NFC battery - AT&T Samsung Galaxy S II SGH-I777

just curious since this has the NFC battery compared to the international SGS2 do we AT&T have to use battery's with NFC !
Also what if the international phone puts in an NFC battery will they have that built in automatically

---

Just inserting the ATT battery will not give anyone the powers of NFC, if thats what you are asking. Apparently someone found four contact tips on the ATT version instead of three in the international, so that the NFC can talk to the rest of the phone. So there needs to be mechanisms like that in place. So, in short no they won't get NFC by putting in your battery.
But you can insert a replacement battery without NFC and it will just work fine but you wont have NFC (if ATT flips the software switch on, that is). Doesn't mean I encourage you to pop in any battery and test.
Experts can correct me on this.

---

thelastjedi said:
Just inserting the ATT battery will not give anyone the powers of NFC, if thats what you are asking. Apparently someone found four contact tips on the ATT version instead of three in the international, so that the NFC can talk to the rest of the phone. So there needs to be mechanisms like that in place. So, in short no they won't get NFC by putting in your battery.
But you can insert a replacement battery without NFC and it will just work fine but you wont have NFC (if ATT flips the software switch on, that is). Doesn't mean I encourage you to pop in any battery and test.
Experts can correct me on this.
Click to expand...
Click to collapse
Actually, no one confirmed the 3 vs 4. I know Infuse had 3 - I don't know what international has. I asked and no one provided the information.
Also international batteries will NOT work fine. They flake out as soon as you connect a charger or USB. connexion2005 spent hours trying to debug constant reboot loops and it turned out to be due to using an I9100 battery.

---

I'm glad I asked because I bought 2 batteries 1 is a off Chinese brand i9199 and the other is a original samsung non NFC battery , I was going to buy the international galaxy but changed my mine and bought the AT&T phone
So your saying the off brand i9100 battery is no good ? just wondering how a battery could cause problems when it's only pushing juice,
I read about the NFC battery someone mentioned it has a chip inside so I guess that isn't true ! so it actually has a extra prong and so does the AT&T phone right!
I wasn't planning to use the i9100 brand battery all the time and charge it in a charger thought I would switch them out when out for the day , So would just using that battery cause problems ?

---

Entropy512 said:
Actually, no one confirmed the 3 vs 4. I know Infuse had 3 - I don't know what international has. I asked and no one provided the information.
Also international batteries will NOT work fine. They flake out as soon as you connect a charger or USB. connexion2005 spent hours trying to debug constant reboot loops and it turned out to be due to using an I9100 battery.
Click to expand...
Click to collapse
In case you still want the info Entropy, I just checked with my roommate's I9100 (I have an i777), and it also has 4 contact tips

---

justabrake said:
So your saying the off brand i9100 battery is no good ? just wondering how a battery could cause problems when it's only pushing juice,
Click to expand...
Click to collapse
The battery has an ID resistor (I think it's the pin between + and -) - if it's the wrong value, the charge circuitry will force a shutdown/reboot when charging.
Actually it's the kernel - but until I'm positive that there is NO risk resulting from removing this protection mechanism, I'm not doing it.

---

The 4th contact is for a antenna in the battery.

---

Entropy512 said:
The battery has an ID resistor (I think it's the pin between + and -) - if it's the wrong value, the charge circuitry will force a shutdown/reboot when charging.
Actually it's the kernel - but until I'm positive that there is NO risk resulting from removing this protection mechanism, I'm not doing it.
Click to expand...
Click to collapse
This is not the first time Samsung did this. The original Galaxy S batteries are the same. You can't use the Verizon version's battery with rest of Galaxy S lines (AT&T, T-Mo and Sprint). If you do, you get the reboot.
agh1701 said:
The 4th contact is for a antenna in the battery.
Click to expand...
Click to collapse
After seeing the information from Verizon about about Galaxy Nexus battery, it seems the NFC chip, not just the antenna, is inside the battery.

---

Entropy512 said:
The battery has an ID resistor (I think it's the pin between + and -) - if it's the wrong value, the charge circuitry will force a shutdown/reboot when charging.
Actually it's the kernel - but until I'm positive that there is NO risk resulting from removing this protection mechanism, I'm not doing it.
Click to expand...
Click to collapse
I'm willing to try it if you are willing to create a special kernel.
I'm currently in Hong Kong and bought an i777 from one of the local small merchants. The phone worked in the store, but did not charge when I got it home. The battery does not say "Near Field Communications" on it, even though it looks like an authentic Samsung battery. So I am assuming that I have a genuine i777 (it says AT&T on the top and has the four buttons on the bottom), and a genuine I9100 battery. I am currently waiting for an NFC battery that I bought on Ebay. I also bought a local noname battery so that I can charge one while using the other. Both batteries have 4 terminals.
If anyone knows what the value of the ID resistor is, I'm also willing to try that route to get the phone to charge.
The phone came with MIUI 10/28 and everything else seems to work, including GPS, WiFi, etc. MIUI 12/2, 12/9, 12/16 (not US version) fix the charging problem, but then the phone audio doesn't work. I tried flashing all the i777 modems, but none fixed the audio problem. I was not able to flash the CM7 converted modems because CWM gave some error messages about "I777" (I think it said failed assertion).
I tried flashing Unnamed, Cognition and CM7, but they don't fix the battery problem, but the phone audio and everything else works.
I am a total noob so if I've used any incorrect terms, please forgive my noobness. The last two weeks have been quite an adventure, searching forums, learning about rooting, recovery, downloading ROMs and kernels and modems, etc. Actually, its been a lot of fun.

---

guilin said:
....
I also bought a local noname battery so that I can charge one while using the other. Both batteries have 4 terminals.
If anyone knows what the value of the ID resistor is, I'm also willing to try that route to get the phone to charge.
Click to expand...
Click to collapse
Most of these batteries will have four contacts, two for power and two for communication to the gas Guage within the battery pack (I2C interface). The few batteries out there with three contacts use a single wire interface, but they are essentially the same - there is a communication path into the cell to talk to the fuel Guage embedded in the pack. The nfc enabled batteries either extend the capabilities of the fuel guage or add an additional part, but in either case, the two functions share the same communication paths - there are no "resistors" to check or change, the ic(s) identify what capabilities they have to the host via the serial interface.
T
Sent from my SAMSUNG-SGH-I777 using XDA App

---

guilin said:
I'm willing to try it if you are willing to create a special kernel.
I'm currently in Hong Kong and bought an i777 from one of the local small merchants. The phone worked in the store, but did not charge when I got it home. The battery does not say "Near Field Communications" on it, even though it looks like an authentic Samsung battery. So I am assuming that I have a genuine i777 (it says AT&T on the top and has the four buttons on the bottom), and a genuine I9100 battery. I am currently waiting for an NFC battery that I bought on Ebay. I also bought a local noname battery so that I can charge one while using the other. Both batteries have 4 terminals.
If anyone knows what the value of the ID resistor is, I'm also willing to try that route to get the phone to charge.
The phone came with MIUI 10/28 and everything else seems to work, including GPS, WiFi, etc. MIUI 12/2, 12/9, 12/16 (not US version) fix the charging problem, but then the phone audio doesn't work. I tried flashing all the i777 modems, but none fixed the audio problem. I was not able to flash the CM7 converted modems because CWM gave some error messages about "I777" (I think it said failed assertion).
I tried flashing Unnamed, Cognition and CM7, but they don't fix the battery problem, but the phone audio and everything else works.
I am a total noob so if I've used any incorrect terms, please forgive my noobness. The last two weeks have been quite an adventure, searching forums, learning about rooting, recovery, downloading ROMs and kernels and modems, etc. Actually, its been a lot of fun.
Click to expand...
Click to collapse
The ID resistor is buried deep within the battery, on the same board as its protection circuit.
Take the unit back to the store and explain that they did not give you a functional I777 - if this merchant were legitimate you should have received a properly matching battery with the unit.
tedkunich said:
Most of these batteries will have four contacts, two for power and two for communication to the gas Guage within the battery pack (I2C interface). The few batteries out there with three contacts use a single wire interface, but they are essentially the same - there is a communication path into the cell to talk to the fuel Guage embedded in the pack. The nfc enabled batteries either extend the capabilities of the fuel guage or add an additional part, but in either case, the two functions share the same communication paths - there are no "resistors" to check or change, the ic(s) identify what capabilities they have to the host via the serial interface.
Click to expand...
Click to collapse
WRONG.
Have you read the kernel source code for the fuel gauge driver? Probably not. I have. I've even fixed bugs in it.
Have you read the datasheet for the fuel gauge chipset's predecessor? (The current one does not have a publically available datasheet, but it's VERY close to the previous version used in the original GalaxyS series of phones) Can you even name the fuel gauge chipset without Googling?
Have you used an ohmmeter to measure the resistance of the ID pin? I have.
The fuel gauge chipset is in the phone itself, NOT in the battery for this device.
The phone checks for a legitimate battery by connecting a constant-current source (looks like around 0.3 mA) to the battery contact between the + and - terminals. It then measures the voltage of this terminal to see what the value is when the device is on AC or USB power, and powers off if it is outside of certain ranges. See sec_bat_check_vf_adc in drivers/power/sec_battery.c
I777 batteries are 4.7k ohms, and the device looks for a range between 1.1 and 1.8 volts on the vf terminal. I9100 update3 makes this stricter - 1.35 to 1.5 volts. I9100 batteries appear to have a 6k resistor approximately, the Vf check looks for a voltage between 2.0 and 2.1 volts.
I'm not sure what is on the terminal adjacent to the - terminal on the other side - this is likely NFC-related, it is not a bare antenna (extremely high resistance when powered off)
I've seen evidence that the NFC solution is distributed between the battery and the device - I9100 kernels have PN544 support, and at least one I9100 user has managed to get a successful PN544 firmware load. However without the battery side of things, their device doesn't work.

---

There is a thread in the "accessories" section that is devoted to batteries. Some seem to work well while others cause the bootloop.

---

Entropy512 said:
WRONG.
Have you read the kernel source code for the fuel gauge driver? Probably not. I have. I've even fixed bugs in it.
Have you read the datasheet for the fuel gauge chipset's predecessor? (The current one does not have a publically available datasheet, but it's VERY close to the previous version used in the original GalaxyS series of phones) Can you even name the fuel gauge chipset without Googling?
Have you used an ohmmeter to measure the resistance of the ID pin? I have.
The fuel gauge chipset is in the phone itself, NOT in the battery for this device.
The phone checks for a legitimate battery by connecting a constant-current source (looks like around 0.3 mA) to the battery contact between the + and - terminals. It then measures the voltage of this terminal to see what the value is when the device is on AC or USB power, and powers off if it is outside of certain ranges. See sec_bat_check_vf_adc in drivers/power/sec_battery.c
I777 batteries are 4.7k ohms, and the device looks for a range between 1.1 and 1.8 volts on the vf terminal. I9100 update3 makes this stricter - 1.35 to 1.5 volts. I9100 batteries appear to have a 6k resistor approximately, the Vf check looks for a voltage between 2.0 and 2.1 volts.
I'm not sure what is on the terminal adjacent to the - terminal on the other side - this is likely NFC-related, it is not a bare antenna (extremely high resistance when powered off)
I've seen evidence that the NFC solution is distributed between the battery and the device - I9100 kernels have PN544 support, and at least one I9100 user has managed to get a successful PN544 firmware load. However without the battery side of things, their device doesn't work.
Click to expand...
Click to collapse
Have not personally looked at the chipset that SAMSUNG uses, but the Texas Instrument line are in-pack gas gages with either I2C or hdq interfaces. The bq245xx series does both flavors of interface.
Are you an electrical engineer? I am, and have been designing this stuff for years - your 4.7k resistor is a pull up on the communication lines. Standard value per the I2C standard.
I have not seen too many designs where the gaging is outside of the pack, it it makes tracking pack changes and giving reasonably accurate results difficult if not impossible. Now, I have no insite into how SAMSUNG designed the hardware, but reading the source code will not give the full picture either. Fuel gaging is a dual discipline function - there is dedicated hardware to measure and track energy going into and out of the cell, and software must controll and track the data.
Peace,
T
Sent from my SAMSUNG-SGH-I777 using XDA App

---

tedkunich said:
Have not personally looked at the chipset that SAMSUNG uses, but the Texas Instrument line are in-pack gas gages with either I2C or hdq interfaces. The bq245xx series does both flavors of interface.
Are you an electrical engineer? I am, and have been designing this stuff for years - your 4.7k resistor is a pull up on the communication lines. Standard value per the I2C standard.
I have not seen too many designs where the gaging is outside of the pack, it it makes tracking pack changes and giving reasonably accurate results difficult if not impossible. Now, I have no insite into how SAMSUNG designed the hardware, but reading the source code will not give the full picture either. Fuel gaging is a dual discipline function - there is dedicated hardware to measure and track energy going into and out of the cell, and software must controll and track the data.
Peace,
T
Sent from my SAMSUNG-SGH-I777 using XDA App
Click to expand...
Click to collapse
Yes, I am an electrical engineer.
And you clearly have not read the MAX17040 datasheet. Otherwise you would know that Maxim designs their fuel gauges not to depend on a coulomb counter function (current sensor) - which is why it does not require any calibration. Yes, it gives up some accuracy because of that, but the advantage is that the pack has no fuel gauge chip itself. We have a MAX17042, which is a close relative of the 17040. It does have current measurement capability, but this is not enabled in our device.
We don't actually have a MAX17042 or 17040 - we have a MAX8997 PMIC, the fuel gauge is embedded inside it but functionally equivalent to the MAX17042. Older devices have a MAX17040 equivalent built into the MAX8998 PMIC.
And tell me, how can a resistor between a terminal and GROUND be a pullUP resistor like you claim it is?
Don't you think it's a little odd that they're IDing the battery by taking ADC readings, when if that contact were for an I2C or one-wire bus, they could just embed a battery identification in whatever chip was in the battery so that it could be read digitally? Oh yeah, and I2C doesn't exactly work on a single wire like other Samsungs such as the Infuse would have to have, given that they have three-terminal batteries. How is it that they're reading an ADC that's part of the PMIC or CPU to identify the battery if that battery terminal is I2C or some sort of one-wire digital bus?

---

Entropy512 said:
Take the unit back to the store and explain that they did not give you a functional I777 - if this merchant were legitimate you should have received a properly matching battery with the unit.
Click to expand...
Click to collapse
Thanks for all the great info. We did't try to exchange the unit because when we discovered that MIUI 12/2 and later fixed the battery charging problem, we knew that the problem could be fixed/circumvented in the kernel or ROM. My wife and I are retired Electrical Engineers and once we discovered the world of rooting, flashing, adb, CWM, ROMs, kernels, modems, we were addicted. We even rooted our Kindle Fire. If the phone had worked correctly when we bought it, we wouldn't be having all this fun.
Like they say, it's the journey, not the destination.
Thanks again (I already clicked the Thanks Meter).

---

so a battery boot loop depends on kernel and/or rom?
i bought the samsung 2000 mah non-nfc international version battery for my i777 thinking it should work no problem because i didnt care about having nfc. when i plug my phone in to charge, it causes a reboot loop. i am running task650 icsdout rom with entropy's latest kernel.

---

polobreaka said:
so a battery boot loop depends on kernel and/or rom?
i bought the samsung 2000 mah non-nfc international version battery for my i777 thinking it should work no problem because i didnt care about having nfc. when i plug my phone in to charge, it causes a reboot loop. i am running task650 icsdout rom with entropy's latest kernel.
Click to expand...
Click to collapse
There is code in the kernel that checks for a proper battery.
Until I know more about what subtle differences might exist between I777 and I9100 - I'm not disabling it.
I'll look into maybe making it a "use at your own risk" optional feature.

---

Entropy512 said:
There is code in the kernel that checks for a proper battery.
Until I know more about what subtle differences might exist between I777 and I9100 - I'm not disabling it.
I'll look into maybe making it a "use at your own risk" optional feature.
Click to expand...
Click to collapse
So after reading almost all of the threads on the 2000mAh battery this one was the deepest and most informative as far as why there are problems.
From what I have gleaned in my readings, the I9100 2000mAh battery can be used in the I777 as long as it is charged externally, and you don't care about the NFC ability. Correct?
And if you plug in the charger at any time with the I9000 battery the phone will reboot and continue in a bootloop?
And perhaps in the future with a kernel change we could use the I9000 2000mAh without the bootloop issue? Correct also?

---

FireRaider said:
So after reading almost all of the threads on the 2000mAh battery this one was the deepest and most informative as far as why there are problems.
From what I have gleaned in my readings, the I9100 2000mAh battery can be used in the I777 as long as it is charged externally, and you don't care about the NFC ability. Correct?
And if you plug in the charger at any time with the I9000 battery the phone will reboot and continue in a bootloop?
And perhaps in the future with a kernel change we could use the I9000 2000mAh without the bootloop issue? Correct also?
Click to expand...
Click to collapse
Correct. (As long as you replace all instances of I9000 with I9100 - I typo that a lot myself.)