Related
Have an SGSII running 2.3.4
When in car and Bluetooth is on, to provide hookup to car kit, all OK ... runs all day no problem. (connects to Parrot in-car kit) ... screen is off unless call is active.
However when I enable Satnav (buit-in Navigation app) even though phone is plugged into car 12V supply the battery drain exceeds its charge rate .... slowly discharges, after about an Hr you get 'battery Low' warning ....
Presumably the display being permanently on is drawing so much current that the USB charge lead cannot keep upl
The phone confirms that plug is active and charging ... but the drain obviously exceeds charge rate.
This is a significant issue ... anybody know a fix for this, can you change the rate of charge ?
need to change your charger, i had this using a 500ma charger, i have a 1000ma one now and dont suffer this problem now
kinjo is right. You may experience this as well when charging off a laptop. Not all chargers are created equal
Ah ... OK I had assumed that there would be not power limit from the DC outlet as they can produce 5A.
I'll have to go loook for a 1A ver
The SGS2 charging rate is limited to a max of 650mA and any current from a charger higher that this will be limited.
Default AC 650mA
USB 450mA
If rooted, there are apps and scripts that can increase the rate of charge through the USB to 650, but you're likely to damage your PC port or encounter other problems if you do this. I would think, but not 100% sure it would be OK from a car USB port, you'd have to check the specs.
Voltage Control
Although the app and scripts have scope to select a higher charge rate than 650mA, any higher setting than that would be purely placebo.
An accurate charge rate will not be shown by most SGS2 kernels. Exception is siyahkernel, where changes have been made to shown accurate charging current with app such as Battery Monitor Widget and Current Widget
For further reading, from Siyahkernel v2.6.9 Changelog;
"added Entropy512′s changes to sec_battery driver to show battery charging current and changed it a little bit to map the adc readings to real charging current miliamps (mapping is done in an unscientific way but it works). you will be able to see how fast your device is being charged and also why those 100%s are not the same. (thanks to Entropy512). since I don’t know how to read discharge current (maybe there is no way to read it) you will only read that data when your device is in charging state"
Your phone will take care of how much current it drains from the USB port or charger. To do this it needs to differentiate between a USB data port and a charger. This is normally done by checking the impedance between the USB data lines D+ and D-. USB data ports will have more or less high impedance since these lines are used to transfer data . Chargers however should have a short between D+ and D-. This is how your phone can tell if it is connected to a charger or USB data port.
It then adjusts its power consumption from USB (less when connected to a USB data port, more when connected to a 'good' charger).
A very simple test to check for a sleazy charger is this: connect the charger to your phone and wait for a moment. If the phone reports a USB device (besides charging ), then change the sleazy charger.
With a 'good' charger your phone should always charge, no matter what Apps your running.
fxrb said:
Your phone will take care of how much current it drains from the USB port or charger. To do this it needs to differentiate between a USB data port and a charger. This is normally done by checking the impedance between the USB data lines D+ and D-. USB data ports will have more or less high impedance since these lines are used to transfer data . Chargers however should have a short between D+ and D-. This is how your phone can tell if it is connected to a charger or USB data port.
It then adjusts its power consumption from USB (less when connected to a USB data port, more when connected to a 'good' charger).
A very simple test to check for a sleazy charger is this: connect the charger to your phone and wait for a moment. If the phone reports a USB device (besides charging ), then change the sleazy charger.
With a 'good' charger your phone should always charge, no matter what Apps your running.
Click to expand...
Click to collapse
I'll give that a try.
Sargan said:
I'll give that a try.
Click to expand...
Click to collapse
It wont make any difference.
The answer of fxrb is controversive. If the charger is physically - by the way - disconnected from D+ & D- (no use) and has a high impedance due to missing connection, it makes no sense to determine if the charger is "done" or not on the phone, because you cant find out this way.
Important is that if your charger has a max. load of 0,7 A (700mA) for example, and you are using it inside your car, it may get hot during exposure to sunlight & loading to the max.
Depending on internal circuitry of the charger it will decrease the output current if the charger gets too hot (thermal protection). The internal transistor will have considerable more losses @ these temperatures (~+80°C) and higher switching times, creating more losses in general.
Usually the cheap chargers have a 34063 (or 33063 for lower temps -40°) Step-Down buck switching controller, which reaches its economical end @ this current - giving all it's got with little room for environment changes.
What will help:
You can try a charger with a higher output current capability (around 1A +) which should compensate with the high temperatures and your current requirement. On the other hand, your phone could limit the current before the accumulator reaches unsafe temperature levels due to sunlight,heavy loading and charging @ the same time. assuming you also have a black sgs II.
Your phone charging current is also not linear, in this case higher @ start and low voltages, and will decrease with charged value.
Furthermore, your phone will automatically limit the charging current in correspondence to the accu charge value. Li-Ion charging is more complicated than an ordinary NICd or NIMH, your "charger" only provides a -as most cheap & stable as possible - current for the actual charging cuircuitry, which is located on the phone bottom.
So it wont adjust to anything except knowing if 650 or 450 is allowed, the rest is managed internally and (important) excluded from the firmware.
I hope i could help. gl
Tskusie said:
It wont make any difference.
>>>
So it wont adjust to anything except knowing if 650 or 450 is allowed, the rest is managed internally and (important) excluded from the firmware.
I hope i could help. gl
Click to expand...
Click to collapse
I have purchased a 3A charger ... and verified that it can give well over 2 A
This is with correct lead as Phone sees this as AC charging not USB.
It still cannot keep up with the discharge rate ..
Phone + GPS + BT means more drain that it replace by charger ............ this seems a significant design flaw.
Phone is being supplied with apps that it can't run without going flat ..
I have purchased a 3A charger ... and verified that it can give well over 2 A
This is with correct lead as Phone sees this as AC charging not USB.
Click to expand...
Click to collapse
Ok, and correct, it is recognized as AC because of missing data leads. I must say it actually works for a lot of people with chargers with an output of 700mA - 1A (5 Volts USB-b) navigating and using standard apps.
this seems a significant design flaw.
Click to expand...
Click to collapse
I dont think there is that design error, honestly. Also, personally never encountered that problem.. It seems that your accumulator if faulty or it gets too hot. How old is your phone? Depending on usage, temperature and charging times it may have reached its end. But this is the last call i think...
Have you:
- Checked the accumulator temperature when performing (with an app like "spare parts" [free] or some code *#0000# or *#0# - battery stats)
it should'nt exceed temperatures above (approx) 54°C or >60°C is critical (for the accu), where the OS should shut down automatically.
- Tried to turn off one or two main things and checked if it is charging?
- Cleaned the contacts from possible metallic objects in between from chewing gum packaging for example (blow a few times)
- Tried to use any power saving mode
--- If you have a rooted kernel (CM9, Fluxi, Siyah etc...)
- Wiped battery stats in CWM & (maybe) clear cache.
- Undervolt the CPU (minor change) & or GPU
-Generally tried another MODEM, LQ5, LPS (depending on your geographic location) http://forum.xda-developers.com/showthread.php?t=1131950
or KERNEL http://forum.xda-developers.com/showthread.php?t=1255790 (fluxxi is not in the thread, which is also a good kernel & tweakable http://forum.xda-developers.com/showthread.php?t=1644648)
Please let me know if it helped you. I know it is an annoying problem. If you have an easy access to try out another accumulator, give it a shot, otherwise i would consider it as last option.
If nothing helped, please let me know which system (firmware) you are using with which kernel and modem in your next post (settings--about phone--). gl
It's about 6 months old.
Sargan said:
It's about 6 months old.
Click to expand...
Click to collapse
Have you read the rest? It is hard to help you if you don't provide information.
Like i said, i think you may have a faulty accumulator. Assuming the circuitry of the car can provide the necessary power, everything should be ok.
Which version of android are you using?
Which modem?
Which kernel?
Custom rom?
Have you tried to stress your phone to the max with the normal AC charger? If yes, results the same?
Android ver 4.0.3
GT-I9100
Kernel -3.0.15
Not tried stress test on AC
So you're on the latest stock ROM?
You should try the AC test, turn on everything, stress it for at least 20 min or longer while charging. If you encounter the same issues it is likely that you have a faulty power source. Clean the contacts before testing.
Let me know when you tried it out.. :highfive:
Tskusie said:
It wont make any difference.
Click to expand...
Click to collapse
It will make a difference if the phone fails to detect the charger as such.
Tskusie said:
The answer of fxrb is controversive. If the charger is physically - by the way - disconnected from D+ & D- (no use) and has a high impedance due to missing connection, it makes no sense to determine if the charger is "done" or not on the phone, because you cant find out this way.
Click to expand...
Click to collapse
I was referring to how a PD (portable device or phone here) detects a dedicated charger. To put it simple: if the PD fails to detect a dedicated charger it will normally limit the charging current at a lower level.
For further information you might want read the Battery Charging v1.2 Spec and Adopters Agreement. Just two quotes from there:
In order for a PD to determine how much current it is allowed to draw from an upstream USB port, there need to be mechanisms that allow the PD to distinguish between a Standard Downstream Port and a Charging Port
Click to expand...
Click to collapse
and
1.4.7 Dedicated Charging Port
A Dedicated Charging Port (DCP) is a downstream port on a device that outputs power through a USB connector, but is not capable of numerating a downstream device. A DCP shall source I DCP at an average voltage of V CHG .
A DCP shall short the D+ line to the D- line.
Click to expand...
Click to collapse
Furthermore there are many devices dedicated to detecting what charging device is connected to a PD. Here are just two examples:
The STUSBCD01B from STMicroelectronics
The FAN3989 from Fairchild Semiconductor
Tskusie said:
Ok, and correct, it is recognized as AC because of missing data leads. I must say it actually works for a lot of people with chargers with an output of 700mA - 1A (5 Volts USB-b) navigating and using standard apps.
>>
Click to expand...
Click to collapse
I use standard SGSII, no custom ROM
Kernel is 3.0.15-1I9100XWLP7-CL3400913
Baseband ver I9100XXLPS
I will try and watch temp on next car test ... one thing I can be sure of it was NOT related to over temp via solar gain in the, as at the time I first noticed this it was snowing.
fxrb said:
It will make a difference if the phone fails to detect the charger as such.
Click to expand...
Click to collapse
The only thing it detects if the connected power source is AC or Hub. For AC it just measures the Impedance (with a current) between D+ and D- which should be <200 Ohm. (doesn't matter if they are connected with low impedance on the AC, because they are not needed for charging).
To detect a hub it will recognize a VDAT_SCR (Data voltage source) on D-, and will start the above detection, or for standard hubs through their pull down resistors on D+/-. If it fails (between D+ and D- >>200), it is a hub charger.
I know what you meant. SOO if the charger itself is limiting the current, which can be possible depending on used charger, it will be detected as dedicated charger but the phone can adjust to everything it likes, it wont make any difference, thats what i meant. The maximum current will be limited to 650mA, and if the charger itself has a current limiting ability and for example only gives 500 because of any reason, how will you be able to destinguish that?
I am not familiar with the detection methods through USB itself, which determines if it is a downstream or charging port. Are you sure that the SGSII supports that? The USB port, doesnt matter which, follow standards, and if there is a device which consumes more than the 500 (HDD's for example) will just have another connector (without data) for parallel switching, (just to double the current sink capability) All of that wont help Sargan either
fxrb said:
For further information you might want read the Battery Charging v1.2 Spec and Adopters Agreement.
Click to expand...
Click to collapse
Thanks, i will have a look into it.:good:
fxrb said:
1.4.7 Dedicated Charging Port
A Dedicated Charging Port (DCP) is a downstream port on a device that outputs power through a USB connector, but is not capable of numerating a downstream device. A DCP shall source I DCP at an average voltage of V CHG .
A DCP shall short the D+ line to the D- line.
Click to expand...
Click to collapse
They are not really shorted, but yes low impedance -- <200 Ohm... . Arent these hybrid ports?
an average voltage of VCHG @ which current?
Yeah, there are a lot of IC's for that, idk what's in his charger. I've seen the DC-DC Converter 34063 in cheap ones...
Sargan said:
I use standard SGSII, no custom ROM
Kernel is 3.0.15-1I9100XWLP7-CL3400913
Baseband ver I9100XXLPS
I will try and watch temp on next car test ... one thing I can be sure of it was NOT related to over temp via solar gain in the, as at the time I first noticed this it was snowing.
Click to expand...
Click to collapse
OK, and you say you have your phone for 6 months? So you must have experienced that at the very beginning already???
Please do the AC test and let us know.
Tskusie said:
Thanks, i will have a look into it.:good:
Click to expand...
Click to collapse
Good idea...
are there any script to increase the usb charging current? I'm using BIFTOR V7.2 with jeeboo V2.1 kernel
A little bit off topic. .
How long does it take to charge ur S2 with a charger doing nothing but leaving it for charging. ?.. It took me 3 hrs to charge mine frm 2% to 100% using a 700mah charger. .
Is this normal or does it take less ? Plz share your opinion. .
Sent from my GT-I9100 using xda premium
Using a New Trent External charger I noticed that the phone was still losing power, just at a way slower rate than without it charging. Phone showed charging. I was using netflix via 3g at the time. Would this be expected? Not super worried as the battery charges just fine when screen is off and not in use.
How many amp output is that charger??
My phone can't keep up with the charge if I'm veiwing a movie using HDMI unless I run incredible kernel with fastcharge enabled.
7000mAh. Think 1V output?
IIRC, the stock A/C charger is 1 amp.
A lot of aftermarket chargers do not have the balls to charge the Rezound.
you may need to modify a usb cable to short the Data wires to get the full charging potential out of it.
or try one of these cables from amazon:
http://www.amazon.com/Naztech-Micro...3?s=electronics&ie=UTF8&qid=1328727460&sr=1-3
thatsricci said:
you may need to modify a usb cable to short the Data wires to get the full charging potential out of it.
or try one of these cables from amazon:
http://www.amazon.com/Naztech-Micro...3?s=electronics&ie=UTF8&qid=1328727460&sr=1-3
Click to expand...
Click to collapse
Does this cable work for sure with the rezound??
I don't need it right now with the incredicontrol running, but I ordered it just as a handy thing to have.
My new Trent works pretty well, but I wouldn't expect it to keep up with streaming vids. I found that it gives a charge to the idle phone at about 800mA, which is about the same as the stock wall charger.
Pick up battery monitor widget and you can track your plus and minus usage pretty well to decide if everything seems OK.
jmorton10 said:
Does this cable work for sure with the rezound??
I don't need it right now with the incredicontrol running, but I ordered it just as a handy thing to have.
Click to expand...
Click to collapse
Well, I'll answer my own question. The cable definitely DOES NOT force higher charging at least when using an HDMI adaptor.
If I use Incredikernel with fast charging enabled with Incredicontrol, it will actually GAIN charge while playing a full length movie through HDMI.
This morning, I flashed back to dsb 1.3 with no frills cpu control & hooked up the new cable for charging. I started with a 100% charged battery & after playing a 1 hour DVD RIP from my card, it was down to 89% charged. Using Incredicontrol, during the exact same test the battery still read 100% at the end.
Thanks for your results jmorton10. Good to know. I wonder if it's the connections inside the HDMI adapter messing with it.
Not sure why HTC can't just get this right to always pull the max possible from whatever it's hooked into!
-j
My understanding is because the MHL adapter needs the USB data pins to work properly and they can't be shorted out as mentioned above, the phone thinks it's on a USB connection and limits it's current draw to 500mA.
mjones73 said:
the phone thinks it's on a USB connection and limits it's current draw to 500mA.
Click to expand...
Click to collapse
That is definitely the problem, the battery monitor widget claims it is charging from USB even though it is plugged directly into an A/C charger.
I guess I will be flashing back to Incredikernel/Incredicontrol tonight. I do like dsb kernel slightly better as it never freezes/bootloops etc. ever on that kernel. It does stop responding occasionally on IC (requiring a battery pull) although it doesn't happen very often.
The Rezound complies with the USB charging specification. If your power source isn't compliant, it won't draw more that 500 mA (and probably less if USB doesn't enumerate, but I haven't checked). That's what the "short the USB data pins" thing is all about. Earlier USB charging spec said they should be shorted, the latest says 200 ohms. Any "USB charger" which doesn't do that isn't compliant, and shouldn't be sold as USB-anything, because it doesn't follow the USB spec. Return it to where you got it, and complain (loudly).
Regardless of how much power a charger can deliver, even if it meets spec you won't see more than about 800 mA go into the battery. There's a limit to how much current both the connector and the battery can safely handle, and the phone takes that into consideration.
Remember, the mA reported by utilities is (always?) what's going in/out of the battery. The charger would be delivering more than that (e.g. 800 mA into the battery, plus 400 mA to power a phone doing video streaming, etc.). AIR, the micro USB connector is limited to ~1500 mA, and I've never seen a battery charge at much more than 800, as reported by the kernel. Conversely, when charging from a non-USB charging spec compliant port, the phone won't draw more than ~500 mA from the port, and the battery only gets what's left after subtracting what's needed to run the phone.
Another thing which can affect charging is the cable. If you have a long USB cable using 28 gauge wire, there will be a significant voltage drop across it. USB specs say the voltage should be between 4.75 and 5.25 V, and Android seems to limit the charging current so the incoming voltage stays above 4.75V. Using 24 gauge USB cables, especially with longer lengths, can increase the charging current. Most vendors don't tell you the wire gauge used in their cables, but Monoprice does (no relation, etc.).
my SGSII takes about 3 hours to fully charge. in japan, we had charging stations, where youd place a battery in a machine, and itd charge it from 10% - 70% in about 8 minutes,. and to 90% in about 15. this was because the machines used a higher current.
ive noticed the stock usb charger that comes with the S2 is a lower voltage of those compared to HTC, or say one youd buy at radio shack. yet even with a higher voltage charger, the phone still charges at the same rate. this is because the kernel controls input levels. the input levels are set to slowly take in a current, there are many reason youd want to have a slower charge.
one is to reduce salt bridge deterioration, which is ware on the battery. while this may be an issue for some who plan on keeping their phone battery for years, without upgrading their phone, or battery, for most of us a year of ware wont make a difference.
a second issue is heat, a faster charging battery generally charges warmer than a slower charging battery. however the phone will still not get as hot as a phone sitting in the sun, or playing a high graphics intensive game.
i would like to adjust the kernel to allow higher frequencies, how much of a higher frequency? well that would require bench testing. but a higher frequency indeed
has anyone played with these settings in the kernel, or done any benchtesting on the matter? id be very interested to hear your findings.
It's not possible.
http://forum.xda-developers.com/showthread.php?t=1468834
This has been asked more times than I've been able to keep count on here over the past month :-/ Answer doesn't change (unless you want to blow **** up).
so then the only way is to use an external battery charger?
because i know for a fact, an external battery charger can charge a battery very fast, ive experienced this first hand
on that link you posted, it said that the sgs2 charging limit is 650mA, but the USB charge is set lower.. has anyone raised the USB charge to 650mA? this would at least help my phone charge faster in the car, or while plugged into my laptop
soraxd said:
so then the only way is to use an external battery charger?
because i know for a fact, an external battery charger can charge a battery very fast, ive experienced this first hand
on that link you posted, it said that the sgs2 charging limit is 650mA, but the USB charge is set lower.. has anyone raised the USB charge to 650mA? this would at least help my phone charge faster in the car, or while plugged into my laptop
Click to expand...
Click to collapse
The USB charge is not set to low, the current given by that of a computer USB port is less than 650mA (500mA if i remember correctly) and also depending on the car charger the output (might) be less than 650mA. I use the same computer USB cable with an external USB charger and i notice 2 things:
1- Since the phone pulls the required current from where ever the USB is plugged in, if it has capability of providing 650mA the phone will take 650mA. And with the external USB Charger (iSound Portable Charger) i can charge the phone in 3 hours.
2- Secondly while plugged in the iSound it shows charging "AC plugged in" this is because it can provide 650mA (which is the same as the wall outlet adapter)
Actually from what I know, and correct me if I'm wrong, even using the charger that comes with the sgs2 isn't the healthiest choice for your battery because the voltage is high to cut down on charging time. The best should be by connecting to your pc
I think it is the amps of the charger what makes a difference in charging speed
Sent from my Galaxy Nexus using XDA
Jetmantrunks said:
Actually from what I know, and correct me if I'm wrong, even using the charger that comes with the sgs2 isn't the healthiest choice for your battery because the voltage is high to cut down on charging time. The best should be by connecting to your pc
Click to expand...
Click to collapse
Mistakenly pressed the thanks button, instead of the quote lol
Anyhow see if the phone is plugged in to the device that can provide as much current from 0.1A to 2.2A (from simple devices to Iphone <- they require 2.2A) the phone automatically draws the amount of current it is made to draw which is SGS2 is 650mA which is coincidentally the same as the wall outlet adapter.
I might be wrong, so correct me if i am.
I as thinking of lower the 650Ma AC rate to 450Ma or less, to allow the phone to slowly charge up overnight, with the aim of finding a rate that acheives a 5-6 hour charge time from empty.
Any thoughts? I was just thinking it should cause less heat and perhaps extend is life accordingly?
I need help... while i am charging in car with AC mode charging, and running gps navigation, it lost charge. I could understand of it is the case in usb mode but why ac mode will also have the problem?
My charger has 2ports, 1 is 1A and another is 2.1A rating.
Please help, i am in a new country and i couldn't do long with gps losing charge. I am on DXELK1 rom
Try another charger and lead, your existing one may not be delivering the rated current. There is also a thread about the S3 deciding to reduce the charging current, with a fix to set it back to 1000mA - search here in Q&A.
I tried different car charger, still the same even those are charged at AC mode. I couldn't find the thread abt any tweak, any hints on keywords?
Probably GPS is draining more than can be supplied, so grab a kernel that has charging limit override and up the max to 1200mA
Usually due to battery temperature. Once it exceeds 45-ish degrees due to sunlight/charging/ambient and so on, the charge rate is reduced considerably to reduce the chance of battery degradation/damage. In other words, it is a safety feature that should never be overridden.
I want to make a bike charger for my S3. I made sure the converter is working and can deliver more than 2A of constant current at 5V.
As long as the phone does not detect the converter as a proper charger, it won't draw more than 500mA which is not enough to maintain charge while navigating or using apps. I have tried to measure out the charger that came with the phone and found out that the data lines are shorted. Apart from this, there seems to be no connection from the data lines to the supply voltage. I tried to replicate this on my selfmade charger but still the phone won't draw more than 500mA. I tried searching the net for info on the charging circuit but can't seem to get anywhere. I would like avoid cracking open the original charger.
Does anyone here have an idea what has to be done so the phone will draw more current from a homemade charging device?
you can always get a Kernel with charging control and up the amperage, but that's more of a workaround
There's a reason why i don't like workarounds like this. It violates usb specifications and may lead to damage to a regular usb port if handled improperly. It may be possible that i am missing something. Maybe i'll get another original charger and dig deeper into this matter. I'll post results as soon as available
Noume said:
I want to make a bike charger for my S3. I made sure the converter is working and can deliver more than 2A of constant current at 5V.
As long as the phone does not detect the converter as a proper charger, it won't draw more than 500mA which is not enough to maintain charge while navigating or using apps. I have tried to measure out the charger that came with the phone and found out that the data lines are shorted. Apart from this, there seems to be no connection from the data lines to the supply voltage. I tried to replicate this on my selfmade charger but still the phone won't draw more than 500mA. I tried searching the net for info on the charging circuit but can't seem to get anywhere. I would like avoid cracking open the original charger.
Does anyone here have an idea what has to be done so the phone will draw more current from a homemade charging device?
Click to expand...
Click to collapse
The key in shorting out the data lines to achieve FAST charge rate is that the Phone sees it as shorted at the Micro Usb connection point of the phone. Anywhere else down the line is irrelevant. Depending on the Kernel, most phones slow down the charging rate after...hmmm...anywhere between 70-100%. The closer to full charge...the slower the charge rate gets...until it becomes a trickle charge...just in case you don't know.
If your still having issues, please post back with a more detailed explanation of your wiring and parts as best as you can and I will help you with it.:good:
[Solved] Charger for S3
Considering your suggestions i was asking myself if i've done everything right and as it turned out, i hadn't.
I do know that chargers current depends largely on the charging state of the battery. Good chargers start initiate completely drained batteries with a low current, which is ramped up to the max at around 5-10%. After reaching 85-90% the current drops down to small trickle pulses as the charging finishes. These values can be adjusted within the firmware/kernel to get a longer battery life/more power. That's why i always do charging tests with a battery that is at least down to 33%.
But back to topic now
Using an oscilloscope i found my switching regulator to have about 150mV ripple voltage which apparently is a bit more than the S3 is willing to take. After sorting it out, ie. resoldering a cold joint, it finally works as expected, the ripple voltage is now around 32mV. The regulator could deliver more than 2A before, but only with about 600mV ripple, which i unfortunately overlooked
Charging the phone's completely drained battery, i get around 320-340mA current draw from the battery(12Vnom LiFePO4) .
The phone takes around 950mA at 5.1V. I wonder why earlier charging circuits had to be so "complicated" like with Apple phones or Galaxy Tab, while most newer phones are happy with just shorted data lines.
The circuit i use is composed of a three-phase voltage doubling rectifier, that takes bike from a generator hub in the front wheel. This one is connected to a battery for storage and buffering. A small switching regulator provides 6V 600mA for the lighting system. A larger LM2596-based regulator (you can get those modules around $1.50) provides power for my S3 and a microcontroller that will later be used for monitoring the systems voltages and rotational speed of the front wheel. EAGLE layout files for circuit and PCB will be finished soon.