- Dec 17, 2012
- 11
- 0
- 0
10 JANUARY 2013 UPDATE
Switching off LTE seems to have worked for me to solve this random battery drop or plummet issue. My phone now drops about 1% per hour when I am not using it. With no supported way to turn off LTE, however, I?ve found that the ?ServiceMode? technique I describe in the 10 January 2013 posting below seems to work best and provides better coverage with no dropouts, as opposed to the APN method I describe in the 5 January 2013 posting below.
06 JANUARY 2013 UDPATE
I was just able to confirm that I live in a location that is on the fringe of service of an LTE tower - and my phone can quickly bounce back and forth between LTE and 4G signals. See my reply of 06 January below. So I think my random and sudden severe battery drain issues are likely caused by the phone constantly trying to make an LTE connection or bouncing back and forth between LTE and 4G.
05 JANUARY 2013 UPDATE
I have now been able to confirm that the issue I describe below is either the same as or closely related to the one in the following thread:
[Help][50% SOLVED] LTE = random heavy battery drain aka "plummet"
In short, I was able to switch to an Access Point Name (APN) that does not support LTE, and I was able to determine that switching to and from the non-LTE APN correlates identically with going from a random/intermittent high idle battery drain of about 7+% per hour to a consistent low idle battery drain of around 1-2% per hour.
See my below post of 05 January 2013 for further details and screenshots.
29 DECEMBER 2012 UPDATE
I've now had the chance to evaluate 3 separate Galaxy S3 phones and have since been able to determine that the idle battery drain issue I describe below occurred on all three devices with similar frequencies, but at somewhat random times and for random durations. Thus, the issue appears not to be hardware specific. Based on my analysis, I would be surprised if these issues are not affecting most Galaxy S3 owners, at least on the AT&T network.
ORIGINAL POST:
I?d like to investigate the prevalence of what appears to be a hardware issue that causes an intermittent 6% per hour draw on an AT&T Galaxy S3 battery while the phone is sitting idle. I think the proper idle draw should be around 1-2% per hour. The problem appears to affect different phones to different degrees and whether it is an issue may be a question of how much or how frequently. Please read below and post a reply if you?re experiencing something similar and to what extent.
Please don?t reply-post various suggestions regarding how to conserve battery power, such as turning off wifi, S-voice, push e-mail, limiting screen brightness, or using Juice Defender. General battery conservation is not the issue here. I?ve got two identically configured phones ? one acting significantly different than the other. I've swapped batteries, factory reset, and upgraded to Android 4.1 to eliminate those as possibilities.
My first AT&T Galaxy S3, let?s call it ?Mostly Bad,? generally draws about 6% of its battery per hour while sitting idle. That means screen off and no user interaction. To me, that?s a problem?
Since I?m in the middle of a mail-in warrantee replacement with AT&T, I also have a second S3, let?s call it ?Mostly Good,? THAT IS CONFIGURED IDENTICALLY, and generally draws down about 1-2% per hour while sitting idle. Now that is the kind of performance I would expect from this phone ? if you leave it on your desk and don?t touch it, you should be able to come back in 24 hrs and see 70+% of your battery remaining, right?
Interestingly, the ?Mostly Bad? phone will behave itself (drawing only 1-2% at idle), somewhat randomly for short periods of time every so often. Also the ?Mostly Good? phone will misbehave, (drawing about 6-7% at idle for relatively short periods of time ? typically for less than 1 hour) every so often. All this, of course, happens while the phones are sitting idle, typically while I?m fast asleep.
On both phones, I?ve been using Battery Spy (free, highly recommended) to graph the remaining battery %, battery voltage and temperature. During my tests, I?m running a relatively light configuration close to the default stock apps with exchange ActiveSync enabled, but the issue seems to be application independent and it persisted through the Android 4.1 firmware upgrade, a new battery, and factory resets. Although these episodes may also start and stop while the phone is in active use with screen on, it?s much harder to detect, since the phone is drawing 10-15% per hour during use.
Here are some characteristics of the bad behavior episodes:
1. Idle draw of 5-7+% per hour (while phone sitting on desk with screen off)
2. Very high frequency fluctuations in battery voltage as graphed by Battery Spy (discussed in more detail below). This seems to be a key characteristic!
3. Episodes randomly start and stop
4. Duration and frequency of episodes seems related to the particular phone (indicates hardware issue or relationship)
5. Will not happen if phone is run without SIM card (even though phone radio is on and able to make emergency calls)
6. Airplane mode suspends behavior (I?m pretty sure about this, but not completely positive)
7. Tendency seems generally independent of applications installed ? I?ve observed this on both straight out of the box and heavily app installed configurations.
8.Can happen in non LTE areas ? I don?t have an LTE signal where I live and there have been postings suggesting an LTE link (perhaps the phone could be looking for an LTE signal in vain?) [I've since learned that I'm on the fringe of an LTE tower and my phone bounces in and out of LTE - which is the problem.]
9. Independent of use of ActiveSync push - occurs on phones with and without Exchange ActiveSync enabled
I?ve attached some screenshots taken using Battery Spy showing episodes of both normal and problematic behavior and I?ve commented in a white box within each screenshot what I?m trying to show. Please pay close attention to the scale on the graph ? some graphs are scaled to 1 or 2 hours while others may be 16 hours and that?s important in observing the battery drawdown rate. Don?t pay attention to the %/hr rate indicated at the top of the Battery Spy graphs as that is just a real-time rate measured at the time the screenshot was taken.
An important feature of these episodes is the very high frequency fluctuations in the battery voltage of the phone that directly correspond to the idle drain. When the phone enters an episode of drain, the Battery Spy graph will show a transition from a very smooth battery voltage to what looks like a seismograph plot when viewed in the fine (1 or 2 hrs) graph modes. I?ve posted some instances where the phone enters and exits an episode from both the Mostly Bad and Mostly Good phones.
I?d like to point out another thread that may be addressing the same problem and which has garnered quite a bit of attention during the past 6 months. The present posting may be of interest to some following the other one:
[Help][50% SOLVED] LTE = random heavy battery drain aka "plummet"
The ultimate question I?d like to answer is whether this is a problem that is pervasive and affects many S3 users or whether it is a relatively rare occurrence. I?ve talked to several other S3 owners who say that they have to charge their phone more than once per day, though I don?t know exactly how heavily they?ve been using it. It?s possible that this is a pervasive hardware problem and many unknowing S3 users may be afflicted but oblivious.
So if you often put your fully charged S3 on your nightstand before sleep and awake 6 or 7 hours later to find it down around 60%, please download and run Battery Spy and see if your battery drain profile matches what I?ve posted here. And then let us know!
Switching off LTE seems to have worked for me to solve this random battery drop or plummet issue. My phone now drops about 1% per hour when I am not using it. With no supported way to turn off LTE, however, I?ve found that the ?ServiceMode? technique I describe in the 10 January 2013 posting below seems to work best and provides better coverage with no dropouts, as opposed to the APN method I describe in the 5 January 2013 posting below.
06 JANUARY 2013 UDPATE
I was just able to confirm that I live in a location that is on the fringe of service of an LTE tower - and my phone can quickly bounce back and forth between LTE and 4G signals. See my reply of 06 January below. So I think my random and sudden severe battery drain issues are likely caused by the phone constantly trying to make an LTE connection or bouncing back and forth between LTE and 4G.
05 JANUARY 2013 UPDATE
I have now been able to confirm that the issue I describe below is either the same as or closely related to the one in the following thread:
[Help][50% SOLVED] LTE = random heavy battery drain aka "plummet"
In short, I was able to switch to an Access Point Name (APN) that does not support LTE, and I was able to determine that switching to and from the non-LTE APN correlates identically with going from a random/intermittent high idle battery drain of about 7+% per hour to a consistent low idle battery drain of around 1-2% per hour.
See my below post of 05 January 2013 for further details and screenshots.
29 DECEMBER 2012 UPDATE
I've now had the chance to evaluate 3 separate Galaxy S3 phones and have since been able to determine that the idle battery drain issue I describe below occurred on all three devices with similar frequencies, but at somewhat random times and for random durations. Thus, the issue appears not to be hardware specific. Based on my analysis, I would be surprised if these issues are not affecting most Galaxy S3 owners, at least on the AT&T network.
ORIGINAL POST:
I?d like to investigate the prevalence of what appears to be a hardware issue that causes an intermittent 6% per hour draw on an AT&T Galaxy S3 battery while the phone is sitting idle. I think the proper idle draw should be around 1-2% per hour. The problem appears to affect different phones to different degrees and whether it is an issue may be a question of how much or how frequently. Please read below and post a reply if you?re experiencing something similar and to what extent.
Please don?t reply-post various suggestions regarding how to conserve battery power, such as turning off wifi, S-voice, push e-mail, limiting screen brightness, or using Juice Defender. General battery conservation is not the issue here. I?ve got two identically configured phones ? one acting significantly different than the other. I've swapped batteries, factory reset, and upgraded to Android 4.1 to eliminate those as possibilities.
My first AT&T Galaxy S3, let?s call it ?Mostly Bad,? generally draws about 6% of its battery per hour while sitting idle. That means screen off and no user interaction. To me, that?s a problem?
Since I?m in the middle of a mail-in warrantee replacement with AT&T, I also have a second S3, let?s call it ?Mostly Good,? THAT IS CONFIGURED IDENTICALLY, and generally draws down about 1-2% per hour while sitting idle. Now that is the kind of performance I would expect from this phone ? if you leave it on your desk and don?t touch it, you should be able to come back in 24 hrs and see 70+% of your battery remaining, right?
Interestingly, the ?Mostly Bad? phone will behave itself (drawing only 1-2% at idle), somewhat randomly for short periods of time every so often. Also the ?Mostly Good? phone will misbehave, (drawing about 6-7% at idle for relatively short periods of time ? typically for less than 1 hour) every so often. All this, of course, happens while the phones are sitting idle, typically while I?m fast asleep.
On both phones, I?ve been using Battery Spy (free, highly recommended) to graph the remaining battery %, battery voltage and temperature. During my tests, I?m running a relatively light configuration close to the default stock apps with exchange ActiveSync enabled, but the issue seems to be application independent and it persisted through the Android 4.1 firmware upgrade, a new battery, and factory resets. Although these episodes may also start and stop while the phone is in active use with screen on, it?s much harder to detect, since the phone is drawing 10-15% per hour during use.
Here are some characteristics of the bad behavior episodes:
1. Idle draw of 5-7+% per hour (while phone sitting on desk with screen off)
2. Very high frequency fluctuations in battery voltage as graphed by Battery Spy (discussed in more detail below). This seems to be a key characteristic!
3. Episodes randomly start and stop
4. Duration and frequency of episodes seems related to the particular phone (indicates hardware issue or relationship)
5. Will not happen if phone is run without SIM card (even though phone radio is on and able to make emergency calls)
6. Airplane mode suspends behavior (I?m pretty sure about this, but not completely positive)
7. Tendency seems generally independent of applications installed ? I?ve observed this on both straight out of the box and heavily app installed configurations.
8.
9. Independent of use of ActiveSync push - occurs on phones with and without Exchange ActiveSync enabled
I?ve attached some screenshots taken using Battery Spy showing episodes of both normal and problematic behavior and I?ve commented in a white box within each screenshot what I?m trying to show. Please pay close attention to the scale on the graph ? some graphs are scaled to 1 or 2 hours while others may be 16 hours and that?s important in observing the battery drawdown rate. Don?t pay attention to the %/hr rate indicated at the top of the Battery Spy graphs as that is just a real-time rate measured at the time the screenshot was taken.
An important feature of these episodes is the very high frequency fluctuations in the battery voltage of the phone that directly correspond to the idle drain. When the phone enters an episode of drain, the Battery Spy graph will show a transition from a very smooth battery voltage to what looks like a seismograph plot when viewed in the fine (1 or 2 hrs) graph modes. I?ve posted some instances where the phone enters and exits an episode from both the Mostly Bad and Mostly Good phones.
I?d like to point out another thread that may be addressing the same problem and which has garnered quite a bit of attention during the past 6 months. The present posting may be of interest to some following the other one:
[Help][50% SOLVED] LTE = random heavy battery drain aka "plummet"
The ultimate question I?d like to answer is whether this is a problem that is pervasive and affects many S3 users or whether it is a relatively rare occurrence. I?ve talked to several other S3 owners who say that they have to charge their phone more than once per day, though I don?t know exactly how heavily they?ve been using it. It?s possible that this is a pervasive hardware problem and many unknowing S3 users may be afflicted but oblivious.
So if you often put your fully charged S3 on your nightstand before sleep and awake 6 or 7 hours later to find it down around 60%, please download and run Battery Spy and see if your battery drain profile matches what I?ve posted here. And then let us know!
Attachments
-
1_Bad_2012-12-16-13-54-49 m1.png144.7 KB · Views: 131
-
8_Good_2012-12-18-01-09-21 m1.png128 KB · Views: 125
-
7_Good_2012-12-18-02-01-50 m1.png143.1 KB · Views: 128
-
6_Good_2012-12-18-02-01-36 m1.png138.4 KB · Views: 126
-
5_ Good_2012-12-17-12-46-45 m1.png149.2 KB · Views: 126
-
4_Bad_2012-12-17-12-46-43 m1.png140.4 KB · Views: 127
-
3_Bad_2012-12-16-13-57-55 m1.png130.6 KB · Views: 127
-
2_Bad_2012-12-16-13-57-35 m1.png130.8 KB · Views: 129
-
9_Good_2012-12-18-15-27-43.png141.4 KB · Views: 114
Last edited: