-
All replies
-
Helpful answers
first
Previous
Page
38
of 104
last
Next
-
Nov 15, 2009 11:55 AM in response to pbraunsby ErvDoggie,Hey Phil:
You're very welcome. I'm glad the information I've shared helps people get a better understanding. I spent 10 years as an I.T. Engineer--specifically with laptops in general, so I'm hopeful that experience can shed some light for others, and give them some things to try until a more permanent solution is made available.
There is no point to having a MacBook if it must be plugged into A/C just to use it. From my recent test, getting 2 hours 45 minutes suggests the cells in my battery are fine--somehow the OS has lost its ability to go to sleep like it should.
Phil, I'm unsure if your MBP has a "charge indicator" built into the battery as with my 2006 Core Duo Black MacBook, but it would come in handy if it does.
On the bottom of my battery, I have a button that I can push that tells me how much charge there is. I'd check that the next time you got a system shut down and also note how long you managed to use it before the shutdown if your battery has that indicator built in too.
After the shutdown, I'd do the SMC reset as described by Apple for your particular model. On mine, I have to hold the power button down for an amount of time before plugging everything back in and letting it sit plugged in on A/C.
Even if you got 30 minutes out of the battery before your system goes out, I'd check the indicator on the battery (if you have one) note the charge, and perform the SMC Reset again and charge it for another 24 hours. This is what I've been doing to get my battery issue to settle down--excessive?? Sure, but at least I can get 2+ hours out of my battery according to my latest home grown battery testing experiment so far without the random shutdowns. More testing required for sure.
If I had your system, I'd do that and download CoconutBattery to help monitor stats over time by taking snapshots of the battery state every day after a "full charge".
Then, I'd find a day when you have a few hours to use it until the battery conks out. If you can get more time out of it, fantastic. I'd use it until it went dead again, then let it sit overnight disconnected (I did for 12 hours and did a 3rd SMC Reset in the morning)
I'd then plug it all back up and let it charge on A/C for 3 days if you can--using it is okay during this time.
I'd check the status with Coconut Battery, shut down, disconnect A/C and then use it on battery again to see if that at least settles the "randomness" of the shutdowns.
As in my case, I got 2 hours 45 minutes after my 3rd reset, and my low battery warning window popped up for the first time in 3 weeks when the battery hit 6% charge remaining. Who knows if its a fluke, but time will tell--perhaps its a learning routine it has to do all over again.
Good luck either way! I still say getting a new battery is a good thing. -
Nov 15, 2009 6:45 PM in response to ErvDoggieby ErvDoggie,Battery Test Number 4 - I'm HIBERNATING again!! No more shutdowns!!
(You'll need another beverage to read this)
Okay, so I've had a lot of time on my hands this weekend to really work through my home grown battery calibration/battery usage testing and results.
I'll spare you all the gory details of how I did the test. My prior post dated Nov 15, 2009 1:02 AM in response to "pbrauns" (Phil) contains a lot of the details of what was done and how I did it. I'll share more of what I noticed and what I believe is happening.
If I continue to get consistent battery performance results over the next few days/weeks, I'll do a more condensed version of my home grown battery calibration fix and post for others to try in an effort to speed the battery calibration process along and hopefully save them some wasted cash and time on a new battery if they don't really need one.
In the meantime, here are the results of my 4th Battery test. Unfortunately this isn't as "coherent" and formally documented/tested as I would have preferred--however given my extreme level of frustration at the battery issue (and hearing those complaints of others) I've been compelled to see if I can't figure out what really is happening based on my many years of I.T. experience, observation, and the way my old trusty 2006 Black Core Due MacBook was behaving before SL.
I'll share my personal "theory" on why we are having this problem at the end. Okay, so on with the results of test number 4 in comparison to test number 3 which I posted earlier in response to Phil.
BACKGROUND
After my update to SL, I had two problems relating to the battery. Significantly reduced runtime on battery power, no low battery warnings, and finally no hibernation--okay, three issues. Other folks have had the "Service Battery" warning, and random shutdowns.
After the conclusion of test number 3, where I responded to Phil, I reported:
1 - I was able to get 2.5 hours of wireless usage out of my system using the defined configuration settings and applications running as an informal benchmark test for myself.
2 - I received the low battery pop up window when my battery reached 6% charge remaining (I haven't seen this warning for weeks)
3 - My system never hibernated like it used to shortly after it issued the low battery warning--it just powered off causing loss of any unsaved data.
TEST #4
So what did test number 4 reveal?
It was fairly late by the time I finished test number 3, so I just plugged the A/C back in and went to bed
The system charged overnight, and with a few hours on my hands the next day, I thought I'd use my system wirelessly again and do test number 4--except that I would put a hurting on the CPUs and battery by doing some more aggressive tasks and really putting some power draw on the battery.
I started up my system wirelessly, logged in, waited the cursory 5 minutes and used all the same softwares/tools I defined in test number 3 so I could have some factual data of what I was noticing.
The battery meter in the toolbar reported 2 hours 16 minutes charge remaining (2 hours 57 minutes in test #3).
So I set the egg timer to match the time remaining so I could monitor the systems ability to "estimate" remaining runtime. I also hit start on the chronograph at the same time to record total runtime once again.
To put high current draw on the battery during this session, I ripped a DVD and forced a Hard Drive Re-Indexing (both power hungry activities) at the same time while doing all my other web surfing, note taking and calendar updating.
Needless to say, with that kind of load on the system, the fans kicked on full blast (expected), Activity Monitor showed CPU usage above 60% until the DVD rip completed and only the Hard Drive indexing remained and had another 90 minutes to go with the fans slowing down a little.
I monitored the battery discharge via the Terminal window with my "pmset -g pslog" command via the terminal window.
Everything went as I was expecting. With test #4, I didn't expect as much runtime as I got in test #3 obviously. In test #4 I got a total runtime of 1 hour 31 minutes with all the new stress I placed on the system ripping a DVD and re-indexing the hard drive.
THE GOAL
I ran test number 4 specifically to look for:
1 - Did the low battery warning pop up, and if so, how much battery charge remained?
2 - Would my system hibernate once the battery got low enough or just cut out as in test #3 and all the weeks prior?
THE RESULTS
To my surprise, I received the low battery warning pop up window which was nice, and much welcomed.
Upon checking the PS Log to see how much juice was left when I received the warning, it showed "8% remaining". In test number 3, I got the low battery warning at "6%".
This would suggest to me the system has/is learning that 6% was too late to hibernate, so it better issue the warning at 8% this time and make sure it has enough power to create the sleep image.
The system also shaved "41" minutes off the battery run time at the beginning of test #4. This also suggests to me that it is "learning" where the limits are on the battery, and when to issue the appropriate warnings and hibernate as it should.
I watched the PS Log window until power remaining hit 1%, then amazingly IT HIBERNATED!!! After the pop up warning, I worked for another 5 minutes before hibernation.
So, I watched as the system created the sleep image file and then powered down with the slowly blinking LED indicating it was sleeping. How fantastic is that?!?
I plugged the A/C power back in, tapped a key on the keyboard to wake it up and everything came right back as it should.
MY THEORY
Okay, so my "theory" as to what is going based on what I'm noticing (at least on my early 2006 Core Due Black MacBook):
SL wipes out the SMC Power Management/battery information previously "learned" by Leopard. Without a frame of reference, it has to "learn" all over again. Given the new more powerful OS, I'd expect lower battery runtime to some degree however, when combined with aging/defective batteries, people not really knowing how long the calibration process will take etc--well insert this forum as to the list of issues.
The SMC isn't anything we can touch or program, we can only "reset it"--which means (to me anyways) to re-learn by setting a new baseline from which to operate, hence the weird results, random shut downs, inability to hibernate when it should, run the battery til its dead--SL has NO idea of where the "limits are" and must find them again, set a baseline and go from there.
My Test #2 (which really wasn't a test) was the 2nd time my system ran the battery completely "flat" after my initial SMC Resets before it, which encouraged me to create a more structured test as #3, followed up by test #4 given my years as an I.T. Engineer.
Test #3 is where I gathered the software tools, timers, and decided to apply a more scientific approach to what I was noticing to make sure I wasn't guessing at what was going on. I had to be able to measure definitively what was going on so I could tell if there was any improvement taking place.
SO WHAT DOES THIS ALL MEAN?
For me, I'll continue testing my system to ensure its properly issuing the low battery warnings and hibernating as it should--a return to normal for me as it relates to the battery.
For everybody else (based on my limited test results), I would recommend you do this if all of you were my clients telling me your laptops had these various challenges--A truly bad battery notwithstanding, I would suggest:
MY BATTERY STABILIZATION PROCESS (preliminary):
1 - Charge your system for 24 hours, get the "Coconut Battery" app, press the blue diamond and hit "save" to take a snapshot of your battery stats as of that moment with a FULL charge, A/C plugged in. May as well grab System Profiler Battery Stats while you're at it.
2 - Shut down and disconnect the A/C when its powered off.
3 - Make sure you have at least 3+ hours (enough time to run the battery out) and get an egg timer and a chronograph so you're not "guessing" and recording actual facts. Perception is reality right?
4 - Power up the system, log in (if you must) to see your desktop, note the system time in the toolbar and wait 5 minutes.
5 - After 5 minutes, hit "start" on the chronograph, and set the egg timer to the battery runtime remaining and hit start.
6 - Use your system however you use it until it powers off, hibernates etc.--doesn't really matter.
7 - Whichever event occurs, hit stop on the egg timer and chronograph (oh, you'll need a sheet of paper)
8 - Record how much runtime you actually got from the Chronograph, and note how much estimated runtime the system indicated you were supposed to have left.
Also note whether it hibernated, or just powered itself completely off, reset, etc.
9 - I'd do an SMC reset (as defined by Apple). On my system, I had to pull the battery and hold down the power button for 5 seconds (I did 10 seconds to be sure)
10 - If you can leave it all disconnected, (battery out) overnight as I did even better.
11 - Put the battery back in and charge again for 24 hours A/C in non-stop (you can use it during this time) and repeat the process using it wirelessly while appropriately recording the times--"NO SMC RESET THIS TIME!".
12 - Continue fully charging and using your system wirelessly at least 3-4 more times, recording the results as outlined above.
Hopefully, if the SMC is truly re-learning, your battery runtime should stabilize and the power loss/resets should go away--a truly bad battery not withstanding.
I believe the SMC has to learn all over again and people aren't being told what to expect or how long the re-learning process could take--just my theory anyhow.
Its a bit premature to be writing this, however, based on what I'm seeing on my MacBook thus far, it appears to be "learning" all over again.
Only subsequent wireless usage tests will confirm the findings. Hopefully others will see improvements as I am seeing.
If nothing else, I've re-calibrated the system and helped it to "re-learn" how much runtime it can get with my present battery, the new power requirements of SL, and to properly shut down/hibernate when the battery reaches its predetermined point before hibernation-- in other words, business as usual
Of course, this is only my experience as it relates to my specific situation and system.
Who knows if any of this will help anybody else. This is by no means absolute--however, if you wanted to recalibrate the way I did, you may get similar improved results.
Good luck!
There is nothing worse than having a laptop that goes dark when it says it has 20% or more battery power remaining.
-Erv -
Nov 16, 2009 7:54 AM in response to blieuxby Jerry Tachoir,I have two batteries for my MacBook Pro. After installing SL, the computer would randomly shut down with no warning. My batteries (both) still had plenty of charge left. Then I noticed the Service Battery warning. This is definitely a Snow Leopard issue. We will have to wait until Apple released an update.
Don't run out and buy a new battery, because it will just happen again with the new battery.
Other option is to downgrade to Leopard.
This *****, but hopefully Apple will have a resolution soon. There are too many people dealing with this and a lot just use their AC power supplies and aren't away of the battery issue.
Be patient! -
Nov 16, 2009 8:09 AM in response to Jerry Tachoirby DiegoRA,Hi, I have a early 2008's macbook pro, y had the same problem after upgrading to SL
but I did this and actually works till today (about 4 days in a row and counting)
- turn off the macbook
- remove the battery
- turn it on with the power supply (without the battery)
- check if any message about battery service appears
then...
- turn it off again
- put the battery, and wait till it's fully charged
- turn it on only with the battery (no aditional power supply)
- check if the message appears again
now
- put the power supply and wait 5 minutes to see if the message appears again
- the message will appear about 2 more times and then it's solved (or kind of)
I believe that is because the OS can't sincronize the info with the battery and when the info from the batt is a "low charge" and the system puts in a hybernation mode to keep the info while u plug the power suply...
this actually worked for me, and I hope it will work 4 u guys too...
(it might take a little more time to fix that, but it worked 4 me...) -
Nov 16, 2009 12:40 PM in response to DiegoRAby pbrauns,Hey Diego,
ran the MBP on battery and it had a sudden shutdown at 1min 15 sec - never had that before, normally at 75% with the "service battery" sign now already gone for a couple of days) ...the battery leds are indicating 4 led (= almost fully charged)!oh well...
then i tried out your routine and this is what happened: i removed the battery, ran it on ac with the battery diconnected, shut it off again, charged the battery, removed the cable, powered it up again only on batt and now i´m running the thing on batt with no "service battery" sign and 3:52 h time remaining. hhm, so quite an improvement (before your routine it said 1h15 remaining)
i wonder, if it´s going to die on me any minute while i´m writing these lines:-) so keep your fingers crossed.
if it´s again going black at about 75%, i´ll try "ervDoggie´s" routine with also resetting the smc and letting the battery rest while being disconnected overnight.
all this battery cr.. is sooo strange after all. in the end, i guess only a software update from apple will really help. there´s noooo way though i´m buying a new batt since i´m veeery convinced mine´s still fine
phil -
Nov 16, 2009 2:38 PM in response to pbraunsby ErvDoggie,Phil:
Surely let us all know how it goes using Diego's routine. From an "I.T." perspective, I'm very fascinated.
I hope that it works--it'd be much less involved than my procedure for sure.
Your results will be very helpful. Thanks for posting the info on the charge status of your battery as of your last "random" outage. I'd be surprised if an SMC reset and focused SMC/battery calibration won't prove beneficial on your system.
Personally, my theory is the SMC/OS on your system still doesn't know how to "play" together just yet because a properly calibrated routine hasn't been established.
From an I.T./test lab perspective, based on my results, Diego's process wouldn't prove successful longterm, as the SMC hasn't been reset and immediately forced to run your batteries completely "flat" with focused charged/discharge sessions to accurately determine the limits of what your batteries charge really is and how long it can run--just a theory from what I noticed on my system. I hope Diego's process works, its much less involved than what I went through.
In my scenario, I started my calibration process with a single SMC reset, which I believe cleared any bad power management data/behavior the system has learned and forced the SMC to relearn.
Because I forced a very methodical charge/discharge/charge/discharge/charge/discharge cycle--the SMC learned faster and more accurately than it otherwise would have without such a focused approach immediately following the SMC reset.
I believe my first calibrated charge/discharge cycle is where my system had to set a baseline and learn what the batteries really could deliver--i.e. run completely to "0" charge.
I believe the 2nd charge/discharge cycle, the system attempted to establish some baseline in which it could safely "hibernate" with reserves, and if it didn't, it would continue this process until it was successful.
The only thing I got in this cycle was my long missed "low battery warning pop up" which displayed at 6% remaining before it went "flat"--no hibernation.
I believe the 3rd charge/discharge cycle the SMC learned/is learning and issued my low battery warning at 8% remaining instead of 6%, in an attempt to hibernate and have some amount of battery reserves left. It did issue my warning and successfully hibernate.
I believe my 4th test was the first in a new series of "repeat tests" to confirm if my battery challenge has truly gone away.
In this session I ripped a DVD, and otherwise put a serious amount of processing power to work to mimic real world usage I would do working wirelessly, and it was successful at issuing the warning and hibernating.
Today, as I'm writing this, I'm running yet another wireless usage test (#5), monitoring total battery run time, low battery notification, and battery charge remaining estimation, and if hibernation is successful--stay tuned.
I'll work wirelessly for most of the next few days making my notes to see if my performance is consistent, and if so, should help others.
Those first few battery charge/discharge cycles I think are the most critical after an SMC reset.
Just like our kids, we want them to learn "good habits" starting out--the first few times
Good luck! -
Nov 16, 2009 5:06 PM in response to DiegoRAby madrich,Everyone seems to have their theory and/or method to solve this problem. After replacing my battery I have had no problems. Right now my battery indicates 3:50 hours of power left. It continues to show Condiition:Normal.
I believe that only way anyone is going to be sure that their battery is good or not is to take it to the Apple store and have it tested with the software the Apple Geniuses have on an iPod. -
Nov 17, 2009 10:23 AM in response to blieuxby Shigglyboo,I'm not entirely sure it's malicious, however, the steady deniers certainly make me think something's up.
My system is back to normal somehow some way. Which says to me... it's DEFINITELY NOT A BAD BATTERY. I haven't updated to 10.6.2, now that my system is working again I'm waiting to hear the the issue has been completely resolved.
I think taking the battery out does help. All I did was run my battery all the way down, then let it bottom out, completely dead, then charge back up and leave it plugged in for a while, then repeat. One day after not using it for a few days I powered up and the warning was gone, battery was reported as normal and battery life was restored.
Explain that one to me all you guys saying a new battery was the solution? -
Nov 17, 2009 3:30 PM in response to blieuxby Dimitrios Varthalitis,I wish i could also discharge my battery, but unfortunately my computer shuts off when the battery is at 85%. When i press the button of the battery 4/5 light turn on which show that it has current, but i can not discharge it completely no matter what. Does anyone have any idea about doing something to discharge it? Maybe connecting the possitive and negative pole (but i do not know which is which) with some resistance (e.g a small lamb or something) in order to discharge it so that i will be able to fully charge it after?
After 10.6.2 the service battery sign disappeared and the system profiler states that my battery is normal:
Battery Information:
Model Information:
Manufacturer: Sony
Device name: ASMB012
Pack Lot Code: 0003
PCB Lot Code: 0000
Firmware Version: 102a
Hardware Revision: 0500
Cell Revision: 0303
Charge Information:
Charge remaining (mAh): 4509
Fully charged: Yes
Charging: No
Full charge capacity (mAh): 4720
Health Information:
Cycle count: 19
Condition: Normal
Battery Installed: Yes
Amperage (mA): 0
Voltage (mV): 12299 -
Nov 17, 2009 4:19 PM in response to Dimitrios Varthalitisby ErvDoggie,Dimitrious:
I wouldn't do what you suggest in your post. I'd consider two other alternatives.
If your system is consistently shutting off (not hibernating) at 85%, I'd try two things first--one less process driven than the other.
1 - Another user reported shutting his system down, removing the battery, plugging in the A/C and turning the system back on.
Do that and log in (if you need to) and once you get to your desktop, the menu bar should report there is "no battery".
I'd run it like that for 5 minutes, then shut the system back down, remove the A/C, plug in the battery, then plug in the A/C and leave it plugged in for 24 hours if you can.
Its okay to use your system during this time.
Once you've got a full charge, I'd try using the system wirelessly again and see if you can get past 85% charge remaining without a power loss.
Report back on your results if you can.
If you still have power shut downs, you can see one of my earlier posts, where I initiated a SMC reset after my battery died and left it all unplugged for 12 hours.
2 - If you look at one of my earlier posts, you can see how I forced a very "guided" calibration immediately following an SMC reset.
My shutdowns stopped occurring as early as they were, and I'm able to get a decent amount of runtime wirelessly. My particular system just doesn't hibernate like it should consistently from the early tests I've run at this point.
My post has the details of what I've done and the results. I'm still recording results of wireless usage to see if my particular issues have subsided.
Apparently a new battery doesn't guarantee the issue goes away according to some customers. Others have reported the problem is solved with a new batter--only time will tell if that is a permanent fix.
Good luck, and let us know which you decide, and what your results are.
-Erv -
Nov 17, 2009 5:40 PM in response to blieuxby AndrewG2009,Just the latest update....
It has been 5 days of steady improvement with my battery and I have done nothing special (no SMC resets, no calibrations). My battery is now back to 93% health (3820 mAh). This is were the battery was PRE - SL. It has been at this level for 2 days now. The only change that has been made to the computer was the Safari update a few days ago.
I have my brand new battery here from apple (they said mine was bad and issued a new one (the system profiler said it was 2068mAh.
I guess I will put the new one in anyway? I have a few days before I have to send the old one back.
Current Status:
Battery Information:
Model Information:
Serial Number: 9G848049Y3NYB
Manufacturer: DP
Device name: bq20z951
Pack Lot Code: 0000
PCB Lot Code: 0000
Firmware Version: 002a
Hardware Revision: 0005
Cell Revision: 0100
Charge Information:
Charge remaining (mAh): 3612
Fully charged: No
Charging: No
Full charge capacity (mAh): 3820
Health Information:
Cycle count: 130
Condition: Normal
Battery Installed: Yes
Amperage (mA): -1074
Voltage (mV): 12112 -
Nov 17, 2009 9:03 PM in response to blieuxby Jon_Doe,AndrewG that's really interesting ... the smc-learning-theory more posts above is interesting as well ...
this issue smells so fishy ...
i've bought a new battery this week (yeah apple you won this round!)
BUT ... i might keep the old one in and see what happens ... i'll keep updating -
Nov 18, 2009 1:39 AM in response to Dimitrios Varthalitisby pbrauns,hey dimitrios,
now, this is EXACTLY the issue i´m having too! I can´t completely discharge the d... thing! the "service battery" sign is now already gone for days, but sudden shutdowns persist (at around 85%)....the routine "diego" posted earlier didn´t help (hey thanx anyways!)
thought i´d share my latest battery stats with all of you guys though.
4592 mAh full charge capacity (of my batt, see below)?!....I guess that´s really true after all..AND since the leds of the battery itself indicate an almost full charge (4 of 5 leds) when checking right after a sudden shutdown, it really HAS 85% capacity left, but SL just forces the system to shut off.
the strange thing is, that i really don´t know, what made the "service battery" stuff go away, since it initially kept being there after installing 10.6.2, smc resetting and all that?!hhm
see ya,
phil
Informationen zum Batteriemodell:
Hersteller: Sony
Gerätename: ASMB012
Pack Lot Code: 0000
PCB Lot Code: 0000
Firmware-Version: 0110
Hardware-Version: 0500
Zellen-Version: 0303
Informationen zum Ladezustand:
Verbleibende Ladung (in mAh): 4592
Vollständig geladen: Nein
Batterie wird geladen: Ja
Volle Ladekapazität (in mAh): 4592
Informationen zum Batteriezustand:
Anzahl der Zyklen: 91
Zustand: Gut
Batterie ist installiert: Ja
Stromverbrauch (in mA): 29
Spannung (in mV): 12619 -
Nov 18, 2009 4:14 AM in response to splatnikGby STCav,splatnikG has an interesting point about error rates and coincidence implying a link to a software upgrade. I have exactly the problems described in this post, but I am *still on Leopard*.
My problems began after I did the last routine Software Update, so I assumed a software issue. However, I have just had a very helpful conversation with Apple, which is supplying a new battery FoC. This is my second replacement battery under Apple Care. The current faulty battery is a Sony ASMB012, as with other posts on this thread. -
Nov 18, 2009 10:56 AM in response to ErvDoggieby STCav,Erv,
Many, many thanks for your helpful and informative posts. As I said in my last post, I have exactly this issue with my MacBook Pro running Leopard 10.5.8, and the symptoms appeared soon after a routine Software Update.
After running a few tests as you suggested, it seems the battery is only about 38% healthy according to CoconutBattery, although it jumped from 38% to 65% when I unplugged from the mains, moved to another room and plugged in again within 60 seconds. It's currently showing a mismatch between the information bar at the top ("Charged"), the System Profiler ("Fully Charged") and CoconutBattery ("96% charged").
Apple is sending me a new battery, having already supplied a new battery in October 2007, but in case it helps your diagnosis stats, I attach the System Profiler battery information. Remember, this is all on Leopard not SL.
Battery Information:
Model Information:
Serial Number: Sony-ASMB012-36ff-75d7
Manufacturer: Sony
Device name: ASMB012
Pack Lot Code: 0003
PCB Lot Code: 0000
Firmware Version: 102a
Hardware Revision: 0500
Cell Revision: 0303
Charge Information:
Charge remaining (mAh): 3526
Fully charged: Yes
Charging: No
Full charge capacity (mAh): 3652
Health Information:
Cycle count: 253
Condition: Check Battery
Battery Installed: Yes
Amperage (mA): -1197
Voltage (mV): 11939
Once again, thanks for the thought and effort you've put in to your posts.
Regards,
Stephen