UPS Shut Off During Power Failure

[Anonymous user]

We had a power failure today totaling about 30 seconds. I don't believe one of the UPSes stayed on during the failure. One of the servers connected to it said that it had been shutdown unexpectedly. The other one had some hard drive errors which we fixed by pulling the drive and doing a chkdsk.

I don't believe this UPS stayed up during the power failure. We have 3 UPSes and all have the AP9617 management card. A UPS that stayed up has the following in the log:

01/23/2009 08:12:42 UPS: No longer on battery power.
01/23/2009 08:12:08 UPS: On battery power in response to an input power problem.

The one that didn't stay on has the following:

01/23/2009 08:13:17 UPS: Restored the local network management interface-to-UPS communication.
01/23/2009 08:12:54 System: Network service started. System IP is from manually configured settings.
01/23/2009 08:12:54 System: Warmstart.

The unit is a SUA750RM2U. It only has about a 60% load. What can I do further to test this unit to figure out why it lost power in a short outage? Can I trust this device anymore? Thanks.

[Anonymous user]

I have a similar issue with a SmartUPS 10000XL

When there is a power failure or self test the power is never transferred out to the PDU's The Unit itself shows that all is well and passes the self test but the Step Down transformer is not showing that there is any load on the unit. The PDU's however show the amps being pulled. Any ideas?

[BillP]

a warmstart indicates the card did not lose power but did reboot. do you have any logs from powerchute network shutdown clients configured on this UPS?

i would suggest doing a pull plug test on the UPS with a non critical load so you can observe the behavior of the UPS to make sure it transfers to battery.

this just means to have a non critical load on the UPS, then pull the plug from AC power, and see if the UPS goes to battery.

if it does, then we can investigate further on your configuration. also, please confirm you have no other USB or serial cables attached to the UPS.

[Anonymous user]

The log from PCNS is:

01/23/2009 08:14:08 Communication has been established
01/23/2009 08:14:05 PowerChute Network Shutdown version 2.2.1 started
12/12/2008 23:20:02 Communication has been established
12/12/2008 23:19:22 PowerChute Network Shutdown version 2.2.1 started


There is nothing about loosing power or a indication of a shutdown.

Using event viewer (The system log) on the same system says:

The previous system shutdown at 8:11:05 AM on 1/23/2009 was unexpected.

For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.

There are a bunch of entries after it about starting services as the machine booted again.








There are no USB or Serial cables attached the unit. Only power cables and a single network cable. I'll do the plug test this weekend on it and let you know the results. Thanks.

Jonathan

[Anonymous user]

I'm planning to do the pull test tonight. However last night I got a lot of emails from the unit saying:


01/24/2009 05:17:51 UPS: The battery is now installed.
01/24/2009 05:17:34 UPS: The battery is not installed properly.
01/24/2009 05:16:25 UPS: The battery is now installed.
01/24/2009 05:16:08 UPS: The battery is not installed properly.
01/24/2009 05:13:52 UPS: The battery is now installed.
01/24/2009 05:13:35 UPS: The battery is not installed properly.


Before I do the pull test I'll see if I can pull the battery out and double check to make sure it is connected well.

[Anonymous user]

I ran the pull test with 3 computers without hard disks plugged to the UPS. 3 out of 5 of the green lights were lit for the load indicator.


Anyway, the unit completely shut off after about 5 seconds without power. No lights, no beeps, nothing.


I checked the voltage on the battery before I ran the test and it came out to 27.5V on the volt meter.


So my question is the battery bad or is the UPS bad?

[BillP · Accepted solution]

i had a similar issue the other day where it turned out to be the battery.

during a self test, the unit runs on battery and compares the voltage drop to a table inside its microprocessor. if that voltage drop is normal, the self test passes. if not, the UPS illuminates the replace battery LED.

it sounds like the voltage drop is normal but the actual battery capacity is shot - meaning the runtime.

are you able to perform a self test on the battery with this load, does it do the same thing and drop the load after 5 seconds? (versus simulating a power outage?)

how old is this battery inside the UPS? based on what you find out, it just sounds like you need a new battery to me.