Apple Silicon Soldered SSD Upgrade Thread

[Rickyandika]

I never tried it on 4 nand system but according to the jc app, it will be programmed as blank nand because they say the UNXXXX(no location specific) so i assume it will work.

Just program all chip with the same files it should work

 

[kiloran]

I never tried it on 4 nand system but according to the jc app, it will be programmed as blank nand because they say the UNXXXX(no location specific) so i assume it will work.

Just program all chip with the same files it should work

In the case of k5a4 in jc app, position is important

 

[Rickyandika]

Oh shoot! Sorry! I remembered wrongly. I thought they had the UNXXXX label. So in that case i dont know if the one in JC app will work…

 

[kiloran]

Oh shoot! Sorry! I remembered wrongly. I thought they had the UNXXXX label. So in that case i dont know if the one in JC app will work…

It turns out that i need to try reading the .bin configuration from a known working NAND, and try write data? Are there any successful cases with read/write data? I looked video on the first page, this method was successful.

 

[Sauve Mon Mac]

It turns out that i need to try reading the .bin configuration from a known working NAND, and try write data? Are there any successful cases with read/write data? I looked video on the first page, this method was successful.

A3112 is two Nand only !
With k5a4 you’ll get 500Go. Format it with iPhone format (it’ll blank them)

 

[kiloran]

A3112 is two Nand only !
With k5a4 you’ll get 500Go. Format it with iPhone format (it’ll blank them)

Sorry i meant a3401. So if i will format all 4 nands k5a4 with iPhone formatting config, i will blank them? has this been tested?

 

[dosdude1]

Sorry i meant a3401. So if i will format all 4 nands k5a4 with iPhone formatting config, i will blank them? has this been tested?

I did try doing this with K5A5, but it did not work, though was likely due to the JCID firmware for the K5A5 just being bad in some way. Just try with K5A4, in theory it should work. Just flash each one by selecting “Format” option in JCID with iPhone 15 selected, once it writes the firmware it should read as being in a “general state”. Once all 4 are programmed this way, they should work fine in any valid configuration, and get programmed appropriately during DFU restore.

 

[kiloran]

I did try doing this with K5A5, but it did not work, though was likely due to the JCID firmware for the K5A5 just being bad in some way. Just try with K5A4, in theory it should work. Just flash each one by selecting “Format” option in JCID with iPhone 15 selected, once it writes the firmware it should read as being in a “general state”. Once all 4 are programmed this way, they should work fine in any valid configuration, and get programmed appropriately during DFU restore.

Got it. I'll let you know when I try it.

 

[Dy1ng4help101]

i have soldered now the original Nands. same behavior. I think the problem will be the pulsing 2.5V Rail. but I am not shure, where to go on a troubleshoot. the rail is like 2-3 seconds on 2.5v than "Break down" to 0.3V (round about) at the same time the USBC voltage meter? restarts again. When booting with (control, option, right shift and power button) the Mac will be charged ist stable 65W at 20V. otherwise its "restarting"....

Hey, so I'm stuck in the EXACT same situation. I tired to upgrade M4 pro MBP. I also checked the logs with original chips and new chips, I noticed that we're both getting
"_AMRUSBDeviceSendDeviceRequestTO_block_invoke: AMRUSBDevice error (-536854449)"

I'm not sure what the issue is but I think at some point one of the capacitors or resistors must've been knocked off. And I think Dosdude mentioned that if a component is knocked off and DFU restore goes bad, usually the chips are no longer usable. I'm not sure if it applies to original chips too though.

Anyways, I did take measurements around the NAND power rails for capacitance and resistance, they were both really low compared to my working M1 Pro MBP. So I reflowed everything and then replaced all the capacitors, resistors and ferrite beads just to be safe. Now all the measurements match between the MBP boards, so I'm assuming either the chips that we tried to format while our boards were faulty, made the chips useless. OR there's something else wrong with it that we're not seeing.

When I search on google "AMRUSBDevice error (-536854449)" I saw somewhere, someone was having a similar issue with M1 MBP 13'. In their case it was a resistor near UL000, which is a wifi/bluetooth chip. I wonder if maybe we have a similar issue? Some random short we're not able to locate.