Thinkpads Forum

I've had this particular W541 for 9 years now, and for those 9 years it's been a rather rock solid machine. Lately I've noticed it's unable to run on batteries, as in it'll go from 100% to 0% in a mere 2 minutes, and request that it's plugged in, or it just powers off instantly.

Today, I finally got a new battery (though going from a 9 cell to a 6 cell), and well, surprisingly I'm running into the SAME battery issue, except the difference being, if I boot back up with the AC adapter, Windows shows that the battery never depleted at all, and is still actually charged. When the ThinkPad just switches off like this, if you attempt to reboot Windows (with the new battery that is), it'll always turn back on, and when the login screen should appear, it just turns off. Trying a Linux USB boot disk, it'll get a few seconds into booting the Linux kernel, and same thing, just switches itself off (but will not power on unless I wait 10 seconds). Thinking originally the battery could be at fault, despite being new, I removed it to check the voltage, but no the voltage is correct, and if I boot to a DOS USB floppy and just play some Monkey Island while on battery power, I DO make it around 2 and a half hours. This indicates to me the issue can't possibly be with the battery. Even Lenovo Diagnostics and Vantage demonstrate that the battery (new one) is healthy and fully functional.

In my troubleshooting, I noticed something rather weird, the EC version is not matched per the BIOS Readme, with the flashed BIOS version number.

What is flashed:
BIOS: GNET94WW 2.42
EC: GMHT29WW 1.14

However, the readme states that 2.42 should have the EC flashed on GMHT30WW 1.15. Thinking that perhaps this could be part of the problem, I re-ran the BIOS update software within Windows, and the utility log DOES state that it is flashing GMHT30WW, but upon reboot, it's still GMHT29WW. 1.15 never actually flashes despite it stating it does. I also notice when it goes through the routine to re-flash the BIOS, it never once on-screen actually states that it is flashing the embedded controller.

I'm at a loss. I've been servicing and administering ThinkPad's for nearly 20 years, but never once ran into this particular issue that the EC simply won't update, and is stuck on 1.14.

Is there any way to forcefully update the EC? /forceit doesn't seem to work here, and I have also ensured that the rollback prevention option in the BIOS is disabled (just in case it thought re-flashing the same BIOS might be a rollback).

(If the EC version mismatch wouldn't cause this particular battery issue, what possibly could? If the battery was faulty I don't see DOS booting and running just fine for a few hours, whereas Linux and Windows immediately switch off. I did also notice that Windows is complaining that the EC is responding when it was not asked to, which is what lead me to check into the EC at all.)

Hopefully all of this was written clearly, I've been at this for 8 hours and I'm banging my head, and about to call it for the night before I stress myself out too much. Of funny interest, it absolutely will *not* update the BIOS from the ISO via CD, everytime it says it's going to reboot and do the update, it just goes to Windows. It only will reboot into the update capsule if it's started from Windows as Admin, but never from the CD.

Ok, so after a bit of sitting down, and fighting with the updater, I decided to just do a downgrade to 2.41 (which supposedly can be done from 2.42, but not under 2.40).

When I did this, after flashing, it then proceeded to actually go to the Lenovo splash screen, and say that it was actually flashing the embedded controller (with the fans full tilt). Upon reboot, the EC was finally on 1.15. I then re-updated the BIOS back to 2.42, and it still was on 1.15.

This then did a few things regarding the battery. It now was able to last for an hour and a half on the new battery, but it shut off at 51% left. This is a major improvement over the previous 2 minutes it was lasting. So the EC firmware *was* indeed causing issues, but now the battery gauge is VERY wrong.

To try and fix the battery gauge, which was shutting off the laptop at 51%, I went into the BIOS and left it there. If I let it boot to Windows, it just shuts off. If I boot to Linux, it just powers off as if the battery died. While in BIOS setup, I was able to sit for an hour, before the LCD started flashing. When this happened I plugged it back in, and let Windows boot. It now stated 5%.

Since then, I opened Lenovo Vantage, and started the battery gauge reset. I hope this works. For the past 45 minutes it's made it to 53%, and it appears to be going at a very linear rate, which is good. If that adds another hour, that brings me to 2 and a half hours of battery, which is within range for a 6 cell. If however the computer still powers off before reaching 0%, and then continues to sit in the BIOS setup for another whole hour as if the battery was never actually dead, I'll keep you all updated.

Shutting down at 51% usually means there's at least one dead cell inside the battery pack.

Yea, after resetting the battery gauge, it no longer acts normal at all again. It's acting like before the EC was forcefully updated (I guess the update reset it). So now it's charging to 100%, and a multimeter shows 12.52v, but as soon as you pull the plug, it goes to 99% and just shuts off.

It appears Lenovo doesn't even sell their own batteries anymore, so I'm stuck with aftermarket batteries, and I guess this "DTK" battery is just garbage. Because now it's random as to when it just shuts off, but it makes no sense, in the BIOS setup it stays running for nearly 3 hours before shutting off, but when booted into Windows or Linux it's immediate, no matter what the percentage is. Vantage keeps insisting the battery is healthy, and doesn't complain it's not genuine either. To me this doesn't quite scream bad cell, but without being able to see what the EC is seeing that it isn't telling me, I just don't know. Generally if a cell is bad, it would shut off in the BIOS setup as well, not such a wild difference.

Without much else I can do, I put the original old 9 cell battery back in, and it went back to 100%, and falling to 10% in a minute. But now with the 6 cell after charging it for hours, it'll just shut off at any percentage basically, except the BIOS (or DOS).

I'm throwing in the towel, I don't know how to win here, if I can't buy genuine batteries anymore, then I don't know what aftermarket I can get. I tried to look for which aftermarket was "better", and DTK on both Bing and Google were supposedly the better of the bunch. Guess not. I just can't get over though that the old 9 cell doesn't last 10 minutes in the BIOS, yet the new battery lasts a few hours, as long as it's not actually booted.

EDIT: I went ahead and this time ordered an iFixit battery. Costs 2x as much as the DTK does, but I'm willing to try this again one more time before the laptop is permanently attached to a charger. I will report back whenever that battery arrives and goes through a re-calibration cycle.

I am beginning to think I had two issues, a misflashed EC, and a new battery with a bad cell.

Since re-flashing the EC to the latest firmware when it was stuck one version behind, the old 9 cell just lasted 3 hours and 20 minutes on the battery gauge reset. It has never lasted that long before. New 6-cell still won't last 2 minutes, or at all if I run prime 95, so I seriously do believe yes, there's a bad cell there.

Now, I had already ordered a new iFixit 6-cell, but it seems I might still have a functional 9-cell after all. I guess I'll have two batteries now (assuming the iFixit one works).

(Admin/mods: I had edited my last reply which I didn't realize was going to put it back in the approval queue, you can go ahead and forget that entire reply since this one just solidified everything, EC was corrupt, new battery has bad cell, and old battery isn't as toast as the corrupt EC was making it out to be).