getting too technical. But the
motors and such that you are talking about in computers, tape equipment,
etc. are actually running on DC and not on AC. As the AC hits the power
supply and the power supply converts it to one or more DC voltages.
There are some odd exceptions and once in awhile saying like a very old
tape deck might actually run the motor off of AC and not from DC. Or the
panel lights are actually 120 volt bulbs or something, but it is rarely
done this way anymore.

Quoted text here. Click to load it

Well there are zillions on devices that run off of AC power. So I can
see some putting out a warning that the wave forms are different and
there is a small chance that an UPS can hurt a very small number of
devices out there. And has I stated above, if there is no rest time in
the output, transformers, motors, and even lamps doesn't get that
current break 120 times a second. Well lamps increases resistance the
hotter they get, so they should be okay. But transformers and AC motors
for sure won't get that time to cool down. Thus they just would get
hotter and hotter and at some point burn out.

Quoted text here. Click to load it

Well UPS and inverter manufactures sometimes use a disclaimer don't use
for other devices. And other devices sometimes have a disclaimer saying
not to use them with inverters or UPS as well. All I can say is there is
no set standard output for inverters and UPS. As they can use whatever
waveform they wanted too. So I can see manufactures getting a little
nervous about this, can't you?

And if you or anybody else gets nervous, I would be checking the
temperature of the device you are worried about. And to be especially
watchful of the temp of motors and transformers. As these would likely
have the most problems. But the waveform that Barry had seen have been
shown to work just fine (people have been using it for decades without
problems). If you want to get fancy, some UPS and inverter manufactures
use tiny stepped square wave output which looks and acts virtually just
like a real sine wave output. But I wouldn't bother unless you have some
device that absolutely won't work without a real sine wave output. And
while there are zillions of things out there that runs off of AC, I
can't think of one actually that would have a problem. But I can't say
100%, but it has to be very close to 100%.

Gateway MX6124 ('06 era) 1 of 3 - Windows XP SP2

Re: Limitations of Recommended UPS Usage

On 3/27/2010 9:20 PM, BillW50 wrote:
Quoted text here. Click to load it
Thanks Bill for those detailed explanations which brought me some +50
years back, when I learnt the basics, and a little more, of electricity.
I am not sure about the English terminology, but I do know the
"efficient" (literally from the French) current intensity for a perfect
sine wave is only a fraction of what you would get simply by applying
nominal voltage (110 in the US) to the impedance of a device. The
difference between the two is what would cause transformers, resistors
and such to overheat is the square wave was perfectly ... square!

I imagine that the output of UPSs has been designed so as to closely
approximate the "efficient" current I was referring to. In some case,
approximating might not be enough, ok, but at least in this forum,
nobody has come forward saying they have experienced of problem. Of
course, this is a little statistical value! I also realize the
imperfection of the wave might be (at least, theoretically) detrimental
to sound and video quality if the filtering behind the power supply
leaves to be desired.

I do not know what prompted Billsey to start this thread. May be he
discovered those warnings, may be he just intended to create some
animation here! Anyway, I gained some additional knowledge, so AFAIAC,
this was worth the bandwidth!


John Doue

Re: Limitations of Recommended UPS Usage

I have oscilloscope photographs as JPEG files.  Send me an E-Mail and I
will send them to you.  It's a square wave with "zero-volts space"
between the positive and negative pulses.  Although I can appreciate
your interest in seeing them ... and I don't have any problem with that
... you should be able to draw it on your own from that description.
And yes, it was an APC (brand) UPS.

[BTW, although it's not my photo, the image at:

is similar to what I saw, although on my UPS, which WAS loaded (with a
100 watt light bulb), the positive and negative square wave pulses were
wider, and there was less "zero voltage" interval in between them.]

As to "What is the fundamental difference between a computer (laptops
excluded for obvious reasons) and most consumer electronic equipments
which would make UPSs acceptable for the former and not for the latter?"

Computers (at least modern ones ... although not my 1970's Sol-20's and
IMSAI's) are run exclusively from switching power supplies.  Switching
power supplies don't give much of a damn as to what you feed them, many
of them would even work from DC.  And while, as you say, "computers do
have motors" ... those are run from the DC output of the switching power
supply.  NOTHING except the power supply itself runs from the incoming
AC line.  That is true of SOME consumer electronics, but other products
have different kinds of power supplies that may not tolerate the square
wave output of a UPS (the operative words here are "MAY not").  I have
no idea how a transformer reacts to such an input, and also many
consumer electronics have devices and/or circuits that are run, in one
way or another, directly from the incoming AC power line.

Of course there ARE sine-wave UPS'; but you will pay more for them ... A
LOT MORE (we are talking about multiplying the total price by integers
here) ... than for square wave UPS (which the UPS manufacturers have
other names for, however).


John Doue wrote:

Quoted text here. Click to load it

Re: Limitations of Recommended UPS Usage

On 3/28/2010 1:02 AM, Barry Watzman wrote:
Quoted text here. Click to load it

Thanks, Barry; I now have a better idea of the shape of the wave from
your description. Thanks for the URL. A picture is decidedly worth
thousand words ...

I think we have covered much of the ground here. I understand that the
equipements using, at least partly, directly AC might have problems, but
I guess that they are becoming a disappearing species.

As opposed to UPS, Pure inverters do cause failures and the warnings on
them are very specific and clear. Any attempt to use them do result in
almost instant failure. But their fonction is totally different.

So I believe, your explanations combined with Bill's make it clear that,
even if the shape of the wave is surprising at first look, the its
design has been proven valid. I also believe that people who own very
special and expensive equipments, which by their conception could be
harmed by UPSs, are probably technically inclined enough to assess the
pros and cons of using UPS.

Of course, it would nice to have a clear cut situation but a modicum of
common sense, education ... and as a last resort, experience helps a lot
avoid dangerous situations.

John Doue

Re: Limitations of Recommended UPS Usage

Re: "As opposed to UPS, Pure inverters do cause failures and ..."

I don't think that there is any difference between what you call a "pure
inverter" and a typical UPS.  A UPS is just a packaging of a number of
components in one device:

-Battery charger
-Surge suppressor & noise filter
-Power fail detect and Transfer switch
-[usually] Computer status reporting & communication

I think that you are suggesting a distinction between the inverter in a
UPS and a general purpose inverter (cigarette lighter to 120v
receptacles) that, for the most part, simply does not exist.  [Does not
apply to true sine-wave UPS', of course.]

[Obviously, there are hundreds of manufacturers of both UPS' and
Inverters, and tens if not hundreds of thousands of designs, so any
discussion of this is going to be a generalization that doesn't apply to
absolutely all devices.]

John Doue wrote:

Quoted text here. Click to load it

Site Timeline