• The reason why this page exists.

• Inside the battery case. (updated on Feb. 22, 2004)

• Opening the battery case

• Oliver Durm's page.

• Ralph Levien's page

• IBM 600 Device Driver Matrix. (Be careful!  A bad BIOS flash can screw up your computer!)

• A Web site dedicated to "smart" batteries and their concept.

• Cliff's post on Slashdot

• Batteries for portables: the dirty little secret

Battery logs

* These batteries don't suffer from any sudden power drop problems (cross fingers).

** These batteries are removed when they aren't needed.  Notice how much longer the batteries last when they aren't left inside the computer on a permanent basis.  Notice also their BIOS is at or below version INET30WW.

*** These batteries USED to be flawless, but now suffer from the power drop-off problem.

Click here to submit your own ThinkPad 600 battery information.  Please note that no more battery reports will be posted at the top of the page.  However, you could still e-mail me information about your particular situation if you wish especially if you've encountered a problem no one else has reported up to now.

Battery News

November 25, 2003.

Although I've received a lot of e-mails in the last few months, not that much new information has come out aside from what follows.  Most of the e-mails were for extra information which I provided to the best of my abilities.

While I'm at it, thanks for all your kind words and gratitude!

I previously reported that the replacement cells at Mega Batteries ( http://www.megabatteries.com ) were on liquidation status.  I did call the company to get more information on the subject and they've informed me that the cells they have in stock were possibly left-overs destined for cordless phones.  And they have a lot of them!  So even if the cells are on liquidation status, Mega Batteries still have plenty of cells left.  They highly recommend the Sanyo cells.

If you're from Europe and you want to buy a few cells, Mega Batteries prefers cash up front including $30 for shipping.  Supposedly, the computer systems belonging to Visa and MasterCard still do not share information over major continents.  As a result, Mega Batteries were left with credit card payments that bounced after the credit card accounts were closed soon after the order was placed.  This is not a problem in North America since the systems here are all wired together.

Scott E. offers this link to a Linux based application that can be loaded onto a bootable CD and will test the battery's circuitry.  http://sweb.cz/Frantisek.Rysanek/battery.html 

David G. offers this link to IBM ThinkPad error codes that might pop up on your system.  http://www-3.ibm.com/pc/support/site.wss/document.do?lndocid=LWIK-3SR4PW&selectarea=SUPPORT&tempselected=5

Paul G. has this information for us.

I'm writing to follow up my earlier Email about the ThinkPad 600 battery pack.  I have replaced the cells in one of my dead packs with three 3000 mAH Kokam Lithium Polymer cells at a total cost of about $75 ($25/cell).  They fit nicely, stacked one on top of the other.  So far they are performing flawlessly, with slightly less run time than the OEM pack.  This is to be expected since they are 3000 mAH (3AH) instead of 3200 mAH (3.2AH) like the OEM packs.

I have been running them since about early July and the three cells are within about 0.10V of each other.  This small imbalance is an indication of one of the real causes of pack failure.  Read on...

I think there are two reasons for premature pack failure.

1.  While the charger monitors the three cell pairs individually, it does not seem to charge the cells individually.  This means that if the cells are not perfectly matched, the voltage on the individual cells will slowly creep out of balance.  When the cells are not in balance, the charger will terminate the charge or discharge before the entire pack has fully charged or discharged because the first cell to reach 3V or 4.2V will terminate the discharge or charge process.  Unlike NiCd or NiMh, lithium cells do not have any self-balancing mechanism during charge.

2.  Some cells in the OEM packs seem to deteriorate prematurely.  This reduces their capacity and makes the cell imbalance problem much more pronounced.

Unfortunately, there is NO cure for either problem without pack disassembly and cell replacement.  And the only thing we can do is replace the cells.  We cannot do anything about the fact that the charger does not appear to be designed to keep the cells in balance.

WARNING:

Discharging any lithium cell below 3V or charging above 4.2V will permanently damage the cell.  Charging above 4.2V also creates the possibility of a cell explosion and/or fire.  DO NOT try to "recondition" a lithium pack by forcing a charge or discharge unless you know EXACTLY what you are doing.  And if you replace cells in a pack, you MUST charge each cell individually until they are all at exactly the same voltage prior to use.  Failure to do this will result in low pack capacity because the charger does not balance the cells.

For more general lithium information, please see my web site at  http://www.pgoelz.com/piccolo_faq.html#70 .  This site is aimed at micro electric helicopters, but the lithium pack information applies to all lithium ion and lithium polymer cells. 

Good luck, guys!

Then he sent me this follow-up.

Since the cells themselves may be the problem, [replacing them with the same make and model cells] would only buy you some time.  I think the real cure is to use different cells.  The 3000 mAH LiPolys I am using are a newer chemistry and hopefully will last a lot longer.  Their smaller cousins are revolutionizing the electric model airplane world and generally stay well balanced.

Also, don't forget that when you replace cells you need to replace ALL cells and you need to charge them individually before you install them to ensure that they are balanced.  If you only replace part of the pack you will almost certainly guarantee that the pack will never be balanced and that the pack capacity will be equal to the weakest cell.

Good luck!  The word I get from a used computer dealer is that the ThinkPad battery is the only one he has ever had trouble with.  Too bad.  I LOVE the computer.  And so far it is fine with the Kokam 3000 mAH cells in the pack.

As a joke, I sent Paul G. a fake picture of what happens when you explore the insides of the battery pack without knowing what you're doing.

Yeah, and not far from the truth!  There is a true story going around the electric airplane circles about a guy who was flying with a big LiPoly pack in his plane.  This was a series connected pack of maybe three or four paralleled cells, and capable of 50-60 amps discharge at 12V.  He crashed the plane and saw that there was some minor damage to the pack.  So he put the plane and the pack in his LEXUS ( ! ) and flew something else.  About five minutes later, someone yelled that his car was on fire.  Total loss!  Apparently, damaged cells can run away and get so hot that the burst into flames.

So PLEASE be careful out there!

As for me, I'm putting my own battery replacement project on hold.  I simply don't have the time to work on the repair job.  Besides, the ThinkPad is really showing its age today.  I'll probably purchase a new notebook some time next year at which point my current ThinkPad will be sent to the recycling bin or to a charity -- after the hard drive has been properly wiped clean of course.

If you don't hear from me for a while and the following applies to your faith, Merry Christmas and Happy New Year!

July 19, 2003.

The story so far.

• We have problems with our battery packs.

• IBM is still behaving as if no problem really exist even though I still get e-mails about whole offices needing to replace the battery packs in every one of their notebooks way too soon.  According to IBM, the battery pack should last from 12 to 16 months before starting to fail.  I know for a fact it only takes three months for the battery to start exhibiting the power drop-off problem.

• Although there are different ways to "fool" the circuitry in the battery pack to recharge the cells to their full capacity, all that was proven by the various e-mails I've received over the last year have shown that no two machines appear to behave in exactly the same way.

• Some people have successfully opened their battery packs only to discover that some cells have completely ceased to function.  One contributor found a way to "reset" the cells by applying pressure inside a hole, but the cell may still be permanently dead after this operation and not merely "mostly dead".

And now the new stuff.

• Bob G. had a suggestion on an alternate portable power source for the ThinkPad.  A motorcycle battery!  They're very small, can be recharged with a proper motorcycle battery recharger, and fits easily inside a backpack.  It's not a "solve all" solution, but it's definitely the easiest solution out there especially if you already have a motorcycle battery recharger.  I'll let you figure out how to connect a motorcycle battery to the ThinkPad.

• Kevin B. has been kind enough to take a few pictures of his own attempts to open his battery case.  They're available in this page for your viewing (and cutting) pleasure.

• Both Bob D. and Bob G. are experimenting with buying a bunch of dirt-cheap expired battery packs on Ebay, open them up, extract the "good" cells, and create fully functional battery packs with the recycled cells.

• Bob G. may have found a source for new Lithium-Ion cells on the Internet.  The site is called Mega Batteries and they carry a cell (part number 18650) which may be a direct replacement for the battery pack cells.  However, I haven't confirmed the accuracy of this information just yet.  Even worse, the cell is now on liquidation status which means they won't be available for very long.

I'm still looking around for a new notebook that will eventually replace my 600X, and I'm quite surprised by the number of models available out there at reasonable prices.  They're not all power user machines, but they will do a decent job for a lot less money than the IBM models.  And if the batteries fail on those models, I won't feel as if I bought an Edsel.

That's it for now.

May 20, 2003.

It's been a long time since the last update.  I was unusually busy working on a special project at work, and there hasn't been that much new information coming in during the last couple of months.  In fact, I've been more busy answering questions than gathering any new information.  But for now, here is the new information I've received at this time.

Duncan L. sent me this e-mail back in March.

I've got some important news about Li-ion batteries and how their chemical properties could be helping to create this problem. It may also explain why some of the Li-ion cells in the batt are being fried.

Its all here:

http://www.buchmann.ca/Chap10-page6.asp

There are more pages but this seemed most interesting!

I've read that some people are freezing their ThinkPad batt. As you saw in the above link, DON'T!

It also seems that although NiCd and NiMh batts have memory they can almost always be reconditioned and so will have a longer life than Li-ion.

Discharging a ThinkPad batt may be more difficult. The link says that Li-ion batts degrade faster with a all-out full discharge especially with high current drain! Discharging the batt FULLY inside the ThinkPad machine while it is on could be doing more damage than good as the ThinkPad will draw too much current! A better method (as far as the health of the batt is concerned) would be to discharge like this:

1. Use the batt normally (don't cover the contacts)
2. Discharge to around 10% (if your batt is already malfunctioning discharge until the machine shuts down)
3. Shut down the machine
4. Cover the usual contacts
5. Boot/restart into dos (if using windows) else stay at bootloader if using GNU/Linux e.t.c. Basically don't boot fully as you may get disk errors (read on)
6. Enter the standby state (close lid etc.)
7. Discharge overnight. (hopefully overnight!)

The batt should now have been discharged at a low steady rate, the way Li-ion batts like. Charge it to 100% when you will use it soon. Make sure you uncover the contacts! Charge it to 40% if you will not use it within the next week. Keep it at 40% for storage.

This may not solve the problem completely as the monitoring hardware in the batt may still be screwing up every now and again.

What may be happening is the circuit is discharging the cells unevenly. Some cells discharging at high current drains thus decreasing the life of the cell. Also the circuit may not know that some cells are fully discharged because it may be measuring the potential difference (voltage) across the entire batt and not cell by cell. When it comes to charging, the discharged cells don't get fully charged and the not so discharged cells get over charged promoting more degradation in the cells when they are discharged in the machine!

It looks like that Li-ion technology is not suited for the high current drain that occurs in laptops. Perhaps we should go back to NiCd or NiMh? Have more, cheaper batts in the pocket, reconditioning them frequently.

Alternatively, we can soon turn to fuel cells:

http://www.theregister.co.uk/content/54/29590.html

Together, we have had enough of IBM's blindness to this problem. Lets all pull together, advertise this site, draw more annoyed IBM customers into our community. Lets push this idea of Toshiba's and make sure that IBM don't stop us from using these fuel cells in our ThinkPads.

Every time I think of fuel cells, it reminds me of the movie "Apollo 13" when the main oxygen tank blows up and all the fuel cell busses fail one after another.  :-)

Here's an e-mail from Richard R.

I have had many problems with my 600X battery - have gone through three.

The last two, I decided to take apart and investigate. As you already know, the batteries are arranged in pairs, in series.

What seems to happen is one of the cells goes open circuit - and I think I have discovered why.

There is a "Pop off" pressure switch built into the top of each cell, which effectively renders the cell useless - BUT it is possible to reset this if you place a thin (jewelers) screwdriver through one of the holes at the top of the cell, and CAREFULLY (You don't want to puncture it) move it up until you hear the valve click.

Editor's note: Richard was kind to send a few pictures.

You need to do this as soon as the battery starts exhibiting the "Flashing yellow all the time" symptoms, otherwise the open circuit cell soon drags its parallel partner down too.

No doubt there will be loads of "Safety Nellies" telling you this is irresponsible, but hey - it works!

You can actually get the batteries to "Click Off" if you have a spare one laying around - just stick a 21w bulb across it for a few minutes - this makes the cell overheat slightly, and you will hear it click.

Also on one of my spare batteries, I managed to replace the cells with some CGR 18650 HG's - slightly higher capacity than the HM's, they were not solder tagged, but with some careful soldering and a big heat sink clip attached, they solder perfectly.

The battery pack I repaired in this way lasts around 2 hours, but the yellow light comes on AND you get the low battery warning after about half an hour, BUT it still lasts for the full 2 hours - I think this is down to the circuitry in the battery not willing to accept the fact that the cells are now fine.

Anyway, I am happy, as I have successfully repaired my battery packs and done IBM out of a not insignificant sum of money!

If only we could find a way to reset the EPROM or whatever in the pack that won't play ball - our mission would be over.

Hope you can make something of this

To anyone who has access to the cells inside the battery pack, check out the existence of the pop-off valve.  But do be careful!  Don't let a battery blow up in your face!

Andrew D. found a battery pack rebuilder on E-Bay, promising to restore dead battery packs for $50-$60.  He didn't give me a link to the company, but a search on E-Bay's site should reveal their existence.  It took three weeks for him to receive back the battery pack, and it does last much longer than before, about an hour and ten minutes.  It's not exactly the battery's original performance, but it's better than nothing.

Magnus P. had this to say.

Thanks for the work you do to keep users informed of the sad state of affairs of the infamous ThinkPad 600 batteries. I have some good news to share with you.

Here is how my "5-Minute-Battery" got into a working state again: I bought my ThinkPad 600 in 1998, and still use the original battery pack. About 2001 the pack started to degrade slowly (the "sudden drop to 3% syndrome) and in 2002 it was down to 5 minutes until shutdown.

In early 2003, the lithium battery that stores the BIOS settings and date/time died. I replaced it, and to my astonishment the battery pack would keep the notebook running for about 1,5 hour once it was fully charged again! I was happy as pie, only to notice that the degrading process started again, and only 2 months later the battery was at 5 minutes "runtime" again.

But: When I recently removed and re-inserted my added so-dimm memory modules, the battery pack "resetted" again, and now the battery keeps its charge for about 1 hour again.

Thus my advice for people with dead batteries would be to either remove the lithium battery, start the notebook, listen to it complaining about it, shut it down and re-attach the battery, or to remove the extra memory (if they have some installed), start the system to see the lower RAM count, shut it down and re-insert the memory.

Repeat as necessary once the pack starts to degrade again...

All I tried on my notebook was disconnect and reconnect the BIOS battery located in the memory compartment.  After a few recycles, my battery no longer works.  Of course it's entirely possible my battery pack was already fatally damaged when I tried this trick meaning whatever I did would not have made a shred of difference.  I could always open up the battery pack and locate the valves.  Maybe all they need is a reset.

Here's an interesting observation by Paul G.

I reached your web page after a search for battery information concerning the failures I am seeing on my ThinkPad 600 batteries. They exactly duplicate your experiences.... the capacity remaining drops off gradually until a particular point is reach and then it drops suddenly to 0% and the machine shuts down.

I am somewhat familiar with LiIon cells... I use them and LiPloy cells in my electric airplanes and helicopters. I have done some reading and have some practical field experience.... and I have an alternate theory to explain the failure modes seen.

I have two defective TP600 batteries now that both perform as expected to approximately the 50% point and then drop like a stone. I have opened one of them and measured the individual cell voltages during discharge. What I found was that while the overall pack voltage remains at an acceptable level through the 50% point, one of the three cell pairs (the pack consists of three pairs of two parallel connected cells) drops to 3V prematurely. 3V is the danger point for LiIon cells, below which they will be permanently damaged. My strong suspicion is that the battery monitor is looking at all three cell pairs separately and if one hits 3V it shuts things off even though the total pack voltage is above the preset cutoff point. This would easily explain the behavior seen where the capacity remaining drops as expected until suddenly the machine activates the battery alarm and shuts down.

To test this theory, I ran the machine until the battery alarm activated and the machine entered standby. I then opened the pack and measured each cell pair and found that they all measured approximately 3.7V. I then applied an external load to the pack and set it to 2A. As the load continued to drain the pack, one of the three cell pairs dropped to 3V in about 3 minutes while the second remained at about 3.4V and the third was at 3.6V. Note that the total pack voltage was therefore about 10V, which is probably the equivalent of about 30% capacity. But the charger would (and should) shut the pack down if any cell pair reached 3V to prevent cell damage.

So that's my theory (aHEM). Comments?

That could explain the sudden power drops we've all been experiencing.  One of the battery pairs falls below 3V and the battery pack shuts down as a precaution.  Here's more...

It occurred to me that I did not fully explain a couple characteristics of LiIon cells.

First, they must NEVER be charged above 4.2V or they will be permanently damaged.

Second, they must NEVER be discharged below 3V or they will be permanently damaged.

With that in mind, you can see why the charger would/should shut the computer down if any single cell pair dropped below 3V even though the total pack voltage was above 9V. The cells are connected in series but the charger monitors each cell pair individually.

You can also see why discharging past the point where any one cell pair dropped below 3V is very bad for the pack. Covering the center pins and allowing the computer to continue discharging the pack will indeed deep discharge the pack.... and probably permanently damage it.

I am considering replacing the cells in one of my defective packs with new technology LiPoly cells and we'll see how long they last.

I think the real culprit here may indeed be the common mission of laptops.... connected to AC power nearly constantly, therefore holding the pack 100% charged permanently. There is some data out there suggesting that this prematurely ages lithium cells.

There's just one catch.  The latest 600 BIOS versions don't recharge the battery pack when the battery is between 95 and 100% of its full load.  Then again, maybe even these occasional recharges may be damaging the battery packs.

Santiago R. sent me an e-mail indicating that even his Ultrabay battery may be experiencing the same fate as the main battery.  He suspects the charging circuit may be responsible for all our fried batteries.  It would not be unusual for the same charging circuit to be used in recharging both the main battery and the Ultrabay battery.

That's it for now.  Have fun dissecting your battery packs!  :-)

February 24, 2003.

Let's start with an e-mail from Bob D.

I have been reading your web page with interest regarding the TP600 batteries and would like to share my story as it may be helpful.  I too had a TP600 with a bad battery (only got a flashing amber led with the charger plugged in).  I proceeded to try all the home remedies, including freezing the battery. I figured I had nothing to lose.  Nothing worked. I proceeded to order a set of replacement cells and opened up the pack.  I successfully replaced the cells and plugged the battery into the system.  After a few seconds I got the amber light to stay on indicating a charge.  After an overnight charge, I booted up the system and found that my battery had a life of about 1:30 using IBM's new power management Fuel Gauge.  I let the battery die and then charged it with the system on and operating.  After a full charge (100% on the Fuel Gauge), I removed the AC charger and after a few minutes the system indicated I had 2:30 hours!!  I didn't believe it so I timed the run down and sure enough, I got about 2:30 hours with the unit playing MP3's and doing normal tasks.

It seems the TP600 version that I am running now (INET36WW) with the new power management software charges to a much higher level with the system on versus being off.  With the system off, I can only get about 1:30 hours off the battery, versus 2:30 hours charging with the system on.

Another bit of information regarding the battery charge.  I attempted to "completely" discharge the pack by covering the 2 middle pins on the TP600 and running the unit until it shut down.  Upon trying to charge the pack I could only get the infamous flashing amber light.  A voltmeter on the pins indicated 0 volts.  I opened the pack and found that the cells had about 9.5 volts across them.  I forced a charge on the cells with a 12 volt power supply enough to get them up to about 10.5 volts.  I then inserted the pack back into the TP600 and was able to charge the pack normally.  It seems the pack shuts off the power to the connector if the pack voltage drops below a certain point.  By force charging them with a supply, I was able to get this circuit to reset and output voltage to the pins.

Also, I have successfully replaced the cells in two packs with no detrimental health effects (exploding batteries).  All the hype about not being able to replace the cells is pure BS as long as you don't take a blow torch to them.  All the cells I replaced had solder tabs pre-attached so soldering to these was very easy.

Brett D. sent me a link to a site which helps explain the interactions between a lithium-ion battery and the smart circuit.  It's at http://www.sbs-forum.org .  He explained to me that it's normally the battery controller that monitors the remaining charge in the battery cells, not the host computer itself.  We may be looking for the bug in the wrong place.  Brett was also asking if anyone has used any of the third-party battery packs out there successfully.  Up to now, all the responses I've received are for battery packs with IBM's own label on them.

Here's an e-mail from Larue and his/her experience with an IBM 600 notebook loaded with Linux.

On this list I've heard differing opinions about the battery problems with the 600s (600, 600E and 600X).  Well, I've gone through many batteries over the past couple of years . . . 

For those who haven't heard about it, the symptom is that the life of the battery degrades way too quickly.  Within a few months, the battery only gives minutes of use.  The consistent observation is that the battery meter has a sudden drop off - it will go from 90% to 80% to 70% and then quickly to 5% and you better get it plugged in or shut down fast!

Well, I was thumbing through the wonderful "Linux on a ThinkPad 600" page by Thomas Hood (http://panopticon.csustan.edu/thood/tp600lnx.htm) and I found a link to a page that discusses this problem (http://www.levien.com/tp600-battery.html).

One of the solutions mentioned there worked for me!  It worked on only one of the two batteries I tried it on, but it definitely worked.  I am going to rerun the experiment on the various "5 minute batteries" I have and see what happens.

The belief is that it is the "brain" of the battery that is buggy.  It believes the battery is out of juice, even though it isn't.  The remedy that was suggested was to prevent the two inner connectors of the battery from communicating with the computer (the ones that provide information rather than power) and discharge the battery completely.  That's it.  Let me review what I did.

1) Shut the machine down
2) Pulled the battery and put a piece of paper around the two inner connectors of the battery to prevent contact.
3) Put the battery back
4) Discharge the battery - you can do it anyway you want, I did so thusly:
4a) Booted into "grub" - a boot loader usually used on Linux.
4b) Printed a "very large file" to the screen - this kept the disk spinning.
5) Removed the battery and the piece of paper and put the battery back
6) Reconnected the AC

********* YOU MUST REMOVE THE PIECE OF PAPER BEFORE RECONNECTING THE POWER OR YOU WILL OVERCHARGE THE BATTERY AND YOU WILL BE SAD ******

I love that last part!  :-)

Here's the clincher!  R.B. had plenty of bad battery packs and this is what he discovered.

I've just been reading through the posts on your site and thought I might be able to at least offer up some more data.  I have been buying up used IBM 600's for about 2 months.  I have accumulated a lot of fried batteries in the process.  (4 out of 6 just gave the orange blinking light.)  I have taken all of these batteries apart after noting spurious voltages on pins 1 and 4.  Some batteries would give a voltage that was in the normal range, but the next minute there is nothing.  After disassembly, all batteries had at least one of the 3 sets with a very low reading.  i.e. 0-1.5v.  After swiping good batteries from one pack and replacing the low voltage cells, the batteries appear to charge correctly and will run my test unit for almost 2 hours.  The center set of cells was bad in at least 3 of the 4 bad batteries I tore down.  However, 2 of the packs had multiple dead cells. Also, these batteries are welded in parallel.  When one battery goes bad, its mate has always been bad.  Even after separation.  The rebuilt batteries are about 85% as good as a brand new battery.

This is the first time anyone has reported which cells were burning out inside the battery pack.  And in R.B.'s case, it's usually the middle ones.  The only question now is why is it always the middle ones?  Or is this just a coincidence?

I've received a flood of e-mails from various people after reading an article by Cliff on the Shashdot Web site (here's the link: http://slashdot.org/article.pl?sid=03/01/30/1744214 ).  Among these e-mails, I've received quite a few from people working in corporations and universities where all their 600 line of notebooks have suffered battery problems.  The exact count wasn't always specified (20 confirmed defective batteries in one location, 500 in another), but it's clearly in the thousands.  If anyone had any doubts on the extent of the problem, this should put them at ease even if the information doesn't actually resolve the battery issue.

Here's a link by Cliff for a site that reveals the "dirty little secrets" of notebook batteries.  http://www.pcbuyersguide.com/hardware/portables/batteries.html .  As they say in the X-Files (I think), "We are not alone."

Even though no one could offer a definite solution to our ongoing problem this time around, all the information everyone has offered up to now has been very useful in at least acknowledging there is a problem somewhere.  Already I've received plenty of e-mails from both ThinkPad and non-ThinkPad users who are grateful to know they're not going crazy.

Thank you everyone!  See you at the next update!

January 2, 2003.

Sorry for being late with the update.  When I go on vacation, I take it very seriously!  :-)  But now that my vacation is almost over, I need to get back on track.

First of all, I would like to announce that I won't be posting any new battery reports in the list.  The latest batch of battery reports I've received in the last few weeks simply repeat what we already know by now: these batteries are incredibly flaky!

Next up is for those whose batteries simply refuse to recharge.  I've received an e-mail from Garry describing the same problem that was previously reported by Olivier a while back.  Both of them on their own have opened up their defective battery packs and have discovered that some of the cells were totally dead.  The cells had no voltage whatsoever and wouldn't hold a charge.  I'm curious to know if anyone else has measured the same results when opening up their own battery packs.  At the same time, could you write down exactly which cells inside the pack are dead?  If the same cells are burning out regularly, this may indicate a pattern in the battery defect.

Please note.  Don't open the battery pack unless you know what you're doing!  I've already been warned that lithium-ion cells can be very dangerous if they're mishandled.  This would explain why you can't purchase the cells on their own at your local Radio Shack.  Unless you're already heavily involved in the electronics industry, I don't recommend playing around inside the battery packs even if you don't care what happens to them.  I'm more concerned about what happens to you!

I've also received reports of people who have tried to isolate the two data contacts in the battery pack in an attempt to force the computer to fully discharge the battery.  Sometimes the battery lasts a bit longer afterwards, sometimes it doesn't.  So anyone who was contemplating trying this out might find the exercise a bit futile.  Also, don't cover the data pins when the battery is recharging!  This might cause the batteries to be overcharged and may even damage the computer itself.

The only reliable solution anyone seems to have found up to now is to swap battery packs when the sudden discharge problem starts to occur.  Even a battery exhibiting the same problem on another computer could be used as a replacement for the time being.  There's no lack of reports from people indicating their supposed "dead" battery packs have suddenly been brought back to life simply by installing them in another notebook.  However, it's still too early to determine if switching the same two battery packs in the same computer will always reset the computer's recharging circuit.  Up to now, no one has reinstalled an original battery pack back in a computer after swapping it with another battery pack for an extended period of time.  Will the computer still recognize battery A after battery B has been used with the computer for an extended period of time?  Or will battery A be seen as a brand new battery and last longer?  No one knows for sure.

As promised, here's a few e-mail links where you can e-mail your battery problems in the hope that one of these media outlets will realize IBM has been stonewalling us with this ongoing problem for way too long, and decides to do a report on the situation.

• Call For Help, Tech TV

• The Screensavers, Tech TV

• Letters to the Editor, PC Magazine

• Jenn DeFeo, Associate Editor, PCs department, PC Magazine

• Editors, PC World

Remember to be brief and diplomatic in your e-mails.  I might even decide to amass all the e-mails I've received in the last year or so and compile them into a single document (with all your names and identifiable information removed of course), then ship it off to the editors and to IBM just to show all of them that this is no joke.  Too many of us are experiencing the exact same problem, eliminating any possibility this is nothing more than an isolated incident.

Finally, I've received an interesting e-mail from Matthijs indicating the battery problem may not be isolated with the IBM 600 series of notebook computers.  His Apple iBook is experiencing the same problem!  But unlike our notebooks which gives us a 5% gap where we can save our work, his notebook goes from 90% to 0% without any warning!  The computer simply switches off and he loses all his data!  Ouch!

We're not alone out there!

December 16, 2002.

We have some interesting information this time around.

First of all, if possible could those of you with more than one battery in their possession try out the following test.  I received e-mails from Markus and Ray describing very similar circumstances where switching the batteries seems to reset the charging circuit inside the computer.  Markus kept his supposed dead battery in his desk for about a year before trying to charge it up again while Ray tried to recharge his dead battery after using a new battery for only a short period of time.  In both cases, the previously dead batteries were being recharged to more or less their full capacity and were running normally until the quick discharge effect started to manifest itself once again.

Next, has anyone tried to change the LCD display setting using the PS2 utility from the default "Normal" setting to the "High" setting and noticed a difference in the lifespan of the battery and its drop-off point?  One submitter named Jim saw his battery lifespan increase from 45 to 60 minutes when he set the LCD monitor from "Normal" to "High".  Even Jim couldn't explain this.  Shouldn't the battery die quicker with the LCD display set to "High"?  It makes no sense that a battery would last longer when theoretically consuming more power.  Does the act of changing the LCD display power from one position to another reset the charging circuit?  Or does it simply confuse it?

Finally, the possible solution of reverting back to BIOS INET30WW for those computers that use the INET series BIOS may not be working.  I've already received one e-mail indicating the BIOS reversal trick didn't work, and no further e-mails that the trick did work.  So that possible solution might be a dud for now until we could figure out what might have caused the recharge circuit in Andrew's computer to reset properly.

Unfortunately, we're not that further ahead than we were a month ago.  However, Chris has suggested that we start sending e-mails to the major computer publishers and explain to them what we're experiencing with our notebook batteries.  I've already sent an e-mail to TechTV's The Screen Savers a while back, but I've never received an answer.  This doesn't mean we should give up.

So the next step should be that we start e-mailing TechTV and major computer magazines a description of our current problems.  They might not be aware of what we're experiencing with these defective notebooks.  So do start e-mailing organizations such as TechTV, PC Magazine, ZDNet, CNet and any other available publications if you happen to have their e-mail addresses handy.  Just remember to be diplomatic in your responses.  There's a good chance they'll take you seriously if they don't need to read through a ton of swear words.

I'll be posting another update later this week with a list of e-mail addresses from a few French and English language publications, and will be asking for more e-mail addresses from magazines published in other countries such as Germany and Australia.  I'll also be adding a few more entries in the battery log once I've finished sorting out the e-mails I've received in the last month.  This problem is far from over.

November 18, 2002.

First of all, I would like to thank all of you who have sent me e-mails over the last few months.  Even though we still don't have a definite solution to this ongoing problem, the information I receive on a regular basis is extremely helpful in exposing the scope of this serious defect in our computers.  At the very least, we can inform the people we know that the problems we've experienced with our notebooks are truly widespread and not just isolated incidents.

A couple more entries have been added to the log, and I've also added a link to Ralph Levien's IBM ThinkPad 600 battery page.  In all, I've received quite a few e-mails since the last update describing different testing techniques performed by various notebook owners along with some new problems.

You'll notice in the logs that R Bena's battery has started to fail.  He had a good run with the battery, coming close to a whopping 22 months of usability.  The only good batteries left are Brian Benson's and Pete's battery packs which are still running strong as far as we know.

New contributor Peter Sauleslega performed two battery drain tests using the APM Monitor program with the screen adjusted first in its power-saving mode, then its full power mode.

• In the power saving mode, the battery lasted 58 minutes and dropped off at 22%.

• In the full power mode, the battery lasted only 36 minutes and dropped off at 36%.

I remember seeing something similar on my own notebook, but I've completely overlooked it.  Somehow these batteries appear to be sensitive to the computer's total power consumption.  In fact, the battery log belonging to Pete (blue line) happens to be a battery that's used with the computer's power consumption reduced to a minimum level.  Up until now, his battery still lasts over two hours with no battery drop-offs.  Does high battery consumption hurt the batteries and/or the computer's battery monitoring circuit?  The only way to confirm this phenomenon is to run a series of controlled tests where a different series of software applications and hardware devices are exploited during each test.  One approach would be to run a computer on its battery first without the CD or DVD drive running in the background, then perform the same test without any spinning drives running in the computer.

More tests could be performed with different software and hardware applications running between each test in order to get more information on the behaviour of the battery in low, medium and high battery draining scenarios.  Based on Peter Sauleslega's tests, the battery should have theoretically dropped off at the same percentage point regardless of the overall power consumption or at the very least be reasonably close.  Instead, there's an overall 18% difference between the two drop-off point readings for a total running time decrease of 32%.  This is as bad as having a faulty gas gauge in your car!  How are you suppose to figure out how much power is left in your battery if you can't even trust the values generated by the battery circuit?

The next item will probably be recognized by many of you out there.  To this day, IBM still denies there are any problems with either the batteries or the power management circuits in their 600 series notebooks.  As far as they're concerned, the case is closed.  This was confirmed once again by Stephan Kien who recently contacted IBM in Scotland about his ongoing battery problem and was pretty much given the brush-off.  At the most, the technician he contacted would only hand out replacement batteries to whoever has a bad battery that was still under warranty, and only if the caller insisted on obtaining a replacement battery.  Aside from that, we're pretty much on our own.

Here's an e-mail by RMB explaining a particular problem he's been having with his notebook.

Problem description: until now, the laptop had performed as expected.  The battery would last for a good while, would discharge progressively and would recharge properly.  In all I probably discharged it about 20 times since I got the computer.  I charged the battery completely around 18 September 2002 and went away on a trip leaving the computer behind.

When I got back 3 weeks later, the battery icon in the System Tray was still showing 95% charge.  A couple of days later, I discharged the battery completely (I thought at the time that it was discharging a bit fast, but did not make anything of it).  I plugged the computer in for recharging: the battery meter was showing 3% charge-critical.  The battery charge light went orange and started flashing and kept flashing.  The next day, the battery meter was showing 5% and later that day it showed 6% and never did better than that: the battery indicator kept flashing orange and still does to this day whenever it is plugged into the wall..

I got out the IBM Hardware Maintenance Manual and started checking the laptop.  The system test showed that there was a problem with the system board and reported that DEV001 (the system board) had ERR91 (never found out what that means) with FRU 3610: the manual says that this FRU number refers to two things: 36 is the battery pack and 10 is the system board.  The manual says that one should try to re-charge the battery and if that fails, replace the battery.  If  a known good battery pack does not recharge, then replace the system board.  I checked other items: the voltage coming out of the AC adapter which reads +16.36V and should be between +15.5 and +17.0: so, that is OK.  I checked the voltage between terminals 1 and 4 of the battery pack: it reads 11.51V which is greater than 11.0 Volts and OK.  The resistance between terminals 3 and 4 of the battery pack should be between 4 and 30 kOhms: mine reads 9.54kOhms at the low end of the range but still OK.

Here's one more e-mail RMB sent me a little later.

One bit of further info which I forgot to give you this morning: on Sunday, I was doing some work with the notebook: it was plugged into the wall (battery dead!, but still in the notebook).  After a while (maybe one hour or so) I noticed that the battery light on the computer was now green (no longer flashing orange!): the battery icon in the system tray showed 6% and charging!  I shut everything down and went home where the same thing happened: battery light steady green but 6% and charging.  I left the notebook plugged into the wall and checked it again the next day (12 hours later): battery light still green but battery indicator in the system tray now showed 7%.  At this point, fearing damage to the notebook circuitry, I removed the battery and have not put it back in.  I took a reading of the battery terminals: 1 and 4 gave +11.86 Volts (a bit more than before) and resistance between 3 and 4 was now 10.92 kOhms, more than before.

Even though RMB's battery problems aren't directly related to our particular situation, It might still help us out later on in case there's a relationship between the behaviour of his computer and our battery problems.

Here's an e-mail from Mike on how he fooled his battery to continue to function even if the power level was at 0% and the analysis he performed on the battery using a MAHA/PowerEx battery charger and analyzer.

First, it's possible to sometimes fool the BIOS' APM shutdown mechanism and force the battery to drain lower.  I do this by plugging in the laptop just long enough to get to the LILO prompt and then immediately unplugging it.

Unfortunately, it doesn't work all the time.  Doing the (un)plug thing at other times works intermittently as well.  Note that I also have a model that no one else seems to have as well: TP 600E model 2645-55U, which takes the 47 BIOS (as of the latest update), so YMMV.

Ok, second is more radical.  I recently bought a MAHA/PowerEx MH-C777PLUS-II universal battery charger and analyzer.  Basically, this is a neat little unit that will charge and discharge Ni-Cd, Ni-MH, and Li-ion battery packs and individual cells from 1.2 volts up to 14.4 volts. 

It's the analysis part that's of concern to us.  The device will measure both the charging capacity (how much current needs to be delivered to fully charge the pack/cell) and true battery capacity (how much current the pack or cell delivers before it reaches its end-of-discharge voltage.  This is conveniently measured in milliamp-hours (mAh).  For Li-ion, the charge current is around 400 mA, and discharge current is around 350 mA (maximum); the analyzer simply measures the time to find the capacity in mAh (and it does it in real time, which is sweet).

Ok, on to the good stuff.  My laptop was made in 4/99, and obviously had the pack replaced at least once because the one that came with it is made by Xtend (yes, the laptop is 2nd hand used).  The date code listed is "01182" which I hope means "year/day of year" but I have no idea--I should probably check Xtend's web site.

Anyway, about a week ago I put the pack on the charger/analyzer and gave it a full cycle.  Total capacity of the pack @ 350 mAh discharge rate was a whopping 3.7 Ah!! (Slightly more than that, but I'm going from memory here.) Now since the laptop will drain the battery at quite a bit more than that (I have yet to measure it), the capacity as the laptop would see it may actually be closer to the 3.2 aH rating on the label.

Now the weird stuff starts: after discharging it completely, I went ahead and charged it back up on the charger.  I then put it in the laptop and attempted to use it.  Guess what: I got less than 10 minutes out of it--the percentage as related by APM fell from 100% down to zero very rapidly (but at least in a linear fashion, not suddenly some many people have reported). Trying to charge it and discharge it while in the laptop didn't fix anything, so after the laptop shut itself off, I stuck it in the charger: plenty of capacity left.  So I had an idea: put the totally (as reported by the analyzer) discharged pack back in the laptop and charge it in that.

DRAMATIC improvement!!  I now got, apparently, an hour and 40 minutes of time out of it!  But something still didn't smell right, so I stuck the apparently discharged battery back in the analyzer and hit the discharge button.  5 hours later it reported that there was still an incredible 1759 mAh left in the battery!! (I did this last night, that's why I remember the number.)

Now I'm trying to figure out what to do next.  Maybe charging while the laptop is running may work.  Maybe screwing with the BIOS.  Playing with the sense pins as someone else suggested may also do something.  But there has GOT to be a way to make this sucker work!

The point here is that, if all else fails, try using an analyzer like I got (you could do it manually, too, using a resister and voltmeter, but Li-ion is known for being very touchy about over-discharging--though not as bad as for over-charging, where lithium metal can form and cause the thing to catch fire! Therefore, I wouldn't recommend the manual way.) to deep-discharge the battery and then charge it in the laptop.  The nice thing about this, too, is that you also know at this point the REAL health of your battery pack!

Thanks for the great page and the comfort (in misery) knowing I'm not alone in this madness.

Don't worry Mike, you're with good company.  If this keeps up, we'll have to start our own self-help group!  :-)

What can we do at this stage?  Honestly, I don't know.  We could try and start a corporate-wide boycott of IBM products, but how many of us here have enough clout in our jobs to convince the IT department to purchase computers from companies other than IBM?  A class action lawsuit may be of no use because based on the e-mails I receive on a regular basis, this is clearly a global problem and not a localized one.  A lawsuit would have to be filed in every country where this problem exists in order to come to some kind of a settlement which may not amount to much for each and every individual.

I can tell you one thing.  As far as I'm concerned, I'll think twice before ordering another IBM notebook for either home or business use.  In fact, finding a decent quality, highly compact and fully equipped notebook for under $2,000 U.S. is no longer a problem.  Have any of these new notebooks suffered from the same kind of battery problems we've had to endure with our IBM 600 notebooks?  I doubt that very much.  So why go through this aggravating problem all over again?  For what our computers originally cost us, you'd think IBM would have taken better care of us.  Instead, we're being stonewalled until our machines become so obsolete that they're no longer usable and we have no choice but to replace them.

Fine then.  The next time an upgrade is due, I'll replace my notebook computer with another company's machine.  It may not be much of a protest, but it's a start.

September 1st, 2002.

There is so much information to communicate!  So many e-mails to sort out!  Among these messages, we have more confirmation that this problem is rampant, one case where the battery has worked for years without any deterioration, and a possible solution for some of you 600 and 600E owners!

We'll start with the solution as described by Andrew of the U.K. (slightly edited by me).

I have some very good news.  After installing INET36WW, the problem was worse, the usable time was shorter and the time after 0% was over an hour also the sudden drop was at 68% now instead of 25%.  So I installed INET30WW and discharged the battery by running at after 0% until it actually turned off.  Now the percentages drops slower (1% in roughly 16 counts compared to 7) and the drop hasn't occurred yet.  Charging takes approx 2 hrs I think, I lost track of time!  Let you know later. So, we could be in business!  The ThinkPad problem is over.  I am definitely getting more usage time.  It took an hour to reach 60% and when fully charged, in idle it registers about 7100secs (Is this good?  What does yours read in secs?  What should a new battery show?)  Also bearing in mind that my ThinkPad is 2nd hand, it must be a few years old and to expect what I am getting is quite excellent.  Also I don't know its history - it might have had a lot of usage!  Thanks very much for your support Oliver, I couldn't have done it without you! The APM Program is a great tool. Let me know of any other users and BIOS versions and their experiences, it interests me. I get very determined when faced with an extremely annoying problem such as this. So:

• Insert a fully charged battery.

• Upgrade to INET30WW - it wont let you without AC Power and a fully charged pack!

• Restart and use the PC until it switches off to fully discharge it (not when it switches off auto).

• Charge with the PC OFF.

• Discharge to 0%.

• Give a few more of these cycles then it can be charged while using it.

So by going from BIOS version INET36WW to INET30WW, Andrew was able to recover much of the full potential of the battery.  However, I should point out that we should wait a few months before reaching any definite conclusion on this solution.  Also, flashing a BIOS can be hazardous to your computer's health.  If the flash operation fails, the main board may need to be replaced.  At the very least, Andrew's BIOS flash reset the APM component in the computer and reinitialized its relationship with the battery.  But for many of us, this is an excellent start, giving us some much needed extra power even if it's only for a short period of time!  Thanks for the info Andrew!

Now what about the other BIOS series?  That'll require some experimentation.  But that might end up being counter-productive especially if you've loaded Windows XP on your computer.  The BIOS versions with Windows XP support were introduced long after the battery problems have started.  Reverting back to an older BIOS may cause problems.  So if any of you out there have loaded Windows XP on a computer with an older BIOS version, please let us know if your machine is working normally.  We'll know then if we should risk flashing our computers with older BIOS versions.

Now what does all this mean for those battery packs where one of the cells has failed completely?  It might mean the problem has extended beyond bad recharge cycles.  When any battery is improperly recharged, the battery runs the risk of dying prematurely and permanently.  So the inconsistent recharge cycles initiated by the latest BIOS versions may actually do more harm than good.  Instead of one problem, we may actually have two of them.

Now here's the horror story of the season.  It's bad enough this lady has had a bad history with IBM batteries.  But unfortunately, she also had plenty of problems with the computer itself!

Greetings,

I've found the webpage on the ThinkPad batteries very interesting, as I've had the same problem with battery life on my ThinkPad 600 2645-35U.
Bios:  IBET54WW
Slave controller: EVET12WW

Windows 95 was factory pre-installed, but I have it upgraded to Win98Se.  Yes, I used all of IBM's website instructions during the upgrade, and have done all the recommended steps for running Win98.

I honestly don't know the approx. number of discharges.  I use it mostly plugged in, but would try to discharge it frequently as recommended.

Here's my history:

• 2/99 - I purchased my laptop. New Li-Ion battery

• 1/00 - 8 mo. battery died - replaced

• 9/00 - 8 mo. battery died - replaced

• 3/01 - 6 mo. battery died - replaced

• 9/01 - 6 mo. battery died - replaced

• 5/02 - 8 mo. battery died

• 7/02 - I ordered a non-IBM replacement battery, and am currently using that.  At this point, it is behaving normally, but it has only been 1 1/2 months, so I would expect it to, even compared to my IBM batteries.

I've had the hard drive upgraded, the system board replaced (twice), the CPU replaced, the bios updated, etc. etc. etc. Nothing seems to extend the battery life beyond 8 months, and yes, I do all the things suggested to me by the IBM techs to extend battery life.

The batteries would be operating normally, and then very quickly, within a matter of a week or two, they would deteriorate to the point that they would not hold a charge. They would say they were at 100% (or sometimes 90-something) and after 10-20 min. (depending on the battery) they would plunge to 5%, then the computer would shut off.

In March of '01, the system board was replaced because of a PCMCIA card slot issue. At that time, here's what they said about the poor battery life (I had 3 dead batteries already and was getting my 4th) "Power management was set for optimum performance when running from battery which reduces battery life."  I don't think the tech understood (even though I sent along a long battery history note, and even included an old dead battery to look at) that I meant the batteries were really dying, and couldn't even be recharged, not just that I was reporting that they didn't last long.  And anyway, even when set at optimum performance, you should get more than 20 minutes of battery life when reading e-mail or using the word processor, which is what my computer is used the most for when on the battery.

In December of 01, they replaced the system board and CPU card.  I had been having problems with the computer freezing several times/day, to the point that I had to unplug and remove the battery to get it to shut down and unfreeze.  I was also still having problems with the PCMCIA slots losing the drivers for the accessory hard drive I had attached through it.  After performing reformats and other various troubleshooting things with the technician over the phone, I was told to send the laptop in.

In Jan. of 02, the system board and modem were replaced because of constant fatal exceptions, even after reformatting the hard drive.

The laptop was in Easy Serve numerous other times, but these are the only ones where they indicated they worked on the System Board and CPU.

I use my laptop at home, mostly using the word processor and Internet.  I practice safe computing, keeping my Anti-virus software and firewall up to date.

Ouch!  Now that's a bad computer!  I certainly hope your computer works properly now.  Do let us know if the aftermarket battery works or not.  Some of us wouldn't mind exchanging the IBM brand batteries with another company's creation.  At this stage of the game, it couldn't get any worse.

August 11, 2002.

Olivier, ton adresse de courriel ne fonctionne plus!  Peux-tu m'envoyer un courriel  indiquant exactement quelle cellule est morte dans ta pile?  Merci!

We may have a breakthrough!  An electronics student in the province of Quebec here in Canada disassembled one of his batteries and took measurements of the individual cells.  One of the six cells in the case was totally dead!  Even his engineer's trick of injecting 60 volts of power in the dead cell in order to rejuvenate the chemistry would not work (don't try this yourself).

He surmised that because the batteries are laid out in pairs, if one cell dies, an extra resistance develops across the terminals and the adjoining cell becomes a charger for the dead cell.  Because the BIOS doesn't recognize this situation, it too tries to recharge the dead cell thinking it's discharged when in fact it's totally dead.  After a while, the adjoining battery may also die because of these excessive discharge/recharge cycles.

This may explain why the batteries lose so much power capacity over time.  It's simply a set of circumstances that appear to get worse after each recharge cycle.  And in my tests involving the last new battery I received, the pack does maintain it's charge for the first few cycles with no signs of any deterioration.  Then later on, the battery starts to gradually lose it's capacity, indicating one of the cells has blown.

I took a closer look at the design of my computer.  One item that sticks out is that the battery pack is located very close to the CPU which can run very hot.  What if all that excessive heat is damaging the cells closest to the processor?  If possible, could those who have opened their batteries tell me exactly which cells are blown inside the battery pack.  If it turns out that most or all the dead cells are located right beside the processor, then the 600 line of IBM Thinkpads may have suffered from a severe design problem for years.  If IBM knew about this flaw right from the beginning, then they may have felt it necessary to bury the problem and simply replace the defective batteries whenever a client complained loud enough instead of going through the expense of replacing thousands if not millions of flawed computers.

It's truly regrettable.  Aside from this one single problem, the 600 line of IBM ThinkPads is still considered to be among the best notebook computers ever developed, offering us a solid design, decent screen size and a keyboard that still hasn't been equaled by any of the notebooks manufactured today in quality, comfort and reliability.  But now that the 600 line is no longer manufactured and the mid-line models are starting to become obsolete, there's less incentive for IBM to correct the battery problem once and for all.  This means we've been denied the portability feature of a notebook computer from day one, a situation that should never had occurred especially on such an expensive machine.

June 24, 2002.

I know I've been asking that people with battery problems to please submit me a battery report, and the information has been very helpful in identifying we all have a common problem.

But what about those who aren't having any problems?  I'd like to obtain a "head count" on the number of people out there whose batteries have not fallen prey to the "sudden discharge syndrome" so many of us has suffered up to now.  So if your IBM 600/E/X notebook is operating normally with the battery, could you please send me an e-mail containing the version of the BIOS that's currently loaded on your computer? Thanks!

If you don't understand what I'm asking, e-mail me that fact too.  I'll know at that moment you've probably never upgraded the BIOS on your computer.

June 5, 2002.

We have a new entry in the battery log.  A 600E running Redhat Linux.  Even here, the battery suffers from the exact same symptoms as under the various flavours of Windows, and one flavour of OS/2.  We can now rule out the operating system as being the possible cause of the battery problems.

For those who were hoping to blame Microsoft for this headache, I'm afraid you're out of luck.  :-)

May 5, 2002.

We have pictures!  A contributor has submitted last week an excellent series of pictures on the internal workings of the battery pack!  Because of the size of the pictures, I moved them on a separate page which you'll find over here.

April 30, 2002.

If you look at the main chart, you'll notice I've added a new column called "Continued operation after 0%".  I've been receiving some interesting feedback that the batteries last much longer than expected once the power reading hits 0%.  To perform this test, you just need to run an application that will prevent the operating system from suspending or shutting down the computer when the Power Management driver issues the suspend or shutdown signal.  Most DVD playback software can easily accomplish this task, and you won't need to change your system's configuration.

Based on the feedback I've received from the current stack of submissions, I may have stumbled on to how exactly the battery may be behaving in the computer.  We've been working along the lines that the battery simply can't be fully recharge and the power reading is way off.  But what about the other end of the scale?  Is it also possible the low end of the scale is being corrupted at the same time as the high end?

If you look at the adjoining image, you'll see what I'm talking about.  As the battery is being discharged and recharged over time, the 100% mark is dropping while the 0% mark is increasing.  The battery arrives at a state where the usable charge is nothing more than a narrow band located in the middle of the battery's true capacity.

This could explain the computer's reporting of sudden power drops and why at the same time we can continue using our computers for such a long period of time even after the battery has reached 0% according to the power management.  It's entirely possible the BIOS is not only miscalculating the true capacity of the battery, it's also shrinking the battery's usable range.

The next step is the crazy one: obtaining the source code for the computer's BIOS directly from IBM.  And no, I'm not kidding!  If this particular bug exists in the BIOS, I want to find it.

Why?

If you reached this page not knowing what this is all about, then you may not be affected by a common aliment that seems to affect an incredibly high number of IBM ThinkPad 600 Series notebook computers including my own.  For many years now, the batteries installed in 600 series notebooks have been suffering from a premature and sudden power loss.  The scenario appears to be the same for most affected 600 owners.  With a new battery, you'd be working on your computer using normal everyday software, and the power meter would indicate a steady drain on the battery just as one would expect.  Once the power level drops down to 5%, an audible beep would indicate the battery is about to die and that you should save your work.

However, after fully discharging and recharging the battery only a few times, the total battery life would slowly decrease after every discharge/recharge cycle.  One moment the battery monitor indicates there's 30% power remaining in the battery, then suddenly the low battery beep is heard and the power meter indicates only 5% power remaining.  With careful observation of the battery drain, many of us 600 owners have noticed the power drop was actually sudden and not at all gradual.  Even worse, the percentage at which the battery would suddenly drop down to 5% was steadily increasing with every discharge/recharge cycle.  Some owners were reporting drop-off points in the 90% range!  Not only does this makes the battery useless for normal everyday operations, the battery is now unreliable as a backup power supply when you're running off the AC.

Despite numerous service calls and battery swaps both in and out of the warranty period, IBM has never isolated the exact cause of this annoying problem no matter how many bad batteries they've received.  Many 600 owners have tried to track down the problem on their own, and the best suggestion up to now is a possible fault with the latest BIOS versions.  But unconfirmed suggestions and off-hand theories are not enough to properly isolate the problem.  To truly track down the exact cause of these sudden power drops, it's necessary to gather as much factual data as possible.  Only then can we start to see a pattern in this defect and come up with a practical solution to this annoyance.

I was contacted by Mr. Oliver Durm, an electronics engineer and amateur radio operator from Germany (you can find his homepage over here) inquiring about the APM Monitor utility I posted on my Web site and asking if I had an OS/2 version of the utility, which I did.  Soon after, Mr. Durm and I started exchanging information about our recent discoveries once we realized both our machines exhibited the exact same symptoms.  After a while, we agreed we needed to gather more information on this problem, so we decided to set up a section on our respective homepages where we can pool our resources.

This page will be used to gather user submitted information on affected ThinkPad 600s.  With enough information, we might finally discover a proper working solution to this problem.

What to look for

Here's a measurement graph taken by Oliver Durm off of his own ThinkPad 600E.  The blue line is the capacity reading as reported by the APM Monitor utility (Windows version) (OS/2 version), and the red line is the voltage read from the battery terminals.  Oliver has made plenty of measurements and posted many more results at this page on his Web site.  No matter how many readings are taken, the behaviour of the different batteries is pretty much consistent.

As the capacity of the battery diminishes, so does the voltage reading.  Suddenly, the capacity drops from a high percentage to 5% almost instantly.  Once the recharge cycle is initiated, the voltage of the battery increases to it's maximum capacity (this could be a residual voltage coming from the A/C adapter) and the battery capacity starts to increase gradually when suddenly, it jumps to 100%.

If you look at the two points where the capacity drops and increases instantly, the height of the drop and the spike is about the same, around 62%.  Other measurements made with different batteries may report different points where the capacity drops and spikes, but the overall behavior of the batteries remains the same.