Page 1 of 2 12 LastLast
Results 1 to 20 of 21
  1.    #1  
    UPDATED (Scroll down for "New Data")

    Treo 650 Battery Drain Log

    In an attempt to create the most boring post ever on TC, I present the following. I decided to do a log of battery usage over a day or so after I had some problems with Ptunes not shutting off all the way.

    I monitored usage using ClockPop 5 which gives both percentage battery life left ("%") and current remaining in volts ("v"). Periodically I double checked this figure with what is reported by FileZ and they were the same.

    I found it interesting how the current numbers do not match up with the percentage battery life left. I started out with 4.17v and 99% and as of this post I am at 3.75v at 35% Obvioulsy the current is much higher still than 35%, but perhaps it is not liniear.

    It is also interesting how the % would sometimes go down without a change in the volts and how sometimes the volts would creep back up after a "rest" from using an app.

    I might as well take this to when the thing goes dead on me so I will update later.

    Set-up Notes:

    using KB lightsoff so keyboard lights were off almost entire time of test except for needing lights for a moment when web browsing.

    Shadowmite patch installed with DUN, Voice and Notepad

    NO BT update installed. BT and Beam recieve turned off.

    Using the Clock Pop battery utililty that shows % and vs

    When plugged into charger, highest v value was 4.20 but often stayed at 4.19

    Was careful to turn off screen (quick red button push) when not doing anything

    Except for when playing ptunes, sound shut off manually.

    Clock Pop 5 and Treo Alarm activated

    Used memopad to jot down the percentages and times.

    ***

    Test 1 Fully charged 99% 4.19v plugged in

    1:45pm 99% 4.17v
    2:02 99% 4.17
    2:20 99% 4.17
    2:47 99% 4.16

    Test 2 Recharged to 4.20v plugged in

    3:11pm 99% 4.17v
    3:29 99% 4.16
    3:45 99% 4.16
    3:57 99% 4.15
    5:31 99% 4.14
    5:54 98% 4.13
    6:17 98% 4.13
    6:34 97% 4.12

    After one phone call 7min 30sec long:
    6:43 94% 4.09

    After 10 min playing Aggression and 20 min on TreoCentral reading posts:
    7:35 90% 4.06

    After 10 min of Streaming music from web via PTunes (screen on whole time):
    7:46 87% 4.04

    After 13 min web browsing on Blazer
    8:08 84% 4.01

    8:38 83%

    After one game of Tetris
    8:56 81% 3.99

    11:15 76% 3.96

    NEXT DAY

    7:30am TreoAlarm goes off, gets weather report.
    Check out news on web for 10 min and 37 sec call.
    8:30 60% 3.84v

    After reading Versamail for 5 min
    8:35 59% 3.84

    9:55 56% 3.89
    10:55 54% 3.81
    11:04 54% 3.91
    11:48 53% 3.81
    12:10 53% 3.81
    1:03 52% 3.80

    After using DA to look up number, then wrong number (dialed wrong!) then 3min phone call:
    1:06 50% 3.78

    1:16 49% 3.79

    After 10 min of reading memos notes to typed this report
    1:27 47% 3.78

    1:34 47% 3.79

    2:36 46% 3.79

    3:28 45% 3.78

    10 min ph call
    3:39 41% 3.74

    After 1:35 phone call
    3:56 40% 3.76

    4:51 40% 3.77

    After 2:47 call
    5:11 38% 3.77

    6:15 37% 3.76
    6:30 36% 3.76
    6:52 34% 3.74

    NEW DATA

    After goofing around some on the internet and checking mail, but no additional phone calls

    11:00pm 19% 3.67v

    Playing Yatzee for 20min

    11:32 14% 3.66v

    Midnight 12% 3.66v

    At this point I turned the phone off by holding down the Red button. Got the Sprint "powering off" screen, made sure the display was off and went to bed. I was hoping that there would still be juice the next day since, with the radio on I had 16% drop overnight the night before.

    At 8am the next morning, however, the phone was totally dead and would not power up.

    Plugged it in just for a second to turn it on, then immediatly unplugged it again. The battery read:

    0% 3.52v
    Display Said my battery was extremely low and should plug into charger

    In just a couple of minutes it then read

    0% 3.47

    And THIS is where display said critical and was shutting down in 30 seconds.

    Interestingly (took a while to get to interesting, didn't it!) according to FileZ my "warning" should come on at 0.10v and the "Critical" level is 0.0v. Further, if I really had (at midnight) a steady 3.66v out of 4.17 I would not be at 12% battery capacity so, obvioulsy, the "volts" being reported by the 650 to the various programs is, in some way, not accurate. Now a battery can certainly have ups and downs depending on temp and drain and I can understand how you can "fool" it for a moment into thinking that there was more current by turning on and off apps or plugging it into the wall for a second, but there is something clearly wrong with the current report on this device.

    Also, with my old 300, if the battery was getting low, I could turn the thing off and there would be VERY little drain of any kind. With the 650, it or apps that dont shut off all the way are draining the battery even when you try and turn everything off.

    "Lessons of the test" reported below
    Last edited by midmofan; 01/07/2005 at 12:42 PM.
  2. #2  
    Quote Originally Posted by midmofan
    Treo 650 Battery Drain Log

    In an attempt to create the most boring post ever on TC, I present the following.
    mission accomplished!
  3. #3  
    Well, I like the info. I was a little worried about my battery life, and this sort of data is quite helpful.

    Thanks!

    Tyler
  4.    #4  
    Quote Originally Posted by midmofan
    UPDATED

    Treo 650 Battery Drain Log

    ***

    NEW DATA

    Midnight 12% 3.66v

    At this point I turned the phone off by holding down the Red button. Got the Sprint "powering off" screen, made sure the display was off and went to bed.

    At 8am the next morning the phone was totally dead and would not power up.
    Plugged it in just for a second to turn it on, then immediatly unplugged it again. The battery read:

    0% 3.52v
    Display Said my battery was extremely low and should plug into charger

    In just a couple of minutes it then read

    0% 3.47

    And THIS is where display said critical and was shutting down in 30 seconds.

    Interestingly (took a while to get to interesting, didn't it!) according to FileZ my "warning" should come on at 0.10v and the "Critical" level is 0.0v. Further, if I really had (at midnight) a steady 3.66v out of 4.17 I would not be at 12% battery capacity so, obvioulsy, the "volts" being reported by the 650 to the various programs is, in some way, not accurate. Now a battery can certainly have ups and downs depending on temp and drain and I can understand how you can "fool" it for a moment into thinking that there was more current by turning on and off apps or plugging it into the wall for a second, but there is something clearly wrong with the current report on this device.

    Also, with my old 300, if the battery was getting low, I could turn the thing off and there would be VERY little drain of any kind. With the 650, it or apps that dont shut off all the way are draining the battery even when you try and turn everything off.
    So what we all learned from this exercise:

    1. You can't trust the current("v") numbers being reported by the 650 to the programs that display them -- something just isnt right. Either the current numbers are reported much higher than they really are, or, the 650 is turning itself off much sooner than it should have to by following the "%" number and not the actual voltage.

    2. I purposfully did not make many phone calls during the test, had I done that the battery would have been used up much faster. Just taking a look at the numbers I got, a short phone call drains the battery much faster than either game playing or internet surfing, or even streaming audio via Ptunes.

    3. With very light phone usage, but playing around with email, the web and some games, I effectively had 1 1/2 days of usage on a single full charge. Had I been out of town or something and known I was in trouble, I could have probably stretched this to 2 days by not doing anything non-essential. For normal use, however, the 650 needs to be plugged in, at least for a while, every single day.

    4. If you are low on battery and don't need the phone, say, overnight, take the battery out since no matter what you do, the thing still draws the battery down. Indeed, the 650 SHOULD have had a "hard" on/off switch that truly turns the power to the thing off, just like you pulled the battery. That is one of the advantages of the non-volitle memory and not having such a switch is really dumb, IMO.

    5. We really need a "resource manager" app that will tell us what programs are running and using battery power at any given time.

    6. I had WAY too much time on my hands this week!
  5. icewing's Avatar
    Posts
    77 Posts
    Global Posts
    78 Global Posts
    #5  
    Quote Originally Posted by midmofan
    4. If you are low on battery and don't need the phone, say, overnight, take the battery out since no matter what you do, the thing still draws the battery down. Indeed, the 650 SHOULD have had a "hard" on/off switch that truly turns the power to the thing off, just like you pulled the battery. That is one of the advantages of the non-volitle memory and not having such a switch is really dumb, IMO.

    5. We really need a "resource manager" app that will tell us what programs are running and using battery power at any given time.

    6. I had WAY too much time on my hands this week!
    4. Have you tried your suggestion, i.e., taking the battery out at night and replacing it in the morning? Should work, just wondering...

    5. Agree

    6. Definately Agree!!!

    Thanks for all the work in documenting your research!
  6. #6  
    Quote Originally Posted by midmofan
    . I had WAY too much time on my hands this week!
    For sure!!


    But thanks for the time, effort, and useful data.
    Nanotechnology Nerd
    i300-->i330-->i500-->6700(1 wk!)-->Sprintt650-->gsm650-->HTC Universal (1 mo.)-->gsm650-->Cing8525(3wks!)-->gsm650
  7.    #7  
    Quote Originally Posted by icewing
    4. Have you tried your suggestion, i.e., taking the battery out at night and replacing it in the morning? Should work, just wondering...

    Thanks for all the work in documenting your research!
    YW, no I havent tried that yet, but will do that tonite or tomorrow. I will note the battery level right before i take it out and when i put it back in.
  8. jkoons's Avatar
    Posts
    215 Posts
    Global Posts
    217 Global Posts
    #8  
    Excellent work Midmofan! What was the setting for the backlight brightness?
  9.    #9  
    Quote Originally Posted by jkoons
    Excellent work Midmofan! What was the setting for the backlight brightness?
    For the display, during the work day it was pretty bright, I would estimate about 75-80% After 6pm or so I was either at home or in a place where adult beverages are served and dropped the brightness to a little below half, around 40%

    As I mentioned, keyboard lights were off the entire time except for a few seconds when I needed them in a dark room.
    Last edited by midmofan; 01/07/2005 at 12:37 PM.
  10.    #10  
    Recharging the 650 from a total drain also shows the "disconnect" between the numbers that are reported for 'current" and '%' charge. The 650s specs, however, that says the battery is fully charged in 4 hours are pretty accurate.

    8:46am 0% 3.51v
    10:20 75% 4.08v
    11:06 92% 4.18v
    11:32 96% 4.19v
    12:30 100% 4.18v
  11. #11  
    The fact that the battery went dead overnight with the phone off is quite bizarre. I am positive that this was caused by one or more applications working the device while it's off. I bet that a hard-reset device with no added apps (and no email setup, etc) would not exhibit this behavior (or at least I would be shocked if it did).

    For the record, KB LightsOff does absolutely nothing when the device is off.
  12.    #12  
    The app that I STILL think is using some juice even when "off" is PTunes. If I do a streaming audio, then turn it off, when I come back to Ptunes there will be the last song/station i was streaming scrolling across the "marquee" on the PTunes dislplay. If I hit play, it reconnects and goes right back to that streaming site. I think Ptunes is staying partially active no matter what you do and draining the thing a bit. The test I am going to do tonite is to turn off the phone when i go to bed, but with a sufficent charge to last the night so I can see for certain what the drain was, then tomorrow night I will first remove PTunes then turn off the radio to see if there is any difference.

    And, btw, totally unsceintific since I was not keeping track of this earlier, but my impression that KB Lightsoff makes a huge difference in saving battery life. That keyboard backlite has got to use a lot of power. Perhaps I will run a test with KB Lightsoff activated and not activated to see what the difference in battery usage is.
  13. #13  
    From what I see, there is a difference between pause and stop in PTunes. You have to touch the screen (and hit the stop button) to stop. Hitting the center button on the 5-way nav only does the pause action. On the surface it looks like pause acutally only pauses the stream, but remains connected, which may explain this quite well.
  14.    #14  
    Quote Originally Posted by silverado
    From what I see, there is a difference between pause and stop in PTunes. You have to touch the screen (and hit the stop button) to stop. Hitting the center button on the 5-way nav only does the pause action. On the surface it looks like pause acutally only pauses the stream, but remains connected, which may explain this quite well.
    Thats the problem I was having at first, which got me into this whole battery drain thing in the first place. Ptunes was really running even though I though I had stopped it.

    Even after hitting stop on the screen though, exiting Ptunes, doing other things, when I come back to Ptunes, there is the scrolling on the marquee of the last address i was streaming from. That is what makes me think it may still be hidden away somewhere using battery power even after truly hitting the "stop" button.
  15. #15  
    Comparing this to the rate my battery discharges during the day indicates having BT causes a significant drain.
  16. #16  
    I'm not sure I understand your issue of voltage vs current battery power remaining. Typically lithium-ion batteries cannot go (significantly) below 3.50 Volts without causing irreversibly damage in which they will no longer recharge. So, it's reasonable to assume that 0% battery power remaining is set at the critical voltage level of about 3.5V

    Additionally, typical battery draw is measured in "milliAmps" (1/1000th of an Amp) and has life reported in thousands of "milliAmphours" which is the number of hours a battery can sustain a fixed current in milliAmps.

    However, I think that the terms "current" and "voltage" are being mixed or reported incorrectly by the utilities. Voltage is similar to water "pressure" while "current" is similar to water "flow". But power is calculated as P=V^2*I (Voltage squared times current). So in other words, the treo is almost always using some amount of power by drawing current, but it is not typically going to cause a linear voltage drop that is used when calculating power remaining.
    Attached Images Attached Images
  17.    #17  
    Quote Originally Posted by Santa8Claws
    I'm not sure I understand your issue of voltage vs current battery power remaining.
    Several issues for me anyway. According to the info inside the 650 itself, the "low battery warning" is supposed to go off at 0.10v and it goes "critical" at 0.0v. This cannot be correct since the device prevents the battery from going down that low.

    Lithium batteries can actually go down to 2.5v or so without damage but, to be on the safe side, most devices cut it off at 3.0v. It appears that the 650 is even more on the safer side by cutting off at 3.5ish.

    The real issue with me on the "v's" vrs "%" is that they are not matching up very well. The percentages in relation to the volts varied quite a bit with no real rhyme or reason that I could detect. Indeed sometimes they went in opposite directions! This was true even though I was careful to keep temp very constant and was the case not just in the short-run but over the whole cycle. I think you may be right as it being a reporting issue, but i am more likely to blame the 650 software than the utility and, in any case, that makes it difficult to accurately determine battery drain per min or whatever.

    A good short battery explination can be found here:

    http://www.batteryuniversity.com/partone-12.htm

    As to the constant battery drain, I know that there will always be some, but since the treo is suppoesed to last 300 hours in "standby" mode (LOL), I shouldn't be losing 16% over an 8 hour period with phone on (and no other usage) and, at least, 14% with the phone radio and everything off. Something more than just the normal slow drain is occurring and the 650 has no way for us to track that. Something more than just normal drain is going on.

    There was most certainly a need for a bigger margin of safety when there was non-volitle memory, but that is no longer the case. The 650 could still be operating significantly below the 3.5v but isn.t. .5 is a lot when the top is 4.2 and the bottom is 3.5.

    Moreover, with the non-volitle memory, it would have been easy to have a "total off" switch that disconnects the battery from the device to really save on any battery drain.

    Ultimatly here is the question people need answered: "Is there something wrong with my 650?" At this point that is a very difficult question to answer because of the way the 650 is reporting battery usage and because of no resource manager app to find out what apps are causing power drains.

    At best what we can do is compare notes with each other to see if one persons rate of drain is significantly different than another's
    Last edited by midmofan; 01/07/2005 at 05:00 PM.
  18. #18  
    Couple of comments:

    Ever since Treo 600 the Treos have not been using the standard battery warning mechanisms of Palm OS. The warning and critical warning voltages are no longer stored in the same fashion AFAIKAFAIKAFAIK.

    The root posting did not take into account temperature and current wattage load. These are the two other metrics critical to measuring a battery performance.
  19.    #19  
    Quote Originally Posted by mwomwom
    Couple of comments:

    The root posting did not take into account temperature and current wattage load. These are the two other metrics critical to measuring a battery performance.
    Temperature was fairly constant throughout the whole period, treo was never in the cold for any length of time and not resting on a hotspot or anything and all my notes were when it was inside, away from hot or cold spots, at room temp. One reason I reported what apps I had been using right before the readings was to get a handle on some of the other variables.

    Again key issues are: "Is something wrong with my 650?" "Is some app I am using causing an excessive drain?" "Is there anything I can do to extend battery usage?" "Is this NORMAL??" Right now those are all hard questions to answer and the best we can do is compare notes as to actual usage to see if someone is having an anomoly or if there is a trend.
  20. #20  
    Make sure you have the newest version 2.0 of BTMute installed. Jeff fixed some issue that was causing battery drain.
Page 1 of 2 12 LastLast

Posting Permissions