When I charge $2000 for Data Recovery - iPhone 15 Pro RAM + CPU Swap
Hi, this is Aaron with iBoard repair, and today I'll be recovering data from an iPhone 15 Pro that suffered a very severe smash. I believe this was run over by a car. And on this motherboard, what I'm seeing is a Ram short, which, if you don't know, the Ram is soldered to the top of the CPU. And this is essentially the hardest type of data recovery that we could possibly see. So let me show you what I'm working with so far.
So this phone has gone to a couple of shops before it got to me. And, they already did a little bit of work. They took off the screen, and they've already even separated this bottom motherboard away from the top one. Here's the top motherboard. And this is the one we're going to be focusing on.
It's the important one where most of the data chips are stored. All of them actually this is the Nand. This is the EEProm and this is the CPU. So the CPU has a layer at the very bottom which is the actual CPU. And most of this package is actually the Ram. And the Ram is where we're seeing a problem on this phone on the last technician looked like they were about to attempt a CPU swap before they figured out that that's not going to solve this issue.
And, they also determined this was most likely a Ram problem before they contacted me. And, let me show you how I figured out that this actually is Ram. So the first thing I'd like to do when I'm working with a phone that is giving me no power is I'll take my DC power supply probes and I will attach them to the battery connector, and I'll prompt this phone to boot, and I will observe the amperage draw I get when I do that. In this case, there's no amperage draw before prompt to boot.
That .002 is normal for my DC power supply, and if I prompt this phone to boot by shorting the power button line, it spikes up here and it will just set at this, this high amperage charge shouldn't be at point three yet. It should be at like, 0.06, and it will move to point two, and then it will eventually start moving up. But it shouldn't go straight to this point. Three six. So that will tell me that I have a short on the motherboard.
And what the last tech did, you showed me a picture of what he was working on. He used his thermal camera. And he just prompted the phone to boot. And he followed where the heat was going, and he was able to see what looked like, the ram heating up. So that's how he determined it was most likely Ram.
And I'll kind of show you if you can see my thermal camera here. So I have my thermal camera here, and I have the CPU and frame, and it won't be the clearest view for you, but you should be able to see the spike in heat. And we can see pretty instantly. This box is heating up. If I move around, there's not much heat elsewhere.
And that box is, you know, like a part of the Ram module. And that's what the last tech was able to determine as the cause of the short. I was hoping for a bit more concrete evidence than just heat, because the heat can be misleading sometimes. So I went ahead and I started measuring all the power bills around the PC, and especially the Ram power rules.
And, I had my 1V8 Ram power rail right here, and that's giving me a normal reading. I had my over six spiral right there that's giving me a normal reading. And my 1V1 power rail right here. If I didn't already suspect a ram, then this may come off as a normal reading. The .02 isn't that far off of what a normal reading should be. I think it should probably be .04.05, but the .02 is not for short.
And I guess right now is probably a good time to compare that to like a donor board and see what that's showing. If I have one. And I do. I have one here that has the Nand off of the board. And if I compare the diode mode reading, I'm not quite sure what it should be. Let's see.
.01 it's even less so. That's interesting. I don't know if this board is actually booting, but let me let me just see this while we have it here. And it won't fully boot with a Nand off the board, but it should give me, some amperage chart that shows it's close to booting.
And, yeah, this is giving me, like, a brain dead, amperage draw, which is normal. So this should have my power rails turning on. So if I were to check the voltage on these lines on my donor board, they should all be correct.
So let's see if that's true or not. Here's my one VR up here. I'm getting 1.8 there. See if I can get access to the OVA. This is over five, not over six. I forgot that they changed on this model, so I should have 0.5 on here.
0.5. And then my one v1 line should be one v1 or 1.11.05 is perfectly fine. In fact that's in other models that's called A1V6 line. So it's definitely perfectly fine. If I do that same test on the board that I'm working with. If I give it power and check these lines.
So we have the 1V8 line up here. And we do in fact get 1.8. And my over five line down here I do get 0.5.
And my 1V1 line right here I'm only getting 0.88. And that was the first bit of proof that I saw that I actually do have a Ram problem here. The next thing I checked was what will happen if I set my voltage to 1.1 and I touched that line. Before I do that. So let's say I set my voltage to 1.8 and I touched the 1.8 voltage line.
What do you think will happen? Basically nothing will happen. I will get no amperage draw. If I touch ground, it will give me a full amperage draw like that. If I touch the 1.8 side, it should be almost nothing. I do get that .03. If I do that on my good board, I wonder if I will also get .03 if or if I will get nothing. .03 so that's that's normal enough. If I set this to 1.1
and I touch the 1.1 line, this is my donor board with the no. Nand, so it should be good. I get that .01, which is, you know, almost nothing. But if I touch my data board here, I'm getting this 0.3. I'm getting a very high amperage draw.
In fact, this looks almost like the same amperage draw you get when you were prompting this phone to boot. So the fact that this is pulling a lot of amperage, tells me that it's partially short. Even though the diode motor reading is fairly fine, it's even higher than the donor board, which I think is a little bit strange. But all that evidence points to Ram definitely being bad. So I have a few ideas of what I can do with this board. Obviously the first one is to remove Ram and set a new one onto here.
Once I remove Ram, I can check to make sure that the voltage goes back to normal on the Ram line. However, that's really difficult, and time consuming, and it's definitely possible. I have a video, other videos of me doing that. But if I can avoid that, that would be much preferable.
The last time I did this, it took me close to four hours. It's a very long job. So what I can do is I can look for other 1.1 lines. I think I have one powering this ship, and I can see if that line is on and if it's on, I can try to see if this line can support that line by connecting them.
Or I can also try to set my DC power supply to 1.1. Put a wire onto there, put a wire on to ground, and then see if I can just support this voltage. And if this stays 1.1 if the phone will turn on,
by by providing that external voltage or not. So I don't know if that will work. That's going to be my first attempt, however, because, either way, that will be a lot easier than having to remove the Ram removing and replacing the Ram is going to be the last, resort isn't.
It's probably the first idea that most people would have. But if I can support this line, I think it's worth it to try. So I think I'll try to do that with my DC power supply first, and let's see what happens. So I have a wire on this read probe. And I have a wire on this black probe. And I'll be attaching them to the 1V1 line.
I do think I have an access point on this top of this board for the for the wire to attach to. So this is 1V1I think wireless chargers. What this line is that's what I was looking at beforehand so I could possibly attach to that, especially if this works and I don't want to keep it on the DC power supply.
But here's the 1V1 line. We can see. It goes into Ram right here. And this is where I was measuring from. But I do have another capacitor on the top of this board right here.
And this is going to be a lot more convenient to attach to. So I'll attach to here. And let's see what happens if I power the board while that's, providing external power while the DC power supply is, supplying external power. So if this actually works, I'll be very happy because, like I said, it's going to be quite a pain to change Ram and I do still have to worry about some of these pads. I don't know exactly which, but I do know that the top board needs to communicate with the bottom board to prevent like a reboot on it. So hopefully.
Hopefully I don't have to repair too many of them. I can just see if it will possibly be how it is, and if it doesn't, then I'll have to start fixing some of those pads. Probably all of them, because I don't know exactly which one is required. I think I'll just attach that charge port first so I don't have to do it once.
This is attached. This is more just a test to see what's going to happen. I don't I don't know if I fully believe it will work, but it's definitely worth trying. First.
I'll probably just attach the ground line to this screw hole. Here. So let's just see if it will turn on. And no luck. But that's fine.
I didn't really think that would work. I just thought it would be worth a shot to try. So, what that does mean is that I'm going to have to do a ram removal and replace this Ram. I don't have this ram already harvested, so I'll also have to harvest one of these rams from a donor board and prepare that as well. So this is going to be a doozy.
This will be a long video. It's going to take me an hour or so to remove the Ram, possibly even up to an hour and a half if I'm a little bit slow with it. Take me another half an hour to an hour to harvest the donor ram, and then, depending on how difficult it is for me to place the Ram, sometimes that takes multiple attempts. Sometimes, if I'm very lucky, I can get it on the first try. So I definitely have a few hours of work in front of me, but we can just go ahead and get started and consider this a Ram swap video, a normal Ram, something video. I guess I'm also curious what will happen if I try to force DFU mode in the state that it's in right now? It doesn't go on its own.
If I tell it to go to mode. So this does go to the FEMA road. If I prompt it to.
I'm not sure if you can see that on your screen or not. window's small, but I do have DFU mode up over here. So that's the first real, like, concrete proof of life that I have on this board. And. I should still have DFU mode in that same fashion.
Once I remove this Ram here. And like I said, I just scrape away the. Ram with the razor blade. And, I'll show you this process here. It's going to take a little bit of time. It is probably time for me to start doing this with a, a CNC, but I don't have one.
And, now, this doesn't take me too long. I really like that first cut. Actually, that's, it looks like it's going to break away pretty easily in this case. I'm right. There is the depth that I'm looking for.
I'm a little bit pleased to see that this actually is scraping away easier than some of the previous models I've worked on. I'll be changing a lot of razor blades. They just don't cut very well. Even after just a couple of cuts on here. So. All right.
Getting that first layer off is usually the hardest part. So I'm glad to see that it didn't take too long. And now I just need to get everything to this layer here that we can see.
Right from data recovery. Okay, I'm getting there. Just a little bit more cleaning, and then I'll do a little bit of testing before I come back and just completely flatten this all the way to make it good enough right now that I can make sure that, like, my voltage should be normal on my 1V1 line at this point. My amperage draw when I prompt a Abou should be a looping like .12 or something like that, a looping amperage
draw, and I should be able to force it into DFU mode. Still. So if all those things are still true, then everything is going according to plan.
Okay. This should be good enough. So this isn't all the way cleaned yet. I'll have to make sure this is as flat as possible. But at this point, I can do some testing.
So first thing I can test is what my DC power supply amperage draw looks like at this point. Like I mentioned a moment ago, I should get a looping current draw and that would indicate normal behavior with Ram off of the board. Should go up to like point one to or something like that, and then just go right back to zero once I prompted to boot. So I prompted the boot by shorting the power button line.
It's a little interesting, and I've never removed Ram off of a 15 Pro before, so staying at point one is a little, abnormal for previous models. We can see that our 0.3 amperage draw, the big spike is gone. This looks more like a braindead mode, but this might be normal behavior on the 15 Pro. Like I said, I haven't removed Ram on a, 15 Pro yet, so they may have changed how it works. For example, like, you don't even need a battery to boot the 15 Pro, and that's not normal behavior for previous models either.
So a few things might be different. I think that's perfectly fine. Probably. But like I said, older models would give me a current loop right there instead of a, steady current draw. So let's check our diode mode measurements real quick. I think they should be. I'm actually not sure what they should be.
Let's just see what they are. So here is my 0.05. That's still point 12. Still the same. My 1.8. That's still also the same.
My one V1 went all the way up to 0.45, and that's, most likely normal. We can probably verify what that is if, if you or, the other board view JK drawing gives us, download measurements. Let's see.
So yeah, we do have diode mode measurements on JK drawing and on my one v1 line, we can see what it should be. If I were to look under the PM, I see this would give me what it was originally, which it actually doesn't have ZB you might have it. It does have it.
So on my one V1 line, I see that .04 is around what it was originally, which is point 02I think is what we saw. But Ram is off of the board, so this is going to change this reading. It'd be better to look at the reading on this end where the CPU is off, and that's 0.47, which is very close to what we saw. I'm not quite ready just for what we saw, but it was just right around there. So download measurement looks good on that line.
But much more important test is whether or not I'm getting correct voltage on here. Now, before we were able to see that I was getting 0.88 on that line. With our multimeter in voltage mode now, we could check the Ram lines.
And first I'll check 1.8 should still be 1.8. Good. And then I'll check the, 0.5, which is good.
And then the one V1. Is still .89. Interesting .89 there,
So let me set my voltage to 1.1 and see if I'm still getting an average draw on that. On that? On that line before we were getting 0.3.
If we try to inject onto this line. So no longer getting an amperage job before we were getting that point. 0.3 I'm not getting any amperage job, so this is pretty strange. I really expected my one v1 line was going to go back to 1.1V. But it's still giving me the .89.
I no longer have the current draw, so that does very much confirm to me that Ram was short. Because otherwise I would have had a, abnormal current draw there. So I really don't know why.
It's not 1.1. But I don't like that at all. It's hard for me to proceed with that being what appears to be wrong. We checked on the donor board, and there's 1.05. So this is just not correct. I can't really just remove Ram on the donor board.
I mean, I could, but that's a whole lot of work to see if this is like normal behavior with Ram off. That's not what other models would do. So I don't expect it would be normal behavior. I do see there's like a really tiny crack.
In this coil next to my line. So I want to see if this is possibly loose. If it was, though, I wouldn't expect the voltage I have on it to be what it is. I'd expect it to be basically zero and.
So I'm just really thrown off by the fact that that didn't change to its to a good voltage. Also, I don't know if this model should behave differently upon prompt to boot with Ram off the board. I know that on other models I should have a boot looping current draw.
And on this model I have a steady current draw. And that's just another like uncertainty I have with this, because I wouldn't really expect that. I mean, it may have changed. I just don't know. And it's not really following the pattern of previous models.
It doesn't really appear to me that this is loose. There was. I think it would have already came loose. So my 1V1 line is not good.
I mean, it's not the correct voltage. It also doesn't take any amperage. If I touch it with 1.1. So it's not sure. The 1V1 line is an output at the ppm.
I see. So. So I guess pma c must be damaged as well.
I mean, that's really the only thing I can think of with this. And I think I'm just going to have to replace the PMI. C I think that's going to be my best course of action, because I can't really justify moving forward if my voltage is not correct on that line.
Because I don't expect the line will work if my voltage is incorrect on that line. Schematics say that line can handle 2.5 amps. So I know I have, like, a kind of a high amperage job. When I first touched it with the voltage, that only got to, like, one point 8 a.m.. So that shouldn't have been a problem either.
So I don't know why this line is not giving me normal voltage on that line. And if I replace PMA C and it's still the same, I'll at least know I ruled that out. And if it's fixed.
And then I could just move forward. There is some small possibility I would have to do a CPU swap to this board anyway, so I hope that's not the case. But that is still an option. I really don't like removing the CPU once.
I've already removed Ram, so I was hoping to avoid that. But, I've done it in the past, so I guess I should be able to do it again. I hope I have some PMI CS ordered already. I'm not sure that I do.
If I don't, I'll have to harvest one from a donor board. And harvesting the mix really sucks. This capacitor is for the crystal, so I'll have to make sure that's on there properly. This whole row is always very hard to keep. Keep on there.
It's just so close. It looks like I accidentally brought up a few capacitors with without. So I have to get all those replaced. I'm seeing some odd. Damage here. Which corresponds to right down here.
So these capacitors up here don't really matter. They're all just remain. On the board will be fine without them. But all the ones that are missing right here, I believe, are going to be important. At least most of them. Yeah, I'll probably just have to put all of these back.
So one, two three, 4 or 5. This one's loose. I really hope I have a donor already ordered because I don't want to harvest this.
If I don't have to. I think I'll just go ahead and flatten these paths before I fix the rest of those capacitors. Just because there's a small chance I could pull them off. I'm flattening these pads.
One. Two. Three. Four. Five. Six. Seven.
One. Two. Three. Four. Five. Six. Okay, let's see if I have a chip in stock or not. Yeah.
It doesn't look like I have ever ordered any 15 chips, so I'm going to have to harvest it from this donor board instead. And it's a pain to harvest under. Chips are pretty hard to clean. I should be able to do it still, but I wish I didn't have to. So now it doesn't matter if I mess up all these capacitors along the side. In fact, I'd rather clear them out to make sure I don't chip my chip.
Instead. I just put a minor scratch on the top of this chip. I think it's okay.
There is, like, a film on the top of the chip. And I think that's all I scratched into. So now clean this chip. The easiest way is going to be to use my hot plate to free up one of my hands. I'll apply some low and low melt solder to the top.
So. Looking okay. Should be able to rebel it now and reuse it. All right. That looks pretty good. I'll just reflow it, and then I'll place it onto the board.
And then we can check and see. If my voltage on my 1V1 line is back to normal or not. I'll reflow the chip once first.
Looks good. It should definitely be attached. Okay.
The very first thing I want to check is what my amperage draw looks like. Now, whether it is the same, which was point 0.1. And steady. Or if it's going to change to something else, I'd prefer to see a looping amperage draw, as is more normal with, previous models.
When Ram is off the board. So it appears to be the same as what it was before I did the PMC swap. So that's interesting. And it will be worth it to check the voltage on the line. And I'll go ahead and check my other lines that I always check first. Let's check this 1.8.
We still got that 1.8. The 0.056. Set up 2.5. So 1.8 or, 1.1 should really be 1.05 or 1.06. It's still 0.89.
So it wasn't the prices for that that was low. This should be a 1.19, I believe. I mean, that's close enough. So 1.6 on my sock line.
Versus .68.62. Point nine. Okay, so I haven't ever used donor board here. I'm going to split it and I'm going to remove Ram. And I'm going to check voltage on that line because I have to know if this is normal voltage or not and what the behavior should look like when Ram is off the board.
And this is going to take forever. So I won't show this on, on camera, but I think it's worth it for me to check this. And I probably won't be checking in with you back until tomorrow.
So I'll be checking this tonight. And, we'll see what happens. I'm disappointed that this line didn't go back to normal voltage. It was for sure partially short, because it would take amperage when I would touch it with 1.1V. And now it won't do that anymore. But it's still not giving me normal voltage.
And I don't know why. And I don't think that the hardest part for me is placing this, this Ram on the board. And I don't want to do that unless I know that this line is normal. So I'll be back once I know what the behavior should look like on this model.
So I went ahead and remove the Ram on this donor board. Here's the original board that I was originally working on. Here's the donor board that I removed. Ram on to see what the, measurements and voltages would be like.
And I did notice some immediate differences. For one, if I take my DC power supply probes and I touch them to the battery connector here, and I prompt to boot, I see what I would more normally expect on an iPhone board that has Ram off, which is a bootloop in current draw. If I short this power button here, it will prompt the phone to boot and we can see this, point 13 sometimes point 12, current loop.
And this is what I would normally see on almost any on any iPhone board with Ram off the board. So that's normal behavior. And that's different than what we saw on our board.
Our board was steady .1. ten. And also aside from that, if I were to measure my 1V1 line itself, I'm getting the point zero, the 1.05, which is what we saw on our other donor board that had Ram on the board. It's the same as I see on this board with Ram off of the board. Put my multimeter into voltage mode, and if we measure that line. We're going to see a looping .1.05.
So that's what we should see with Ram off of the board. I'll also just make sure this is going into DFU mode, just as the last, little check that I usually make. And the board automatically goes to DFU mode. So this board is behaving as I would normally expect. A board that has no, Ram to behave. And if we take all those same checks, we did just check this already.
But this is the the data board. And if we take all those same checks we're going to see some pretty different behavior. First being if I prompt this phone to boot after, touching with the DC power supply probes here, we don't get that looping current draw. We're getting like a point ten steady draw.
That's different. If I check the voltage on this, 1V1 line. I'm getting 0.89.
I believe. And both of those behaviors are incorrect. If I prompt this tabu, or if I try to prompt DFU mode. I don't remember if I've even tried this yet on this board, so I don't know if it will go or not. But we can see.
I suppose. It actually looks like we're not even getting DFU mode on it. So all bad behaviors. And, that tells me there's really no point in moving forward with the CPU swap as we are right now.
I mean, I'm sorry with the Ram swap as it is, right now, I'm going to have to do a CPU swap and see if these voltages will normalize. If they don't, there's no point in moving forward. It's just not going to work.
So it's possible the CPU is also damaged from the impact. But we do know for sure the Ram was bad because, before I removed it, if I were to touch the 1V1 line with 1.1V, it would give me a 0.3 amperage draw and it will only draw amperage like that when the line is partially short. So we know it was partially short. But in this case, I'm going to go ahead and flatten this as much as I can while I have it on the board and all original.
And after I do that, I'm going to go ahead and remove this CPU and I might as well check while I have the CPU removed if these voltages go back to normal or not, because even with the CPU off, I should still have normal voltages on my power rails. And if the voltages go back to normal after I remove the CPU, that's kind of a bad sign, because that will indicate that my CPU is the problem. And, we'll have to make a decision of what to do.
But first, let's just go ahead and see what happens. Once I remove the CPU, and I want to make sure it's completely flat on top before I do that, because once I place it on to another board, any type of flaws I have underneath it are going to make it a little bit uneven, and it'll be harder to get like a true flatness on top. So for now, I'll just flatten this and then we can move forward.
And I'll probably try to continue to touch this up once it's, on the other board. But for now, I think I'll just go ahead and remove this. Like I said at the beginning of this video, the previous tech didn't remove the under fill along the outside of this, CPU here, but they didn't actually remove the CPU.
Like, this has never been rebuilt before. And I don't particularly like removing CPUs once the Ram is already off the board, because the CPU is kind of, flimsy, like the Ram provides a lot of structural support to it while you're removing it. We don't really have a choice in this. In this case. I think I'm going to want to move some of these out of the way.
To make it as easy as possible to get underneath. I forgot that I said I wanted to check the voltages once. This is off.
And hopefully none of these components are going to stop it from giving me at least the braindead mode that I'm looking for. So. Yeah. As this board is right now, I'm going to. Test the temperature.
So a normal iPhone with a CPU off of the board will give us, like, a 0.0 to steady amperage draw. So let's see if that's what we get or not. So that's very normal amperage with, CPU off of the board. So, because that is the case, health. Let's see if my one v1 line has gone back to the normal voltage.
The point. The 1.05. So that's that small, steady amperage when we prompt the phone to boot, is just indicating that my secondary power rails are turning on.
Now, you know, I have one V1 up here, so I'll just test it right there. So if they're turning on, I really want to know if I'm still stuck with that 0.89 or if it's gone back to the normal 1.05. I almost want to see the stuck .89, because that will mean that my CPU has no problems whatsoever, and it was just the board that was bad. If I see 1.05, that's not necessarily
game over because my CPU, it did take a really hard impact. So it may have been partially detached and that could have also been, the cause of the abnormal amperage draw. But there we go.
So this is the one V1 line. And it's no longer .89 as the other one was, as it was stuck when the CPU was on the board. So like I said, that's, that proves that the low voltage was because of the CPU, and it's just whether or not when I put it back onto a new board, is it still going to have that low voltage or not, or will it go back to normal? It's possible that it had low voltage, because of like a loose pad because of the hard impact. But it's also possible that my CPU itself is damaged. So hopefully, hopefully, hopefully we don't find that our CPU is dead.
If that voltage is stuck low, there's not going to be much I'm going to be able to do. So I think I'll just, swap this to the, the donor board that I had used to check the Ram behavior. I'm pretty sure this board should be fine as well. So we might as well use this one.
So let's just start with that process. This is going to be quite a long video at this point. Two Ram swap plus CPU swap is always, quite a doozy.
So, first things first. I'll remove the CPU. I'll clean up the pads. I remove the Nand, I remove the EEProm. Once I have all that removed, I will, rip all the data board CPU. I will place that on the board.
And then I'll finally be able to check if my one v1 voltage is normal at that point. If I do all of this work. And I find the one V1 voltage is still only 0.9 or 0.89.
Unfortunately, I think I'll just call it and say this is unsuccessful because I don't believe that I will be able to get this board to work if I don't have the proper voltage on that line. I recently released a video. Well, at the point of filming this, I don't know when I release this, but I recently filmed a video or released a video titled jobs that threatened my sanity. And it's a four hour long video of a Ram swap on an iPhone 13, iPhone 13 Pro. And, I have another video for that series. That's another Ram swap, and I have another video that's a CPU.
Trace repair. But this video will most likely go into that that series as well. And at this point, I almost may as well change this series to just know the Ram swaps that are in my sanity, because it's always these Ram jobs that take, the most time.
It's like sitting through the most intense, Like patience inducing activity that you can imagine to just sit here and work on the same board for hours and hours and hours. Now clean up these pads. Get rid of all this under fill. Next, I will remove the Nand memory chip.
I'm not going to bother to clean these pads right away. I want to get the CPU on to the board first, just so I can, check that voltage to see if it's going to go back to normal or not. Lastly, I will remove EEProm right here. This. Okay.
This board is fully prepared. Ready for the transplant. CPU's cleaned and ready to go. I do need to clean. The Nand area still of a EEProm is also removed. Before I do anything else, I want to clean up my CPU rebel and get it on this board and measure my 1V1 line and see if I'm still getting that 0.89, or if it's gone back to normal.
I'm very glad to see there's not very much, Underflow left on this CPU, so it'll be very easy to clean and rebel. Looks like I missed a little bit. And I don't have a ram for this set yet. I will have to harvest one. But you know this video just getting longer and longer, like, I don't even know, but, I will put a different.
The iPhone 14 Ram in place here just so I can get this chip to sit at the proper level. As you can see, this is like a little bit skinnier than that chip. But up here it will sit at the proper height now. So the stencil is, pressing down properly. I didn't notice before, but I do have another missing pad here.
And I don't know what that one is, but it's probably a CPU power rail or something like that, so I think I'm just going to mask that off just in case. I don't want anything too short to ground on accident. Okay.
Looks pretty good. I'll just reflow it now. All right.
That looks pretty good. I'll go ahead and put this onto the donor board. Now, and we can see if it will. Give us our normal voltage or not.
This. So let's get the CPU under here and see if we can get our voltage back to normal. I hope we see normal voltage so we can continue. But if we don't, then, I'm going to have to call it and say this is unsuccessful. All right. That should be on there. So.
I'll wait for this to cool. Clean it up a bit, and then we can test the 1V1 voltage and see if it's back to normal or if it's still staying at that point. Eight nine. Revo looks pretty good. So, my diode mode measurement on my 1V1 line should be back to like that .02 or .04 or something like that.
Let's see what it is. Oh, sorry. It shouldn't be a point. 4.9 is still normal because Ram's not on the board. I don't know why I said that. So fingers crossed.
Or C measure the voltage. Hopefully it's back to normal. I really hope to see the 1.05 here. If not, then we are no good. Multimeter to voltage mode. All right, 1.05.
We're back to normal voltage. All right. We got a great chance at this still, so we're going to move forward. Still a lot of still a lot of work to do.
I'm glad to see that. But still a lot of work to do. I'll have to harvest a Ram chip. Still, which is a whole ordeal. I have to clean Nand and pull that from the donor board and place it on here.
I have to pull the EEProm in, place it on to here. I'm after those two things are done. You know, we should be good to go. I kind of wish I would have reflow these top balls while I was on the other board at this point now, because I'll have to do that. And I also have to put solder paste on to here. But now I'll also have to worry about it flowing down underneath into the CPU board.
So I have to avoid that and be very careful. I think the very next steps I should take is probably to get the Nand onto the board and the EEProm, and then we can finish up with the Ram harvest and place afterwards. I think at this point I'll make sure it's going into DFU mode. It should at this point go straight to the Fue mode with with Nand off of the board.
I believe. So I shouldn't even have to prompt it to go into DFU mode. I should just go straight to DFU mode as soon as I plug it into a computer. And it does. It's going to do a Q mode on its own.
So iPhone 15 Protea film mode looks good. So yeah. So now I will clean up the Nand area. I'll have to rebuild Nand and place it on to here, place EEProm. And then we can continue to move forward. This is another doozy of a video.
This one is going to take a very long time. In total. All right. That's the name of the game. Sometimes nothing I can really do about that.
Sometimes data recovery just takes a lot of work. And this customer is willing to pay a lot of money for this recovery. Obviously, a job that takes this long isn't under standard pricing. It has to go to my severe tier, pricing. So when the customer is willing to pay, I'm willing to give it my very best shot to recover the data. When it comes down to it, I do have enough volume that it's probably more beneficial for me to just work on normal jobs, rather than spending multiple days on on one job.
I do do some work in between. I sometimes like to just knock out a normal data job while I'm working on this one, just because I. I don't like to get like too caught up on one thing for too long. I can probably finish, like, you know, who knows how many data jobs in the time it takes me to complete this one job? But I do pride myself on being able to recover data on the hardest of cases when nobody else can.
I am almost curious now what would have happened if I never removed Ram? And I just did a CPU swap. I for sure had a partial short because the line itself would take, like, amperage when I touched it with voltage. If I touched it with 1.1V,
it would take point three amps, and it shouldn't do that. But once I removed Ram, the actual voltage that was showing on the line, the point nine volts did not change. And it's possible that if I just did a CPU swap, the MC could have powered through that partial short and it would have drawn a little bit of extra, amperage from the battery.
A little bit of extra power from the battery, but it may have been able to just straight power through the short without me changing the Ram at all. It may have needed the Ram change, but I won't really know that ever, because I removed Ram first. I did see evidence that I could, definitely have been the problem. But it does make me wonder if I could have gotten away with not removing Ram. I guess when it comes down to it, I probably should have known to just try CPU swap because of how damaged that board was.
So I just reflow the Nand pads with low melt solder. And now I will remove and find those pads. I'll remove this order and flatten those pads. That looks good. So I'll go ahead and pull the the chip from the donor board. Now. I mean, from the data board.
Here's the important memory chip from our data board. Here's our important EEProm, which is used. And decrypting the data chip. We'll have to transfer both of these over.
So, for whatever reason, they gave two size Nand chips. As we can see, this is not the same size. And it's okay, but I'm going to have to remove all of these, resistors along the outside. These are all just ground.
The ground on both sides. I'll show you that on the x w but, this chip is going to be a bigger footprint, so that's a little bit annoying. As you can see, like, the actual footprint is the same. The chip is just, bigger. If we look at, the board view software, we can see these are just all ground resistors. And same thing out here. It's just.
Except for these two. So this will be on both boards. But other than that, I'm going to be missing all of these other ground resistors. And these ones, they don't need to be here either way. It's just, I believe it's just for, like, thermal support.
I will reflow Nand one more time. Sure. On. The soccer balls are connected. Well. The chip is almost ready. But I do have to clear out all of these resistors.
All right. That looks just fine. I'll go ahead and place them on the board. Now.
All right. Good. That should be on now. Last part of the CPU swap will be to transfer you from ship over.
And first, I'll try to do a clean pull. I don't really want to have to rip it if I don't have to. And I just place that onto our donor board. And there we go. About a second move to replenish the fine.
All right, CPU swap. Part of the recovery is now finished. I just need to complete the Ram swap now. And like I mentioned earlier, I don't actually have Ram in stock, so I'm going to have to harvest one. And, not very many people show the hardness of the ram.
So I guess that would be interesting. You can buy Ram off of AliExpress from China. But they don't tend to be perfectly flat.
For my experience. They tend to be a little bit bold and makes them very, very hard to use. So I much prefer to use the harvested Ram.
Okay. I verified this, donor board was working, and, I split it, and now I'm going to remove the CPU and start the harvesting process from it. Basically, I will just, remove the CPU like, normal, and then, and then I'll scrape away the whole CPU, leaving only the Ram.
As my donor Ram chip. So let's pop the CPU off now. And harvesting.
The ram is a lot like the removal of the ram, except the opposite. Now I'll be scraping away the CPU. Leaving only the Ram left over. And that's all you need to scrape. It's a very thin layer. This, this black part right here is what I'm looking for.
Okay. That's pretty good. That's good enough that I can. Rework these pads now. It's always a little bit tough to get those pads to take some solder, but they look pretty good. Now.
And it's getting pretty late, so I will finish rebuilding this tomorrow and setting it tomorrow. And hopefully we will have access to data. Okay. So I'm going to go ahead and rebuild the Ram now. It's the next day and I'm ready to finish this up. But I did just realize I don't have the right stencil.
I have to use an iPhone 13 Ram stencil, which is very inconvenient. It doesn't fit exactly. There's like this awkward space on this stencil right here that makes it to, like makes it not fit properly. And this one has an awkward space.
The opposite, column here, which which makes it not fit properly. And if I skip that very last one with the awkward space, I could almost get all of the balls to fit. I just missed one row at the very end, so it's a little inconvenient. To right here. I'll miss those ones, but besides that, it gets all of them. I can probably do the top and the bottom at the same time, but it's it's actually a little bit awkward.
I think I'm just going to do one at a time, so I'll just do this one first. Then I have to do the three separately. I have to do the side separately. I have to do a bunch of them separately. Let's just see what I can do here. It's definitely not an ideal situation.
I assume that my other stencil that has the CPU stencil on it would have Ram and I just never checked and it doesn't. So such a move on I guess. I should still be able to make this work, but I'm a little afraid that my solder balls will be uneven. I won't have nice even balls. That. It doesn't look too bad.
Let's just see if I can get these ones to form properly. I had a bit of office. A bit of an issue right there. Kind of hard to see what happened.
One ball didn't form, but I'm not quite sure where the solder went. Because nothing seems too uneven around it. If it is a little bit uneven, I will just go ahead and scrape this chip. When I'm done with it, I'll show you that technique. And it should, to make things as even as I can get them.
For now, I'll just go ahead and fill in more solder right there. It's a little bit strange. It doesn't look very uneven. There. I guess things don't look particularly even in this area at all.
But I have a method to even things. I'll show you that when I come to it. For now, I'll just go ahead and probably grab this solder ball right here.
Throw it on there. Not too bad. Could be worse.
I think for those corners, I'll just actually take some of these other solder balls as well. I think they may not be big enough. So. Yeah, I'll just wait for right now. Okay. Let's try to.
Do the other side now. I guess. I guess I'll reflow these balls first to make sure that they're on here pretty steady. I was going to do that at the end, but the stencil is going to, like, push down on them. And I don't know if the stencil is going to ruin them or not. So we'll have to see what happens.
Again. Not too bad. I'm a little bit concerned with the sides, because I may smash the solder balls I've already made. When I have the stencil on there. I'm not quite sure how I'm going to avoid that.
I could individually place each ball on here. Because, yeah, you can see some of these already were getting smashed. It's going to be a real pain, but I think I'm going to have to individually just place all of these balls on here. I think that's that's the course of action.
Otherwise, I think I'm going to ruin the solder balls I've already created. Be interesting to see how many different sizes I got. Obviously, some of them just popped up along the sides. Some of them are probably okay to use.
But. Oh my God! I don't think I have the patience to continue doing it this way. Let's just give it my best shot. That's good enough to work with. Let's just do this last side and then see if I can even out everything. The solder balls are very uneven at this point, but I should be able to get them somewhat.
Even. Okay, so every pad has a solder ball. They're wildly uneven. But like I said, I have a method to deal with that. Basically, I'll just, flip this chip over, put it on a piece of paper flat, and then I'll just, you know, put a little bit of pressure on my finger and I'll grind it down. And that's going to make the, the biggest balls have a flat top and or remove a little bit of the solder and then I'll reflow it.
And I'll continue to do that over and over until I feel that everything's even for something like this. It's quite uneven in my opinion. I think I'll probably have to do that.
I don't know, 4 or 5 times before I'm satisfied with it. We'll just have to see how it goes. But I'll reflow once to make sure everything is formed into a nice, nice ball. And I solder ball, some nice balls. So that already doesn't look too bad.
But like I said, it's wildly uneven, so I'll grind it down a bit. And that was like, by far the most scuffed free ball I've ever probably done. But it was not the ideal way to do this.
So I'm going to have to definitely order a new. Stencil as soon as his videos don't. I never have to do this again. So you see how everything is a solder ball? I'll show you what happens once I grind down a bit of them. I've got my paper here. Let me go to a flat, flat surface.
And I just grinded it down a little bit. And now we can see, especially those ones on the sides. They have a flat top on them. Some of the solder has been ground away.
I'll just reform it into a ball and do that again. I just basically, you know, put my finger on it and go like that on the paper. And after enough times it should make the solder balls about as even as I can get them.
Okay. That's about as much as I have the patience for. It's pretty even. I think I'll be okay. That is not an ideal way to rebel a chip. But, I'm on a time frame here, and I don't think I really have time to order a stencil.
If I. If I mess up the placement of this and I have to rebuild it again, at that point, I'm going to have to tell the customer she's going to have no choice but to wait. But if I can get this done today, I would much prefer to do so. Almost every ball has, like, somewhat of a flat surface. Whether it's not as flat as the one next to it or it is, at least it seems like it was touching, which is really all that really matters. I don't have a ton of, er, room in between these solder balls.
They're very close together and Ram likes to squish down, so I do want it to be as close as possible, but I think this is what I can do. So this is looking good. Let's take a look at the board one more time. Honestly, the top looks pretty good. I almost don't want to have to run solder over it, but I am concerned that once I apply heat to it, that they might change a little bit.
I know that these, like the tops of the ram, like the solder likes to kind of like go downwards once you heat it up. So where there wasn't a hole after you heat it, there's a little bit of like a concave hole. All the holes are concave.
These holes aren't very deep, but some of them are deeper than others. So I think I'm definitely going to have to go ahead and apply a solder paste to these holes and make sure everything is as flat as I can get it to go. I really would like to get this on the first shot, which is historically very hard for me to do with Ram. I tend to not get it. The first shot.
Goodnight, baby. I love it. Okay.
I think we're here. Let's check the boot sequence one more time before we put the ram on. Make sure it's still normal.
I did see a few solder balls around the PMI. See, from when I did the EEProm change. I knew I should use a smaller nozzle for that, but I thought it was fine.
Looks pretty normal to me. Well, I can't believe that. But I just missed the most important part. I just set this on here. I'm not quite sure why I pause the video.
I'll stop and see. Yeah. So, Really dumb of me. This is like the the culmination of the whole video was me setting this ram onto the board. But I didn't actually record that.
It was very, very simple. All I did was place it on to here and then heat it up. And now I'm waiting for it to cool back down.
I didn't mention how nervous I was just because this is a culmination of a lot of work, and I'm really hoping it will go the first try. This is probably almost about cool. Now, and we can take a look to see how it looks. And then I'll be measuring, fast Ram lines to see if they went short or not. I mean, I don't think anyone would ever think I'm faking it, but.
But if you did want to, like, double check, you can see that these two capacitors had these marks on the on the end of them for the whole video. I really don't. You know, it's really hard to all this chip. So if this doesn't go the first try, there's a few things I can try to do, but, I may end up having to just wait and order the proper stencil. Let's take a look at see how it looks.
I really hope I don't see any shorts. Honestly, that looks beautiful. That's one of my best placements. You can still see the CPU balls. You can see the rambles. Looks very even.
The thin side is the one that tends to short more often, though. Which would be like this side. My microscope ones zoom out quite far enough to see. But again, that looks very good. I think the placement seems so far very nice. So my hopes are high.
Now. Short in a short. In a short. Okay. .06 is perfect. That's what I want to see.
One up here. Perfect. I have movie five right here. Thank you. And. Everything looks good.
Everything looks great. Okay, let's test the amperage. Draw. It looks like it's moving. Let's grab some parts and see. Take one last quick look.
But I already saw the amperage draw, and it looks really good. Let's grab these parts. I don't know how long this video is or how much footage I have, but if this is booting, basically nothing went wrong. Everything just takes this long. It's just so many steps.
I didn't have the stencil, which really slowed things down, and I didn't have a harvested Ram which slowed things down. I didn't have a prepared CPU board. We have an Apple logo. This is beautiful.
Oh my goodness, I'm very happy. Let's make sure it gets all the way to boot. Then I have to grab that jig and make sure it fits.
Long Apple logo. What is that? I honestly don't know what this could indicate. If I get this far and it's getting stuck on the Apple logo, obviously. I'll probably have to try a update to see what's going on. Maybe it just needs an update. But if it's getting to the stage where it's showing an Apple logo that's so far along the boot process, it's basically done.
Unless I have like data corruption, the Nand chip, I use a lot of heat to remove it. It was like very stuck on there. The under felt was kind of hard. It's possible something went wrong there. But let's just put this in the jig and try an update because this is this is this is going to be heartbreaking if it doesn't go.
It's in there. I'll go ahead and attach the, proximity sensor and also the wireless charging coil for this one. Choose as my hands are shaky from from all of this work. still have an Apple logo. Let's just wait one time, first again, and see if it does anything different.
If not, I'll try to get it to, go to recovery mode, Yeah. It doesn't look like it's moving. So. The only way I can really get this one into recovery mode is if I, hold the power button, either while it's rebooting or while I boot it up. After plugging the the battery at the same time that I hold this power button, or that I short the power button. Plug in the turned for.
Okay. It's in recovery mode. Let's plug it into my main computer.
Okay, let's give this an update. See what it does. That was a very long update.
It stayed on that 24% for, I don't know, ten minutes is took forever. It's on the Apple logo. Let's see if it actually boots up or if it's still going to, like, give me a reboot Swipe up to recover. Okay. I don't have power button plugged in, so let me just press it once to turn the screen on. Here we go.
Swipe up to recover. Attempting data recovery. So let's let it go through the screen. Thank freaking God. I'm again very hopeful, I think.
I think we're in the clear now. So the the, the NFC chip, the important one, the big one is on the top board. And I never transferred the NFC chip in on this phone.
So it's possible that the NFC mismatch was causing that issue. I'm not quite sure normally wouldn't do that behavior. It would just go to a swipe up to recovery screen. Maybe it's different on the 15 Pro I don't know. A lot of a lot of, iPhones will have the NFC chip on the bottom board.
So you won't really see that issue with, like a CPU swap. Anyways, I think we're probably going to be okay. Let's just wait for this to finish. It will take a few minutes, and then, I'll be right back once it's done and we can finish this out. Okay, so it looks like I'm in. Let me let me double check.
So having me go through the network settings for some reason. But it's still ask for the passcode, so I know it has data on it. Anyways, it seems to be sitting on the screen for a little bit.
I'm sure it's going to go through in just a moment, but I'm going to plug this into a computer. As soon as it's ready, I'm going to pull all the photos from it, and, all the rest of the data as well. The customer is going to be very happy.
This was quite a long video and nothing even really went wrong. Just all parts of the process took a very long time. The fact that I didn't have a 15 pro donor board already, like, ready, like CPU removed and removed, caused this to take a little bit longer.
The fact that I didn't have the proper stencil for the Ram, that took quite a lot of time. And the fact that I had to harvest the ram itself all took a lot of time. In the future, now I won't have to harvest Ram again. I already have a good one. If I ever have to do a Ram swap on the 15 Pro again.
And, I already have a good, donor CPU swap board as well. So once I'm done taking the data, I'll take the CPU off of this board so I can reuse this again. And, you know, like I said, everything went fairly smoothly.
Nothing really went wrong. Just all the processes took a very long time. And just to recap, I do have the phone plugged into the computer now, and it's taking data.
And I kind of just want to go over everything that I notice on this board. So the first thing that we saw was that it appeared to have a Ram short on the CPU. And aside from having the Ram short, I also noticed that the 1V1 voltage line was low. It was only giving me 0.89, volts on there. And even after changing the pin, I see that didn't fix the voltage.
It was still stuck at .89 even after removing the Ram, it was still stuck at 0.89. And a note on that is that the 1V1 line was actually drawing amperage. If I were to touch it with 1V1 or 1.1V with my DC power supply, it would draw 0.3 amps if I were to touch the line directly.
And when it does that, that indicates that the line is partially short. It won't draw anything. If the line is not sure or if it does draw something is something very low. Point three was abnormal. Once I swap the CPU over to a new donor board, the Ram was already off of the board and I noticed that the one v1 line went back to normal is the.
Finally, it was finally the normal voltage, which is actually 1.05 on that line. And once I saw that I was able to move forward with the Ram swap, I harvested the Ram, rebuilt it and set that on the board. And once everything was settled, we have access to data. So once again, thanks for stopping by. Have a good one. Bye.
2025-02-25 07:39
