Tuesday, November 1, 2016

Tips on extending battery life; from a battery guru.

Probably best to get this out of the way right upfront: The battery guru referred to is yours faithfully.  Sorry to disappoint.  

Regular readers of this blog (all seven of you) know my rather famous (!) blog post titled “Battery rules”, where, inspired by Michael Pollan, I penned a poem that went,

Don’t charge them too high
Don’t swing them too wide
Keep the temperature low to extend their life.

This post was a follow on to my other tremendously popular (read:  eight total views) post titled “Pull the plug.  Your battery will thank you.”  The posts were meant to provide an understanding on why Li-ion batteries fail and how one can use this understanding to extend the life of batteries.  

Just to be clear: These rules are for Li-ion batteries; for phones, laptops, cars etc.  Not for other kinds of batteries.  And not for every Li-ion battery known either; but the most popular ones.  

The reason’s for these rules are simple.  Li-ion batteries don’t like to be at the top of charge (because of side reactions that consume lithium).  They don’t like being charged and discharged completely (because of the volume change associated with moving a lot of lithium back and forth and the associated stresses).  And higher the temperature, more the side reactions that impact battery life. 

When the blog post was written, the whole fast charging of batteries was not a big deal.  But now, we are seeing more and more emphasis on this.  Fast charging can also be bad.  More on this at the end of the post. 

The simple statement “Pull the plug” was really a way to implement the 1st rule “Don’t charge them too high” without much thought.  In other words, don’t use your laptop like a desktop and keep it plugged in.  All. The. Time.  This only makes the side reactions worse.  Stop charging.  Let the battery discharge a bit. 

Every time I walk into a meeting and see the inevitable dive under the table to find the charging plug, followed by attaching the charging cable to the Mac laptop, only to see that the charging light is green, my blood pressure increases.  I then go thru the sermon explaining why they need to pull the plug. 

They comply.  For 15 mins. After which they are back to plugging it in.  Blood pressure increases.  Cycle repeats.

And No.  No real breakthroughs have occurred since the post that makes the rules obsolete. 

Clearly, the rules are not working.  I’m sure it is not the messenger (?).  I’m sure it is not the message.  So it must be the way the message is delivered. 

So this post is battery rules redux.  I will explain how I charge and use my battery.  Hopefully, this gives folks a sense for how they can maximize their battery. 

My track record:  I have a computer that lost 4% capacity in 2 years.  I have a phone that lost 5% capacity in 2 years.

In this post, I will focus on the phone and how I baby the battery.

First some basics:  On any typical day I only discharge my battery around 30-40%. There may be an occasional day when I will discharge the battery by 70%, but those are rare.  In other words, I could use my battery for two days without needing to recharge it. 

If my usage seems rather minimal, it is not (although I never got into the pokemon revolution).  Rather it is because of two attributes that many of you probably share. 

First, I have an iPhone 6 plus (the YUGE one that does not fit in any normal size pocket).  This means that the phone has a pretty large battery.  And although the screen size is larger, the bigger battery more than compensates for the power draw of the screen.

Second, the operating system on the phone, iOS, is very battery friendly.  I moved to the 6 plus from an Android phone (with, I believe, the KitKat version).  That phone had to be charged after an eleven hour day, despite having a reasonably large battery.  It would drain 5% of the battery capacity every hour, even when the phone was, supposedly, idle.  iOS appears to lead to less battery drain, especially during rest.

Based on my usage profile and the attributes of my phone, here is what I do:

- I charge my phone in the evening close to when I’m about to go to bed.  My battery is around 40% state of charge (SOC).  I charge it to around 80-85% SOC.  Takes me ½ hour or so.  I don’t agonize over the exact SOC, I pull the plug when I remember to. 

- I sleep and when I wake, the battery has lost maybe 3% of so of its capacity (No. I don’t sleep for 1 h.  I sleep the normal 7 h).  I don’t plug it in. Rather, I go about my day, make calls, check email, listen to podcasts, etc.   I use the phone as much as the next person.  When I get back home I’m at approximately 50% SOC and a couple of hours later, when I’m ready for bed, at 40%.  I repeat the charging cycle.

In effect, I don’t let the battery charge too high.  I don’t let it swing too wide.  In other words, I follow my rules.  I don’t leave it charging all night (i.e., where the battery is fully charged a couple of hours after we hit the bed and is sitting at the top of charge for the next 5-6 hours).  I don’t discharge it all the way down each time before charging it.

- When I feel like my day may be particular long with a lot of phone use, which for me typically means I’m on travel, I charge the phone upto 100% SOC, typically in the morning before the day starts.  Meaning, it is fully charged, but is not sitting at the top of charge for hours on end. 

- Every few months, when I’m not on travel and have a predictable week, I fully charge the battery and drain it down all the way before recharging it.  This should allow the phone to calibrate the SOC by performing a discharge capacity check. 

This may sound counter intuitive:  I have a battery that should last 2 days.  But, I choose to charge it every day.  In effect I’m racking in cycles, twice as fast as I need to.  Is that not bad? 

I would posit that it is not.  The number of times the battery is cycled has to be put in context of the SOC that the battery swings, and how much time it spends at places where side reactions can dominate.  We know that batteries can cycle 300,000 times (yes.  Three hundred thousand!) when the SOC swing is kept rather small (say 3%, like what occurs in a hybrid electric vehicle).  In these situations, the calendar life of the battery (i.e., how many years it lasts) is more critical than the number of cycles.

As we widen the SOC swing, the battery life decreases.  The worst is fully charging and discharging the battery.  Even worse is going to the top of charge where we have side reactions.  My scheme minimizes the time spent at the top of charge, without overly swinging the battery. 

And it keeps me on a reasonably manageable daily schedule for charging without having to think too deeply about it.

Frankly, it would be better for me to charge the battery upto say 50-60% SOC and let us drain down to 10-20% each day.  But, that, to me, is living dangerously.  Having that extra 20%-30%  of buffer capacity gives me peace of mind (What if I get lured into Pokemon Go on my way home?).

Irrespective of what phone you have, and how you use it, the two rules (don’t charge them too high and don’t swing the too wide) will help you.

As far as the rule to keep the temperature low, I don’t keep my phone on my car dash where it is hot. My phone happens to be sitting somewhere in front of the air vents, which helps during summer.  I don’t keep my phone on hot metal tables in summer. 
I have been known to go so far as to contemplate living in frigid conditions just to avoid the high temperatures in some parts of the world.  Talk about taking ones job seriously… 

Now for the whole fast charge issue. 

In my blog post titled “The Hero with four faces: Part 1” I explored the question of fast charge and the issues surrounding it.

Short summary:  Lithium can plate (which makes the battery go boom).   If you swing the SOC too much and too fast, you can break the particles (although this may not be as big a deal as folks think it is).  And the battery tends to get hot when charged fast , violating the rule to keep temperatures low. 

Turns out that one does not want to go too low in temperature either.  We know that lithium plating is much more of a problem at low temperatures; i.e., Golidlocks effect. 

Of these, lithium plating is the one to avoid at all cost; the rest only degrade the battery and are not dangerous (unless the battery really starts to cook).  All it does it kill the battery capacity rapidly, which, for a device that one changes every 2 years, is not really a big consideration.

Having said all this, the phone maker meters in the charge from the charger.  Meaning, just because you get yourself a beefier charger does not mean that your phone is going to accept the charge.  The phone maker is supposed to watch out for the battery. 

But there is a push to use fast charge as a differentiator in consumer electronic devices.  This means is that unless the company is performing some sort of smart charging to avoid the problems above, the battery is probably not going to do well. 

So, unless you hear it from this blog, best to avoid fast charging your phone. Assuming your phone maker gives you such an option. 

In the meantime, assuming poems are not your thing, can I remind you all that that friends don’t let friends keep their batteries plugged in.  So, be a nice buddy, and ask your neighbor to pull the plug. 


Venkat