Related
Hi all,
just after I had my new Kaiser month agao I was collecting all kind of
software to fit my needs. And short after that I realized - my Kaiser sucks
the battery within hours
A battery tool that shows the temperature and the current showed me:
250 to 350 mA all the time. For a 1300mA/h battery this means 4 to 5 hours
or less.
The other day I removed all installed goodies - and surprise! Sometimes the
current drain was down to 10mA.
To make the story short: remove all timer software.They eat up your battery!
I had installed two of them! One to alert if park-tickets are expired and one
timer to remind me for meetings! Those little bastards apparently run in tight
loops, check the time constantly and use up the most power.
Power consumption is now low as 15mA in idle times.
good luck with your Kaiser / Tilt
wkm1
Hi!
so this would also apply to alarms set in Outlook for meetings? My meetings are often same time, same people every week, so better to not use alarms for those....
I think my Kaiser needs a charge every 36 hours......
Ofiaich
ofiaich said:
Hi!
so this would also apply to alarms set in Outlook for meetings? My meetings are often same time, same people every week, so better to not use alarms for those....
I think my Kaiser needs a charge every 36 hours......
Ofiaich
Click to expand...
Click to collapse
No, I think you'll find he's only referring to third party apps not Windows Mobile functions themselves.
He has a good point. people in other threads have found many fancy Today screen apps (always running) that show memory usage, comms, and battery status are polling the equipment on a regular basis also using lots of battery power.
Thanks Farsquidge!
I will leave my Outlook as it is then !
I think my screen is quite simple as compared with some in the post your today screen thread !
Ofiaich
Possibly stating the obvious here but i never knew so here we go...
Change the settings so the data connection is not always on, and only connects when looking for email, web browsing etc.
Its doubled my battery life.
What tool did you use? I'd like to optimize my apps too.
wkm1 said:
Hi all,
just after I had my new Kaiser month agao I was collecting all kind of
software to fit my needs. And short after that I realized - my Kaiser sucks
the battery within hours
A battery tool that shows the temperature and the current showed me:
250 to 350 mA all the time. For a 1300mA/h battery this means 4 to 5 hours
or less.
The other day I removed all installed goodies - and surprise! Sometimes the
current drain was down to 10mA.
To make the story short: remove all timer software.They eat up your battery!
I had installed two of them! One to alert if park-tickets are expired and one
timer to remind me for meetings! Those little bastards apparently run in tight
loops, check the time constantly and use up the most power.
Power consumption is now low as 15mA in idle times.
good luck with your Kaiser / Tilt
wkm1
Click to expand...
Click to collapse
[email protected] said:
What tool did you use? I'd like to optimize my apps too.
Click to expand...
Click to collapse
And what battery life are people expecting. I have SPB's GPRS monitor installed and it is claiming I will get a little under 6 hours (which compares to 8 on my Hermes).
Schooleydoo said:
Possibly stating the obvious here but i never knew so here we go...
Change the settings so the data connection is not always on, and only connects when looking for email, web browsing etc.
Its doubled my battery life.
Click to expand...
Click to collapse
I second this. I moved from direct push to polling every 10 minutes, and set the connection to drop after 30 seconds (annoying for browsing the web) and the phone loses less than 2% battery overnight.
Surur
how do you modify the internet connection to disconnect after a certain period of inactivity? i can't seem to figure it out.....
dmb129 said:
how do you modify the internet connection to disconnect after a certain period of inactivity? i can't seem to figure it out.....
Click to expand...
Click to collapse
Use KaiserTweak. It's in a sticky.
do you have any idea where i could find the individual .cab for this function? rather than using KaiserTweak?
hi,
i tryed to "i guess disable the wifi hardware", and all animations in order to have a greater battery life (with kaiser tweak) , but my phone needs to be charged for 6 to 6 hours...
this phone has a battery problem, for sure, i hope someday, HTC solves this problem....
sorry about the english....
I had an alarm clock to ring every Saturday morning, on for the whole week...I took it off today, following this thread. It appears to have made a huge positive difference in the length of the battery! Many thanks for a good tip.
michalopoulosgk said:
I had an alarm clock to ring every Saturday morning, on for the whole week...I took it off today, following this thread. It appears to have made a huge positive difference in the length of the battery! Many thanks for a good tip.
Click to expand...
Click to collapse
What alarm application did you use? I'll test it with mine the following days and let you know the result.
I had my office outlook active sync basically "always on". I just switched it to every 10 min during peak and every 30 min off peak. We'll see how much better that goes.
I've definitely noticed that keeping the wifi turned off until necessary helps a lot. I'm using Tmobile here in the US, so no 3G yet. I've been told that 3G sucks battery as well.
3G consumes more power. Use GSM network only.
start -> settings -> Phone -> Band Tab -> Select network type as GSM
wkm1 said:
Hi all,
just after I had my new Kaiser month agao I was collecting all kind of
software to fit my needs. And short after that I realized - my Kaiser sucks
the battery within hours
A battery tool that shows the temperature and the current showed me:
250 to 350 mA all the time. For a 1300mA/h battery this means 4 to 5 hours
or less.
wkm1
Click to expand...
Click to collapse
Please let me know what battery tool...... want to check mu Tytn II too.... battery only lasts for HOURS!!!!
Many thanks,
Optimistisch_nl said:
Please let me know what battery tool...... want to check mu Tytn II too.... battery only lasts for HOURS!!!!
Many thanks,
Click to expand...
Click to collapse
FYI, I had to return my Tytn2 yesterday. It lost 100% battery power within 4-5 hours. It was warm to the touch and when charging would become very hot. I put Batterystatus on and it read a draw of 300-400mA when idle. This was also after a Hard reset to check that nothing else I had added to the unit was causing it.
My replacement draws between 20-80mA when idle and 220-250mA when under load i.e. using Wi-Fi, browsing web etc. Battery went from 54% when I got it (1pm) to 26% by the time I went to bed at midnight last night
It sounds like you have a similar problem?
I've set 2 alarm applications to run last night: the alarm in Mortplayer and Pocket Wakeup.
I charged the battery to 100% before I went to sleep. In the morning, after about 6 hours of sleep, my battery meter showed 94%.
This gives an easy calculation of about 1% per hour (with those 2 alarms active).
I don't know if it makes a difference to uninstall the applications or just not using them in terms of battery usage. Anyone an idea?
smads said:
FYI, I had to return my Tytn2 yesterday. It lost 100% battery power within 4-5 hours. It was warm to the touch and when charging would become very hot. I put Batterystatus on and it read a draw of 300-400mA when idle. This was also after a Hard reset to check that nothing else I had added to the unit was causing it.
My replacement draws between 20-80mA when idle and 220-250mA when under load i.e. using Wi-Fi, browsing web etc. Battery went from 54% when I got it (1pm) to 26% by the time I went to bed at midnight last night
It sounds like you have a similar problem?
Click to expand...
Click to collapse
Is this 'Battery Status' part of an app??? Please let us know as this itself can draw power if it is constantly polling the battery. (Heisenberg principle??)
This thread refers to cases where the HTC Kaiser (and probably other HTC phone models with the same built-in GPS chip, such as Polaris) sucks the battery empty within a single day, particularly during standby, particularly if all features of the phone are DISABLED, or after low usage. Another symptom may be a phone which unexpectedly did not come back to life, but operated normally after recharging. Yet another symptom is unusually short battery life through normal use. Note, this thread is not about limited battery endurance due to heavy usage.
*** Battery drain FAQ ***
How can I see whether my Kaiser is affected?
- A large percentage of HTC phones with Qualcomm chipset and internal GPS seems to suffer from the bug described here, some appear to be robust enough, likely depending on chip revisions. If you never ever use GPS, you are probably out of risk. If you use GPS at least occasionally, you should have a look at your phone, and closely. It is possible that an otherwise normal to heavy use of the phone conceals a basic, added current consumption which is what we are talking about here. In such a situation you may have got used to a certain battery endurance, which may be much higher under normal circumstances. So far I observed currents between 28 and 78 mA, depending on DUT and OS. Some users reported currents in excess of 100mA. Note, this added current does nothing for you, except accelerate your battery drain!
How to check this?
- Check the true standby current (see the following description). That simple.
How to measure the standby current?
- Clean up: switch all phone features off (GPS, Phone, Bluetooth, Media Player, really everything). Terminate all applications, use Task Manager to verify this. Make sure no processor intensive background tasks are running (standard installations should meet this requirement). Do NOT soft reset at this point!
- Put the phone in standby, and wait approx. 10 seconds. The phone needs a while to complete entering the standby though it appears to be off immediately.
- Now measure the current. Best and fastest way of doing it would be to have a current meter connected between your battery and your phone which gives you realtime readouts. Second to that is a suitable battery tool, such as "BatteryStatus", but you have to get used to the delayed current display (see post #4 for details). Using the software battery gauge, you should wait an additional 20 seconds or so to allow for the current capture, then reactivate from standby and take the lowest possible current readout.
- If the phone is in a good condition, the standby current must be in the range 1..3 mA, roughly. If you see a repeatable current well in excess of 20mA, your phone is in the BAD condition !!
- Another simple method is to leave the cleaned up phone in standby overnight. Next morning, soft reset your phone because the phone may have lost track of the battery capacity. Check whether the capacity dropped dramatically. And also check whether the phone feels warm to the touch.
How to reproduce the problem?
- Activate GPS until you get a fix. Probably receiving the first NMEA strings is good enough, but I have not verified this. GPSTest, HTC's GPS Tool or any navigation software does the job.
- Deactivate GPS. Just to be safe, terminate the GPS software, too.
- Check the standby current.
- The fault does not pop up always, so you may have to repeat these steps several times until it appears. The phones I tested usually catch fever after only one or two tries, but it is possible that you need to cycle through this procedure 5 or 10 times. Which is in the nature of intermittent bugs.
How to reset this nasty condition?
- Fault recovery is possible by continued on/off cycling of the GPS unit, similar to what provoked this fault.
- Activating the cellphone unit does also seem to cause the phone to return to a low consumption, but maybe not in all cases.
- Try to soft reset your phone, or to cycle the power to the phone (long press of the power button).
- If you really cannot get rid of the problem, back up your phone data, then execute a hard reset. BEFORE restoring the phone, load a battery gauge software and see whether the consumption is gone. If yes, some application is likely to cause your headaches.
After performing one of the above steps you may repeat the described current measurement, to confirm that the standby current is back in the normal range.
***********************
Original text:
I have reason to believe that all Kaiser models are prone to the "excessive standby current consumption" problem. I tried it on two original "VPA compact V" by Vodafone (= HTC Kaiser), using the original WM6, ROM 1.56.162.5, Radio 1.27.12.11. I gave a **** on warranty now, stepped up through HardSPL to "Duttys_Official_WM6.1_Rom_5.2.19716_UC.zip", Radio 1.64.08.21. The results are all the same: an excessive current consumption which can vary between 25 and 75 mA. Interestingly, the current is always the same in a specific setup, but varies between phones and operating systems. In the latest case, the same phone took 28mA under WM6, went up to 78mA under WM61. Consequence being, the battery will be sucked empty within a day or so, without obvious reason. The only mitigation is a reset, or a complete switch-off.
There is no application which could cause the current consumption, at least none I installed. It does not even matter which GPS application you used. The only active processes are (according to TaskManager, latest WM6.1 version, ".exe" omitted for the sake of ease): filesys, device, biotouch, gwes, shell32, cprog, services, quickdial, connmgr, mediahubmini, taskmgr, htcactionscreen, sapsettings, aplauncher, quickmenu, nk. I repeat: ALL applications properly terminated, ALL internal units are OFF (WLAN, Bluetooth, Phone, GPRS, GPS, Camera, Media Player etc. - really NOTHING).
Before anybody prematurely states that these findings can not be reproduced: the problem is unlikely to appear if you switch on the GPS for a couple of NMEA strings, then off again (though it did already). If need be, you have to leave it on for a while, play a bit, walk around a bit and so on. Take your time testing it _thoroughly_, really. I cannot tell when the fault actually appears, and it may not come up immediately because it seems somewhat sporadic in nature - but take my word, it will, I observed this issue for a long enough time. I can only repeat myself: I am sure there is something wrong with the power management in the GPS driver.
Anybody, feel free to contribute, but PLEASE avoid funny statements like "you have to shut down all programs", "WLAN can take up lots of energy" and so on. And before you express doubts just because there are not quite many people out there who come to the same conclusions, think again. Without going into details, most users are simply not in the position to come to the correct conclusions.
BusterTyTN
OK, making it sticky; depending on the feedback, I (or other mods) may unstick it in the future.
How do you measure current while in standby? When it's running, I can see current usage in the battery monitor of AE Button plus. Just after wake up from standby, it shows 16ma. A few seconds later it jumps to 385ma and then settles at 140ma.
I'm using 3.02.DKv0.0 6.1 Lite Rom from akadonny.
@ Menneisyys: thanx!
@ tdusen:
there's two ways of doing it. First, I do not know whether "AE Button plus" works in a similar way, but "BatteryStatus" (also included in Dutty's newest WM61 release) has some seconds of a delay in the current display. I assume this is because of an averaging process running in the background which collects current samples over a certain period of time, or soemthing to this effect. Anyway; you have to keep your in standby for at least 10..20 seconds, then reactivate the PDA. If it's still showing a high current, tip on the current/power display a few times, in most cases this helps getting the low standby current. You simply have to try to catch the lowest reading, right after reactivation from standby. If everything is OK, you will get a current reading of a few mA, maximum (in most cases 1..3 mA), assuming you shut every feature of your phone off.
The other way is also the verification of the "BatteryStatus" method. I built a current shunt test probe and measured the battery current directly. Attached is a screenshot taken with my digital scope, which shows typical results (see above -- moved to post #1).
In your case - just to be safe, give your PDA a soft reset, and leave everything off. Check your current reading. If it's still at 16mA as you wrote, I wouldn't bother. Later you may fire up a GPS application like GPSTest, the HTC GPSTool, Tomtom, OziExplorerCE or whatever, and play with it for a while. Running a circle or two around your house should do it. Switch GPS off again, make sure all tasks have ended (if you want to make it perfect). Repeat checking the current as described above. It is also possible that you have to redo the test one or two times until the problem appears. You will see!
BusterTyTN
Addendum to the previous post: explanation of the screenshot.
The vertical center is zero current, discharge currents are going down. To the very left you see the PDA coming out of a soft reset, then I switched it into standby (the short flat line right before the 200sec division, ~0..1mA). Directly at the 200sec division I reactivated the phone, started GPSTest, waited until GPS was all up and running, stopped GPS and terminated the program. I also checked that no other applications were running at that time. After a little while the current settled to a discharge current of approx. 48 mA (the wide track in the center; the exact level can vary between devices). It continues drawing this amount of current until the battery is empty, or until you soft reset the device. I did the latter to the right of the screenshot, which shows another boot sequence followed by a standby, which in turn returned to a very low current consumption.
Also give a try to acbTaskMan - it's a very nice meter tool, see my related articles
I don't know anything about current measurement, but my kaiser usually lasts 2-3 days of intensive use (1-2 hours calls/day + some GPRS data + bluetooth always on + GPS once a week) and it seems quite OK for such device.
Rumcajs_tr said:
I don't know anything about current measurement, but my kaiser usually lasts 2-3 days of intensive use (1-2 hours calls/day + some GPRS data + bluetooth always on + GPS once a week) and it seems quite OK for such device.
Click to expand...
Click to collapse
Seriously thats gr8... What ROM / Radio are you using??
Rumcajs_tr said:
I don't know anything about current measurement, but my kaiser usually lasts 2-3 days of intensive use (1-2 hours calls/day + some GPRS data + bluetooth always on + GPS once a week) and it seems quite OK for such device.
Click to expand...
Click to collapse
That's intensive????????
How about 5-6 HOURS of calls a day
4-5 HOURS a day of HSDPA
checking emails every 15 minutes on 3 email accounts
sending emails 60-75 times a day
WI-FI 1-2 hours a day
150 SMS/Texts a day
10-20 MMS a day
and then a few misc. apps ran.
To me, my usage is not really intensive, but I do have to have a car charger and a couple of wall chargers because I can run a battery dead in about 5 hours.
i played doom for 30 minutes starting at 100% battery, and the battery was at 75% afterwards. i had to charge it for about 30+ minutes to get it back up.
On a daily basis, i will go to school, leave my phone in my pocket on standby, then by the end of school its at about 80%. i did nothing all day and it dropped 20%.
I would agree, 5-6 hours is probably the max i can get out of it.
BusterTyTN said:
...
I have reason to believe that all Kaiser models are prone to the "excessive standby current consumption" problem.
Click to expand...
Click to collapse
At least two devices work nicely as far as battey power is concerned.
I still enjoy more than 10 days standby with phone switched on.
http://forum.xda-developers.com/showthread.php?p=1652208#post1652208
I hope you'll find a solution.
Regards,
V
T-Mobile Germany
VARIO III rom 1.56.111.4
radio 1.27.12.11
When my phone is on, but not running any active apps, it consumes about 145 mA - 150 mA with nothing but my phone on and backlight lower 3rd bar.
I've read some others guys on the forum have a power consumption of about 100 mA with higher backlight settings.
indeed excessive consumption
I must confirm the OP's finding. Though I assume it is simply related to the Radio used (see comparison in another thread where est. time of usage went from 227 to +400 minutes)
When I used my stock rom with radio 1.27, I wouls still have 40-50% cap using all day Blutooth on, no 3G, still GPS enabled for 3 hours.
When I falshed to DCS 1.7 and changed radio to 1.64, power consumption doubled, draining the battery at incredible speed (full chare morning, less than 10% in late afternoon)
I had to disable BT radio and set backlight low to let me go through the end of the day.
Looking forward to read a solution here.
With stock orange radio and rom, my device would last circa 36 hrs with very light usage and about 6/7 hours with heavy usage.
With Dutty V3 and orange stock radio (1.27) I got very similar, maybe slightly faster drain.
I have today flashed dutty 6.1 and a 1.64 radio and will monitor battery usage tomorrow.
There's a lot of anecdotal information here...
To get this to baseline information I suggest the following starting point:
1. Device (Kaiser, tilt, etc)
2. Which battery are you using (brand, mAh rating, etc)
3. Which Rom and radio version
4. What application are you using to measure current draw? (Great spot for a recommendation or a .CAB file)
5. Provide typical usage description and corresponding battery life
6. At what point do you charge the battery (i.e. top off every night, charge whenever power's available, recharge only when battery hits low level (i.e. 10%)
To effectively analyze the data, there needs to be a consistent method of capture.
Be careful not to mix up the problems here!!!
1. I assume you will not be confronted with this problem when you never use GPS at all, though you should not treat this as a final statement - has to be investigated further. So far I have not found any increased standby consumption if I did not touch GPS (well, at least after the last soft reset).
2. I am not talking about short battery uptime under heavy use, either. For instance, if you have GPS permanently on, you can suck the battery empty within 4~5 hours, max.. However, it is well possible that a suspiciously short uptime is somewhat concealed by an overall heavy use of the phone. To figure this out, you simply have to check the standby current consumption of a suspicious phone, with all its features (temporarily) disabled, as described earlier.
3. Note, this problem is all about current consumption during standby !
@ Liquidsilver:
"anecdotal information", sure. Please read again, particularly the start of the thread which will answer most of your questions. Add the thread "TyTN II / Kaiser issue: GPS & battery drain" for the remaining ones.
Please try to understand. You can get the phone into a condition where it sucks the battery dead empty within 24 hours give or take, during standby, all features of the phone OFF. ZERO usage. PROVEN, some 10 times, on different Kaiser's, using different OS's.
@ Stay0Puft:
if your battery capacity does not drop much overnight, you may look at an issue other than the one described here.
I would like to add my observation to high battery drain. Normally on my stock Kaiser rom that I used for about 4 months, normal overnight drain is 4% with radio in standby and BT On, WiFi Off. I use GPS with iGuidance almost daily and seldom soft reset.
I have witnessed extremely high power drain only a few times and they were always involved somehow with WiFi. If I have WiFi On and forget to turn it off and just put the phone in standby or let it auto power off, I have found the phone hot to the touch and dead or close to dead in only a couple of hours. The phone is usually locked up and non responsive at this point, needing a soft reset. I haven't found what causes this, it's been at home on my wireless LAN but with no active application, WiFi was simply on. I make it a point to keep WiFi Off when not actively needed and never have issues in many months of use.
If you haven't checked this out, I'd be curious to see what the current meter would show. At some point I'll drag out the scope and make a shunt as well, just no time these days...
AT&T Tilt...
I used to get 2-3 days with mediocre usage when i first bought it... and now i'm lucky if it lasts an entire day... I cannot trust it with my morning alarm anymore... I dont think its the ROM, because i even tried flashing AT&T Tilt rom back on and even the minor update on HTC... but still no luck... I think i need like a calibrator program...
Yo
heres a brutally simple battery monitor software thingy that graphs your usage and battery level - top graph is mAh usage (15000ms refresh (os updates info every 30 secs)) and bottom graph is battery life (60000ms refresh)
i wrote this thing to monitor my batt usage cause my bat life is utter crap, barely lasts 8 hours with my level of use/cellular climate, which sucks its just a straight up .exe, just copy to where-ever and run it
this version here is so pre-alpha-uber-dev-debug-build its not funny but i thought id put it up for you lot anyway as i have found it useful just to glance at from time to time - there are no user controls, i spazzed this out in a hurry but i intend to improve on it and implement any features you lot might think useful.
best i can get out of my raph with screen on with celluar, gprs and bt ad2p connected is -63mAh - whats the best you can get?
just tryin to put smt back into the community <3
p.s. this is a debug build (i kept getting microsoft error reporting crap on the emulator, so i dunno how itll go on other devices)
p.p.s. i take no responsibility if this bricks your device, sleeps with your wife, sets your house on fire or kills your cat etc etc (but it works ok on my raph
***UPDATED 25.03.09***
SEE POST #9 IN THIS THREAD BELOW
going to try this out soon, loving the simple graph
now i can see whats happening when 40% of my battery dies overnight as i sleep >:O
works good, any way of making it show the current time on the x-axis? or able to scroll left and rigth to see what happened earlier?
and whats the bottom graph for?
07accordEX said:
any way of making it show the current time on the x-axis? or able to scroll left and right to see what happened earlier?
Click to expand...
Click to collapse
both of these features are on the todo list (which ill put up when i get a moment)
and whats the bottom graph for?
Click to expand...
Click to collapse
read the initial post.
the top graph shows the past hour of activity (ish, its actually more like 57.5 mins), the bottom graph shows the past 8 hours (ish, more like 7.7hrs).
no data is collected when the device is asleep - and the graph doesnt get updated during these times, so you only see 'active' use in the graphs - i plan to implement some sort of visual cue system to signify times of sleep and also to keep the temporal coherency of the graph more contiguous (lol, e.g. a graphed minute will be 8 pixels wide no matter what).
i believe its impossible to collect data while the device is asleep as afaik when the processor goes to sleep the only part of it that functions is the bit that listens for interrupts (from i.e. the phone management cpu, etc) (and maybe a timer (for alarms etc?)) - if im wrong on these details please enlighten me.
fusi
Nice app I would love to see it becoming more mature, will probably end up with my standard equipment, thanks 12
12aon said:
Nice app I would love to see it becoming more mature, will probably end up with my standard equipment, thanks 12
Click to expand...
Click to collapse
thanks, watch this space should have something new up by the weekend
This opens up a blank white page on my phone. Did I do something wrong?
behrouz said:
This opens up a blank white page on my phone. Did I do something wrong?
Click to expand...
Click to collapse
the graphs dont refresh as soon as you open the app - youll hve to wait for the first update (15 and 60seconds) before anything is rendered - this will be fixed soon.
ok, a bit later than advertised but here is an updated version - fixed most things, still working on scrolling through the history - will add more features in coming days (colour customisation, toggling graph display, scrolling through history, exporting graphs as images, user customisable update intervals, etc)
btw ive added in support for unattended mode - this keeps the cpu alive when the device is asleep - i dont recommend keeping this on unless you are trying to debug unexplained power drain in standby - unattended mode keeps my raph ticking along at -21mAh, when im pretty sure true idle for this handset is more like -4mAh. so yah, not recommended to leave it on
im pretty sure there arent bugs but if you find one please let me know!
enjoy
fusi
fusi said:
ok, a bit later than advertised but here is an updated version - fixed most things, still working on scrolling through the history - will add more features in coming days (colour customisation, toggling graph display, scrolling through history, exporting graphs as images, user customisable update intervals, etc)
btw ive added in support for unattended mode - this keeps the cpu alive when the device is asleep - i dont recommend keeping this on unless you are trying to debug unexplained power drain in standby - unattended mode keeps my raph ticking along at -21mAh, when im pretty sure true idle for this handset is more like -4mAh. so yah, not recommended to leave it on
im pretty sure there arent bugs but if you find one please let me know!
enjoy
fusi
Click to expand...
Click to collapse
I'm curious to know what you think the power consumption caused by this program is. Like you, I get about 21mA drain with the phone screen off - but I'm getting that with your old version. I use S2U2 with the "but only blank screen" option and turned off standby in Windows power management so I think that's effectively putting the phone in unattended mode. Normally (without your program running) I see pretty modest battery drain with the phone in unattended mode. Just off the top of my head I'd say that 5-15% battery drain per hour would be about what I'm used to with zero usage - phone just sitting in my pocket or on a desk in "unattended mode" but just now while running your older EXE my phone went from 65% to about 40% in under an hour while "unattended" (locked, screen off). Looking at the graph for that time period I see a consistent 21mA drain. If 21mAh=25% capacity, I'm in trouble. I know my battery isn't that far out of whack because I get what I'd consider (based on reading other's experiences) pretty normal life out of it. Maybe the drain characteristics for that "portion" of the battery are a little different - I'm not sure whether the capacity is judged solely on capacity minus drain or it figures voltage levels in as well...?
To be clear, I'm basing my sense of normal battery consumption on nothing more than "took it off the charger in the morning, didn't use it all day at work and have ~50% when I get home 9 or ten hours later" so it's anecdotal at best.
I'll keep playing - I love this tool. Maybe I'll be a little more scientific about my testing and see if I'm imagining things.
Hi thanks for posting
if your just turning the screen off, i dont think thats unattended mode, unattended mode actually powers down non essential parts of the device.
if unattended mode is off in the program and your phone goes into standby, it wont consume any battery as when the phone goes into standby everything is shutdown and the cpu pretty much stops functioning (apart from a very small part) - but this also means the historical data and graphs dont get updated . if unattended mode is on and your phone hits that standby timeout, it wont go into standby but unattended mode instead where it just turns off a lot of the non-essential stuff, like voltage to the sd card, the screen, etc - but it keeps the cpu running and the operating system pumping its messages.
i think a chunk of that 21mAh is going to be the operating system - but not all of it - currently the program registers 2 timers with the os, one has a 15second timeout and the other has a 60 second timeout. timers do drain battery, as the cpu keeps having to go 'is it time yet? is it time yet?' ad infinitum.
btw, when i said 'true idle' in my prev post i meant that that is the current used in true standby.
one way to calculate how much current its drawing is to leave your phone running it for 24hours (airplane mode on etc) and see how much its depleted over that time, then do it again without it running. using that information you can figure out how much current its using - i havent done this yet, i cant put the thing down for 5 minutes let alone 24 hours
im currently working on a system that doesnt use timers, it might work, might not if it does i hope to see that 21mAh reduce, but i dont imagine itll go down by that much. there are however loads of optimisations that can be done, and im investigating them atm.
im no battery scientist but as far as battery capacity, i think li-ion batteries judge their capacity by their voltage, from something like 3v up to 4.2v (0% -> 100% [wiki link below]) - i could be wrong on that though.
http://en.wikipedia.org/wiki/Lithium_ion_battery <-- this is a good informative read.
i dont think the battery is very capacious tbh - any amount of drain and it seems to go down very fast - i think the saving grace of the device is that is consumes minimal power when in standby - on days where im constantly using it heavily, im practically tethered to a power outlet as the battery just drains far too fast - i think the manufacters banked on the assumption that users would mostly have the phone in standby mode for most of the day. i usually use it for music while im at work and have putty sessions open etc, so its always on - i barely get 8 hours use without a charge .
ive noticed with my battery that even though the drain is constant (i.e. 21mAh or 63mAh) the charge level of the battery does not decrease linearly - e.g. ive noticed my phone suddenly start going down and lose 10% in 5 minutes before, and then the charge level would level out and stay at the same capacity for like 10 minutes before starting to go down again - this to me seems a little funny, as if the circuitry reporting the capacity isnt quite accurate - i think there a many factors, including battery temperature, the drain on the battery in mAh (i think a spike of higher drain may cause the battery do the nose-dive-then-level-out thing) - but i also think that the battery just doesnt deplete linearly, it seems to always have lumps and bumps in the graph, no matter how smooth the drain is.
glad you like the app ill be posting an update to it in a week or so.
peace
fusi
I'm not sure what exactly S2U2's "but only blank screen" does, but it leaves me with a 21 mA draw, so it can't be much different from unattended mode. I haven't tried to monitor it in that mode with (for example) an active Wifi connection or a program accessing the SD card to see if that changes things, but as it is it seems to be running the same things your set is.
The only reason I think the capacity calculation must not run off of only the voltage is that it has a definite "learning" capability. I think it must be doing some more complex things behind the scenes in terms of monitoring consumption and recalibrating periodically based on charge and discharge rates. I know I've seen it sit at 99% charging for a (very) disproportionately long time. The only good explanation for that is that it wasn't perfectly calibrated and the battery is still accepting charge at a higher rate than an almost full battery would. I've seen the same behavior in a lot of charge monitoring systems as they "learn" the characteristics of a battery. I guess it could be basing that purely on voltage, but I doubt it. While connected to a charge, the system will report a substantially higher voltage than it will as soon as the charging voltage is disconnected.
I don't know, I'm just kind of thinking aloud here. It would be great if there were a way to poll the power consumption without affecting the power consumption - sort of a Heisenberg's uncertainty problem... I'd really like to know what the power consumption is in unattended mode as well as true standby. Even if you do the 24 hour test you have to assume a lot of things about the battery's initial condition and the accuracy of the meter to arrive at a consumption number (and, like you, I'll never have a day I don't want to putter around with this phone). I suppose if you were really into it you could stick an ammeter between the battery and phone and control for the added loss
It's definitely true that the discharge isn't linear, even given a constant discharge rate. This has to be an artifact of including battery voltage in the capacity calculation. If it was using purely capacity minus usage the relationship would have to be linear.
Anyways, keep up the good work. I love stuff like this, just from an academic standpoint. Practically speaking I'm pretty much tethered to my chargers, too...
hyachts said:
I'm not sure what exactly S2U2's "but only blank screen" does, but it leaves me with a 21 mA draw, so it can't be much different from unattended mode.
Click to expand...
Click to collapse
That's exactly the case, this setting causes your device to run in unattended mode (backlight off, audio off, wi-fi off, ...)
hyachts said:
I think it must be doing some more complex things behind the scenes in terms of monitoring consumption and recalibrating periodically based on charge and discharge rates. I guess it could be basing that purely on voltage, but I doubt it. While connected to a charge, the system will report a substantially higher voltage than it will as soon as the charging voltage is disconnected.
Click to expand...
Click to collapse
Microsoft's reference battery driver for PXA270 CPU is calculating battery percentage based on voltage exclusively. HTC's custom driver for Qualcomm cores seems to have some kind of more advanced logic behind it, but it doesn't work well/smooth.
hyachts said:
I don't know, I'm just kind of thinking aloud here. It would be great if there were a way to poll the power consumption without affecting the power consumption - sort of a Heisenberg's uncertainty problem... I'd really like to know what the power consumption is in unattended mode as well as true standby.
Click to expand...
Click to collapse
Not without external tools, it should be possible to hook up an external multimeter to the battery directly to measure the voltage accurately in standby. Using the battery driver information you can only evaluate it and the evaluation is likely to be pretty rough.
Attached are some charts from my Touch HD...
The battery level graph seems to follow consumption graph very closely.
You can barely notice the resemblance with the voltage graph though.
These charts depict my device's utilization over 2 days, assuming 0 mah consumption when in standby. This assumption is apparently not accurate, as the battery capacity is 1350 mah and only about 1000 mah were utilized. Out of these 48 hours the device was in standby for about 40 hours, so my estimated standby utilization is -350mah/40h => -8.75mah. This doesn't sound reasonable, as according to manufacturers specs, the device is capable of 450h of standby, with 1350mah battery standby consumption should be -1350mah/450h => -3.375mah. I would agree with fusi, that taking manufacturer's estimation errors it should be safe to assume standby consupmption is around -4mah.
40*-4mah=160mah, so I have about -190mah remaining unaccounted for. I blame measurement accuracy for this... IMO, with this degree of inaccuracy, standby consumption can not be accurately estimated and has to be assumed based on manufacturers specs only.
I'm running more tests to reconfirm these observations, but I doubt the results will be extremely different.
For your reference, I get -69mah on Touch HD when idling with minimal backlight level and -27mah in suspended mode.
I noticed, a few others, but should be minority has the same problem as mine.
My N1 powers off itself when my battery shows around 15% or so.. sometimes 17% sometimes 14%, never lower than 10%.
I already tried to wipe battery data at recovery or deleting the battstat.bin file etc etc but it is still like that.
Is there a solution for this and any reason for this?
Majority of people's phone only power downs below 5%.
Many thanks
Use the battery calibrator, it's in the development section.
Sent from my Nexus One using Tapatalk
baseballfanz said:
Use the battery calibrator, it's in the development section.
Sent from my Nexus One using Tapatalk
Click to expand...
Click to collapse
thanks man..
there are two with the same name.. i'll try them one by one
Try bump-charging* then clearing the battery stats
*Charge phone to 100%, unplug it, turn it off, charge again till indicator goes green
Rusty! said:
Try bump-charging* then clearing the battery stats
*Charge phone to 100%, unplug it, turn it off, charge again till indicator goes green
Click to expand...
Click to collapse
actually this i have tried a few times already but it never helped.
My phone turns off at around 19% now after I played with the battery calibrator app. (I set the life at 100%, was lazy to run the full cycle calibration)
Its powerful and probably can fix your problem.
Link: https://market.android.com/details?id=net.jonrichards.batterycalibrator.ui
britoso said:
My phone turns off at around 19% now after I played with the battery calibrator app. (I set the life at 100%, was lazy to run the full cycle calibration)
Its powerful and probably can fix your problem.
Link: https://market.android.com/details?id=net.jonrichards.batterycalibrator.ui
Click to expand...
Click to collapse
i am trying to do this too.
so you are saying it wont help coz yours got even worse (turning off at 19%)?
britoso said:
My phone turns off at around 19% now after I played with the battery calibrator app. (I set the life at 100%, was lazy to run the full cycle calibration)
Its powerful and probably can fix your problem.
Link: https://market.android.com/details?id=net.jonrichards.batterycalibrator.ui
Click to expand...
Click to collapse
To calibrate your battery and have it reliably die at 0% every time, follow the instructions that I wrote in post #3 here:
http://forum.xda-developers.com/showpost.php?p=7850480&postcount=3
schizophrenia said:
i am trying to do this too.
so you are saying it wont help coz yours got even worse (turning off at 19%)?
Click to expand...
Click to collapse
He said he was too lazy to complete the learn cycle, which is part of the calibration technique. There is no sense in trying to pick "pieces" of the process and hope they work, they won't. Start with step 1 and follow the directions to the letter, don't skip anything. I would suggest reading the instructions a couple of times prior to starting and ask questions in the thread if you aren't sure. There are several of us that monitor the thread and will respond quickly...myself during US daylight hours and Temasek in the evening (he's in asia, I think).
mtw4991 said:
To calibrate your battery and have it reliably die at 0% every time, follow the instructions that I wrote in post #3 here:
http://forum.xda-developers.com/showpost.php?p=7850480&postcount=3
Click to expand...
Click to collapse
am doing this right now and hope I understand everything and it works.
thanks a lot
schizophrenia said:
i am trying to do this too.
so you are saying it wont help coz yours got even worse (turning off at 19%)?
Click to expand...
Click to collapse
It will work if you follow the steps and run the full calibration (long time)
mtw4991 said:
To calibrate your battery and have it reliably die at 0% every time, follow the instructions that I wrote in post #3 here:
http://forum.xda-developers.com/showpost.php?p=7850480&postcount=3
Click to expand...
Click to collapse
Thanks. Wish there was a quicker/easier way.
I'm currently testing a direct way using simple math:
If 1452mAh is 100% how much is 19% ?
i.e. 1452/100*19 = 275mAh (for me)
I've set "full40" to 1452-275= 1177 (the calibrator app's minimum is 1200 so I just set it to that.)
britoso said:
It will work if you follow the steps and run the full calibration (long time)
Thanks. Wish there was a quicker/easier way.
I'm currently testing a direct way using simple math:
If 1452mAh is 100% how much is 19% ?
i.e. 1452/100*19 = 275mAh (for me)
I've set "full40" to 1452-275= 1177 (the calibrator app's minimum is 1200 so I just set it to that.)
Click to expand...
Click to collapse
That may work as a guess-timate, but the stock battery comes rated at 1452mAh and stock age is set to 94%, so you would really be setting to 1128mAh unless you reset age to 100%. It really doesn't take that long to set up the application for learn mode...just a few minutes. It does take 31/2-4 hrs to fully charge the phone and complete learn mode if you set the min. charge current to <20mA. To speed things up, just set it to <40mA as you won't miss the little bit of extra juice.
britoso said:
It will work if you follow the steps and run the full calibration (long time)
Thanks. Wish there was a quicker/easier way.
I'm currently testing a direct way using simple math:
If 1452mAh is 100% how much is 19% ?
i.e. 1452/100*19 = 275mAh (for me)
I've set "full40" to 1452-275= 1177 (the calibrator app's minimum is 1200 so I just set it to that.)
Click to expand...
Click to collapse
hmmm good point, i never thought of that lower limit. we'll see if we can update the app.
but to answer this thread's question, the age register in your phone is now out of whack, and you need to set it accordinly to allow the battery to go down to 0%. the learn cycle determine's the proper "age" value and sets it accordingly.
you could also guess an age value, say 83%, would be close. but really there's no reason to not do the learn cycle and figure it out.
BTW all these values are stored in your battery's chip, and it needs to be updated to new value.
Quick question:
1) Sometimes even after I turned on the "Detect Learn Mode" my Current occassionally dropped below -200mA (e.g. 160mA) for a quick while... will it affect the result?
2) "Turn off" means turning off the Detect Learn Mode right?
3) where can I see the "Charging current"? Capa.?
Thanks
What if I rotate among three different batteries on a regular basis? I have an external battery charger too. Will the calibration be in any way useful or will it get confused by the changes?
I have my original January 2010 battery, a December 2011 replacement from HTC, and a MOMAX battery that also holds up a pretty good charge.
Sent from my Nexus One using XDA App
cmstlist said:
What if I rotate among three different batteries on a regular basis? I have an external battery charger too. Will the calibration be in any way useful or will it get confused by the changes?
I have my original January 2010 battery, a December 2011 replacement from HTC, and a MOMAX battery that also holds up a pretty good charge.
Sent from my Nexus One using XDA App
Click to expand...
Click to collapse
you can rotate batteries no problem. when you do a learn cycle, the age value gets written to the chip inside the individual battery. so you can swap batteries no problem. just make sure that the correct age value shows up when you swap, cause sometimes the age resets when removed from the phone. just set it manually back to your proper age if so.
we were trying to add "profiles" support to the app so you could swap batterys and it remembers the parameters for each and sets them accordingly when you swap the battery. but i dont know how long that will take...
Okay good to know. I bought this charger:
http://www.amazon.com/PowerGen-Mult...CAT6/ref=sr_1_3?ie=UTF8&qid=1295658169&sr=8-3
And it's pretty neat because I can just position the pins myself to charge any battery I want. I've tried it with camera batteries, Huawei U1250, Moto Milestone and RAZR, and Nexus One batteries. It works really well and it's somewhat future-proof because it will probably work with my next phone's batteries too =)
Hey all,
Someone on reddit just posted a link to an apk file that let's you set a custom charging limit. Should be a great tool to limit battery wear.
Here you go: https://www.reddit.com/r/Nexus6P/comments/5x9xax/set_changing_limit_requires_root/
See Battery Warner : https://forum.xda-developers.com/android/apps-games/app-battery-warner-t3560920 :good:
swa100 said:
Hey all,
Someone on reddit just posted a link to an apk file that let's you set a custom charging limit. Should be a great tool to limit battery wear.
Here you go: https://www.reddit.com/r/Nexus6P/comments/5x9xax/set_changing_limit_requires_root/
Click to expand...
Click to collapse
I highly recommend accubattery. It lets you check lots of things like discharging, charging speed, battery health, what apps use how much charge an hour etc and tells you how long you have sot and deep sleep and hibernation etc. You can set an alarm for it to tell you when you get to a certain charge. When mine gets to 80% it tells me in a notification that makes noise to disconnect the charger. Really useful and works great.
DEVILOPS 007 said:
I highly recommend accubattery. It lets you check lots of things like discharging, charging speed, battery health, what apps use how much charge an hour etc and tells you how long you have sot and deep sleep and hibernation etc. You can set an alarm for it to tell you when you get to a certain charge. When mine gets to 80% it tells me in a notification that makes noise to disconnect the charger. Really useful and works great.
Click to expand...
Click to collapse
It doesn't help when you're charging overnight. The app linked in OP let's the user define a threshold, such that the phone doesn't get charged beyond this threshold, regarding if it's plugged in
Download automate, theres a flow that lets you choose what to limit it to. you will need the premium version of automate though
https://forum.xda-developers.com/android/apps-games/root-battery-charge-limit-t3557002