Options for replacing an early Li-Ion main battery

[beachycove]

A question has been troubling me of late.

As many of us know, it is becoming increasingly difficult to find main batteries for the first PowerBooks taking Li-Ion batteries: the 2400c and 3400c/ Kanga PowerBooks are particularly affected, but soon enough, it will be the turn of Wallstreets and Pismos, clamshell iBooks, and so on.

My 2400c main battery is working at present, but whether I use the machine or not, one day before too long the battery will surely expire, as unfortunately, this is part and parcel of the battery chemistry used. I have read and duly noted here and elsewhere many dire warnings about trying to re-cell a Li-Ion battery like the one in the 2400c, so I will not be attempting it myself 'experimentally'. Blindness, housefire or serious skin burns are not things anyone should risk to save on pocket money, or for the sake of curiosity. Thus I will ask a question of the cognoscendi.

Is it possible to get a Lithium-based main battery re-celled professionally, and if so, does it make sense to do so?

To be specific, there are two or three specific issues here that I don't properly understand. 1) Is the circuitry inside compatible with new cells with higher capacity? 2) Is the circuitry capable of working with repeat renewals of Lithium-Ion cells, even if properly and professionally installed? [i have read somewhere or other that the controlling chip limits the total number of charge cycles that the cells will accept - which, if correct, would appear to imply that re-celling would be a waste of money.] 3) Would any new controller board have to be specifically designed for the 2400c, or are generic controllers used – in which case, questions ##1 and 2 are effectively answered?

In the longer term, the answer to these questions might have an impact on my own collection: if no main Li-Ion battery can be had, I for one would most likely collect only portable machines capable of taking NiMh or NiCd cells, since a portable without a main battery is basically non-functional, and I have no interest in non-functional machines.

Would anyone knowledgeable about the technical issues involved care to comment?

 

[Bunsen]

http://lowendmac.com/ed/hatchett/08jh/rebuilt-laptop-batteries.html

http://lowendmac.com/ed/hatchett/08jh/inside-powerbook-battery.html

 

[tomlee59]

You might also want to search this forum for Wally's posts on this subject. He's done the most thorough experimentation and research of anyone I know. I'll briefly share my experience: I've swapped out cells to rejuvenate battery packs for 'books as well as for Dells and VAIOs. I've had mixed results. Some packs worked great right away, others never did. Most packs store data about the battery capacity, and can retain that information even with new cells. Unless you have a utility to reset that type of information, the recelled pack may not function any better. There are other factors that can get in the way as well, as Wally has shown. The full science of it remains to be revealed (at least to me).

 

[wally]

Best I can tell the controller boards are all custom but the heart of the intelligence is in chips that can be custom or commercial parts. The intelligent battery controller in the Lombard/Pismo battery packs has some Apple programmed values in a serial memory chip that are used by the associated commercial TI bq2040 battery management chip. With an I2C programming interface the serial memory chip 24LC01A contents could in principle be read out, interpreted then reprogrammed for higher capacity cells using the bq2040 datasheet and app note as guides. No cycle counting limit that I know of, but should the battery management board itself detect certain internal board failures it could blow a fuse to protect against shorted series pass transistors. So not all Lombard/Pismo battery packs can be recelled, but I think the majority can unless the series pass transistors are shorted (could happen by ESD or shorting the external terminals).

For other the PowerBooks mentioned, should high resolution photos of the intelligent battery board inside their Li-Ion batteries become available on the Internet, it may be possible to identify the management chip and memory system for each as either a commercial chip or custom proprietary, and infer from that if reprogramming can be done from commercial data alone.

I have not yet had time to examine the trend over time to see if Li-Ion cell current maximums track increased capacity over time as new cell models evolve. In the case of the Lombard/Pismo as an example, since the shunt in the battery manager board is used both to measure mAH as well as to ensure that max current is not approached let alone exceeded (much more important with Li-Ions than with prior technologies!), the shunt can be hacked downwards in value only if safety considerations allow. If, for example, a new candidate cell had 1.3x increased mAH capacity but also tolerated 1.3x increased charge and discharge maximum current values, then the shunt could be adjusted to 1/1.3x the original resistance and the programmed values could remain as original and the new cells used to full capacity (do the math with ratios, NOT percentages!).

The indicated starting mAH would still be reported as that of the original battery, but the scaled shunt would magically cause the PowerBook to appear be drawing less current, and during charge appear to be charging at a slower rate for longer. The reported gas gauge percentage and time remaining numbers would be directly usable without any mental scaling by the PowerBook user.

Each battery model needs to be examined to see if it is a good higher capacity recell candidate, and a given PowerBook may have multiple versions of the battery manager board in the installed base (Lombard/Pismo has at least two and I have only examined one of the two PC layout types). Because of the Li-Ion safety considerations these battery hacks are something best done by engineers and battery recelling organizations with the know-how and necessary equipment.

 

[beachycove]

Thank you all. This is very helpful.

The gist of it in my judgment is that the re-celling of a 2400c battery would be slightly speculative, but worth the price of speculation in the case of something like a functional 2400c. I would not myself bother to spend the money on something like a Wallstreet in a few years time, but for my 2400c, I probably will.

The other option would be to buy new. I noticed after posting that powerbookguy has unopened/ unused 2400c batteries advertised, but he states that these are Apple-branded, which presumably means that they are old, old stock. The 2400c was, however, as I understand the matter, mostly sold in Japan, and is something of a cult item for Mac fans there.

We have, or have had, very knowledgeable Japan-based members. Are there new 2400c batteries (with updated management chips) available in that 2400c-blessed land, by any chance, in the way that there are new batteries with new management chips still available for Pismos and Wallstreets? I have googled but came up empty on this one, the language being such a barrier.

 

[beachycove]

This evening I read with some surprise that the cousin of the 2400c, the 3400c, can use and charge the battery from a PB 190/ 5300, which is, of course, a NiMh battery rather than the 3400c's native Lithium Ion.

The 2400c and the 3400c are very similar machines technologically.... I am now wondering whether a 2400c would accept NiMh cells also. Could I damage the machine by trying? Or merely catch it, me and the house afire?

Mind you, cramming enough NiMh cells into the space occupied by the six Lithium Ion cells (presumably wired in series) would be a challenge, to say the least.

 

[Unknown_K]

Have you looked into professional places like www.batteryrebuild.com to rebuild your old batteries?

You can have major problems trying it on your own.

 

[beachycove]

I have a 2hr battery, so do not need a replacement just now. 2 years from now, maybe. This latest post was made purely with a view to the possibility of experimentation on a dead battery in the meantime — but only if it would not fry the machine or blow me up.