Li-Ion trickle or not?

Candle Power Forums

Help Support Candle Power:

AilSnail

Flashlight Enthusiast
Joined
Jul 11, 2003
Messages
1,299
City & State/Province
Bergen, No
I am thinking of putting some li-ion (likely 2s3p 18x68mm cells) from camera packs into the base of one of George's Flexileds.

I am thinking that the three stacks could share a single protection circuit?

The battery is supposed to kick in automatically when the grid goes down.

I can think of some schemes to do this, but how should I go about to keep the cells fresh. It would be neat to use one of those little chips to start charging whenever the battery is low, and then when it is finished do what is best to keep the cells healthy. Then again, what is the best way to keep it healthy? and, what is a good trigger for such a smart charger? voltage? Finally, is there any preferred chips?

Edit: I searched some more and found that li-ions should not be trickled continously. So I need something that kicks in when the capacity gets below 90% or whatever.
 
I'm currently working on a variation of the above, and am having the same snag. The only battery topology that lends itself to long-term simple float is SLA (Lead-acid) or somewhat, Ni-Cad. Li and Nimh do not work well without a 'smart' charging system. These are available as ICs with 95% of the work done, but you still need to set it up with the rest of the bits. Li cells are very picky, and each cell in a series stack is monitored. Parallel cells are ok.

Here's the schematic for my CCFT lantern's charge/powerfail section, if it can give you any ideas. These have been in use for years here, and have held up well. (See emergency lantern thread in homemade/mod forum.)

(Helps if I actually put in the URL..)
http://www.minespot.com/cpfstuff/CCFTemeg.JPG

Adapting this to drive Flexileds should be easy. Let me know if you have questions.

Good luck!!

Yours, drs the crazed.
 
since it will be carried to the cabin and back, sla or nicd won't cut it. I wanted li-ion for the weight saving and cold temp functionality, but nimh would work too. If it was for me I could manually charge it, but I need something idiot proof. (not to imply anything about the recipient)
 
Sounds like a fun project. You will need only one protection circuit for the pack. And it would have to be from a 7.4v pack. (You can't parallel or series protection circuits, just the cells themselves).

As you've noted, you'll need a charger designed for proper Li-Ion operation. You could build one. Just about all of the major chip manufacturers (TI, NS, Maxum, Linear Tech.) offer single chip chargers that are designed for Li-Ion. Another approach would be to buy any 7.4 Li-Ion charger and place it's circuit into the base of your lamp. Have fun! (And be sure to show us your mod.)
 
so is it enough to just set a voltage sensor on the pack, with a comparator threshold to start the charger? then shut it down when it reaches 8.8v? [edit: 8.8 seems too much. look down.]

Guess I'll look into those chips, thanks.
You know where I will come and show off any new mod ;-)
 
As an Amazon Associate we earn from qualifying purchases. Product prices and availability are accurate as of the date/time indicated and are subject to change.
thanks drs, i don't know how big these are, but I know I have space for six or maybe eight 18x68mm cells. 1.8AH on ebay for $3.5 or less on ebay in packs.

Found no specs on the charger. If you buy it be sure to report on how it works and if it is suitable for standby charging.
 
Okay I like TI's sample policy, and I found this chip, bq2057, for two cells. There is a 8.2v and a 8.4v version, but you can set the voltage if you want.

I think it starts a charge each time power is connected, without checking if it is needed. I don't know if this is harming the cells if say the lamp is moved around in the house a lot it will start charging all the time. When it has terminated the charge, it starts a new charge when the battery voltage drops below Vo(reg)-0.2v, which means 8v 0r 8.2v respectively. At least this is how I interpret the sheet. I'll see if I can find any others too. Does anybody know approx how much % capacity is left in a pack that reads 8v or 8.2v?

http://focus.ti.com/lit/ds/slus025f/slus025f.pdf
 
[ QUOTE ]
Does anybody know approx how much % capacity is left in a pack that reads 8v or 8.2v?



[/ QUOTE ]
Assuming cells rated at 4.20V, you will have about 90% at 4.1V/cell and 70-75% at 4.0v/cell. You will get much longer shelf life if you maintain them at 4.1V/cell rather than 4.2V/cell. Some Li-ion in longterm standby are maintained at 4.0V/cell.
 
Doug's absolutely right about maintaining the cells at 90% to get long life. I use a linear tech IC that stops charging at 4.15V and kicks in at 4.05V. These have worked quite well in single cell charging.

CM
 
ok please tell me more about voltage rating. How do I know what the ebay cells are rated for, and if not, how do i find out, and if not, what voltage should i design for in order not to shorten their life too much?
 
[ QUOTE ]
ok please tell me more about voltage rating. How do I know what the ebay cells are rated for, and if not, how do i find out, and if not, what voltage should i design for in order not to shorten their life too much?

[/ QUOTE ]
*Almost* all are 4.2V so just assume this. BTW, perhaps an easier alternative for your application is to not have a built-in charger. Just refresh them once a year. I have found that Li-ion self-discharge is low enough for this to be OK.
 
Won't an accurate 4V (to 4.1) voltage regulator adequately keep Li-ions "float charged" without "fancy" circuitry? IIRC, the tricky part of Li-ion chargers is going from CC mode (when the cell voltage is lower) to CV mode, and being careful about putting too much current into cells that are "boken"...

billw
 
[ QUOTE ]
billw said:
Won't an accurate 4V (to 4.1) voltage regulator adequately keep Li-ions "float charged" without "fancy" circuitry? IIRC, the tricky part of Li-ion chargers is going from CC mode (when the cell voltage is lower) to CV mode, and being careful about putting too much current into cells that are "boken"...

billw

[/ QUOTE ]
IMHO, maybe. It should of course be current limited to the max recommended charging rate but this is easily done. For a single cell or group in parallel, I see absolutely no problem. With series strings, however, if the cells in the series string have differing self discharge rates one will eventually become charged above the /cell charging voltage and the other will fall below the /cell charging voltage. One could protect against this by adding some type of shunt voltage regulator across each cell or cell group in the series string.
 
Back
Top