A solution to one cause of poor smartphone battery life

[curiousfunk]

I'm not sure if this qualifies as badassity, but I thought this might benefit others.

Poor battery life is a common critique of smartphones in online phone reviews and forums.  This post explains one reason for this, and a workaround to extend battery life.

I bought a used HTC Sensation in December 2012 after avoiding cell phones for as long as possible.  The phone was officially released May 2011, but is still a well-spec'd device in 1Q13.  I had done plenty of research prior to picking this phone, and picked up on the poor battery life theme.  So I had low expectations, and sadly those expectations were met.

To clarify: I expected the phone to consume battery during actual use, but I was not prepared for high battery consumption when the phone sat idle.  In my ignorance, I initially assumed this was due to the age of the phone & battery.  I received a new battery for Christmas, but the idle drain persisted.

WiFi calling is a feature of T-Mobile and Republic Wireless (maybe others).  In my case, I have poor T-Mobile reception inside my house so WiFi calling is great.  Other carriers offer similar functionality through use of their femto cells.

I have WiFi at home and work, and my phone will automatically disable its cellular radios when connected to WiFi.  The behavior is like this:

1) get within known WiFi AP,
2) associate with WiFi network,
3) disable cellular radio.

This works automatically at home and at work, which is nice.  Except that I noticed a pattern of drastically different battery consumption rates at the two locations.  Looking at Android's own battery power info (Settings -> Power) as well as Battery Battery Stats, I noticed that my main battery drain at work was WiFi.  I'd lose 60% battery in a 9-hour shift.  With this knowledge, I disabled WiFi at work - leaving just the cellular radios on.  This has cured my at-work idle battery drain!

The cause turns out to be related to the distance between my phone and our wireless access point.  At home, I'm never more than 20 feet from my AP.  At work, my desk is 50 feet from the AP, through two cinder block walls and two metal stud framed walls.  At home, I have a "loud" WiFi signal.  At work, my phone struggles to "hear" our access point and frequently drops & reacquires its connection.

This applies to any device that automatically associates with known wireless networks - if your signal sucks, it'll drain your battery quick.

I now manually enable WiFi at home and disable it at work.  I'm going on over 3 days since my last charge, and I still have 25% life remaining - image attached.

Based on my reading, there are two explanations for this.

1) It takes a lot of power to "hunt" for a RF base station.  This is as true for WiFi as it is for cellular base stations.  Here's an interesting look at how WiFi affects laptop battery life.  In short, while using WiFi resulted in only a 2% decrease in battery life, constantly dropping & reconnecting to an AP caused a 6-7% drain.

2) The inverse square law applies to all RF wireless transmission through clear airspace - radio signals attenuate with the square of distance.  Add obstacles and other losses, and attenuation increases.  The RF amplifiers in cell phones compensate for this by automatically adjusting output power depending on distance and signal quality.  If distances are great and/or signal quality is poor, output power will increase.

[Daley]

Sounds about right, and stands up to logic and reason: wireless reception will impact battery life. Holds true to all phones. Glad to hear you've found a way in your usage pattern to maximize battery life through the day. Definitely something here for others to learn from.

Unfortunately, it still doesn't change the fact that battery life on smartphones is just straight up worse than feature phones, or that letting a lithium ion battery deep discharge like you are will shorten its lifespan.

[Travolta]

I highly recommend an app called JuiceDefender. What it does is shut off both the data network AND wifi while the phone is in sleep mode (screen off), and re-enables it when you turn the screen back on. You will still receive calls and texts in real-time, and for everything else (like email) it syncs every fifteen minutes, or whatever you decide. The battery savings are extraordinary. It effectively doubles the life of my Galaxy S2 battery.

[sol]

I had mixed results with JuiceDefender, and can't really recommend it.  It works sometimes, but the lags are annoying and it interfered with other things I wanted my phone to do on a schedule by disabling the radios at times when it wasn't supposed to.

The biggest battery suck on all of our android phones has been the GPS and location services.  If you have those enabled so that it tracks your whereabouts, battery life suffers.  My old evo was only good for about 20 minutes of GPS navigation if it wasn't plugged in.  Second biggest suck is 4G service, which I've decided is a waste.  So much power for such a slight increase in speed.

With CM7 and several of the alternative ROMs, you can underclock your phone to preserve battery power.  Bigger faster phones with faster processors need more power, and if you tolerate the slight lags and delays you can lengthen the battery life some.

I've found that I get way better battery life on my current phone (LG Optimus Elite) than I got with my old HTC Evo, which was a much better phone.  The Optimus comes with a slightly larger battery (1520 mah vs only 1500 for the evo), but is a much slower chip, with a much smaller screen, and no 4g.  The difference is night and day.  My old Evo basically had to stay plugged in at work all day in order to last me through the evening.  The Optimus only gets charged at night, and hasn't died on me yet.

[NumberJohnny5]

In general, using wifi will HELP battery life.  In fact, you'd want to go in the settings and make sure wifi is NOT set to be turned off when the phone is idle.  This is because, usually, using wifi is less of a battery drain than maintaining a mobile data connection.  Note, I know this is true for 3g/4g, but not so sure about 2g, any cdma technology, etc.  Now, if you're getting a very weak wifi signal but a strong cellular signal, I can see where the inverse may be true.  If your work is open to the idea, you could install a wireless repeater to boost your wifi signal.

Another idea if you're using a gsm provider, is to force 2g mode.  2g uses much less power than 3g.  In many cases, you'll get an even stronger signal than before (many phones will do everything they can to use a 3g signal, even if there's a much stronger 2g signal available to use).

As for using a battery defender app to enable/disable mobile data, be careful.  If you have a small allotment of data and it's billed per megabyte instead of per kilobyte, you're chewing through a lot of extra data each time your data connection is reset.  Basically, the mobile service rounds up your data usage per session.  A session can last for many hours (even many days, though often it'll be reset at least once a day for billing purposes).  Let's say every 30 minutes your phone uses 1.5kb of data (background data for checking email, facebook messages, whatever).  If your phone maintains a data session for the entire 24 hours, that's 72 kilobytes that's used.  If your mobile provider rounds up to the nearest kb, that's still 72kb a day, or 2,160kb in a 30-day month (just over 2 megabytes).  If you use a battery defender app and the data session is reset every 15 minutes, then that's a minimum of 1kb billed every 15 minutes, 144kb billed per day, 4,320kb in a month (just over 4mb).  Ok, if your provider bills per megabyte, then each day you're using 72kb in one single session, it's rounded up to 1mb, or 30mb per month.  But if you're using a new session every 15 minutes (and they always round up...so 0.0001mb is rounded up to 1mb), that's 96mb per day, or 2,880mb per month (about 3 gigabytes).  THAT could provide for a huge bill shock.  Side note...every time your phone connects to wifi, it closes the mobile data session; so if you connect/disconnect from wifi many times a day, the same phenomenon happens.

[curiousfunk]

Unfortunately, it still doesn't change the fact that battery life on smartphones is just straight up worse than feature phones, or that letting a lithium ion battery deep discharge like you are will shorten its lifespan.

Thanks for the link.  I've been staring at Table 2:  "Cycle life as a function of depth of discharge", and it seems to me it's at best a wash for lifespan.  In fact, the table data argues in favor of deep(er) discharge.  For the sake of easy math, let's say a battery is 100% depleted in 10 hours.  That means I can get 1 hour of use and 3750-4700 discharge cycles, or 2.5 hours of use and 2000-2500 discharge cycles, or 5 hours of use and 1200-1500 discharge cycles, etc.

Recreating Tabe 2:

Code: [Select]

Depth of discharge      Discharge cycles
--------------------------------------------------------------------
100% DoD                  300 –   500
 50% DoD                1,200 – 1,500
 25% DoD                2,000 – 2,500
 10% DoD                3,750 – 4,700
In all those cases, the increase in use-time (depth of discharge) is greater than the decrease in discharge cycles.  So I can technically get more "life" through deeper discharge.  See what I mean?  Example using the 10% and 50% data: I can get 5x use-time at the cost of 3.13x decrease in discharge cycles.

In any case, if I recharge every other day at say 50% discharge, I'm still looking at 6.5 years worth of discharge cycles.  If I realistically see even half of that, it doesn't matter - I'll likely have a different phone in 3 years.

[curiousfunk]

If your work is open to the idea, you could install a wireless repeater to boost your wifi signal.

I just picked up two used WRT54G routers for $10 (gotta love CL!).  One is for my mother; the other is coming to work with me on Monday to be used as an AP on the opposite side of our building.

[chucklesmcgee]

I had mixed results with JuiceDefender, and can't really recommend it.  It works sometimes, but the lags are annoying and it interfered with other things I wanted my phone to do on a schedule by disabling the radios at times when it wasn't supposed to.

You just need to set the application to less aggressive settings. For instance, force it to keep the radios on after inactivity or the screen being off, but still keep gmail and other apps to only sync every 30 minutes or hour or more. Also set the phone to turn off any data during whatever hours your bedtime is.

I've been very impressed with my Galaxy S3's battery life. Despite the enormous screen and the fact that I'm using it for about 4 hours a day or so, I have no trouble getting through the day.

[Daley]

Thanks for the link.  I've been staring at Table 2:  "Cycle life as a function of depth of discharge", and it seems to me it's at best a wash for lifespan.  In fact, the table data argues in favor of deep(er) discharge.  For the sake of easy math, let's say a battery is 100% depleted in 10 hours.  That means I can get 1 hour of use and 3750-4700 discharge cycles, or 2.5 hours of use and 2000-2500 discharge cycles, or 5 hours of use and 1200-1500 discharge cycles, etc.

Recreating Tabe 2:

Code: [Select]

Depth of discharge      Discharge cycles
--------------------------------------------------------------------
100% DoD                  300 –   500
 50% DoD                1,200 – 1,500
 25% DoD                2,000 – 2,500
 10% DoD                3,750 – 4,700
In all those cases, the increase in use-time (depth of discharge) is greater than the decrease in discharge cycles.  So I can technically get more "life" through deeper discharge.  See what I mean?  Example using the 10% and 50% data: I can get 5x use-time at the cost of 3.13x decrease in discharge cycles.

In any case, if I recharge every other day at say 50% discharge, I'm still looking at 6.5 years worth of discharge cycles.

I don't think you're reading that table quite right. Also, that logic doesn't factor in all the other environmental issues that help shorten battery life further. Most cell phone dealers I've talked with say that the average person is lucky to get about 12-18 months out of most smartphone batteries before the lifespan gets noticeably shortened, and is frequently an excuse used to justify buying a new phone despite it being a $10-20 part replacement.

If I realistically see even half of that, it doesn't matter - I'll likely have a different phone in 3 years.

How very consumerist treadmill of you.

[Daley]

I had mixed results with JuiceDefender, and can't really recommend it.  It works sometimes, but the lags are annoying and it interfered with other things I wanted my phone to do on a schedule by disabling the radios at times when it wasn't supposed to.

The biggest battery suck on all of our android phones has been the GPS and location services.  If you have those enabled so that it tracks your whereabouts, battery life suffers.  My old evo was only good for about 20 minutes of GPS navigation if it wasn't plugged in.  Second biggest suck is 4G service, which I've decided is a waste.  So much power for such a slight increase in speed.

With CM7 and several of the alternative ROMs, you can underclock your phone to preserve battery power.  Bigger faster phones with faster processors need more power, and if you tolerate the slight lags and delays you can lengthen the battery life some.

I've found that I get way better battery life on my current phone (LG Optimus Elite) than I got with my old HTC Evo, which was a much better phone.  The Optimus comes with a slightly larger battery (1520 mah vs only 1500 for the evo), but is a much slower chip, with a much smaller screen, and no 4g.  The difference is night and day.  My old Evo basically had to stay plugged in at work all day in order to last me through the evening.  The Optimus only gets charged at night, and hasn't died on me yet.

QFT.

I also can't recommend JuiceDefender for similar reasons. I did get slightly better performance out of Easy Battery Saver over JuiceDefender, but I also got the same (if not marginally better) battery life and better overall phone performance without having EITHER installed. These sorts of apps only seem to significantly improve battery life on monster high-end phones that have terrible power management due to the manufacturer's configuration of the OS.

Josetann also made some very valid points regarding data billing with these apps as well that shouldn't be ignored.

In hindsight, battery saver applications seem like they're in the same camp as app management apps... panacea cures for crappy OEM Android handset configurations. At best, they fix what shouldn't be broken, and at worst can potentially cost you more money depending on your carrier.

[curiousfunk]

I don't think you're reading that table quite right.

Well, of course I think I am. ;)  How do you read the table?

Several people in that battery university link commented on the same pattern.  And "Eske Rahn" took the next step, adding degradation due to cycling to Table 2's DoD data:

On May 7, 2012 at 5:48pm Eske Rahn wrote:

But I wonder if the conclusion drawn on table 2 are absolutely accurate?

Here I have extended it with a third column, taking the deterioration into account.

  DoD   Cycles   Total Usabilty
 100%    500    435
  50%    1500   638
  25%    2500   531
  10%    4700   400

The added last column says how many times of "full charge" you will get out of the battery in total during its life time with each approach. Assuming a linear deterioration of the battery from 100% to 70% of its original capacity (This assumption is backed by a graph in the article) So the integration is simple, and can be replaced by a mean factor of 85% * to the pure product.

This would yield that the optimal total capacity gained from the battery over its lifetime is a DOD of around 50% (Plotting the four points and combining with a smooth curve indicates that the maximum is around 60-50% DoD)

*
More precisely
Sum{n=0 to N-1 of DoD*(1-30%*n/N) }
=DoD*(N - 30%*Sum{n=0 to N-1 of n/N})
=DoD*(N - 30%(N-1)/2)
=DoD*(N*85%-15%)
~DoD* N*85%

Best Regards
Eske Rahn

Also, that logic doesn't factor in all the other environmental issues that help shorten battery life further.

My argument doesn't include other factors on purpose - you wrote specifically about deep discharge in your first reply.  I would argue that while depth of discharge is a factor, it can be irrelevant in light of other factors.  In my hypothetical case (recharging every 2 days @ 50% DoD), age/temperature/overcharging will kill my battery well before I reach the theoretical 1500-cycle "limit".

Most cell phone dealers I've talked with say that the average person is lucky to get about 12-18 months out of most smartphone batteries before the lifespan gets noticeably shortened...

This fits my experience as well, but I only have a single anecdote.  My wife's HTC Aria is 2.5 years old.  Her original battery almost made it to 2 years.  One day, she noticed that she couldn't keep the phone's back cover closed.  She also complained about poor battery life.  I took the case off her phone and the back literally fell off.  Her battery had swelled to the point that the phone's TPU case was the only thing holding the phone together.  Swelling seems to be most frequently a consequence of overcharging or temperature extremes.  I have no way to determine the cause of her battery's bloat.  If it's overcharging, then the fault is within the charging circuitry.  It may be due to temp extremes - no doubt the phone has occasionally been forgotten in a glove box on a hot sunny day.  Or keeping her phone on charge all the time when home.  Or aging: consensus says most li-ion packs fail in 2-3 years, even if unused.  Most likely a combination of the above.  Except deep discharging, which doesn't describe her usage.

How very consumerist treadmill of you.

This is simply unbecoming.  If this is the level of discourse I can expect here, I'm done.

[Daley]

My argument doesn't include other factors on purpose - you wrote specifically about deep discharge in your first reply.  I would argue that while depth of discharge is a factor, it can be irrelevant in light of other factors.  In my hypothetical case (recharging every 2 days @ 50% DoD), age/temperature/overcharging will kill my battery well before I reach the theoretical 1500-cycle "limit".

No, I tried gently point out that discharging 75% of your battery like you're doing is going to shorten its lifespan further, nothing more, nothing less. I wasn't trying to pick an argument, I was trying to help you get the most life out of your cell phone's battery. Instead, you try to argue semantics about limited scope factors on battery life and defend abusing your battery out of general principle because "it doesn't matter - I'll likely have a different phone in 3 years."

If you hadn't noticed, that's a pretty weak excuse around these parts. This is the MMM forums, not Miss Polly Prissypants' Credit Boo-Boo Hugbox Forums. We are a hardcore community of frugal, anti-consumerist, tough-love, make-do badasses, and so long as Canadians are getting berated for their irrational boat attachments, upgrading your car every five years and your cellphone every two-three like the rest of the first world is going to earn you a figurative punch in the face around these parts.

Even still, the numbers you threw back to defend your point still places the sweet spot on Li-ion lifespan discharge at between 25-50%. This is still reasonably faulty for calculating total effective usage hours as Li-ion discharge isn't linear (why I said you aren't quite reading the stats correctly), as the first 50% lasts longer than the last 50%, and doesn't change the fact that discharging your battery as much as you are is shortening its lifespan more than necessary. If you're fine with that, more power to you, but don't be shocked for being called out for it for the reasons you cited.

This is simply unbecoming.  If this is the level of discourse I can expect here, I'm done.