UPS 3000 XLM strange behaviour

[Anonymous user]

Hello,

APC SMART UPS 3 KVA 1phase in / 1 phase out with add. 48V battery pack. (APC-SUM3000RMXLI2U +  2 * APC-SUM48RMXLBP2U) . Purchased and installed  in 2013 on remote unmanned site. External power is very unstable.  After some time we noticed the following situation (taken from SNMP messages  log) :

Sep 12 15:26:50 UPS: On battery power in response to an input power problem. 0x0109

Sep 12 15:26:52 UPS: No longer on battery power. 0x010A

Sep 12 15:27:52 UPS: In bypass in response to an internal hardware fault. 0x011B

Sep 12 15:27:53 UPS: The internal battery temperature exceeds the critical threshold. 0x012C

Sep 12 15:27:54 UPS: Refused a self-test; UPS is overloaded. 0x0106

at the same time diagnostic showed

  Internal temperature: 27° C,

  Enviroment temperature: 24° C

  UPS Load: 12%, 1,7A

  Input voltage: 228V

  Input frequency: 50Hz

After switching to bypass diagnostic had been showing continuous discharge of batteries. In one hour  UPS turned off and site was   deenergized.      When we arrived on site and manualy swithed UPS on it returned to normal operation but in several days the situation repeated. Question : is it 100% hardware problem or could be some solutions?

Thanks in advance and look forward for any help!

[BillP]

It sounds like you have a battery problem. Are the UPS internal batteries and external battery pack batteries all the same age? If so, it sounds like they are failing. If they are different ages (which isn't good since they should all be replaced at the same time), then we'd need to narrow down the culprit. A failing battery can damage good batteries on the same bus though.

The "refused a self test; UPS is overloaded" message can also indicate failing batteries that overload the inverter when on battery.

Can you provide the full event.txt, data.txt and config.ini files? Instructions are here -> How can I download Event, Data, Configuration, and Debug files from my Network Management Card? | FA...

Anyway, I'd probably also visually inspect all batteries and beware they may be warm to the touch if they are damaged and/failing. There is also an event in the log about a high internal temperature. If you find problems, it might be best to disconnect all batteries from the system to prevent them from charging further and causing more problems.

[Anonymous user]

Hello,

Thanks for quick answer! But I don't think that its the battery problem. All batteries and the UPS are new. Installation and setup was made about two month ago.

We tryed to restart UPS (both through WEB interface and through SNMP) - no any effect, it simply refused to restart. We restarted management card - it's rebooted, but no any affect again. In some time the UPS turned output power off and we've lost connection because a network equipment is powered from that UPS .

When our staff arrived on site, the UPS was in turned off state. The UPS was turned on by power button and showed that it worked on batteries, at the same time, through WEB interface, it showed that the UPS charged the battery, which is strange. After that, power button was pressed again and UPS went on main power.

Our staff checked all batteries and they seemed are OK - no any damages and its voltage is OK too. So they cleaned all electrical contacts and checked all mounting and connection, reassembled all UPS parts, fully charged batteries and producedrecalibration. Setup the sensitivity to Low. Now we are watching this UPS. In the near future we are planing to install automatic voltage regulator with hope that it help to make work of equipment more stable.

I've attached log files and config.ini, may be they will be helpful. As I can see - after the UPS had turned off the output power - it worked and collected information about input power, but it didn't turned  the output power on because it thought that it had some fault. The only fault it had (by log)  - "The internal battery temperature exceeds the critical threshold", but on UPS diagnostic - internal temperature never does not exceed 28 degress of C. So it is intresting - how does the UPS determine battery temperature?

Sorry for long post, hope that it gives some clues for finding solution to our problem.

[BillP · Accepted solution]

Can you tell me any specific things you want me to look at in the event log which corresponds to what you did and what happened to the UPS on its own? I see a lot of on battery events earlier in the log.

I see the battery voltage on the unit very low and it seemed to be because the UPS was in bypass at some point around 8/21 but we don't have a data log that far back. I also see runtime calibrations started - this should only be done when you have a problem with the reported runtime - is someone issuing runtime calibrations?

A couple of comments:

• When you restart the management card, it has nothing to do with restarting the UPS. You'd need to do that under the UPS control section of the management card to reboot the output. To do a full restart on the UPS and restart it completely, you'll need to braindead it. To braindead, you turn the attached load off and turn the UPS off. Remove any cables attached (which if using a management card, there should be no USB or serial cable attached). Then unplug the UPS from AC power and hold down the off (o) button on the front of the UPS until the UPS clicks and/or flashes its LEDs to discharge completely. Then you can reconnect it to AC power, turn it back on, and plug your attached load back in.
• I need to better understand what events in the log are caused by you or your team versus which that you believe the UPS was logging on its own.
• I think its possible that the battery temperature (versus the UPS internal temperature) is getting too hot and maybe the UPS is putting itself in bypass to prevent further charging and damage to the battery. It could be that the battery is just faulty from the factory and could require a warranty replacement.
• Someone or something (10.32.3.4 and 10.32.1.19) keeps trying to do runtime calibrations and this is concerning. Why are we doing that??