T22 Power / Battery Issues - Looking for Advice

[geenoh]

I've seen this topic pop up more than once, so I'm hoping that one of the advanced users can give me some direction or assistance...

I have a used T22 that I got from a client who was using it as a dust collector... at first, it wouldn't even power on, but after a quick disassembly and cleaning, it powered up just fine. Well, almost fine.

Here's the breakdown so far:

(1) On line power (with the IBM adapter) and no battery installed, the system will boot - after about 15 seconds, a 0190 error code (critical battery level) will pop up and the system will shut down.

(2) With a battery installed and line power applied, the system will not boot - the status LEDs will flash momentarily, but then the system locks up and cannot be restarted until the DC adapter is unplugged and plugged back in.

(3) When the system is started up with DC power only and a battery is inserted before the 0190 error pops up, the system will run perfectly on line power, but will not charge or recognize the battery. The amber battery light flashes, power miser shows unknown remaining and gives no battery status.

(4) I've swapped out the original battery with 2 known-good original IBM replacement packs with exactly the same results.

(5) Per the IBM service PDF, I've tested for resistance on the packs and tested for voltage at the connectors on the main system board and all seem to be within spec.

(6) The inductor labeled "150" on the system board seems to be seated and soldered with no breaks.

Is there anything else that I need to look for (either obvious or really esoteric?) in my quest to restore some portability to this otherwise decent freebie? I've had the system running on line power for almost a month continuous without one fault, but the lack of battery sensing/charging is driving me a little nuts.

My geek-fu is pretty good, but my board-level repair skills are only fair to middlin' - so if there's a test procedure that I need to follow, I'd ask that you be slightly explicit in your description...

Any input would be helpful - solutions will earn extra bonus karma points! Thanks in advance for the assistance, folks!

[RealBlackStuff]

See if replacing the CMOS battery helps. You can measure it, to see if it still gives 3V, if not it may help the BIOS to overcome the problem.
Do your Date and Time still indicate correctly?

[geenoh]

See if replacing the CMOS battery helps. You can measure it, to see if it still gives 3V, if not it may help the BIOS to overcome the problem.
Do your Date and Time still indicate correctly?

Strangely enough, the system date/time have been rock solid so far, so I hadn't thought to test the CMOS batt - I'll have to try that for kicks n' giggles.

As an aside, I'm also stuck at a BIOS that's two removed from the last upgrade - without a charged battery pack, the update utility won't allow me to flash the latest release - and I don't have access to a floppy to do the offline utility.

Thanks for the reply - and the idea...

peace,

.g.

[geenoh]

Update: Well, I've tried a few of the suggestions mentioned previously in other threads (ie. reboot w/o battery pack multiple times, power switch hold to clear NVRAM, boot w/o battery and install "on the fly", etc.) with no change in the status.

CMOS battery tested fine @ 2.97V...

the blinky light is still blinky - and I'm all out of ideas...

anyone have anything else to suggest before I go out and order a replacement system board?

thanks all.

.g.

[bonestonne]

http://www.rkawakami.net/ibm_t2x/t21_mo ... legend.jpg

the large regulator in D5 popped off my T22's original board, caused similar symptoms, only i didn't have a working battery at the time.

resoldering the component to the board did nothing to help.

ended up being forced into a replacement board, that had a catch...that involved some workaroundsthatshallnotbenamed.

there would be times when the board would post, but only after furiously pushing the power button a ton of times, and even then, it would cut off before i got an error message.

[geenoh]

http://www.rkawakami.net/ibm_t2x/t21_mo ... legend.jpg
there would be times when the board would post, but only after furiously pushing the power button a ton of times, and even then, it would cut off before i got an error message.

Fortunately, I've never had an issue with it not booting - with the exception being when the battery is installed when powered on...

In that case, I get about 1/2 second of "go" - and then the system locks up and can't be powered back up until the battery is removed and the DC cord is unplugged from the back of the unit.

When I power up with the battery not making contact with the internal contacts, the system boots just fine - and as long as I seat the battery pack within 10 to 15 seconds (or before I get the 0190 error) - I can use the system without a problem on line power... but the amber battery indicator just continues to flash.

From this, I'm gathering that the system "knows" a battery is installed - because BattMiser doesn't show it "not installed" - but I'm obviously not getting any charging action, or any readback as to the condition of the battery at all... just black bars all the way.

I can't imagine that I have three bad battery packs - and as an aside, they are all IBM packs - not "made fors" or "compatible withs"...

What trips me up is why the system won't power on with a battery (even if fully discharged) installed - and why I have to remove same and disconnect DC power before I can get the system to boot again. Unless there's some built-in circuit protection feature that cuts the power when that condition exists, I have to assume that might be key to the problem with this unit - alas, I haven't much experience with TPads on the bench, so I'm at ends here for a solution...

thanks for the input, tho...

peace,

.g.

[RealBlackStuff]

It's not the most likely case, but you still MAY have 3 duff batteries.
I think a T30 battery might also fit, so that would be another option.

Another possibility, there are fuses on the mobo, that may have popped. Check this: http://forum.thinkpads.com/viewtopic.php?t=31162

[geenoh]

Another possibility, there are fuses on the mobo, that may have popped. Check this: http://forum.thinkpads.com/viewtopic.php?t=31162

OK... thanks for that... without some sort of block or logic diagram, it's gonna be hit or miss, but it won't hurt to run down some of the fuses and make sure that they're all intact.

re: that thread - it seems to me that what Ray was implying is that a T2x should be able to run from the adapter alone - even if a battery is not installed... or am I reading too much into that remark? If that is indeed the case, then I'm going to have to backtrack - this unit will only run on the DC adapter for about 10 seconds or so - then it will shut down after the 0190 critical battery error pops up. If I snap a pack in before the error, it'll run for weeks on DC without a complaint.

Back to the multimeter, I suppose.

Thanks again, bro.

.g.

[RealBlackStuff]

Thanks again, bro.

FYI, RealBlackStuff is The Real Black Stuff = Guinness!

[geenoh]

Thanks again, bro.

FYI, RealBlackStuff is The Real Black Stuff = Guinness!

Apropos of nothing, I was indulging in a few home-poured black and tans this weekend - Harp Lager and Edmund Fitzgerald Porter - it seemed to make the saga of the non-charging Thinkpaddy a bit more tolerable.

peace,

.g.

[geenoh]

Well... I've had a lot of reads on this thread, but no change in the status of my project machine, so before I consider ordering a replacement system board for my freebie, I'm going to pose a few more questions to the Geekpadders at large... indulge me if they seem rather basic, but I'm much more familiar with Compaq & HP notebooks - this is my first foray into reclaiming a TPad...

(1) the HD was low-level formatted when I got the machine - no OS installed - and no recovery partition. I installed 2000 Pro without a hitch, updated the drivers - but don't have access to PC Doctor (either DOS or the 4.x release). Question is: would it help in diagnosing a power/battery issue like I'm experiencing or is it too limited for that type of task?

(2) If it wouldn't hurt to run, does anyone have a link to a download? All I can find on the IBM/Lenovo sites are links to the DOS version or the 5.x release (which is not indicated for the T2x series of Thinkpads, or so it says!)

(3) Is there any info on the specs I should be reading when testing the battery connector on the system board? I get nothing on searches for the IBM site - and I'm not certain I'm willing to trust some of the returns I got from da Google...

I've sourced a few replacement boards in the off-chance I can't get this one to work, but any suggestions for quality replacements would also be welcome...

thanks once again, folks...

.g.

[geenoh]

Same T22 - Same power issues - Few new results...

After a bit of searching, I found a reference that helped lead me to a few of the fuse locations in the power circuit on the system board...

all checked out just fine.

Next, I ripped out all of the previously installed drivers and software and did a clean install with everything to do with system and battery right from the Lenovo/IBM site...

to my great shock, now the battery miser will actually show the status of all 3 battery packs as "discharging" - but none of the battery data (make, model, cycles, etc) shows up.

PC Doctor 5 for Win2K reports the batteries as "IBM Li-Ion" - but all fail the test when run - no real shock there, I guess.

Last, since the reinstall, I can now pull out a battery pack after Win2K is loaded and the system will run fine on DC without a battery installed - a change of sorts, but not much help if I want to go fully portable.

Still running BIOS ver 1.10 - without a charged pack, I can't upgrade yet - and the embedded controller is 1.05, which I take to be the latest and greatest (hah!) for the T22...

which puts me out of fresh ideas. I'm trying to locate a store in my area that may either have a compatible system or battery that I can take this unit to and do either a ROM upgrade and, hopefully, a battery test with a charged pack...

That's all I've got - so if there's anyone else who has a clue, hint, fix or a good story to tell - feel free to drop me a note. Otherwise, I'll have to assume that my former Midas touch with broken systems has finally lost power, and I'll have to go slink off to an actual (cough - hack) computer repair person for help.

Enjoy the day, folks...

peace,

.g.

[RealBlackStuff]

Even if you get a new/used mobo, a new battery should be sourced as well (on eBay). No need for an original IBM one, who knows how long that has been on the shelves!

[geenoh]

Well, I see that a lot of people have read this thread since I posted last, but it doesn't seem to be an issue that anyone has (a) any fixes or suggestions for, or (b) has experienced in the same way. Yea gods, I hate to be a trailblazer!

I'm fairly certain that this T22 isn't experiencing the dreaded BoD - but I've tried every one of the fixes listed in that thread just to be sure... and still can't get the machine to charge batteries or boot into Windows from the AC adapter alone... every time I try, the system will shut down with the critical battery error 0190 before it even starts to load the OS. When I do start it and insert a battery before the error pops up, the OS loads just fine - and once the system is running, I can remove or insert batteries as I please - the T22 will continue to run from the adapter without a hitch and will recognize the removal and/or insertion of a new battery - it just won't charge them!

So what I'm wondering is, do any of you circuit hounds out there have any idea why the system might not boot with a battery in the circuit - won't even run off AC, in fact - but will run without a fault once it gets into the OS? Except for the BIOS, every driver and utility is up to date - and the only reason the BIOS is lagging is that I can't update it without a charged battery in the system. Seeing how many people have replaced mainboards out there only to get a defective or partially working one, I'm a little leery of doing a swapout, but I'd really like to get this rugged little machine working off batteries if possible. Any hints? I'm willing to send a 12-pack of some of the best microbrews in the region to the first person that can help me work thru this issue - so put on your thinking caps and get thirsty - because I'm confident that someone out there has an answer for this. And thanks again in advance, folks!

peace,

.g.

[rkawakami]

Sorry for not replying any sooner. I've seen this thread pop up a couple of times and thought I couldn't add anything to what already was being suggested.

With none of the advice appearing to help so far, I'd throw out this WAG... pull out the MiniPCI card and try powering up. I've seen some weird power problems associated with a bum modem or combo card. Nothing like getting a boot error and battery issues, but it can't hurt and it's free to try.

Most likely there's a power system issue on the motherboard. It might be as simple as a blown fuse or a loose component but anything other than that would take some good troubleshooting skills and test equipment to pin down. For the fuses, check this scan:

http://www.kawakami-ca.com/ibm_t2x/t22_ ... legend.jpg

and look in the area of G8/H8 for the huge white ceramic fuse (which is probably fine) and a tiny one which should be labeled with an Fn number (like F8 or F11 - "F" indicating a fuse). I believe that the smaller fuse protects the interface which reads the battery status; i.e., the "smarts" inside the pack which keeps track of the number of cycles, the serial number, etc.

To answer a couple of earlier questions:

- PC Doctor for DOS does contain tests which are supposed to check both the AC adapter and battery. What they check and how well they cover ALL of the possible problems, I don't know.
- The three diskette set for the T2x can be downloaded from here:

http://www-307.ibm.com/pc/support/site. ... IGR-4YZM3F

or if you have a CD burner, you can use this .ISO file that somebody found a couple of years ago and that I've hosted on my site:

http://www.kawakami-ca.com/ibm_t2x/ibm_t22_pcdiag.iso

- A properly working system can be operated off AC power and without a battery pack installed
- If you had a BoD issue you wouldn't even be able to boot the system

[geenoh]

You get serious props and my thanks for taking the time to try and run down the issue that I'm having... drop me a PM and I'll make sure you get a tasty treat from my local brew-house for your troubles

I'm much more a code monkey than a circuit guy, so I tend to think on that level first... and that's where the disconnect probably happens for me - I look at the the error conditions and try to walk it back from there. But without any logic flowchart to go by, I'm stuck trying to isolate what I need to troubleshoot. Your reply did give me some help and some hints to look at.

It sure would be nice if an error table was out there that would let us walk through and isolate the failed component/circuit... but I take it that Thinkpads are pretty closed in that respect. Too bad, really - they're nice and robust when they work!

I did the standard fuse and inductor check - tested battery and power supply - and tried all of the BoD fixes per that thread... but your hint on the circuit fuses dealing with the " smarts" (F8-F11 et al) are next on the list - because that's exactly where I have issues - the T22 "knows" when a battery is in or out - but it can't read any data from the pack, let alone charge it... I get no cycle count, power info, etc. - but the system does know when it is and isn't in the circuit.

I can meter the fuses, but the gods know that I'm not going to try to solder up a new SMD fuse - so if that's the solution, it's a board swap for me and this machine... but once again, thanks for the help tracing it down... and let me know where to send the beer!

peace,

.g.

[rkawakami]

I'll take the props; no beer is necessary . While I do enjoy an occasional taste of the suds, my current vice is Double Chocolate Chip Frappuccinos (with extra Hershey's chocolate syrup added and then frozen for a couple of hours).

System problems like this interest me as well. Since there doesn't appear to be any circuit schematics (official or un-official) available for any Thinkpad, nor have I seen any diagnostic flowcharts, troubleshooting down to the component level is fairly hard. Sure, you can spot the simple problems - loose parts, cratered packages, etc., but if one of those tiny surface mounted resistors happens to "open up", you'd be hard-pressed to find it, and even if you did, there's usually no markings on them to tell you what the resistance value should be. We're left with checking the easy things, fuses being one of them. Even still, if you do figure out that a component has joined the choir invisible, what do you replace it with? The obvious answer is to pull one off a similar motherboard. The harder route would be to try and identify who makes the part, what the value/rating/specification is, and finally, who sells just ONE at a reasonable price. I've been able to do exactly this a few times but it does take some work and plain, simple luck. Thus, your justification for keeping it simple and doing a motherboard swap.

[geenoh]

I'll take the props; no beer is necessary .

With all due respect, sir... beer is almost always necessary!

troubleshooting down to the component level is fairly hard. Sure, you can spot the simple problems - loose parts, cratered packages, etc., but if one of those tiny surface mounted resistors happens to "open up", you'd be hard-pressed to find it,

and I'd be even more hard-pressed to fix it even if I could find it, considering my woeful skills with the soldering iron...

Thus, your justification for keeping it simple and doing a motherboard swap.

Well, I've had pretty good luck finding working "pulls" for other laptops from a company that I've done business with for a few years, so with fingers crossed, I'm going to try and go the swap route on this beast. I may even jump up the CPU a few mhz while I have everything apart on the bench... and I'll be certain to post the results of my labors once I'm done. And if worse comes to worse, I still have a limited (but serviceable) pseudo-desktop that I can slap a dock onto to use as a LinuxMCE plaything for the spare room.

Thanks once again for the input and the advice, Ray. All the best.

peace,

.g.