Power Bank Endurance Test – Hillo Power Jin Gangxia (Part 19)

This is one of those ongoing experiments which just seems to keep dragging on and on. Initially, I thought I’d torture a relatively unknown power bank by cycling it over and over on the test rig, expecting to see it fail not too far past 500 cycles. Now, I really don’t know who’s being tortured, since it’s now reached 950 cycles since the experiment started, or 966 cycles since manufacture.

Results

effective-capacity-graph38

This time around, the results do have something interesting to show. From the last result, it seems that the degradation took a good dive for a number of cycles, before recovering towards the accelerating degradation trend established earlier. The reason for this dive may be related to early charging termination, as the recovery may have had to do with the charging being interrupted after a full charge to do another top-up. But the fact the battery capacity didn’t recover further than about 2560mAh tells us that the cell capacity is really diminishing on a cycle by cycle basis, and is accelerating. The narrower range of the data with less spikes seems to reflect the fact I now hold my room at a fairly well-regulated 21 degrees due to other temperature-sensitive testing which is also going on at this time.

effective-capacity-graph37

In the grand scheme of things, the degradation is becoming more noticeable. Only 65% of the initial capacity remains, however, this is still a useful amount (2500mAh).

Conclusion

It seems some strange reversal of degradation occurred in this set of data which may have been charging circuitry related, but otherwise, the degradation almost appears to be accelerating. Around 2500mAh remains, which is about 65% of initial capacity and still somewhat serviceable.

Unfortunately, due to changes in circumstances in the near future, it seems I won’t be able to continue this experiment much longer, so the next report at cycle 1000 (or 1016) might have to mark the end of the experiment, just when things start to get interesting.

About lui_gough

I'm a bit of a nut for electronics, computing, photography, radio, satellite and other technical hobbies. Click for more about me!
This entry was posted in Electronics, Power Bank and tagged , , , . Bookmark the permalink.

5 Responses to Power Bank Endurance Test – Hillo Power Jin Gangxia (Part 19)

  1. Dan Gauthier says:

    Can you generate a capacity graph vs. age of the pack (days) rather than number of cycles?

    • lui_gough says:

      I might do that for the final installment, as it requires that I go through all the individual run-sheets one-by-one … or (more likely) alternatively write a script using a few tools to try and extract it automatically from the set of files. Either way, best left until I throw in the towel.

      – Gough

      • Dan Gauthier says:

        I can do that (analyze, locate, extract, output), probably with a one-liner if you can send me the data. (I dont graph as I only use text, but I can send you a .csv with any extraction you like). If there isnt a reason to not send it (proprietary agreements or funding related),then you can email me anything (though you may have to temporarily change your SMTP server if yours limits size.

        Dan

        • lui_gough says:

          The data is almost a gigabyte in size, and I’m on a 1Mbit/s link, so I’d rather get it done myself. It’s some text extraction and date format conversion so I can manage that just fine – it’s just a matter of wanting to do it once at the end of the project, rather than do it piecemeal. Don’t worry – the final part will be coming within a month or so.

          – Gough

        • lui_gough says:

          Well, since you were so insistent, I managed to work out a two-part command to do it. The first part extracts the date information between brackets in the input file, the second gets rid of the stupid white space that gets thrown in by the app, removes a — delimiter between date and time and replaces it with a space, and then removes all the ‘/’ in the date field and prefixes 20 in front to turn it into an ISO date.

          sed -n ‘s/.*(//;s/).*//p’ | sed -e ‘s/ //g;s/–/ /g;s/\///g;s/^/20/’

          Now I just have to use a shell script to iterate … but I’ll still leave the posting to the last and final part :).

          – Gough

Leave a Reply to Dan Gauthier Cancel reply