Discharged batteries not recognized

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

Discharged batteries not recognized

Postby JGSchubert » Mon Aug 26, 2013 3:07 am

Just got back working with my charger board after last weekend's visit from relatives (yes, it was fun!).

I put my original two batteries back into the charger, which were in a discharged state (these are the one Duracell and one Energizer ones - I need to go back and label them also, probably will go with 0A and 0B - that's 'zero' here). With the newest 0506 code loaded, I at first saw the normal startup, and the log showed the version info and 'waiting for battery to be removed' as expected.

When I removed the batteries, the white light went out also as expected. However, reinsertion in either order or even one alone never caused any lights to come back on, and no further entries in the log. It is as if the charger does not recognize the battery is there at all (which is wierd since it knew enough to have me remove them).

Tried this will several different switch settings (both switches) and no go with any of them. I was starting to think I had somehow broken the board - except the initialization seemed to be working as well as the logging.

So, in desparation I put those two batteries in a comercial Energizer charger and it started charging them! (I tested them before inserting them there, and they showed around 0.8 V ea. - defninitely discharged as I left them a couple weeks ago.) I put two other batteries in on discharge mode and the charger responded appropriately.

After a few minutes, I pulled the 'make sure the board still works' already charged batteries out and put the other two which were not recognized before, but now had been charging for about 10 minutes in the commercial charger - and it indeed started an 'analyse' phase (the switch is on 'cycle'). Is this wierd or what? What happens when other end users throw in deeply discharged batteries? Will the same effect happen? I can try it with other 'deeply discharged' batteries as well, to make sure this is not a function of those two particular batteries behavior.

Time for bed now, but I will edit this message in the morning with a copy of the log to see what additional data it may have.
UPDATE: Now attached log file!
UPDATE2: Did not notice before... board prevented me from uploading file due to being over 3 MiB! Panic!

-- Geoffrey
Attachments
0A0B_DATA_prt2.CSV
0A and 0B first run PART 2 of 2
(975.1 KiB) Downloaded 264 times
0A0B_DATA_prt1.CSV
0A and 0B First fun new code PART 1 of 2
(2.67 MiB) Downloaded 272 times
Last edited by JGSchubert on Wed Sep 04, 2013 3:01 am, edited 2 times in total.

Mark
Site Admin
Posts: 415
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Discharged batteries not recognized

Postby Mark » Mon Aug 26, 2013 5:58 am

If a cell is in really poor shape, my code may not detect it. If the voltage is under 0.5V it will act as though the cell isn't there.

For some reason, the charger pulls 1-2 mA from inserted cells even if it's not doing anything - I'm not sure what's causing this since there's no voltage divider. 0.8V no load can easily drop below 0.5V with a very small load, so I'd say that's what has happened here.

I'm thinking about adding in a feature so that with the new version of the hardware, a long press of the button when set to charging mode will initiate a charge even if no cell is detected.

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

Re: Discharged batteries not recognized

Postby JGSchubert » Tue Aug 27, 2013 2:38 am

Mark,

Ahh, your mentioning of the code not recognizing batteries at 0.5V or less, and it drawing a bit of current even when officially doing 'nothing' certainly sounds like what happened here. I like the idea to add a 'forced charge' function! Perhaps we can get Paul to add in a diode somewhere to prevent current draw when nothing is supposed to be happening? Do we know if anything like that was included in his latest revised boards he says work so well?!

Have a look please at the logs, I should graph it out myself also to see how it worked out (the older Duracell or 0B - in slot 1 vice 0) took much longer to charge oddly.

Back to testing (and sleep again - funny how that keeps coming up, eh?)

- Geoffrey

Mark
Site Admin
Posts: 415
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Discharged batteries not recognized

Postby Mark » Tue Aug 27, 2013 5:23 am

I don't think Paul has made any changes to prevent the small drain when not doing anything. I sent him an email last night asking him about it amongst other things but I haven't got a response from him yet - I know he's pretty busy with his paid work at the moment, so hopefully he'll get back to me soon.

I can't see any log files on any of your posts in this topic...

Mark
Site Admin
Posts: 415
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Discharged batteries not recognized

Postby Mark » Sat Aug 31, 2013 5:27 am

I've just found that I was wrong on this - the code wasn't detecting any cells with a voltage below 0.9V

I've made a change for the next version so that it will detect any cells with at least 0.5V

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

Re: Discharged batteries not recognized

Postby Paul.Allen » Sat Aug 31, 2013 5:48 am

In my code I set the min Voltage as .3V

Mark
Site Admin
Posts: 415
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Discharged batteries not recognized

Postby Mark » Sat Aug 31, 2013 5:55 am

OK, I've adjusted my code to be the same 0.3V

I don't think it matters too much - if a cell is reading less than 0.9V it's already in pretty poor shape - less than 0.5V would be even worse.

I'll still include code in a future version to force charging on cells below the cutoff by holding the push button down for a few seconds.

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

Re: Discharged batteries not recognized

Postby Paul.Allen » Sat Aug 31, 2013 6:39 am

I agree, anything below 0.9V, especially after resting, and you should probably throw it away. My only reasoning in going that low was to detect really bad batteries so it could tell the person "oh I detect them, and there Bad!" That being said, maybe we should make it even lower and then do something like, even a long button press won't force a charge below 0.5V or something.

Mark
Site Admin
Posts: 415
Joined: Sun Jul 28, 2013 6:47 am
Location: Brisbane, Australia

Re: Discharged batteries not recognized

Postby Mark » Sat Aug 31, 2013 6:50 am

I don't see any harm in letting them try to charge low voltage batteries. It's not like we're dealing with lithium cells that are likely to explode (or "vent with flames") if they've been discharged too far.

Most likely, a cell that has a really low voltage has just developed a very high self discharge rate, or has been left in a device that has drawn it down very low.

I've got some cells that have been left in things that were left switched on and taken the voltage down to quite a low level - e.g. one Eneloop was reading about 0.2V when I found it. I previously tested it on a C9000 and found that it had reduced capacity. I recently tested the same cell with this charger and although the capacity is still reduced, it still retained a good percentage of the charge that had been last given to it maybe 6 or 12 months ago.

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

Re: Discharged batteries not recognized

Postby JGSchubert » Wed Sep 04, 2013 3:19 am

Mark et al,

Glad to see some conversation going! I like the idea of 0.3V threshold. Gives severely drained batts a chance to be resurrected through some cycles in this charger before being declared 'really dead'/bad.

I finally figured out why my upload did not work the first time on the original post (tried again and this time noticed the error about being over 3 MiB). I am also uploading a full cycling (almost a day in real time) of those same two batteries to this post. It will have to be in three files as the original is over 7 MB!

This is done with Mark's 0508 code, not sure if I have the absolute latest incorporating the 0.3V threshold, etc. But should be good data to crunch anyway.

Mark, is there any particular part of your code you would like me to look at for optimization or any issues you have? Are we working on any display code yet? I saw Paul is planning on adding a second ATMEL MCU to offload most of the work from the main unit on the final boards.

-- Geoffrey
Attachments
0A0B_CycleData_p3.CSV
full cycle - part 3 of 3
(2.75 MiB) Downloaded 282 times
0A0B_CycleData_p2.CSV
full cycle - part 2 of 3
(2.46 MiB) Downloaded 242 times
0A0B_CycleData_p1.CSV
Full cycle - part 1 of 3
(2.32 MiB) Downloaded 261 times


Return to “Beta Testers”

Who is online

Users browsing this forum: No registered users and 1 guest