Full Cycling data

Forum for announcements and discussion of beta firmware.
JGSchubert
Posts: 25
Joined: Sun Jul 28, 2013 7:59 pm

Re: Full Cycling data

Post by JGSchubert » Fri Sep 27, 2013 3:37 am

Here's my latest full cycle data.

This is with batteries #1 and #2 on my new (thank's Paul!) v7 board and the v0.510 software.

Board was set on 'high' for the charge rate, and on 'analyze' with the short button push used.

Batteries #3 and #4 are on now. Should have results tomorrow. (Still working on user manual draft, sorry for the delay!)

-- Geoffrey
Attachments
v7_an1_b1b2.zip
board rev7 - short analyze - b1 and b2
(1.5 MiB) Downloaded 461 times

Mark
Site Admin
Posts: 418
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Full Cycling data

Post by Mark » Fri Sep 27, 2013 1:36 pm

I've only had a quick look at the data file, but the results look fairly good to me.

I'd suggest updating to 0.512 when you can since there are some problems with 0.510

JGSchubert
Posts: 25
Joined: Sun Jul 28, 2013 7:59 pm

Re: Full Cycling data

Post by JGSchubert » Sun Sep 29, 2013 5:16 pm

All,

Here is the batts #3 and #4 as promised. It is still with version 0.510 of the software since I had started it before I saw Mark's note about issues with it. I will be starting another with 0.512 today.

-- Geoffrey
Attachments
v7an2_b3_b4_v510.zip
v7 board - 0.510 software - batts 3 and 4 cycling (short button press)
(1.5 MiB) Downloaded 459 times

Mark
Site Admin
Posts: 418
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Full Cycling data

Post by Mark » Sun Sep 29, 2013 11:17 pm

Thanks for that.

It will be good to get some test results from the latest firmware just to confirm that it's working properly - not that I expect any problems, but problems do tend to crop up when you least expect them!

JGSchubert
Posts: 25
Joined: Sun Jul 28, 2013 7:59 pm

Re: Full Cycling data

Post by JGSchubert » Sun Oct 06, 2013 10:16 pm

All,

Here is one of two sets of data from my version 7 board and using version 0.512 software. This was using additional batteries (#5 and #6, both 2450 mAh Energizers). The other set, using batts #7 and #8 on the v5 board with the v0.512 software also (recompiled with #devine Board_ver7 changed to ..._ver5) seems to have had an issue and the SD card will not read in my computer. It was one of the two 2 GB cards I mentioned I just bought. Perhaps it was not formatted and the charger writing to it caused the corruption? My windows 7 machine now says it is a RAW 30.1 MB partition and it will not succeed on formatting to any size, with any file system type - with either the Windows formatter or the official SDCard.org formatter I downloaded. Any ideas on this one? (and no, the locking tab on the card is not in 'lock' position!)

Anyway, here is the data from the batts #5 and #6 run with my working-just-fine 8 GB card.

-- Geoffrey
Attachments
v7_an1_b5b6_v512.zip
V7 board with v0.512 code and new batts 5 and 6
(722.21 KiB) Downloaded 448 times

Mark
Site Admin
Posts: 418
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Full Cycling data

Post by Mark » Mon Oct 07, 2013 12:05 am

Both of these cells are in poor shape. Internal resistance is getting quite at over 200 milliohms on both. #6 is the worst - it has higher resistance, it start out at only 0.93V (which is a bad sign in itself) and terminated because it reached the maximum temperature limit.

Despite #6 being completely flat, #5 was still about 70% charged - it only took about 600 mAh to charge it initially. #5 did fail in the end with high resistance - the discharge resistance came out significantly higher than the charging resistance - this seems to be an indication of damage that cells take when they've been overdischarged. With the latest version of the firmware, I've made a change so that it only checks the charging resistance to determine whether to allow charging, so this cell shouldn't fail with the next firmware. The firmware will still do a discharge to determine resistance that way and log the result, but that value will no longer be used to determine the health of the cell.

Mark
Site Admin
Posts: 418
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Full Cycling data

Post by Mark » Mon Oct 07, 2013 12:08 am

With regard to your SD card, I'd say that it was either faulty out of the packet, or it has been corrupted by the charger. This can certainly happen if the power is removed whilst the charger is writing to the card.

If you can get an imaging tool, you could probably copy an image of the good card over to the bad card. The one that I use is:
http://www.terabyteunlimited.com/downlo ... indows.htm

There's a free 30 day trial available which should be fine for doing this at least once. There's probably free alternatives out there as well that will do the same job.

If you're not sure how to do it, let me know!

JGSchubert
Posts: 25
Joined: Sun Jul 28, 2013 7:59 pm

Re: Full Cycling data

Post by JGSchubert » Tue Oct 08, 2013 3:16 am

Mark,

Thanks for the tip. I tried a formatter (the one from SDcard.org) and an imagewriter (to copy the good card's layout, etc. to the bad) and neither would work. Oh well. Lesson learned on impulse buys of tech.

Attached is another run on my v7 board with batts 5 and 6. The board was set to charge and was on medium speed for mA. Interestingly, even though these had just been discharged from a prior cycle, the charger did not want to start a charging cycle, giving me the purple signal in both short and medium button presses. It finally got going with the long button press.

Not sure why batt 1 (#6) showed 'capacity reached, aborted' at the end of the log - this showed a purple light also. If the capacity is reached, should that not be a good thing and show a normal 'complete' with a green light? A bit confused here. I bet it is my misunderstanding rather than a code bug though.

-- Geoffrey
Attachments
v7_ch1_b5b6_v512.zip
v7 board - charge cycle 1 - batts 5 and 6
(1.21 MiB) Downloaded 456 times

Mark
Site Admin
Posts: 418
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Full Cycling data

Post by Mark » Tue Oct 08, 2013 4:28 am

I'd say that you should take the cards back under warranty and get a refund!

The internal resistance of cell 5 was a bit on the high side, so that's why it refused to charge it initially.

Looking at the charging graphs, something really weird was going on with cell 6. Looking at the graph for cell 5, it looks quite normal:
Cell 5 - Normal Charge.png
Cell 5 - Normal Charge.png (21.14 KiB) Viewed 11854 times
Cell 6 on the other hand had a voltage graph that was really weird - it looks like it's just going up and down somewhat randomly - it's actually rather flat though - I've adjusted the scaling so that it's possible to see more detail:
Cell 6 - Abnormal Charge.png
Cell 6 - Abnormal Charge.png (25.36 KiB) Viewed 11854 times
Checking the voltage and current under charge doesn't really reveal any answers as to what was going on:
Cell 6 - Charging Voltage and Current.png
Cell 6 - Charging Voltage and Current.png (25.4 KiB) Viewed 11854 times
At this stage, I'm really puzzled!

To answer your last question - no, it's not a good thing for it to hit the capacity limit. The capacity limit is there as a safety feature so that the charger doesn't keep going on forever if the normal charge termination isn't detected. The capacity limit should never be reached normally - if it is hit, it's normally because termination has been missed and the cell has been overcharged (which isn't good for it!)

Paul.Allen
Site Admin
Posts: 98
Joined: Tue Aug 06, 2013 5:33 am
Location: Utah, United States

Re: Full Cycling data

Post by Paul.Allen » Tue Oct 15, 2013 6:13 am

this is now the outdated Version 0.512. I ran some almost brand new eneloops on analyze and one of them showed up as bad. Not sure if it is a bug or if it really thinks it is bad. If it is setting a threshold such that one of these is bad I suggest we raise that threshold. I noticed lots of updates in your latest code 0.514, i will run the same cells on it and post the data.
Attachments
almost new eneloops.zip
(3.24 MiB) Downloaded 477 times

Post Reply