Square waves from UPS into transformer increase heating?

The harmonics due to these little jumps are going to be very weak. How many pulses are being used to define the sine wave? I assumed

100 pulses and calculated the MSE between the pulsed sine wave and an ideal sine wave over a single cycle to yield roughly 6.5e-4. I don't think a transformer is going to heat up due to 0.065% of the input waveform energy being in higher order harmonics, at least for the example of the 100 pulse sine wave.

If your laptop is like mine, the output from the ac adapter is dc, not ac. My adapter is a Toshiba Model PA245OU with 100-240 vac input and 15 vdc output. While the adapter may run a little hotter with a high harmonic content source like your UPS, I doubt if this will have a significant effect on the life. If you have a good battery on your laptop, I see no need to use a UPS, however.

Highly unlikely it is the tiny harmonics doing this.

More likely it is your line voltage is low while the UPS always puts out

117-120 volts.

A quick measurement of the line voltage versus the UPS output voltage will confirm or refute.

However, when I plug it from my APC brand UPS, I am

Any waveform other than sinewave will stress transformers & other parts in the power supply more (thus extra heat generated). UL 1778 is the standard which applies to UPS systems, and for that reason, no longer than 30 minutes of operation in stepped wave (modified squarewave) is allowed under UL 1778.

This 30 minute limit is somewhat arbitrary since some power supplies will work fine for hours whereas other power supplies might be damaged even after 5 minutes. So, my recommendation would be to just operate it on UPS for 1 minute or so for orderly shutdown and wait until utility AC is restored. Then, UPS will send the utility power (which is sinewave AC) to your power supply & everything will be OK.

To run any device for over 1 or 2 minutes without overheating (& thus damaging the power supply in the equipment), you need to use UPS with Sinewave Output Inverter - so that you get sinewave output even if UPS is running on battery.

Yes, it is a well known phenomena. It doesn't have to be that way, but to save money, space and weight, power transformers are run very close to saturation. The few extra volt seconds in the square wave may go ahead and saturate the core. At that point, it is only the winding resistance that limits current and you get a nasty current spike through the primary winding.

Thanks to all for their replies. However, you don't all agree, and most of the others are saying that there is a noticeable effect. One says it's from the UPS giving out a bit more voltage than I get from the socket, some others say it is indeed the stepping in some way. I'm not sure how you calculated exactly, but doesn't it matter how fast the step rises, not just how many steps there are? I'm not sure how to evaluate the theory and opinions, but the bottom line for me is, I am sure the transformer gets hotter.

On Wed, 21 Jan 2004 16:14:36 +0000 (UTC), snipped-for-privacy@specsol-spam-sux.com Gave us:

If they were indeed tiny.

That shouldn't change the behavior of a device plugged into the UPS. Otherwise, the UPS is out-of-spec.

That would or could confirm that he is feeding a different voltage to the module. This WOULD be a factor. If the UPS put out more voltage than the line, the module would likely waste more than at nominal or low line conditions.

Besides, it isn't the UPS transformer getting hot, it is the small laptop module that he states is heating more. Input line sags would not cause this, if as you stated, the UPS is always supplying 120VAC or whatever constant voltage..

Also, the harmonic content of square waves through a transformer made for 50 - 60Hz is not "tiny".

Which they are in a modern UPS.

Say the line voltage is 109 and the UPS output is 119 just for arguement sake. Why would you not think a device plugged into the UPS is dissipating more power than the same device plugged into the line?

That is the whole point.

Sigh. The module gets hotter plugged into the UPS than it does plugged into the line. When your are plugged into the line, you see the sags.

True, but modern UPSs don't put out square waves, they put out stepped sines.

There is no way to tell for sure without some measurements and/or experiments.

If you really want to know, measure the temperature rise. Put an AC line filter between the UPS and the transformer and measure the temperature rise again.

Measure the line voltage and the UPS output voltage. Get a variac and put it between the transformer and the line. Measure the temperature rise at nominal line voltage and at nominal UPS voltage.

Report back with your findings.

Doing this as a calculation that has any validity would require a hell of a lot more information, and equipment to get that information, than the simple experiment outlined above.

On Thu, 22 Jan 2004 02:10:13 +0000 (UTC), snipped-for-privacy@specsol-spam-sux.com Gave us:

In your own words, you state that the UPS voltage is constant, and refer to the line voltage being low, not the UPS output voltage.

UPSs by their nature, put out a constant AC output voltage from fed-thru-ac power to full-on-battery power, all the way down to the low line spec for the battery voltage as it discharges. That entire time, the output AC voltage won't move more than 1.2 volts.

The device dissipates the same power on it's output side either way. The question is what are the costs of operation.

At low line, the device should usually be the least lossy (read most efficient), so your argument fails.

If the UPS puts out a higher voltage than the line does, the device might be warmer on the UPS.

On Thu, 22 Jan 2004 02:10:13 +0000 (UTC), snipped-for-privacy@specsol-spam-sux.com Gave us:

What frequency does it get quantisized / step modulated at?

Yes.

Yes.

Yes.

How?

Exactly the point.

The device gets hotter plugged into the UPS than it does plugged into the line.

Now do you get it?

Depends on the UPS model and who made it, you tell me.

In general, a bit higher than 60 Hz.

I think this is one place I might have 1 up on this subject....

Its depend... On type of UPS.. Switching UPS that uses a offline inverter use generally a SCR or triac switches that directly feed the load side with a bit of a filtered output.

Best Product and Sola brands are the system most often that use the offline method.. They have a output inductor that is resonant with the line freq. This allows the inductor to maintain output voltage for the one or two cycles it takes it switch the inverter on and transfer the load to the battery powered inverter.

In this method is very cost efficient requiring no power to run the inverter until required.. But it means that the inverter Must run at really close to input line voltage because you would see a voltage and current spike when the system switched from line to inverter.

The second method for solid state UPS is isolated input where the input voltage is rectified to DC ... That voltage is tapped to trickle charge the batteries and contentiously run a inverter that supplies the load...

This type ( Exide, Deltec,Leibert, Seimens ect.) have no switching time because the battery plant is parallel with the Input DC.. And care's little where the DC source comes from...

This means you can have a very high input voltage ( smaller wires , cheaper input systems like breakers ect) and have a much lower output voltage.. Say a 480vac in and a

120vac out.. The two voltages have no relationship ...

There are also Rotary UPS ( General Power, PSE and some others) Where there is a AC motor and DC motor and a Generator... These also isolate the input from the output ( even completely electrically isolating ) That's why the are preferred for secure high value data centers because data can't be captured from the RF feed back of the data being process out on the power grid.

Paul Mays

By the time the load see it... Its a sine wave.. at least in most systems.. the output inductor filter and cap banks smooth out the step waves that are produced but the inverter which most are pulse width modulated.

Yup... The faster the step freqency the more steps per full wave at 60 or 50 hertz ... So the more steps the less filtering required to clean up the wave form... So you can have any freq ya want..

So has to be much higher at 60hz you would get a square wave at 120hz you get 1 step each 45deg of rotation of the wave form and so on.....

On Thu, 22 Jan 2004 03:32:09 +0000 (UTC), snipped-for-privacy@specsol-spam-sux.com Gave us:

Look, ASSWIPE, you said that fluctuating line voltage would change the module temp plugged into the UPS.

It will not. When will you get it?

On Thu, 22 Jan 2004 03:44:17 +0000 (UTC), snipped-for-privacy@specsol-spam-sux.com Gave us:

Like duh, ya friggin dope.

What I said was the module sees one voltage plugged into the line and another plugged into the UPS.

Learn to read.

This is patently untrue. I'm sure you can find me devices that really do object to so-called 'modified' sine waves, and I can imagine the UL has enough spare time on their hands that they write up standards that are useful for certain applications, but I can also guarantee you that there's tons of really cheap inverters out there making pretty nasty looking pseudo-sine waves that's used to power lots of dirt cheap equipment, and the vast majority of the time nothing 'bad' happens.

Although you probably could convince me that statistically there might be some impact on the life time of the equipment... but still, for many applications, a true sinewave output inverter isn't worth the (significant) additional cost.

---Joel Kolstad