Page 1 of 1

Charging bug ?

Posted: Sat Dec 23, 2006 6:58 pm
by bigbear2007
I dont want to cycle my battery too many times so I set (in the power manager) that only recharge when the batter is less than 20%.

But if I plug in the AC without the t60 being on, it would still charges my t60 even the battery level is over 20%. It works fine if I plug the AC in when the t60 is on.

Posted: Sat Dec 23, 2006 7:24 pm
by GomJabbar
When the laptop is off, the Embedded Controller chip on the motherboard controls the charging. Any settings you make in Windows will have no effect on the Embedded Controller. Once Windows boots up, it takes control over the Embedded Controller for charging the battery. In short, there is nothing you can do to make any difference when the laptop is turned off.

Well, I say it's the Embedded Controller - at least I am pretty sure it is - but I suppose it could be the circuitry in the battery. At any rate, it is controlled via the hardware and firmware, it is not controlled by software settings.

Posted: Sat Dec 23, 2006 7:24 pm
by christopher_wolf
I don't think that the Power Manager thresholds that are set while the Thinkpad is online affect the charging behavior while the Thinkpad is offline; unless the logic controller in the battery has the ability to dynamically set those charging thresholds from the Power Manager. :)

At least, I have never seen that behavoir before.

EDIT: That post slipped in while I was typing this one; no, it is *not* the Embedded Controller.

Posted: Sat Dec 23, 2006 7:50 pm
by jdhurst
Beyond what has been said, the battery is a Lithium Ion battery, does not have a memory like NiCad's and can be charged any time. There is no need to run it down. My T41 battery is now fully three years old and is dying of old age, not charging or lack of it. It is only good for about 90 minutes now, but that is the chemical death of any battery. Use as you wish and enjoy while it is young. ... JD Hurst

Posted: Sat Dec 23, 2006 8:30 pm
by GomJabbar
Interestingly, a quick Google search brought up the following information.

Summary of changes for X30 Embedded Controller Program Update Utility wrote:Version 1.02 (1KHT14WW)
Note: This version of Embedded Controller Program will only work with BIOS Version 1.00c or higher.
(Fix) (Windows XP and Windows 2000) If the computer is attached to the UltraBase X3 with a battery installed and goes into standby mode, warm-undocking the system from the UltraBase X3 may be failed.
(Fix) (Windows NT 4.0 and DOS) If the computer is attached to the UltraBase X3 with a battery installed, beeps may sound continuously when the system resumes normal operation from suspend mode.
(Fix) The battery may not be charged while the computer is turned off.

Embedded Controller Program (Non-Diskette) - ThinkPad X30

It appears as though the Embedded Controller does have something to do with battery charging when the laptop is turned off. I am not going to claim (with any confidence) much more than that regarding the relationship between the Embedded Controller and the battery.

Posted: Sat Dec 23, 2006 9:32 pm
by christopher_wolf
The Embedded Controller acts to merely relay information to the logic circuitry in the battery; it communicates with the logic circuitry that is the battery's way of self-regulating itself...That is the "something" it has to do with it (as I noted in my first post, if care was taken reading it). It serves, as its name states, as a controller for almost all of the hardware on the laptop to interface to. The logic circuitry on the battery is what does the work of managing the lowest level power states of the cells. Having a bug in the EC simply break the offline charging of the battery is not a logically sound diagnostic conclusion. "Something" is not good enough to say that it has or hasn't the ability to store charging thresholds from the Power Manager for implementation offline.

Although that is off-topic and, now, settled.

Posted: Sun Dec 24, 2006 12:46 am
by gator
I have the exact same issue as the OP. Is there a way to set things such that the battery thresholds remain as set in the power manager?

Posted: Sun Dec 24, 2006 1:05 am
by christopher_wolf
gator wrote:I have the exact same issue as the OP. Is there a way to set things such that the battery thresholds remain as set in the power manager?
When the system is on or off? :)

There was an issue with the Power Manager where the user-defined charge thresholds would be ignored and the default charging thresholds followed. That has been resolved, if I recall correctly.

I don't know of any way to set the thresholds for it to follow when it is offline, however. I don't even think that is supported with the current hardware.

Posted: Sun Dec 24, 2006 1:12 am
by gator
christopher_wolf wrote: When the system is on or off? :)
When the system is off. Thresholds remain correct when the laptop is on **if** I drain the battery to the levels I want (since after switching it off and leaving it connected will always bring it back to 100%). I did that once thinking that I forgot to save my settings or something, and then realized that leavin it connected overnight brings it back to 100%. I don't think its much of a problem, but it'd be nice to know if it is possible to set thresholds the 'right' way.
christopher_wolf wrote: I don't know of any way to set the thresholds for it to follow when it is offline, however. I don't even think that is supported with the current hardware.
What is the point of having thresholds, then? If it is going to reset everytime you switch it off, I'd bee better off sticking to the 'optimum battery setting' right?

Posted: Sun Dec 24, 2006 1:17 am
by GomJabbar
After doing a little ferretting online, I came across the following pdf document Smart Battery System Manager Specification.

There are some good diagrams in the pdf that show the relationships between the various components. Following is an excerpt. I refer to pages 6, 7, and 8.
The following diagram is a block diagram of a SBSM, Smart Battery Charger and SMBus host as added functionality to the EC (embedded controller). Smart Battery A and/or B are available to power the system.

<snip>

It is expected that the SBSM will operate in an entirely autonomous manner, independent of any highlevel control such as that provided by an application or system BIOS. This autonomy allows the system to charge multiple batteries while the host intelligence is not operational (e.g., when the system is off or suspended). Since the SBSM operates autonomously, it is totally responsible for the battery system's safe operation and for maintaining the power integrity of the system.
The diagram shows an Embedded Controller as part of the Smart Battery System Manager (SBSM). The document does not make clear if the Embedded Controller is active when the power is off (as far as I could find). In the end, it doesn't really matter. It is the Smart Battery System Manager (SBSM) that controls the charging of the batteries when the system is powered off. This is separate from the circuitry in the batteries themselves as the diagram in the above document shows. Since this is independent of the OS, it would appear that any settings made in the OS would have no effect here. What I should have said in my original post (if I had understood the details more clearly) is that the Smart Battery System Manager in a notebook computer is what controls the charging of the battery when the power is off.

Posted: Sun Dec 24, 2006 1:26 am
by christopher_wolf
Yeah, it does limit the threshold functionality to whenever the system is online. Whenever I want to charge my Thinkpads to full, I either just turn off the thresholds or let them charge offline. Else, I don't know of another way to implement those thresholds when powered off. :|

PS: Embedded Controller stuff is getting off-topic fast as nobody said it wasn't involved in battery charging and power management. That's enough for now.

Posted: Sun Dec 24, 2006 6:35 am
by GomJabbar
PS: The OP's question was answered sometime back.
It's the OP prerogative to steer the thread.
Thread fuzz not needed.

Posted: Sun Dec 24, 2006 10:56 am
by christopher_wolf
GomJabbar wrote: Thread fuzz not needed.
Then don't add it, please. If the question was answered, leave it at that as nothing that could help has been added thus far.