File:Battery-drain-logs.tar.gz

From OLPC
Jump to: navigation, search
Battery-drain-logs.tar.gz(file size: 258 KB, MIME type: application/x-gzip)
Warning: This file type may contain malicious code. By executing it, your system may be compromised.

Battery drain test logs for Tests/other/Battery, build 760, run 9/17/2008 by User:Mchua.

Summary

When an XO running 8.2-760 does not go into suspend mode, it takes approximately 3 hours to drain a battery from full charge.

Contents

the 3 subfolders are:

  • testbed-1: 15 logs that started with full battery, ran until battery was dead.
  • early-shutdown: 1 log that started with full battery, ran until battery was dead, and expired considerably earlier than the other XOs in this test.
  • 45min-delay: 4 logs of XOs that started with full battery, but didn't start olpc-pwr-log until about 45 minutes into the discharge.

Analysis

All the XOs had the olpc-pwr-log script run in VT (so they wouldn't go on standby) and were plugged in and reading as fully charged before they were unplugged. As per the olpc-pwr-log wiki documentation, http://wiki.laptop.org/go/XO_power_draw#Output, I'm assuming the first column of numbers is a timestamp (seconds), and the second is the percentage of battery charge.

The 4 laptops by Wad's office (in folder 45min-delay) ran all the way down to an average of 3% (2%, 2%, 4%, 6%), taking an average of 2 hours and 14 minutes to do so (8529, 7527, 9214, and 6890 seconds) in the logs - however, since the olpc-pwr-log script was started approximately 45 minutes after the XOs were unplugged, these laptops really took about 3 hours to discharge completely.

The 16 laptops in the bookshelf by Adam's office started logging before they were unplugged.

1 (in folder early-shutdown, in the results file) shut down (due to what we thought was no battery) far earlier than the rest, within the first hour. Logs show that it went down to 95% charge from 99% and this took 490 seconds (8 minutes) before the logging stopped.

11 (in folder testbed-1) have logs that show them going down to less than 20% charge (average 9.5%, but they can be split into a group of 5 that discharged to a mean of 14.8% and a group of 6 that discharged to a mean of 5%) before the logs stop. These took an average of 164.6 minutes (2 hours 44.6 minutes) to discharge. Short data for these 11 below: first column is percentage charge when log stopped, second is total time in seconds.

15, 9314
17, 8034
14, 9477
14, 9586
14, 9744
5, 10142
4, 11483
4, 11290
6, 10051
7, 10209
4, 9338

I'm confused by the last 4 (also in folder testbed-1), which have logs that show them going from a high charge (in the 90's) down to a fairly high charge (70%, 82%, 24%, 58%) before the logs stop. These took 3110, 1995, 7540, and 4609 seconds respectively.

Mchua 21:03, 18 September 2008 (UTC)

File history

Click on a date/time to view the file as it appeared at that time.

Date/TimeDimensionsUserComment
current16:20, 18 September 2008 (258 KB)Mchua (talk | contribs)Battery drain test logs for Tests/other/Battery, build 760, run 9/17/2008 by User:Mchua.

The following page links to this file: