Ups output voltage anomaly

[sbwattam]

I have been looking at an active UPS system in a lab feeding a small consumer unit and carrying out a periodic inspection of the fixed wiring (much help on avoiding problems with Zs values thanks to other posts on here). I did however come across an anomaly on the actual output voltages under its various operating modes.
Under the internal bypass mode the voltage was 230v to earth, 230v live to neutral and 0v neutral to earth as was the reading when the UPS was in its normal mode (ac to rectified dc to inverter to output), however when running on battery under supply fail the live to neutral voltage was still 230v but live to earth was 160v and neutral to earth was 60v.
Has anyone else come across this before?

 

[cossie400]

What instrument were you using to record the voltages?

You need to use a True RMS voltmeter when testing UPS voltage when it is in "battery mode", if you are using a non True RMS voltmeter it cannot read the voltage properly. But if it is a True RMS voltmeter, it will read all the voltage points and would tell you that the voltage is.

 

[sbwattam]

I was using a megger bmm2500 insulation / continuity tester on voltage setting but also a Fluke 1653b multitester, a fluke clamp meter and a dilog voltage indicator! I did wonder if it might be an anomaly on one of the testers, so tried the others however the readings were pretty much identical on all meters. Not sure if any of these are true rms. The unit is an Xtreme power conversion NXTi unit and is a double conversion unit so unless it is in bypass mode then the output inverter is being supplied with rectified dc normally or batteries under power fail but why the difference under battery mode?

 

[Golden_Boy]

I have been looking at an active UPS system in a lab feeding a small consumer unit and carrying out a periodic inspection of the fixed wiring (much help on avoiding problems with Zs values thanks to other posts on here). I did however come across an anomaly on the actual output voltages under its various operating modes.
Under the internal bypass mode the voltage was 230v to earth, 230v live to neutral and 0v neutral to earth as was the reading when the UPS was in its normal mode (ac to rectified dc to inverter to output), however when running on battery under supply fail the live to neutral voltage was still 230v but live to earth was 160v and neutral to earth was 60v.
Has anyone else come across this before?

have you access to an oscilloscope? have a look at the output waveforms. maybe a clue there ?

 

[UKMeterman]

I suspect that in battery mode it no longer refers the output to the supply neutral, thus the L-N voltage is correct and the L-E and N-E voltage ratios are only due to stray capaciance, or other leakage paths. It might be interesting to see what happens with an AVO 8 on this job...!

 

[sbwattam]

Many thanks for all your assistance guys. I've managed to get to the bottom of it after a useful conversation with a guy in the technical department for the UK representatives. He told me that this unit and many others still use the supply neutral as a reference point even when it is in battery mode. Unfortunately when testing it in battery mode I was using the double pole isolator to simulate load failure so was also removing the neutral reference at the same time causing the voltage to float in relation to earth.
He advised me that they always recommend that the supply neutral to the UPS is always connected through and that the supply neutral should only be isolated when maintenance is being carried out on the unit and never if the unit is supplying load in battery mode. Would have been helpful if the installation manual contained this information!