Slow charging apparently related to temperature. Warm = slow

[RedOctobyr]

I did a quick test last night, comparing charge current with my case installed, and then I removed the case, and the phone's back, and had a fan blowing across it. I observed higher charge currents as the temperature decreased.

So today I repeated my test. This time using an air conditioner, instead of just a fan, to get the battery cooler.

Ambient temp before starting is 80F. Brightness up fully, Auto-brightness disabled, screen on constantly, WiFi on, Waze running, to try and increase the current draw as much as possible. I used the stock LG charger (1.8A) and cable. I did not unplug the charger, or turn the screen off, during the test. My phone is a Verizon model.

Charge current is *definitely* related to temperature. Presumably the phone's measured battery temperature.

I had the phone sitting around, with the Spigen Slim case installed. It was charging around my usual 280mA. The battery was around 98F.

Then I removed the case, and the phone's back. Turned the window air conditioner on high, and held the phone in front of the A/C's output, to cool it down.

As the battery temperature dropped (eventually reaching 57F), the charge current increased. At 86F, charge current was 600mA. At 74F, current was 1570mA. It stayed around 1500mA even as the temperature dropped further.

It appears that 90F may be the temperature cutoff for dropping to 280mA, since the current changed as it passed that temperature. I don't know when exactly it lets the phone increase to roughly 1500mA.

Then I put the cover & case back on, and turned off the A/C. As the battery temperature increased again, the current dropped off again.

I wish the History tab of Battery Monitor Widget was working better for me. It sometimes doesn't record. I don't have a good chart that shows current, battery %, and temperature all together.

See attached screenshots. In the graph views, the test started with the solid-green line that indicates AC charging. Starting with no cooling, about halfway across the graph, then I turned on the A/C when you see the current spike up, and the battery temperature drop.

So if the phone stays cooler, it will charge faster. Alternately, if the software perhaps allowed a higher battery temperature before reducing the charge current, it would also charge faster.

 

[anon(782252)]

Very nice. I'll see if I can try some sort of test later with my QC charger.

 

[erwaso]

That's some fancy testing right there!

 

[Fred Wilson2]

That would be expected with any lithium smart charger.
The charging current should always be in reverse proportion to the battery temperature.
One of the fastest ways to really quickly decrease the capacity of a lithium battery is to charge it when hot.


Fred

 

[belodion]

Thank you for your thorough and scientific testing, and for sharing the findings.

Posted via the Android Central App

 

[RedOctobyr]

That would be expected with any lithium smart charger.
The charging current should always be in reverse proportion to the battery temperature.
One of the fastest ways to really quickly decrease the capacity of a lithium battery is to charge it when hot.

Fred

I use lithium-polymer (LiPo) batteries in a hobby, so I'm familiar with some of the lithium-battery charging considerations. Some of those batteries are several hundred dollars each, as are the chargers, so taking care of them is wise. The user typically manages temperature issues themselves, however, as there is no standard temperature interface.

But with that said, if the phone is set to reduce the charge current to ~0.3A when the battery reaches 90F, for discussion (those #'s may not be quite right), maybe that is simply too-conservative. If the phone is working and generating some heat, which already pushes the battery's temperature higher, then charging will be throttled back so far that it barely does anything.

For discussion, maybe raising the temperature threshold limits by 10F might at least allow 0.6A charging more often, vs 0.3A, which would at least help.

I had navigation and audio going in the car this morning, using a 2.1A charger, and my charge percentage stayed level. Then it actually *dropped* by 1%. To me, that's not really acceptable. I've had my charge level actually drop overnight, while on a 2.1A charger, with the screen off. I suspect because something was running and warming up the phone, which then reduced the charge current. Some people report warm-running phones, and that they got replacements which ran cooler. So maybe my results are skewed, maybe I have a warm-running example.

Now realizing that the phone has the *ability* to charge at 1.5A even with the screen on, and at full brightness, it makes my usual 0.3A charging (screen-on, Auto brightness at 6%) even more frustrating. I'm realizing my charging speed is "phone-limited", not input-power-limited.

And thinking about Quick Charge 2.0 chargers, I wonder how QC 2.0 can do much to speed up the process, if the phone often doesn't take advantage of the power already available from the 1.8A stock charger?

 

[anon(782252)]

So I played around with temps a little with my Tenergy 2.0 charger. No matter how low I tried to push the temp, the phone wouldn't do any more than 2050 mA with screen off. It was nice to see the phone charge at 1500 mA with the screen on when the temps are under control. Once the temp approached 90F, it would only do about 330 mA.

I even tried putting an ice pack on the battery (paper towel in between to control condensation) but still couldn't get anymore out of it.

 

[dpham00]

Even at over 90 degrees f, I was still getting around 1.8a

Btw my charging profile is here for both the stock and QC 2.0 charger.
http://forums.androidcentral.com/lg-g4/536936-charging-profile.html


Sent from my Verizon LG G4

 

[RedOctobyr]

Very interesting, thanks for sharing that.

Was this with the screen on, or off? If I can get Battery Widget Monitor taking History readings consistently, I can try the test again with the screen off.

Of course, it's entirely possible that more than one factor is considered when the phone picks a charging speed. It may be looking at more than just battery temperature, including whether the screen is on/off, etc. As other people share their findings, maybe we'll get a better understanding of what's going on. We can't really change it, of course. But it's encouraging to see some people getting higher currents, maybe something is not working quite right on some phones. Mine is Verizon.

 

[dpham00]

Also keep in mind that the temperature in the graph is battery temperature not cpu temperature. Cpu temperature usually is why charging is reduced.

Sent from my Verizon LG G4

 

[RedOctobyr]

I wouldn't have guessed that it would so CPU-temperature dependent, but I'll take your word for it. I guess it's less clear to me how CPU temp should impact battery charging behavior, but that's OK.

Three CPU-Z screenshots are attached. I don't remember what the phone was doing during the first two. And I have no context for what these temps should be, or what's "normal". But in the first, the #s are higher than I would have expected, I suppose, though that means nothing. In the second, they are even higher, and an app is giving a high-battery-temp alert.

The third was taken during my first charging-current test yesterday, when I had the back off the phone, and had a fan blowing air across it. So it's probably about as low as my temps get. I did not get any CPU-Z screenshots during the air-conditioner test today. BTW, the "battery" temperature entry is not the same value as other apps show for the battery temperature. I don't know what this is actually measuring, but my battery is not at 140F.

The fourth screenshot is when the phone was charging a few minutes ago, with the screen off, and having just been rebooted. Again using the LG charger & cable. When the battery is still at 88F, the current ranges from 1200 to 2100mA. In the next entry, 3 minutes later, the battery is at 95F, and the current dropped to 180mA.

 

[dpham00]

I should clarify that that is my opinion based on my observation. Sometimes I have seen battery temperature hit 100 and still get 1a or more charging.

Cpu z has something fishy with regards to the celsius to Fahrenheit conversion. Fwiw, BWM reports 99F

For example 40.7 C~105 F. CpuZ reports 130.9f

Sent from my Verizon LG G4

 

[RedOctobyr]

I should clarify that that is my opinion based on my observation. Sometimes I have seen battery temperature hit 100 and still get 1a or more charging.

Cpu z has something fishy with regards to the celsius to Fahrenheit conversion. Fwiw, BWM reports 99F

For example 40.7 C~105 F. CpuZ reports 130.9f

Hmmm, very odd Your Celsius screenshot of CPU-Z shows 53C, then the F version shows 162F. Except that 53C is 127.4F, not 162F. So there is a conversion error going on.

Interestingly, for me, the Battery tab in CPU-Z shows the correct temperature.

 

[RedOctobyr]

I see the apparent C->F conversion problem on my S3 as well as my G4. I've only noticed it in the Thermal tab. Thank you for pointing that out, I hadn't tried switching between units before to compare. I emailed the CPU-Z developer to let him know.

I had my phone on the LG charger overnight, screen off, WiFi off. This morning the battery was showing 102F, and the phone was warm near the camera, when I went to unplug it. Something is generating a fair bit of heat just sitting there, after the charge is complete. The house was at 73F this morning.

I rebooted the phone after unplugging it, to start fresh. In the car (set to 72F, and it's rainy here today, so no bright-sun issues), the battery showed 94F after about 40 minutes of navigation & playing audio, with the screen on, set to Auto, 6%. I was not charging it in the car. So just sitting around, full, on the charger, it was running warmer than when not charging but actually doing stuff.