UPS voltage reading

[fixit7]

I have a UPS for my computer.

With no power supplied to the battery backup so it operates only with it's internal battery, the output voltage showed around 80 volts when I was expecting around 115 volts AC.

Is that because I do not have an RMS meter?

[ejeffrey]

That also means your UPS has a square wave output rather than modified sine wave.  Modified sine wave inverters use three levels (+V, 0, -V) to maintain the same RMS and peak value as a true sine wave, so they will read approximately correctly with either a peak detection or true-RMS multimeter.

Where did you find that relic?  I hvaen't seen a square wave inverter as part of a UPS... ever.  I thought they only ever showed up in super janky "car camping" inverters, not any sort of UPS.

[fixit7]

Is that because I do not have an RMS meter?

You just answered your question.

A peak-detection multimeter infers RMS voltage by its peak value, so if it detects 170V, it says 120V.

Your UPS outputs 120V peak square wave, so the peak is only 120V, so the meter reads 120V*(120/170)=85V.

I do not understand your second sentence.

When I measure my wall current it is around 115 volts.

My computer runs on 115 volts. I don't think it can run on 85 Volts.

[fixit7]

That also means your UPS has a square wave output rather than modified sine wave.  Modified sine wave inverters use three levels (+V, 0, -V) to maintain the same RMS and peak value as a true sine wave, so they will read approximately correctly with either a peak detection or true-RMS multimeter.

Where did you find that relic?  I hvaen't seen a square wave inverter as part of a UPS... ever.  I thought they only ever showed up in super janky "car camping" inverters, not any sort of UPS.

It is not a relic as it is only about 2 yrs. old.

Are you saying that other UPSes would should 115 volts on my meter?

Bought it a Frys.

I am thinking of getting a better meter. How would I know if it is a true RMS meter.

I was looking at this one.

https://lygte-info.dk/review/DMMUNI-T%20UT191T%20UK.html

[wraper]

Is that because I do not have an RMS meter?

You just answered your question.

A peak-detection multimeter infers RMS voltage by its peak value, so if it detects 170V, it says 120V.

Your UPS outputs 120V peak square wave, so the peak is only 120V, so the meter reads 120V*(120/170)=85V.

Actually nope. Non true RMS multimeters measure average rectified value (ARV), not peak value. Nonetheless they don't properly measure modified sine. But relation between peak and measured value you wrote is wrong.

That also means your UPS has a square wave output rather than modified sine wave.  Modified sine wave inverters use three levels (+V, 0, -V) to maintain the same RMS and peak value as a true sine wave, so they will read approximately correctly with either a peak detection or true-RMS multimeter.

Nope it does not mean that. Because of what's written above. BTW modified sine UPS does not really maintain proper peak voltage. It drops a lot with increased load and battery discharge, so UPS increases duty cycle to compensate for that. With such waveform change, true RMS multimeter should maintain about the same output voltage readings but ARV multimeter readings should significantly change (given that modified sine UPS was properly designed).

[bob91343]

The problem is indeed related to waveform.  The computer, at least if it's somewhat recently made, uses a switching power supply derived from a rectifier off the power line.  So worrying about power line waveform being sinusoidal isn't the important thing.  The wave needs to be such that the required 170 Volts or so comes out of the filter to the switching circuit.  Many kinds of waves will produce that but unless you have a UPS designed specifically for recent computer power supply circuits, you can't do much.

If you are really concerned, especially if the computer seems not to operate properly, you will need to measure the rectified voltage inside the computer power supply and verify that it's within specification.

In addition, since a UPS is a switching device, it will generate noise that may interfere with computer operation or, for that matter, other devices in the vicinity.  Unless the UPS feature is vital to your operation, I'd suggest you remove it from the system until such time as it's needed (power line issues).  Of course that would mean the computer shuts off or something and you could lose information, but for most people it's not an important thing.

[wraper]

In addition, since a UPS is a switching device, it will generate noise that may interfere with computer operation or, for that matter, other devices in the vicinity.  Unless the UPS feature is vital to your operation, I'd suggest you remove it from the system until such time as it's needed (power line issues).  Of course that would mean the computer shuts off or something and you could lose information, but for most people it's not an important thing.

Rather strange suggestion. If there is no mains voltage loss, they either work in bypass mode or connect load through transformer to increase/decrease mains voltage. No switching is happening. Only expensive online UPS with proper sine output can work all of the time. But even those have bypass mode which you can turn on/off.

[bob91343]

Sure if they function that way it's fine.  But I don't know that the cheaper ones do.

In any case, a UPS is, in my opinion, a solution to a problem that occurs so rarely these days that it's not really needed.  Like surge suppressors.  And cellphones that only call 911.  Most of us are Eskimos who don't need refrigerators.  My computer freezes and has to be rebooted perhaps once a week - my last power outage was over a year ago and lasted about 20 seconds.

Pardon my irreverence; much is tongue in cheek.

[james_s]

I've had a number of "modified sine wave" inverters and with every one of them the output was a square wave with some dead time. Maybe some use multiple levels but none that I've owned. The RMS voltage is 120V but it's not a sine wave so a meter that is not true-RMS will not read the right voltage.

Lower cost averaging meters assume a 50-60Hz sine wave so if you try to measure something else the meter will lie to you. A true-RMS meter will correctly measure the RMS voltage which is what the load sees.

[james_s]

Sure if they function that way it's fine.  But I don't know that the cheaper ones do.

In any case, a UPS is, in my opinion, a solution to a problem that occurs so rarely these days that it's not really needed.  Like surge suppressors.  And cellphones that only call 911.  Most of us are Eskimos who don't need refrigerators.  My computer freezes and has to be rebooted perhaps once a week - my last power outage was over a year ago and lasted about 20 seconds.

Pardon my irreverence; much is tongue in cheek.

That depends on where you are. Where I am there are a ton of trees and every time a wind storm rolls through in the winter there are numerous momentary power glitches, some of them just long enough to reboot a PC. When I was using a desktop as my primary machine I had a UPS for exactly this sort of situation. Now that I'm mostly using laptops it doesn't matter because the laptop effectively has it's own built in UPS.

If your computer freezes and has to be rebooted once a week there's something wrong. I reboot mine on average once every 4-8 months and most of the time that's to install something that requires a reboot.

[windsmurf]

Interesting look at the various waveforms generated by UPS systems

https://www.hardwareinsights.com/database-of-ups-output-waveforms/

[bob91343]

Yes there is definitely something wrong with my computer that freezes.  But I have no idea what or what to do about it.  It's an occasional minor nuisance and has been acting this way for years so I presume it won't get worse.

It's not a power issue, as it might even do it while I am typing a message.  So I hit the power switch, wait the obligatory couple of minutes, and then it asks me if I want to start from scratch or from where I left off.  It doesn't save my incomplete message but otherwise it's not been a serious problem.

I have considered designing and building a UPS but after giving it some thought I decided the need for a battery makes it impractical.  For the same reason I don't use cordless tools; just when I need them I find the battery has expired.  I have about three of them that all need new batteries but my power line driven tools still work every time.  Even my cellphone battery gets run down after a full day of being off the charger.

[wraper]

Yes there is definitely something wrong with my computer that freezes.  But I have no idea what or what to do about it.  It's an occasional minor nuisance and has been acting this way for years so I presume it won't get worse.

Try running RAM tests, if nothing wrong shows up, just try setting lower memory clock if bios allows it.

[bob91343]

This is off topic but I will try what you suggest if possible.