Home-made Pila (Sort of)

[CM]

Courtesey of Chop, I got a hold of a dead Pila Li-Ion and proceeded to take it apart /ubbthreads/images/graemlins/grin.gif /ubbthreads/images/graemlins/grin.gif Not the cell but the protection circuit at the tail end. No big surprise, looks like a Seiko part (SOT-23 size) and a pair of SO-8 FET's and R's and C's. Looks like a "canned" solution using the Seiko parts as I long suspected. Anyone that wants to make their own protection circuit could do it if they want to cut a PCB. Next step was to put it on a bare 18650 Panasonic 2000mAH cell for some serious testing after playing around. That was two weeks ago and no surprises. It works just like a -- well -- a Li-Ion. I tested the charge and discharge protection and the circuit cuts in and out properly. My only concern is that the circuit cuts charge at a terminal voltage of 4.3V just like my other Pila's and that's the absolulte upper limit on Li-Ion. I've resorted to using a home-made charger (Linear LT4054.4.2) which cuts out current to the cell once it reaches 4.18V terminal voltage which should help maintain the life of the Li-Ion. I didn't measure the overdischarge cut-off but using Pila incandescents, the light was still reasonably bright so I'm assuming it cuts out near 3V. I'd post pictures but I've given up on image station. Anyone that wants to see a picture, I'll have to email it.

CM

 

[SilverFox]

Hello CM,

Just how big is that protection circuit?

With a single cell light, is there enough room to add it to a 123 rechargeable cell?

Tom

 

[LitFuse]

Nice work Cris. Can I send you all my bare cells for "Pila-izing"? /ubbthreads/images/graemlins/grin.gif

Imagestation has indeed gone done the toilet. Try one of these instead, both allow direct linking.

Photobucket

logicnerd411 Gallery host


Peter

 

[Mark_Paulus]

Could you post some specifics on creating the protection components, such as specific parts, and maybe, where to get them? That's the hardest part for me, is figuring out where to get the pieces I need.

Thanks.

 

[CM]

Silverfox,

It adds 3.1mm to the overall length. If you have a spring loaded battery compartment, it should work fine.

Peter,

Thanks for the info. I need to dig those photos out and post it. I don't intend on making the circuit myself but if you have a dead Pila, there's a good chance you can salvage the circuit and slap it on a cell.

Mark

Here's a link to the data sheet Li-Ion protection IC

If you look on page 16, it shows a complete schematic for a single cell protection circuit. The biggest parts by far are the two FET's. Copia's circuit uses looks like it uses three resistors and a cap (compare to the data sheet)


FYI, I ran a runtime test on an L5 using Panasonic 2000mAH 18650's and got an hour and 15 minutes. Those 18650's probably have more energy content than 2x123's. Li-Ions /ubbthreads/images/graemlins/buttrock.gif Now if we can convince flashlight makers to adopt these darn things and we can save those 123's for emergency use. I know they're cheap but Li-Ions are waaaaay cheaper in the long run.

CM

 

[CM]

Picture:

CM

 

[Chop]

CM,

The dead Pilas were donated by Tim B. Sadly, I saved the envelope that the cells came in, but I don't remember his CPF handle. I'd just like credit to go where it is due. I had him send me the cells because you were on a break, so to speak, at the time that they became available.

BTW, I've been using fototime.com for image hosting. It's like $20 a year and they haven't let me down yet.

 

[CM]

Chop, thanks should go to both of you. Yeah, that photo looks more like a thumbnail doesn't it?

These cells are just getting a real workout. The L5 just doesn't stress them enough so I'm using my M2 with the Pila 3.7V lamp assembly to really put it through the wringer (>2A current draw). So far, it's been flawless after about 7 "discharge till shutdown" cycles. Mostly, I've been topping off though. I did run them in the L5 as well but spare cells are necessary since the regulated light will just shut down without dimming like the incandescent does.

CM

 

[Mark_Paulus]

Since these are now "protected" 18650 cells, can you/have you tried charging them in a Pila charger?

 

[CM]

Mark,

The 7 full discharge cycles above have been charged on the Copia (same as Pila) charger. It behaves just like the Copia's that I have. I don't like the way the charger gets them all the way up to 4.3V so I've been using the Linear Tech 4054 for charging them. However, I know I can use the Copia charger for faster charging.

CM

 

[jamaica]

CM wrote: [ QUOTE ]
I tested the charge and discharge protection and the circuit cuts in and out properly. My only concern is that the circuit cuts charge at a terminal voltage of 4.3V just like my other Pila's and that's the absolulte upper limit on Li-Ion. I've resorted to using a home-made charger (Linear LT4054.4.2) which cuts out current to the cell once it reaches 4.18V terminal voltage which should help maintain the life of the Li-Ion. I didn't measure the overdischarge cut-off but using Pila incandescents, the light was still reasonably bright so I'm assuming it cuts out near 3V.

[/ QUOTE ]

CM I've done the same with some LiIon packs of my own, and had the same worries: 4.3V is bad (too high) and will take a nasty toll on the cell's lifespan, and the cut-out voltage was surprisingly low -- 2.5v if I recall correctly. Both levels are well outside of the manufacturers' recommended maximums. My interpretation is that this is a redundant system to ensure safety, but the charger / device are expected to avoid these levels in normal operation.

I have a quite a LiIon collection & love the things, especially the part about them still being full a few weeks later, by which time the lesser of my NiMH cells have self-discharged to nada. I charge my cells with a lab supply ;-)

jamaica

 

[CM]

Jamaica,

The 4.3V and 2.5V is right at the limits (not outside) of the Panasonic's recommendation for protection. However, I tend to be conservative and try to stay at between 3V and 4.2V. The cell did go through 7 cycles at the limits so we'll see how much this diminishes the overall life. I suspect that I probably won't notice the difference in the end since I probably would be using 3AH 18650's on my next switchover /ubbthreads/images/graemlins/grin.gif Heck, they have 2150mAH (and probably higher) capacities already.

 

[jamaica]

[ QUOTE ]
The 4.3V and 2.5V is right at the limits (not outside) of the Panasonic's recommendation for protection. However, I tend to be conservative and try to stay at between 3V and 4.2V. The cell did go through 7 cycles at the limits so we'll see how much this diminishes the overall life.

[/ QUOTE ]

Greetings CM! You're quite right about these being the limits for cell *protection*, as in "don't blow me up!" protection. Still, for best lifetime Panasonic recommends avoiding discharge below 3.0v and charging above 4.2v. http://www.panasonic.com/industrial/battery/oem/images/pdf/Panasonic_LiIon_Charging.pdf has a diagram showing the recommended operating range as 3.0-4.2v. The points where safety circuits should kick in are spec'd at 2.3v & 4.3v, just like the LiIon pack circuits both of us have.

I've read elsewhere that even 4.2v reduces the available cycles considerably compared to 4.1v, but what exactly "considerably" means I don't know.

Regarding overdischarge: I've charged even a few 0v cells and had them work & verified them to possess at least 50-75% of rated capacity, but others had failed "open," gone to that great lithium repository in the sky.

Best,
jamaica

"Precautions and safety notes for Li-ion Batteries in PDF Format - 29KB"
http://www.panasonic.com/industrial/battery/oem/images/pdf/Panasonic_LiIon_Precautions.pdf and more stuff at the main page:
http://www.panasonic.com/industrial/battery/oem/images/pdf/Panasonic_LiIon_Precautions.pdf

 

[CM]

jamaica,

I'm quoting from the same PDF's info for the Panasonics. Yes I agree with 3V and 4.2V as acceptable. My home brew charger cuts out at around 4.15V which is perfect for me. I also top off my cells quite often such that the open circuit voltage never falls below 3.6V. So these things should last quite a while. Still, I like the idea of on-the-cell protection as in the Pilas primarily for discharge protection, especially in regulated lights where you can't really guage the cell level based on output.

As far as cells blowing up at 4.3V, I respectfully disagree that it is even remotely possible. If it was Copia and Pila would have some serious liability issues. True, you would diminish the cell life but "blow me up" is highly unlikely. The cell remains cool when charged with the Pila chargers so the dangers of explosion is pretty low. Of course I could be wrong and everyone using Pila's are subjecting themselves to some serious bodily harm.

CM

 

[jamaica]

Yes indeed CM, no disagreement here. 4.3v isn't going to blow anything up, and I should have said that. No criticism of the Pila intended either.

I meant only to underscore what we both already knew: that the protection circuit is a feature for extra safety, that safety is ensured by both charger and protection circuit, and I'd be loathe to lose one of those layers. I would not want to, for examle, routinely hang one of these on one of my 5.3v NiMH chargers and rely on the cell's protection circuit to terminate charging (I have a bunch of *really* cute little 5.3v 400mA chargers from the 99cent store). Even so, I must hang my head and admit to charging a cell phone pack or three with an itty bitty solar panel up to the protection circuit's safety cutoff voltage. Gotta brew me up a clamp of some sort for this setup...

Your homebrew LT4054 charger and 4.15v sounds like a sweetheart -- I'm still in the stone ages with my lab supply -- and 3.6v is also very conservative.

I too like the Pila over-discharge protection; that's smart.

I also certainly wasn't trying to nitpick or lecture -- you obviously know your stuff, and that was very plain in your postings; the specs and stuff were for others, in case they might be interested.

No offense intended.

Best,
Jamaica

 

[dukeleto]

As to what "considerably" means, in terms of life reduction with overly high voltage charging, Klaus posted some interesting graphs in the rechargeable CR123 review thread
here is the page...

Olivier

 

[Mark_Paulus]

I have a question. I was stumbling around the web, and ran across this Li-Ion protection pill. This looks like it would just sit on top of any 18mm cell (18500/18650), and make them protected cells. The question then becomes, is this an accurate assessment, and if so, where can one find these little jewels. Would this be something that might need to be done in a group buy?

 

[jamaica]

Hey thanks Olivier! That's nice data. I'll have to ask Klaus where he found those graphs.

Jamaica

 

[jamaica]

Nice find Mark. Alas, these are Klixon-based devices, offering short-circuit (current-limiting) and thermal protection only. Over- and undervoltage conditions are not handled.

Jamaica

 

[CM]

Jamaica,

Didn't want to sound defensive-I apologize if I did. You sound like you too know your Li-Ions quite well and we all benefit from that. As far as lab supplies, I've used those in the past and there ain't nuttin wrong with them /ubbthreads/images/graemlins/smile.gif I just played with the Linear IC's and once I got it working, just kept on using it. I ocassionally use a bench supply if I need to charge more than two cells.

Mark_Paulus,

What jamaica said /ubbthreads/images/graemlins/grin.gif These are really stopgap protection for when everything else fails. Would be nice if someone made one with over/under voltage protection as well.