I'm trying to repair a dead Samsung Grand Neo (GT-i9060) which was given to me by my ex-employer to manage their club. I have some important contacts/WhatsApp data and I was stupid enough to not back it up. I tried getting it fixed at a couple of places but no one seems to be able to get it working or apparently isn't interested enough (common reasons being that parts aren't available or its a fried motherboard)
In light of this, I have decided to troubleshoot and fix it myself. Apart from the satisfaction, I feel it's a good opportunity for me to learn how these things work even if I'm a novice in electronics and micro-circuitry. I tried asking for help online on a few other forums but none of it has been of great help so far. Stumbling upon NF's videos on YouTube has been inspirational and informative and I feel this would be the right place to ask for help 😇
Samsung GT-i9060
•No boot, no charging, no port
•Battery not in optimum condition; holds around 4.14V when charged but discharges rather quickly.
•On pressing power button, phone doesn't turn on but area around the processor (also opposite side of motherboard) gets really hot.
Kindly excuse the long post but I really am looking hopeful to get this phone working again.
Cheers
Hi Rizzi87,
Step 1: Do a visual inspection - check for hints of liquid damage or corrosion. If anything is found make note of areas and clean with alhocol and a brush.
Step 2: Use a multimeter in diode mode to check between the positive and negative of the battery connector on the baord (make note of correct orientation of the leads). If the voltage drop is low then you definitely have a short and needs to be investigated. If not low then proceed to step 3.
Step 3: Hook the device up to a bench power supply at 3.8-4v and see if it is drawing current in the off state.
If it is drawing current when in the off postition then it's time to look at what is actually getting hot and tracing the short.
If it is not then try to power on and see if it will boot.
Although these phones are old, I have and do repair much older (symbian based as well as early android devices).
Based on what you are saying I would take a look at the pmic (usually found on the opposite side of the board to the CPU).
Thank you for your reply. Your suggestions toward approaching this repair is really helpful.
I have dared to disassemble the phone already and inspect the motherboard. Besides some whitish stains on some areas (according to me they dont seem important), I haven't found any heavy liquid damage or corrosion. Initially, I didn't have IPA so I'd used petrol to lightly clean the board. I'm yet to inspect the components under the emi shields but I doubt I would find anything unusual there.
As for the readings in diode mode, this is what I got:
- 1.462 (don't know what bias/orientation it is called but the red probe was on positive battery connector and black probe on negative battery connector or ground)
- .349 (black probe on positive and red probe on ground)
Apart from this, I have checked all the lines for VMBAT & VBUS using the schematic diagrams. Voltages were found in the normal expected range. I tested this with the battery connected and the charger plugged in but did not switch the phone on.
Prior to the above, I had also checked the visible caps and resistors on the board in continuity mode on the multimeter and found 3 capacitors which do not show continuity on either end (that is, the meter did not beep when check on either end). I have attached a pic for reference as to what I observed.
I currently have not invested in a power supply unit, hot-air station, microscope and a few other tools so I cannot perform any tests using voltage injection.
I will reply fully later on, but have you ran through the boot sequence given in the service manual? Making sure you have all the correct voltages when switching on? If you had a bench power supply you could monitor the current draw and see for sure if it's booting or not.
I will reply fully later on, but have you ran through the boot sequence given in the service manual? Making sure you have all the correct voltages when switching on? If you had a bench power supply you could monitor the current draw and see for sure if it's booting or not.
I couldn't fully test the boot sequence according to the manual since it asks to measure the frequency of an oscillator and I don't have any equipment to do that. Besides, it also mentions voltage checks on a few capacitors after the phone is switched on which is difficult for me since the battery needs to be connected and the said components are either on the underside of the motherboard or under the emi shields.
I tried hooking up the battery to the motherboard with the latter removed from the phone body using a rather flimsy arrangement but powering it on made the board hot enough to burn my finger and I had to back out fearing it might cause additional damage.
I just want to add that from the schematics, I see that the seemingly 'bad' caps I have found are connected to the microphone; could that be the cause of the problem?
Once again, thanks for your reply. This is an amazing community and I'm glad to partake in these insightful conversations <insert happy face with thumbs up smiley>
Update: I tried measuring the voltages again with the battery connected and phone switched on(using the same flimsy arrangement!). Please refer the attached image and advise.
@rizzi87
I am sure you have checked all of these voltages, but I just wanted to check (please see attached picture)
If it is getting hot around those capacitors you have mentioned in your last post then I would go to resistance mode on the multimeter and measure the resistance in that area.
@matrs
Do you want me to measure the voltages again across the caps that you have mentioned in your post? Because the ones that I measured didn't last long, i.e. the meter read 0v across them after some time. Probably because the phone may have shut itself down by then. The processor/emmc were still a bit more than warm to the touch even in that condition though.
When you say to measure resistance, should I do that across the caps or relative to ground? And which caps/components do I measure?
Kindly excuse me if that is a lot of burning(pun intended) questions.
@rizzi87
If you are messing the caps for resistance then you will need to measure across the component itself. A normal cap should read OL/L/1 (depending on meter). If you are getting any low reading then the cap has shorted and will need to be removed/replaced.
Funnily enough I am currently working on an i9300 with the same symptoms as your device. I will let you know what I find on this. As it showed itself to be the max chip then I found a couple of shorted caps around the emmc and that has resolved my issue
@matrs
I don't have an auto-ranging multimeter and the diode/continuity settings on it are also different as opposed to a single setting in some meters. What mode should I use to measure resistance? I might have answered this question myself as I found the reading to be OL for all the caps* in continuity mode and I get a reading if I set the meter in kilo-ohms or mega-ohms range.
Which brought me to another predicament. What should be orientation of the probes be to measure resistance? Sorry if this is a noob question.
*I only measured the resistance of the capacitors which were mentioned for voltage checks in the service manual and the ones which I had deemed faulty.
@rizzi87
The testing of continuity doesn't really matter for most circumstances. The voltage used for continuity is not high enough to forward bias any silicon/germanium/other differential junctions so if you can read anything on them (in the lowest ohm range) then you know it's that component completing the circuit.
As you haven't found any shorts on the caps you have tested and thought it might be; I would start testing caps around the areas that have gotten hot. But it might be a short that's gone to the CPU which wouldn't be an easy fix as they are usually stacked with lpddr and CPU in one underfilled tower.
If you are competent at running fly wires you can do a direct dump of the emmc and then either decode it or dump it onto another working i9060
@rizzi87
The orientation of the leads for testing of continuity doesn't really matter for most circumstances. The voltage used for continuity is not high enough to forward bias any silicon/germanium/other differential junctions so if you can read anything on them (in the lowest ohm range) then you know it's that component completing the circuit.
As you haven't found any shorts on the caps you have tested and thought it might be; I would start testing caps around the areas that have gotten hot. But it might be a short that's gone to the CPU which wouldn't be an easy fix as they are usually stacked with lpddr and CPU in one underfilled tower.
If you are competent at running fly wires you can do a direct dump of the emmc and then either decode it or dump it onto another working i9060
@rizzi87
The testing of continuity doesn't really matter for most circumstances. The voltage used for continuity is not high enough to forward bias any silicon/germanium/other differential junctions so if you can read anything on them (in the lowest ohm range) then you know it's that component completing the circuit.
So, any component, or namely, a cap giving out a reading in continuity mode translates to it having low resistance and therefore is faulty. Am I right?
As you haven't found any shorts on the caps you have tested and thought it might be; I would start testing caps around the areas that have gotten hot. But it might be a short that's gone to the CPU which wouldn't be an easy fix as they are usually stacked with lpddr and CPU in one underfilled tower.
What about the caps that aren't showing any continuity relative to ground? These are ones that I thought may have gone kaput. In diode mode, these buggers also exhibit readings on both sides.
If you are competent at running fly wires you can do a direct dump of the emmc and then either decode it or dump it onto another working i9060
I wouldn't say that I'm competent enough to carry out microsoldering, BGA or chip repair tasks but can definitely give it a go as a last resort. With practice, of course!
As a means to further this curiosity, I have also undertaken repair of a Samsung A10, a Galaxy S7 and an iPhone 5S all of which were just lying around at home, ignored just because they had stopped working.
@rizzi87
So, any component, or namely, a cap giving out a reading in continuity mode translates to it having low resistance and therefore is faulty. Am I right?
I will be talking about capacitors here:
As with any rule there are exceptions. On a normal day if a capacitor reads the same value as you touching your probes together then it is dead/at least on a line that is going to ground. The exception is when the capacitor is connected to ground on both sides (EMI reduction), or isn't connected to ground directly at all (DC removal from microphonesignal, using it as a "battery")
The rule of thumb is if it runs perpendicular to the chip then it will have one end on ground and one on a power/data (anything with some sort of resistance). If it runs parallel to the chip then it will be tied to ground on both ends. The best way to see this is to see the traces that come from under the chip in question as not all components surrounding the chip in question actually relates to the chip itself.
As in one of Alex's many great videos, you will normally find that resistors are used anywhere but ground. Again, there are many exceptions, but they do not tie directly to ground.
Chokes/inductors are the same where they will have a low resistance across them but will not normally be tied directly to earth/ground.
What about the caps that aren't showing any continuity relative to ground? These are ones that I thought may have gone kaput. In diode mode, these buggers also exhibit readings on both sides
If I remember correctly they were linked to the microphone? If so, they would be there to remove the DC power from your voice/sound as sound is an AC wave but you need voltage to carry the wave on as you'd be looking at mV or less for a microphone input.
I wouldn't say that I'm competent enough to carry out microsoldering, BGA or chip repair tasks but can definitely give it a go as a last resort. With practice, of course!
As a means to further this curiosity, I have also undertaken repair of a Samsung A10, a Galaxy S7 and an iPhone 5S all of which were just lying around at home, ignored just because they had stopped working.
Try the obvious charge/battery change on those first. If you are semi serious about this sort of stuff I would invest in a decent bench tester. The current pull at various stages will become invaluable for your diagnostic skills.
I have a pin out diagram for where to get your various voltages and Data lines from for that device if you need to just pull the data off.
If you have a spare 30 mins or so. Put your multimeter in continuity mode, touch the ends of your leads togehter and note the value. Go over every capacitor on the board and note down and if you can photograph the ones that are reading the same value as previously noted.
Please don't think I have ignored you. I have replied, the message is still awaiting moderation.