Nut upsuser - Feb 2009 - APC SMART-UPS 1500 always reporting LB

Using nut v2.2.2, and the APC serial cable.

I have a client who's APC SMART-UPS 1500 (circa 2000) is always reporting
LB. The batteries were old so they were replaced two weeks ago with brand new
batteries. UPS still reported low batteries...

I ran a calibration and a test on the unit (using the nut upscmd utility)

During those tests, the battery charge on the unit goes to zero and slowly
starts making its way back up 100%. Then, the "LB" ups.status will
come and go every few hours until it finally stays on. All battery leds on the
unit are flashing also indicating that the UPS thinks there is a low battery
condition - which corresponds to the ups.status "LB" reported by nut.

Here's the problem: The load on the unit is only reported as 27% by nut, and
when I pull the AP power on the UPS, the system runs on batteries for ... well
currently it has already been 50 minutes and the battery voltage is only down to
23.28 from about 27.x when I pulled the AC.   With upsmon running, they system
will immediately shuts down do to the "OB" and "LB"
ups.status condition so I needed to disable upsmon during this test.

Also, I did the "braindead" thing to the UPS that I found in the APC
forum. Not sure what help this was.

Here is a current upsc output: (on batteries for about 50 mins now)

--[snip]--
# upsc smartups at localhost
battery.alarm.threshold: 0
battery.charge: 000.0
battery.charge.restart: 00
battery.date: 09/20/00
battery.packs: 000
battery.runtime: 0
battery.runtime.low: 120
battery.voltage: 23.15
battery.voltage.nominal: 024
driver.name: apcsmart
driver.parameter.pollinterval: 2
driver.parameter.port: /dev/ttyS0
driver.version: 2.2.2
driver.version.internal: 1.99.8
input.frequency: 60.00
input.quality: 00
input.sensitivity: H
input.transfer.high: 132
input.transfer.low: 103
input.transfer.reason: T
input.voltage: 000.0
input.voltage.maximum: 000.0
input.voltage.minimum: 000.0
output.voltage: 119.0
output.voltage.nominal: 115
ups.delay.shutdown: 020
ups.delay.start: 000
ups.firmware: 70.11.D
ups.id: UPS_IDEN
ups.load: 028.6
ups.mfr: APC
ups.mfr.date: 09/20/00
ups.model: SMART-UPS 1400
ups.serial: XXXXXXXXXXXXX
ups.status: OB LB
ups.temperature: 013.9
ups.test.interval: 1209600
ups.test.result: NO
--[snip]--

Could the APC SPMART-UPS 1500 be defective? What other tests can I perform on
it? What other information can I provide that might be helpful?

Thanks for any help or ideas. 

Oh!  And thanks for NUT... Great utility, especially in some multi-server
environments I have it running in!

--
Bill Arlofski
Reverse Polarity, LLC

l?rdag 28 februari 2009 20:37:44 skrev  waa-nut at revpol.com:
Hi Bill,
8<------------------snip-------------------------
> Here is a current upsc output: (on batteries for about 50 mins now)
>
> --[snip]--
> # upsc smartups at localhost
> battery.alarm.threshold: 0
> battery.charge: 000.0
> battery.charge.restart: 00
> battery.date: 09/20/00
> battery.packs: 000
> battery.runtime: 0
> battery.runtime.low: 120
> battery.voltage: 23.15
> battery.voltage.nominal: 024
Yep, it seems like you have a broken ups.  As the battery runtime and battery 
charge is dropping to zero. You still have 23.15 volt on the battery (If we 
can trust this measurement).

Regards
Kjell

Hi Again Bill,

Checked up the manual for the APC.
It looks like you have to do some configuration with the POWERCHUTE SOFTWARE.
Reset the battery date and make a runtime calibration test.

If this is not done it may as it say in the "Problem and/possible
cause"

THE LEDS IN THE BATTERY BAR GRAPH ARE BLINKING TOGETHER

Problem
While online or on battery, the
expected remaining runtime is
lower than the configured low
battery warning interval.

Note: The number of blinking
LEDs still indicates the rela-
tive state of charge of the
connected batteries.

It may be:
The battery runtime meter has
fallen out of calibration and
needs to be recalibrated.

This is worth testing before calling the APC support.

Kjell