All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 134 posts ]  Go to page Previous  1 ... 3, 4, 5, 6, 7
Author Message
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: April 23rd, 2024, 20:52 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
good. thanks for confirmation. so nothing wrong here, but there is definitely something wrong elsewhere. "C1" capacitor which is not a capacitor, but resistor
i looked for photos around and found a good one. it also explains why i measured negative capacitance one time before i lost it

Attachment:
s-l1600.jpg
s-l1600.jpg [ 423.15 KiB | Viewed 1305 times ]


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: April 26th, 2024, 17:58 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
based on initial diagnosis i suspected the resistor is 50ohm. took apart 2.5" drive and traced the pin to a 50ohm resistor. soldering resistor did not change anything.
remembered the DRAM moved slightly when i was taking off faulty CPU, and a bit of solder came out. quite possible a pin lost contact there.
lifted DRAM chip and indeed there was a pin with hardly a bit of solder on PCB. found pinout, half of the DQ pins to CPU are open. damn i thought either the CPU is faulty or not made good contact. a closer look at datasheet made me realize these are actually x16 chips, not x32. so half of the pins are unused. and indeed DMM check on all DQ0-DQ15 pins showed identical values and all DQ16-DQ32 plus some other x32 specific pins are open. the pin that was missing solder is unused. so nothing wrong here.
reballed DRAM and put back, nothing changes (i was looking if there is something wrong with this since i got the message "configured for UDMA/133 (device error ignored)" while in case where they recovered it with JTAG there was only "configured for UDMA/133")

there was also idea to check crystal, because with damaged CPU there was -3.63V, could this affect crystal stability? i knew after repair i had 0.63V on both crystal pins and 0.92V on nearby test point. meanwhile also flashed initially reconstructed BIOS, with only bad block table additions drive added automatically.
this again resulted in no drive detected in OS, and at this point i checked crystal pins again - one pin is good at 0.63V, another went down 0.48V, also 0.92V test point shows 0.64V. multiple restarts and connecting drive through USB adapter did not change a thing.

next i change BIOS again to the one drive initally tried to reconstruct, and on first boot both crystal pins are good at 0.63V, however 0.92V pin remained at 0.64V - it seems that one of these tables in BIOS contains CPU config register, and with wrong values sets up hw incorrectly.

not knowing enough about the BIOS structure i suspect on table at 0x150000, because if i flash dump that would boot drive to "unallocated" state quick, after messing with MU03 loader, this code gets overwritten with same values each time, boot is slow takes several restarts to get the drive detected in disk utility, in general unstable and unpredictable behavior. likely the MU03 loader looked for data at different offset and fooled the CPU somehow

Attachment:
Untitled.png
Untitled.png [ 394.52 KiB | Viewed 1222 times ]


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 6th, 2024, 23:15 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
after so much bullshit with this drive, i took off MSP430 to check it again.

22 pins have identical readings on MSP430V402 and new unused MSP430G2353 (about -10mV on new)

pin 3 on V402 starts at 580mV climbs quickly to 630mV and stays like that
same pin on G2353 starts at 530mV and climbs slowly to 550mV (did not check further as it was slowing down with time)

version difference? or there is something wrong with V402? the pin is connected to PG output of 5505


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 7th, 2024, 17:55 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
problem identified. all 3 inductors are bad, looks like shit quality components, one is even chipped off in the corner - 0.63uH
the other two are around 0.7 - 0.75 uH


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 7th, 2024, 18:27 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15606
Location: Australia
If the supply voltages are correct, then the inductors are OK. They would be rated for operation at 1MHz or thereabouts. If you are testing them in the kHz range, then you will get unreliable results.

_________________
A backup a day keeps DR away.


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 8th, 2024, 1:35 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
wrong. the voltage may look good, but the inductor value goes down causing all sorts of instabilities in circuit. i've seen this while repairing wifi card, there were 1.2V and 3.3V outputs from dual converter in single package both through a 2.2uH ind.
the one on 1.2V went as low as 0.4uH, the card worked but throughput was low and it often crashed unless used very short usb cable. everything worked great after changing it

true it might be difficult to test these, especially if testing only single piece (that's why i thought the chiped one was good when checking it first time) but testing 5-6 of these and comparing results clearly confirms which ones are out of spec

from the wifi card mentioned above at 3.3V output - 2.21uH
from miwifi mini router near 5ghz chip - 2.12uH
from samsung tablet board - 2.16uH
from gl-ar150 router - 0.95uH
from gl-ar150 router - 0.99uH
from m600 at 1.8V - 0.63uH
from m600 at 1.2V - 0.71uH
from m600 at 1V - 0.75uH


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 8th, 2024, 2:41 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15606
Location: Australia
The MP5505 chip has a switching frequency of 500kHz whereas the SLO and RAU chips are designed for 2MHz.

The MP5505 application circuit uses a 4.7uH inductor whereas the others use 1uH.

The difference in the physical sizes of the inductors on your PCB would be consistent with their inductance specs.

Have you compared the switching frequencies of the DC-DC converters in those other devices?

_________________
A backup a day keeps DR away.


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 8th, 2024, 12:29 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
devices are old and broken, i use them for parts, so not interested in their specifics.

all these inductors are of approximately same size. the two from ar150 routers fit the board. managed to find one more on mr13u router, it's a bit larger, will see if i can solder it without shorting to nearby converter pins


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 8th, 2024, 21:42 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
fzabkar wrote:
The MP5505 application circuit uses a 4.7uH inductor whereas the others use 1uH.


yes, application circuit. did anyone confirm this on actual hw ?

RAU application also suggests 10uF and 22uF in/out caps, but on board there are actually 4.7? (no idea what the hell is value, at 100Hz gave 5.5uF at 40kHz gave 3.9uF) and 10uF


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 9th, 2024, 21:28 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
storage voltage is indeed 17.2V on this device, while checking 5505 straps i found damaged resistor on FBB pin, one of its terminals is cracked and it seemed at first it is fully open but after some poking with DMM on it i managed to get inconsistent readings, like 1.5Mohm, 337kohm, and few times is settled at 220K. which appears to be its value.
while at it, i submitted a question on MPS about formula, once they reply i will confirm if the big inductor is good


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 12th, 2024, 11:20 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
FBB resistor was good. well good enough until desoldered once. if there is no connection in its place nothing starts (rattle noise coming from 5505), storage voltage is at 3.3V

few pages back someone said voltage on crucial drive was 12.7V, if the release voltage is 4.2V in that case inductor ends up at 2.43uH (assuming peak to peak ripple is 40%)
here is storage at 17.2V with release of 2.75V and the inductor value 2.5uH
i measured 2.42uH on mine - probably same part is used on all versions of PCB

after replacing small inductors consumption from 5V USB port through adapter went up by almost 100mA (400-450mA)
one would think there is a short somewhere based on all the current it draws, when i removed crystal and two caps on it and tried to power on the board it took 40mA (all the voltages were present so this was consumption of converters and other components on PCB without controller active)

but that's not all

on the test point near crystal there is still 0.64V
i traced this pin under the controller on 2.5" drive it has 0.97V
looks like some sort of Vcore test


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 14th, 2024, 6:46 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
connected drive through usb adapter and left for a while. after all error messages and timeouts appeared [hardware error] message with some IDs like manufacturer

soldered UART wires. silent

entered service mode. shell prompt appeared in putty. same hardware error message appeared after 6 seconds through usb connection. through SATA connection no device detected (probably need to remove wires i used to enable service mode, did not want to just short it because they suppose to bridge through a resistor)

0.64V on test point remains

what i did when this test point lost 0.9 voltage? nothing particular, changing BIOS - hot air was blown only on that chip, rest of the board has been exposed to about 140 celsius from bottom

what else is missing? 3.3V disappeared from MSP430 pins 4,5 and 16

tried changing crystal, same result, nothing wrong with it


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 15th, 2024, 17:17 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
i just don't believe what am seeing with this device right now!

thought at first, a small transistor on msp pin16 is a supervisor to reset the mcu. one of its pins is on gnd, another 3.3V. so if it failed the msp would stuck in reset mode which would explain missing voltage on its other pins. but the transistor looks fine, same reading as when i soldered it a while ago.

then came the idea to re-check dram, and at the same time load switches since they are right next to it and one of them gave me problems in the beginning. a small 'resistor' on msp pins 4-5 as well since its shape looked more like a capacitor or bead. turned out to be NTC.

if the above suspect pin on msp had anything to do with reset the culprit would be on 3.3V so i check all the resistances after removal of components, and immediatelly caught my attention is 3.3V after the 5505. used to be around 78kohm, with removed parts went down to 25kohm

no problem, it will go back when parts are soldered again. checked dram, everything as before (680-710mV on DQxx, lower on power pins)
check load switches, both are the same - 0.5V to output, 1.7V to gnd, 1.4V to enable. cleaned them and put on 0.3mm balls

soldered everything. 3.3V after 5505 remain at 25kohm. i'm thinking either something made no connection or there was short under load switch.
already tired of this, i decide to just plug it in, holding fingers on nand and controller chips, if it starts burning i'll unplug it.
nothing. heats up slowly and boots slowly just as before.

check voltages, surprise, there is no longer 3.38 or 3.4V, it returned to standard 3.3V, 3.32V on msp pins (probably input too)
what the fuck did i just do i have no idea, by just removing and soldering few components back, to have Vcc fixed into normal value.
even more strange, before anything was touched (after initial failure few years ago) there was already 3.4V and i did not replace anything today, out of a sudden it is 3.3V


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: May 17th, 2024, 17:01 
Offline

Joined: June 27th, 2020, 8:58
Posts: 39
Location: europe
serial access in service mode appears to be useless, typing some letters return short hex codes that are probably known only to manufacturer and guys at ace, otherwise the return would always be just 'E'
typing 'j' and hitting enter would freeze the console

although it was silent in normal mode, i noticed it accepts input, but whatever typed it returned "??? cmd" or cmd too long
tried bunch of stuff i could think of, one command appeared to be valid "SATA" but it returned just a several blank lines


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 134 posts ]  Go to page Previous  1 ... 3, 4, 5, 6, 7

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 7 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group