"Runtime Calibration" option missing in PowerChute Serial Shutdown

[i3v]

I've just noticed that my APC Back-UPS Pro 1600 does not seem to have "Runtime calibration" option anywhere.

In the "PowerChute Serial Shutdown" (PowerChute Serial Shutdown Agent Version 1.3.0.301),
the "Select diagnostic action" drop-down menu only provides "None selected", "Self Test", "Test Alarm" options.

The built-in help ( https://localhost:6547/help/english/BackUPS/index.html#page/PCSS_UserGuide%2FDeviceModel.html ) explicitly says that only these two option are available:

Spoiler

There are two types of diagnostics:

•a self test performs internal UPS diagnostics
NOTE: If the UPS is On Battery, a self test cannot be performed. Restore power to the UPS to run a self test.

an alarm test of your UPS (using a drop-down box in the Initiate Diagnostics section).You can initiate an immediate self testa calibration or a UPS alarm test by selecting an option in the drop-down box under the Initiate Diagnostics section and clicking on Apply.

1. Note two typos in this small text fragment:
   • the "second list item" is not actually formatted as a list item.
   • "testa"
2. This looks like there's no "Runtime calibration" for anyone now (provided that the "PowerChute Business Edition" is "Discontinued" and "PowerChute Serial Shutdown has replaced PowerChute Business Edition")? Even though this feature was available before? Why?
   • This video (about PowerChute Business Edition V9.5) clearly shows that the "Runtime Calibration" option was there at some point.
   • I've tried to install the "PowerChute Business Edition V9.5" but it seem to not recognize my UPS.
   • I'm aware of that "Back-UPS" series is effectively the most basic, and I was not expecting to see any fancy features here. However, I'm still surprised to see such a basic feature (present in all noname cheap UPS models I know) missing. I've considered "manual calibration" (the one that involves physically unplugging the UPS) the thing of the past.
3. The "UpsDiag" tool is unable to connect to this UPS too.
4. Am I just missing something? Is there some way to do the "Runtime Calibration"?

[Teken]

The Back-UPS Series was supported by PCPE not PCBE. You may try the last version of PCPE to see if the battery calibration is present.

If it’s not than obviously PCSS would not support the same. You may contact APC Technical Support to obtain the last copy of PCPE if available.

Otherwise, you’ll need to find a copy you trust to download.

Another option is to use third party software like APCUPSD / NUT to invoke the battery calibration if present. 

Lastly, a battery calibration should only be performed when the cartridge has been replaced or when the expected runtime displayed is wildly off.

As invoking the battery calibration impacts the 260 cycle life which affects and shortens the service life of the cell. 

Questions Ask . . . 👍

[Teken]

PCPE: https://www.apc.com/us/en/download/document/APC_PCPE_310/

[fredkm]

Please refer to this post for how to run a manual runtime calibration on your Back-UPS. A software-initiated runtime calibration for Back-UPS is not supported in PowerChute Serial Shutdown and also not in the older PowerChute Personal Edition software. Most modern Back-UPS do seem to support the runtime calibration USB HID usage. Thus support for runtime calibration for Back-UPS may be added in a future PowerChute Serial Shutdown release. The typos in the on-line help will be addressed in next release.

[i3v]

@Teken , 

1. I've tried PCPE_3.1.0 and I can confirm that it does not provide any "battery calibration" option.
2. Thanks for mentioning the Apcupsd! I've not tried it yet, but I'll run some tests later and report back.
3. About what I'd like to achieve... 
   1. The built-in standard "self-test" is unreliable. The "Back-UPS Pro 1600" UPS only started to complain about the (original) battery health after it degraded way too much --- it was unable to support the 900W load for even a second at that point (the nominal rating is 960W). Maybe the battery would have still worked not that bad for ~30% load or so, but this is not the "replacement threshold" I want.
   2. The methodology I'd rather use would be:
      1. ~50% discharge once a year, under full practical load
      2. Warn to replace the battery if ~50% discharge takes less than 15 seconds (or whatever preset).
      3. A stable full practical load could be provided by running some CPU+GPU benchmarks.
      4. Without much risk to get an unexpected complete power loss and/or a need to mess with cables.
      5. The yearly log of the results of this test could be easily extrapolated (provided that the load is about the same) and the battery replacement could be easily planned in advance.
      6. Same for the new battery --- I'd like to run a similar test with minimal hassle to make sure I do get the expected on-battery time.
4. BTW, in my view, not only "Back-UPS Pro 1600" suffers from a lack of useful battery health monitoring. Symmetra PX 48 units also report almost nothing useful about their battery health stats via their web interface.
5. I'm well aware of the limited number of the discharge cycle of such batteries. I'm clearly not talking about running a "deep discharge" test every day. I doubt that running the aforementioned test once a year would significantly reduce the usable battery lifespan.

@fredkm ,

1. It would be great if the support for the runtime calibration would be added in some future version of PCSS!
2. Thanks for correcting the typos!

[Teken]

At a high level in a enterprise environment all of the batteries are replaced on a scheduled interval based on each sites unique conditions such as power quality, usage, environmental, and mission critical status.

The average service life of the batteries are 2-5 years based on usage, temperature, and environmental conditions.

The vast majority of stringent ISO sites replace the RBC at the three year interval. While others are replaced at the maximum of four year interval regardless of any pass / fail tests conducted.

The APC Self Test is a very crude method to get an idea of the RBC health. It in no way is a true indicator of the RBC health or operational runtime of the cartridge.

A yearly lights out / grid down validation test is one of the major benchmarks & indicators when compared to the APC Runtime Calculator as to battery health.

This is coupled by measuring every cell for internal resistance and comparing it to the manufacturers stated specifications / data sheet.

Other validation tests are at rest voltage readings after being fully charged and what the values are after 24 ~ 72 hours. Along with carbon pile load testing at a defined wattage for a duration called out by the manufacturers data sheet.

All RBC / individual cells are monitored by a thermal imaging camera to record expected temperature rise or out of band elevated internal temperatures.

All of the above insures the RBC is providing the expected performance and operational runtime needed for the infrastructure.

A battery calibration is only required upon a RBC being replaced or (IF) the displayed / calculated runtime is wildly off.

As noted earlier an annual lights out / grid down validation test is the equivalent to a battery calibration. Both are extremely stressful on the RBC and goes against the 260 cycle life of the cells. ☝️

Lastly, regardless of all the battery validation testing both static & dynamic testing of the system must be performed especially if the system is hot swappable and also offers cold start capabilities.

Questions Ask . . . 👍