Dead iPhone 4 - GSM-Forum

jegerjon · 08-07-2013, 05:52

Hello guys.

I hope that someone can help me here.

I have two iPhone 4 on my table, with the exact same problem. They are both dead after heavily water damage. I have managed to clean up the boards very good, replaced several components and made everything shine.

There is no shorts around the PMIC.

When I press the ON button short (like 0.5sec), all the voltages related to the CPU rises, but does not come back again! All voltages are fine:

PP1V35: 1.115V
PP1V8: 1.815V
PP1V2: 1.2V
PP3V0_IO: 3.0V
PP1V8_SDRAM: 1.815V
PP1V1: 1.108V

I have compared these voltages with a working board, and they are all correct. The problem is that they never fall back to 0V, and they should! (I can see this on a working board).

So what could be my problem?! Burned PMIC? Corroded balls under the CPU?
What makes the PMIC lower its buck voltages again? Does it need some kind of signal from the CPU? I can not find this in schematic.

Hope you guys can help me! Would really appreciate it.
Cheers.

jegerjon · 08-07-2013, 06:10

How does the PMIC and the CPU communicate?

I see that we have a RESET_IN line. This seems to be controlled by the CPU. I removed R93 on a working board, and things are fine without this line. So this is not the problem

Does anyone know what happens if you removed the CPU? Would the PMIC still generate the buck voltages when the ON button is pressed?
If so, then my problem clearly is the PMIC!
Maybe it is burned from shortage of voltage lines?

I will try and change the PMIC tomorrow. It is very late in Denmark now. 7AM

I need some sleep.

kostelectronics · 08-07-2013, 08:20

Hey,

good work so far. If the voltages stay and doesn't go back to 0 after you press the power-button for a very short time, then is it not so an easy Problem.

At first, i don't think it's the PMIC. (Expierience)! Did you heat the board before you tested all the voltages? The PMIC get's often loose after heating the board.

If the Voltages doesn't go back to 0, there is a data communication-problem between the CPU, NAND, PMIC. I give you a short list what all can cause this Problem:

Caused by Water-Damage without heating anything before:
- blocked i2c-bus (CPU, PMIC or somewhere else). this can be a corroded pull-up resistor or a testpoint is corroded with another track.
- Missing voltages at the NAND or CPU
- One Output from the PMIC is shorted.
- Corrosion inside the CPU

Caused by heating the board too much:
- CPU is loosen or shorted somewhere.
- RAM is loosen or shorted somewhere.
- NAND is loosen or shorted somewhere.
- PMIC is loosen or shorted somehwere.

First, remove the soldered metal-shield to Access all parts and check for more corrosion. (Take care and don't heat too much)!!

good luck, this is hard bread!

BR Matthias

jegerjon · 08-07-2013, 15:54

Matthias, thank you very much, your are the man!

The board has not been heated, at least not by me.
I have already removed the EMI shield covering the NAND and PMIC, but I did this without heat gun (taught by earlier experience with dead board

).

Quote:

Originally Posted by kostelectronics

- Missing voltages at the NAND or CPU

All CPU voltage lines are good, and the PMIC power these up. I will check the supply voltages for the NAND.

Quote:

Originally Posted by kostelectronics

- One Output from the PMIC is shorted.

This i have checked, and everything seems fine.

Quote:

Originally Posted by kostelectronics

- Corrosion inside the CPU

This is what i fear

It will be last resort, but i hate doing CPU reballs, it is so time consuming!

Quote:

Originally Posted by kostelectronics

- blocked i2c-bus (CPU, PMIC or somewhere else). this can be a corroded pull-up resistor or a testpoint is corroded with another track.

I will now go over all resistors surounding the CPU, and resolder/change everything that looks the least suspicious. I hope for this solution

Thank you so much Mathias, excellent reply. I know you are very skilled in this area, so I defiantly listen to you.

Good day, and I will let you know when I have news!
Cheers.

kostelectronics · 08-07-2013, 16:40

Thank you

Then i whould try anyways to reball the PMIC. Maybe the phone was dropped before landed in water. It doesn't take much time

jegerjon · 08-07-2013, 18:03

Quote:

Originally Posted by kostelectronics

Thank you

Then i whould try anyways to reball the PMIC. Maybe the phone was dropped before landed in water. It doesn't take much time

The first one is now alive! I am so happy about this

Thank you for pointing me in the right direction friend.

I remembered that I had an area close to U52 with VERY bad water damage. I decided to change C181, C203, C23, C19, R369, C8, R24 and C264.

I don't know which of the components that saved me. I am guessing either R17 (pullup on the RESET_3V0_L line) or R24.
Or maybe if both C181 and C203 had lost contact (and I think they had), the accelerometer couldn't start up because of unstable supply. I don't know if this would cause any problems....

Anyways, I have this one working, and that makes me happy

The second one the owner told me that the LED was on allways before the phone eventually died completely. The FPC connector for the camera was badly damaged, so could the CPU be damaged from high currents running through the data patch?

I am going to do the same routine on this one however. That is change every component that looks just the least water damaged in the CPU area. Even if they are less damaged than I would usually care for. The communication lines for the CPU are apparently very sensitive

kostelectronics · 08-07-2013, 18:24

Hey!! Congratulations!!

only to your Information: The LED is driven by the camera on the CAM_STROBE_EN line. It goes trough L7. So, if the camera connector is corroded, it's a good Chance that a leak currend driven this line and enabled the LED permanent. Also U17 (LED-Driver) dies often after water-damage. You can see this when you have about 10-50 ohms direct at the battery connector. If you remove L9 and the short is gone, you can be sure that you have to replace the U17. This is only by the way. I wish you good luck with the 2nd board

BR Matthias

jegerjon · 08-07-2013, 18:39

Ah, you are probably right about the leak current to CAM_STROBE_EN, I see how that would cause it to permanently be on!

Yes yes, I know this, but the LED chip is just fine. I replaced the FPC connector, and everything seems to be fine in regard to the camera. No shorts, and I have continuity between LED ic and FPC.

Thank you for your input anyways, I will get cracking on the second board. First thing first, lets replace some bad looking resistors

jegerjon · 08-10-2013, 19:52

Ok, so an update on my second phone, just in case someone find this thread in the future:

I changed a bunch of components, but no luck, then I reballed the PMIC, still no luck. I put away the phone, and when i came back some days later, I changed R27, R77, R419 and R424 even though I did not see ANY damage to these components (at this point I had already set my mind on an A4 reball). Guess what? After this the voltage lines drops down to zero as they are supposed to

I am glad that I didn't reball the A4, this saved me half a day!

Anyways, after this the phone hanged in DFU, but no big deal. I measured only 0.9V on the PP1V2_SDRAM line, so I changed R76, after this U7 and then the phone was back to life. Excellent!

Big thanks to you Matthias! Hit me up for a beer if you ever visit Denmark

Cheers.

08-07-2013, 05:52	#1 (permalink)
jegerjon Freak Poster Join Date: Nov 2012 Location: Denmark Posts: 172 Member: 1839857 Status: Offline Thanks Meter: 46	Dead iPhone 4 Hello guys. I hope that someone can help me here. I have two iPhone 4 on my table, with the exact same problem. They are both dead after heavily water damage. I have managed to clean up the boards very good, replaced several components and made everything shine. There is no shorts around the PMIC. When I press the ON button short (like 0.5sec), all the voltages related to the CPU rises, but does not come back again! All voltages are fine: PP1V35: 1.115V PP1V8: 1.815V PP1V2: 1.2V PP3V0_IO: 3.0V PP1V8_SDRAM: 1.815V PP1V1: 1.108V I have compared these voltages with a working board, and they are all correct. The problem is that they never fall back to 0V, and they should! (I can see this on a working board). So what could be my problem?! Burned PMIC? Corroded balls under the CPU? What makes the PMIC lower its buck voltages again? Does it need some kind of signal from the CPU? I can not find this in schematic. Hope you guys can help me! Would really appreciate it. Cheers.

08-07-2013, 06:10	#2 (permalink)
jegerjon Freak Poster Join Date: Nov 2012 Location: Denmark Posts: 172 Member: 1839857 Status: Offline Thanks Meter: 46	How does the PMIC and the CPU communicate? I see that we have a RESET_IN line. This seems to be controlled by the CPU. I removed R93 on a working board, and things are fine without this line. So this is not the problem Does anyone know what happens if you removed the CPU? Would the PMIC still generate the buck voltages when the ON button is pressed? If so, then my problem clearly is the PMIC! Maybe it is burned from shortage of voltage lines? I will try and change the PMIC tomorrow. It is very late in Denmark now. 7AM I need some sleep.

08-07-2013, 08:20	#3 (permalink)
kostelectronics No Life Poster Join Date: Apr 2009 Location: Wikon, Switzerland Posts: 941 Member: 1012778 Status: Offline Thanks Meter: 245	Hey, good work so far. If the voltages stay and doesn't go back to 0 after you press the power-button for a very short time, then is it not so an easy Problem. At first, i don't think it's the PMIC. (Expierience)! Did you heat the board before you tested all the voltages? The PMIC get's often loose after heating the board. If the Voltages doesn't go back to 0, there is a data communication-problem between the CPU, NAND, PMIC. I give you a short list what all can cause this Problem: Caused by Water-Damage without heating anything before: - blocked i2c-bus (CPU, PMIC or somewhere else). this can be a corroded pull-up resistor or a testpoint is corroded with another track. - Missing voltages at the NAND or CPU - One Output from the PMIC is shorted. - Corrosion inside the CPU Caused by heating the board too much: - CPU is loosen or shorted somewhere. - RAM is loosen or shorted somewhere. - NAND is loosen or shorted somewhere. - PMIC is loosen or shorted somehwere. First, remove the soldered metal-shield to Access all parts and check for more corrosion. (Take care and don't heat too much)!! good luck, this is hard bread! BR Matthias

08-07-2013, 16:40	#5 (permalink)
kostelectronics No Life Poster Join Date: Apr 2009 Location: Wikon, Switzerland Posts: 941 Member: 1012778 Status: Offline Thanks Meter: 245	Thank you Then i whould try anyways to reball the PMIC. Maybe the phone was dropped before landed in water. It doesn't take much time

08-07-2013, 18:24	#7 (permalink)
kostelectronics No Life Poster Join Date: Apr 2009 Location: Wikon, Switzerland Posts: 941 Member: 1012778 Status: Offline Thanks Meter: 245	Hey!! Congratulations!! only to your Information: The LED is driven by the camera on the CAM_STROBE_EN line. It goes trough L7. So, if the camera connector is corroded, it's a good Chance that a leak currend driven this line and enabled the LED permanent. Also U17 (LED-Driver) dies often after water-damage. You can see this when you have about 10-50 ohms direct at the battery connector. If you remove L9 and the short is gone, you can be sure that you have to replace the U17. This is only by the way. I wish you good luck with the 2nd board BR Matthias

08-07-2013, 18:39	#8 (permalink)
jegerjon Freak Poster Join Date: Nov 2012 Location: Denmark Posts: 172 Member: 1839857 Status: Offline Thanks Meter: 46	Ah, you are probably right about the leak current to CAM_STROBE_EN, I see how that would cause it to permanently be on! Yes yes, I know this, but the LED chip is just fine. I replaced the FPC connector, and everything seems to be fine in regard to the camera. No shorts, and I have continuity between LED ic and FPC. Thank you for your input anyways, I will get cracking on the second board. First thing first, lets replace some bad looking resistors

08-10-2013, 19:52	#9 (permalink)
jegerjon Freak Poster Join Date: Nov 2012 Location: Denmark Posts: 172 Member: 1839857 Status: Offline Thanks Meter: 46	Ok, so an update on my second phone, just in case someone find this thread in the future: I changed a bunch of components, but no luck, then I reballed the PMIC, still no luck. I put away the phone, and when i came back some days later, I changed R27, R77, R419 and R424 even though I did not see ANY damage to these components (at this point I had already set my mind on an A4 reball). Guess what? After this the voltage lines drops down to zero as they are supposed to I am glad that I didn't reball the A4, this saved me half a day! Anyways, after this the phone hanged in DFU, but no big deal. I measured only 0.9V on the PP1V2_SDRAM line, so I changed R76, after this U7 and then the phone was back to life. Excellent! Big thanks to you Matthias! Hit me up for a beer if you ever visit Denmark Cheers.