Why do laptop batteries disconnect their output?

[AmmoJammo]

So I now have two laptop batteries that have discharged below their normal level.

As a result, the internal protection circuit disconnects the output terminals from the cells, no worries...

The problem is, the laptops will then no longer charge them, and even by charging the cells with an external changer (opening the battery up) the protection circuit still refuses to connect them to the output terminals.

Why is it so?!?

[Simon]

Are the batteries maintaining a charge. It could be that the protection circuits lock out permanently and need resetting.

[Jeroen3]

Two genuine batteries, or two aftermarket third party batteries?

[AmmoJammo]

Genuine.

One had a cell at under 2 volts.

The other I don't remember as I found it to not be working a few months ago now.

One I've only just gotten today, so not sure if it holds charge.

The other one does.

[peter.mitchell]

this is pretty common, when you over discharge the cells, the controller permanently disconnects them, because if you start charging/discharging  a damaged cell there could be some serious trouble.

[AmmoJammo]

So how does one get the controller to reconnect them?

[SeanB]

Replace the bad cells, replace the safety fuse and reset the controller. Or buy a new battery and get a dozen of those cheap power bank cases to use the good cells in. must look on eBay next time and order a few more, they are kind of handy, especially at 99c incl shipping.

Edit, just ordered 10 at the expensive price of 99c CAD each.... Incl free shipping from Shenzen. Different prices per colour, so of course chose the cheapest, as that likely is the one they have the most of.

[AmmoJammo]

how does one reset the controller?

[SeanB]

There are methods, but you need to find the model of the charge controller and see how it does failed battery detection. Then you either have t replace the safety fuse it blew to disconnect the pack, or how to reprogram it using I2C or JTAG to reconnect the pack. If you are lucky and it is just locked out try disconnecting all connections from the cells and then reconnecting them, so the BMS powers up and it might then work.

[Kjelt]

this is pretty common, when you over discharge the cells, the controller permanently disconnects them,

Weird that the controller would allow to over-discharge the cells, it should protect the cells from over discharging.

[AmmoJammo]

this is pretty common, when you over discharge the cells, the controller permanently disconnects them,

Weird that the controller would allow to over-discharge the cells, it should protect the cells from over discharging.

It most likely happens if the battery is left flat, then the cells self discharge slightly, enough for the protection to then disconnect them.

[AmmoJammo]

This is someone elses picture, but my battery has the same two bigger ICs.

[wraper]

Controller can be reset but not with free tools. Controller is password protected. Software will cost like a few batteries so repairing less than a handful of batteries is financially prohibiting. http://be2works.com/Home.aspx You would need to pay $200 to be able to reset BQ8030.

[Kjelt]

So if you have such a 3yr old battery and want to preventively replace the cells will it block asap you remove the old cells?

[mikerj]

So if you have such a 3yr old battery and want to preventively replace the cells will it block asap you remove the old cells?

It's quite likely.  I have replaced cells on older laptop batteries without the safety fuse blowing, but that was probably more luck than judgement. 

I have also salvaged a a few over-discharged packs without needing to reset the controller, but you need to be very careful with the charge regime.  The cells should be trickle charged until the voltage is about 3.3v and then charge at a low-ish rate under supervision (I use 0.5amps), and in an area that won't matter if something bad happens (i.e. not in your bedroom).

[Audioguru]

A "protection circuit" in a battery is no good if it allows its battery voltage to discharge too low. It is supposed to disconnect the load, not prevent charging. The charger circuit prevents charging a lithium battery that has been discharged too low. A lithium battery that self-discharges too low is also no good.

I have the Sanyo Lithium battery from my daughter's first cell phone. It was last used and charged about 19 years ago and it measures 3.68V today with plenty of power. Its self-discharge is almost nothing. It charges perfectly.

[stmdude]

I used to work for one of the largest home-electronics manufacturers, and I dealt quite a bit with the batteries and charging, and I can explain how we reasoned when it came to protection circuitry.

1. Once a cell has been discharged beyond a certain point (varies between cells, even within the same chemistry), it's _bad_. The tolerances have been exceeded, and you're working outside of what the cell-manufacturer has tested and what the datasheet says. Ergo, as a manufacturer that takes responsibility, and actually care about our customer (and fears lawsuits), we have to make sure that the cell is not used again, as I could cause (in worst case) fires.
Dell, Fujitsu, etc know this all too well...
2. Replacing the cells in a pack is _way_ beyond what their intended use is, and to be honest, I wouldn't even do it myself..  Why? Am I chicken/boring/inept?  No, because, the likelyhood of me finding the _exact_ same cell to replace the failed ones is close to impossible. No, having the same formfactor and chemistry is not enough. The charge algorithm is tuned _exactly_ for the cell that it shipped with. And the cells that we shipped with were pretty darn good, so sourcing the same from a random internet-site is near impossible. Meaning, if you put a lesser specced cell in there, it will overcharge, over discharge, overheat, depending on how much the cells specs are off.
We easily spent a month _per product_, just fine-tuning the parameters. And that's not counting the hundreds of cells that spend a few months verifying the end-result.

And, frankly, if you do change the cells in a pack, and get it working, and it burns down your house, your insurance will find out.  We did forensics for things like that, as the insurance-companies liked to go after us (usually chargers though, but the occasional batteries as well). As we don't like getting blamed for things we didn't do (both us engineers, as well as our lawyers), it was important for us to figure out. Most of the time it comes down to "this wasn't ours", but we saw "poor repair jobs" as well, for parts that we simply didn't perform repairs on (packs and chargers are both such items, as it's more economical to trash them and send new ones).

[mikerj]

A "protection circuit" in a battery is no good if it allows its battery voltage to discharge too low.

Where do you think a protection circuit draws its power from?  How can a protection circuit prevent the natural self discharge of a cell (which can be quite high on aged Li-Ion cells)?

And, frankly, if you do change the cells in a pack, and get it working, and it burns down your house, your insurance will find out.

That is simply scaremongering.  How many insurance policies prevent the repair of battery packs, either explicitly or implicitly? Unless you deliberately set out to cause a fire with a battery pack then you would be covered (at least in the UK).

[Simon]

I'd be curious to know how they find out given how well lithium burns. If you do it yourself and your not qualified technically it may breach your insurance, assume they find out.

[Audioguru]

Anything that has a name ending with "ium" burns well: magnesium (used in flares), titanium (used in jet engines and it burns if it gets too hot).
Brits call out aluminum "aluminium" so maybe British aluminum also burns well but ours doesn't.

[wraper]

Brits call out aluminum "aluminium" so maybe British aluminum also burns well but ours doesn't.

It burns well, actually you can make some explosives with it.

[Galenbo]

... so maybe British aluminum also burns well but ours doesn't.

If you ignite Diesel Fuel the way you tried on aluminium, it also doesn't burn.

Both burn well in the same way: When pulverized, mixed with air.

[mikerj]

so maybe British aluminum also burns well but ours doesn't.

A basic thermite compound is made from iron oxide and aluminium powder, and that burns extremely well.  Perhaps thermite doesn't work in the USA...

[stmdude]

I'd be curious to know how they find out given how well lithium burns.

They sent it to us, with a note basically saying "your shit (nearly|completely) burned down our clients house. WTF?".

Once the engineers get their hands on it, everything is inspected, and compared to a known original from the same production-run. The folds on Lithum-Ion casing are almost always intact. Stamping on the same.
The PCB might be burnt to a crisp, but most of the time you can still see the placement and approximate size of the chips that used to be on it. This catches at least 99% of the clone batteries.

Analysis of the LiIon casing would most likely reveal any replaced cells.

We also did forensic analysis in some extreme cases. Ever tried to dump NAND-flash that has been through a bombing?  I have..

[Nerull]

Metals in general burn pretty well. Steel wool can be set on fire, and most powdered metals are extremely reactive. Magnesium burns well in strips, but is difficult to light as a solid block - but if you manage it, good luck putting it out again.

You may have seen these big white things, strapped to the shuttle: http://www.spd.org/images/blog/Shuttle-Layout-3L.jpg

They burn aluminium as fuel (with ammonium perchlorate as the oxidizer).