All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 113 posts ]  Go to page 1, 2, 3, 4, 5, 6  Next
Author Message
 Post subject: Crucial MX100 repair after PSU failure
PostPosted: August 4th, 2016, 17:43 
Offline

Joined: March 28th, 2014, 9:11
Posts: 44
Location: Nederland
My server PSU failed and seems to have taken the SSD containing the OS with it. Now while I do have backups of the important bits of this OS drive, rebuilding the system from these backups would likely take about a day so I'm willing to invest some time into finding out if the SSD can be repaired.

Symptoms are simple and clear: BIOS doesn't detect the drive. POST isn't delayed, so there's no hang or failed init there.

I opened the drive and took some pictures (attached). Now I need some help determining what might be defective. The resistors marked 0-ohm are fine, but a lot of them are marked with a yellow/orange square and usually measure a different resistance over them. I also see some diodes but they don't appear to be TVS, at any rate they're not close-circuited.
Any points on where to go from here?


Attachments:
IMG_9834.JPG
IMG_9834.JPG [ 3.45 MiB | Viewed 38966 times ]
IMG_9832.JPG
IMG_9832.JPG [ 3.39 MiB | Viewed 38966 times ]
IMG_9827.JPG
IMG_9827.JPG [ 3.83 MiB | Viewed 38966 times ]
IMG_9824.JPG
IMG_9824.JPG [ 4.07 MiB | Viewed 38966 times ]
Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 4th, 2016, 22:32 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15440
Location: Australia
I'm not sure of some of the markings, but I have identified at least some of the supply voltages.

FYI I have compiled a list of datasheets for SSD ICs here:

Datasheets for SSD and Flash Drive ICs:
http://www.hddoracle.com/viewtopic.php?f=100&t=1577

MP5505, Monolithic Power Systems, 7V, 4A, High-Efficiency Energy Storage & Management Unit, 2.7V-to-7V in, marking 5505, QFN20L:
https://www.monolithicpower.com/Desktop ... nt?id=2477


Attachments:
V3.jpg
V3.jpg [ 50.37 KiB | Viewed 38952 times ]
V2.jpg
V2.jpg [ 55.71 KiB | Viewed 38952 times ]
V1.jpg
V1.jpg [ 123.44 KiB | Viewed 38952 times ]
MP5505.jpg
MP5505.jpg [ 162.96 KiB | Viewed 38952 times ]
LDOs.jpg
LDOs.jpg [ 150.69 KiB | Viewed 38952 times ]
5V_TVS.jpg
5V_TVS.jpg [ 120.18 KiB | Viewed 38952 times ]
5V_power_switch.jpg
5V_power_switch.jpg [ 203.12 KiB | Viewed 38952 times ]

_________________
A backup a day keeps DR away.
Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 5th, 2016, 9:25 
Offline

Joined: March 28th, 2014, 9:11
Posts: 44
Location: Nederland
Thank you fzabkar, that's very helpful and valuable. I'll post back with my findings later tonight.


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 5th, 2016, 16:16 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15440
Location: Australia
Oops, I have incorrectly identified two of the "RAU" chips. They should all be TLV62080 devices, not CSD25401.

_________________
A backup a day keeps DR away.


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 6th, 2016, 13:13 
Offline

Joined: March 28th, 2014, 9:11
Posts: 44
Location: Nederland
I've performed the measurements (with only power cable attached, no data cable). The measurements are in the attached pictures.
Not sure if there's a plausible conclusion to be taken from this. I'll find out of the TLV chips are okay next.


Attachments:
5V_power_switch.jpg
5V_power_switch.jpg [ 208.9 KiB | Viewed 38863 times ]
5V_TVS.jpg
5V_TVS.jpg [ 102.16 KiB | Viewed 38863 times ]
LDOs.jpg
LDOs.jpg [ 130.9 KiB | Viewed 38863 times ]
MP5505.jpg
MP5505.jpg [ 67.4 KiB | Viewed 38863 times ]
V1.jpg
V1.jpg [ 103.45 KiB | Viewed 38863 times ]
V2.jpg
V2.jpg [ 50.4 KiB | Viewed 38863 times ]
V3.jpg
V3.jpg [ 47.71 KiB | Viewed 38863 times ]
Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 6th, 2016, 14:03 
Offline

Joined: March 28th, 2014, 9:11
Posts: 44
Location: Nederland
Note to self, images appear in the reverse order of uploading.

The TLV62080 near V3 seems okay while the V1 and V2 ones might not be.
Does the 0.0V measured at Vstorage indicate the MP5505 is likely to require replacement as well?


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 6th, 2016, 15:46 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15440
Location: Australia
The pins I have marked as 12V could be the problem (they should measure 3.3V). I suspected that you may have switched modular cables between PSUs, in which case the wrong voltages would have appeared on the SATA power pins.

_________________
A backup a day keeps DR away.


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 6th, 2016, 17:44 
Offline

Joined: March 28th, 2014, 9:11
Posts: 44
Location: Nederland
Ouch. After checking, it appears you're indeed correct. The same-brand (be quiet) replacement PSU has similar modular cables (including matching notches etc.) but pin assignment is rearranged.
Guess I'm lucky only the one drive was connected.
Is there hope for this SSD?


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 6th, 2016, 18:12 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15440
Location: Australia
Do you get the same measurements with the correct PSU inputs?

The resistors near the +3.3V pins would suggest that the 3.3V supply is being used to switch on the +5V supply via the P-channel MOSFET. The gate voltage at the MOSFET (+4.95V) would suggest that the MOSFET should be switched off, yet its output voltage (+5V) indicates that it is on. This in turn would suggest that there is a short between the D-S pins of the MOSFET. That said, this short should not pose a problem -- it just adds to my confusion. :?


Attachments:
SATA_power.jpg
SATA_power.jpg [ 127.76 KiB | Viewed 38822 times ]

_________________
A backup a day keeps DR away.
Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 6th, 2016, 18:17 
Offline

Joined: March 28th, 2014, 9:11
Posts: 44
Location: Nederland
The measurements I did were performed with another setup (non-modular PSU). The sata connector voltages in your last picture match. So I performed the measurements while it was correctly hooked up. But after replacing the other dead PSU it appears I used an incompatible modular cable.


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 6th, 2016, 18:24 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15440
Location: Australia
Please give me some time to digest the measurements. The fact that the 5V TVS diode survived seems like a good sign. Hopefully there is a single common factor that explains why the various regulators are not working.

_________________
A backup a day keeps DR away.


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 6th, 2016, 18:41 
Offline

Joined: March 28th, 2014, 9:11
Posts: 44
Location: Nederland
Absolutely, thanks again for looking into this. I really lack the expertise to do anything more than the most basic diagnosing so I hope to learn during the process.

I've reverified my measurements and it appears I've made one mistake, the V3 voltage is actually also 0.0V just as V1 and V2, and not 5.0V like I put in the picture.


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 6th, 2016, 21:47 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15440
Location: Australia
Check the EN(able) and ENCH inputs of the various chips. EN should be high if the chip is enabled, or low (0V) if the chip is disabled. I expect (and hope) that EN will be low, as this would point to a common factor.

There is a fourth regulator (V4) which is shrouded in darkness in one corner of the board.

The OAH LDO regulator has an input voltage (Vin) of only 2.75V. IIUC, this voltage is too low (Vout should be 3.3V), in which case the source of Vin must be suspect.

Can you measure the resistance between the Drain and Source pins of the MOSFET switch? As mentioned before, the Gate voltage is not consistent with an ON state.

Can you identify the markings on ICx and Dy (does Dy have a polarity stripe?) ?

Can you measure the voltages on the capacitors adjacent to ICy? Is the IC marking CY or QY? I think I may have incorrectly identified it.


Attachments:
V4.jpg
V4.jpg [ 78.02 KiB | Viewed 38786 times ]
TPS62067_ODH.gif
TPS62067_ODH.gif [ 9.86 KiB | Viewed 38786 times ]
TLV62080_RAU.gif
TLV62080_RAU.gif [ 10.98 KiB | Viewed 38786 times ]
MP5505_circuit.gif
MP5505_circuit.gif [ 22.16 KiB | Viewed 38786 times ]
MP5505.gif
MP5505.gif [ 39.53 KiB | Viewed 38786 times ]
ICy.jpg
ICy.jpg [ 73.02 KiB | Viewed 38786 times ]
ICx.jpg
ICx.jpg [ 314.72 KiB | Viewed 38786 times ]

_________________
A backup a day keeps DR away.
Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 7th, 2016, 8:03 
Offline

Joined: March 28th, 2014, 9:11
Posts: 44
Location: Nederland
Quote:
Check the EN(able) and ENCH inputs of the various chips. EN should be high if the chip is enabled, or low (0V) if the chip is disabled. I expect (and hope) that EN will be low, as this would point to a common factor.

Each of the 4 (3x RAU, 1x ODH) enable pins are low.

Quote:
There is a fourth regulator (V4) which is shrouded in darkness in one corner of the board.

V4 is at 0.0V.

Quote:
The OAH LDO regulator has an input voltage (Vin) of only 2.75V. IIUC, this voltage is too low (Vout should be 3.3V), in which case the source of Vin must be suspect.

There's some heat generation in this IC (see flir attachment). I'm guessing this is the primary suspect.

Quote:
Can you measure the resistance between the Drain and Source pins of the MOSFET switch? As mentioned before, the Gate voltage is not consistent with an ON state.

With the power connector attached it's 25.2 ohms. With no power it's 24.8 ohms. I also re-checked the gate voltage, it's indeed 4.95V as reported earlier.

Quote:
Can you identify the markings on ICx and Dy (does Dy have a polarity stripe?) ?

ICx is 43V6ZI I'm fairly certain. Unlikely, V may be Y, 6 may be G or 5, I may be 1 or lowercase L but I'm 95% confident my reading is correct.
Dy is much harder to read, it might be '5jw' but I'm only remotely certain about the 5. Cathode marking is on the SATA connector side. Voltage drop is 0.247V so I'm guessing it's a schottky.

Quote:
Can you measure the voltages on the capacitors adjacent to ICy? Is the IC marking CY or QY? I think I may have incorrectly identified it.

Voltages are 0.0V, ICy marking is definitely QY.


Attachments:
flir_20160807T134151.jpg
flir_20160807T134151.jpg [ 379.81 KiB | Viewed 38759 times ]
V4.jpg
V4.jpg [ 66.21 KiB | Viewed 38759 times ]
ICy.jpg
ICy.jpg [ 68.48 KiB | Viewed 38759 times ]
ICx.jpg
ICx.jpg [ 130.4 KiB | Viewed 38759 times ]
Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 7th, 2016, 8:40 
Offline

Joined: December 8th, 2010, 11:37
Posts: 738
Location: Ottawa, Canada
@zzattack,
Quote:
There's some heat generation in this IC (see flir attachment). I'm guessing this is the primary suspect.

Nice Flir image. May I ask what make/model of Flir imager you used?

_________________
Sabo Computer Repairs & Data Recovery


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 7th, 2016, 8:59 
Offline

Joined: March 28th, 2014, 9:11
Posts: 44
Location: Nederland
Simple Flir ONE android. They've been sold out everywhere or offered at an extreme premium in Europe, but I got a chance to buy one in the US for $240.


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 7th, 2016, 10:54 
Offline

Joined: December 8th, 2010, 11:37
Posts: 738
Location: Ottawa, Canada
Thanks! It's for sale for $240 USD here. Very tempting. Unfortunately, my phone is not compatible.

_________________
Sabo Computer Repairs & Data Recovery


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 7th, 2016, 15:20 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15440
Location: Australia
Short version:

ISTM that the problem area is around the OAH regulator and possibly the MSP430. We need to be patient and take enough measurements to determine how each of the supplies are connected and controlled.


Long version:

Please DO NOT measure resistance while the device is powered on. The meter injects its own small current into the circuit being tested. Under some circumstances damage may occur to the circuit. No harm appears to have resulted in this case, though.

The QY device is a 2.5V regulator. Its input voltage is 0V. Can you check whether the RAU, ODH, and MP5505 parts are also missing their input voltages?

The OAH part is enabled whereas all the other regulators appear to be disabled. It could be that the EN signals are coming from another chip (MSP430 ?), or it may be that the EN pins are pulled up to the regulator's own Vin, in which case Vin would be missing. In short, the regulators require a supply voltage plus a high signal on their EN pins.

I think it would be prudent to determine where each of the supply voltages is coming from, and where the EN signals are being generated. We can do this with resistance measurements, but first we need to see your voltage measurements.

I suspect that the MSP430V952 microcontroller may control the power sequencing for the PCB. It would be interesting to see whether it is getting hot and whether it is powered from the OAH LDO regulator. The MCU's supply voltage spec is 1.8V - 3.6V, so this seems a likely culprit. I would measure the resistance between the Vout pin of OAH and the pins of the MSP430 MCU to determine which pin is the MCU's supply pin. (There are datasheets for various flavours of the MCU, but I'm not certain if all have the same pinout.)

One other thing to do would be to measure the resistance between ground and the outputs of each of the regulators. This will tell us if the flash controller or any of the NANDs are shorted. If any of these ICs are damaged, then there would be no point in continuing.

Can you measure resistances Rx and Ry, and voltage Vy, in the MSP430 image? I suspect that Vy should be +5V. If so, then could you test the resistance between this point and the SATA +5V pins and the Drain pins (output side) of the power MOSFET? This will tell us if this 3.3V supply is our always-on standby supply. I'm not certain, but the resistor-like components with dots on their bodies look like zero-ohm links. It would be interesting to measure the resistances of other similar components for comparison purposes.


Attachments:
MSP430.jpg
MSP430.jpg [ 94.73 KiB | Viewed 38716 times ]
TLV70025_QY.gif
TLV70025_QY.gif [ 5.82 KiB | Viewed 38716 times ]

_________________
A backup a day keeps DR away.
Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 7th, 2016, 15:32 
Offline

Joined: March 28th, 2014, 9:11
Posts: 44
Location: Nederland
I did some more measurements near the 3.3V LDO. See attached picture, I believe there's a 10 Ohms resistor nearby with 4.5V coming into it and only 2.75V getting into the regulator.



Attachment:
LDOs.jpg
LDOs.jpg [ 89.8 KiB | Viewed 38716 times ]


I removed the component. It's indeed a 10Ohm resistor. Now I measure 5.0V on the left side and 0.0V on the input on the regulator.


Top
 Profile  
 
 Post subject: Re: Crucial MX100 repair after PSU failure
PostPosted: August 7th, 2016, 15:54 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15440
Location: Australia
FWIW, here is a datasheet for the MSP430F1122 chip:
http://www.ti.com/lit/ds/symlink/msp430f1122.pdf

It appears to have a different pinout, but its max current rating in active mode at Vcc=3V is 350uA. However, a voltage drop of 1.75V across a 10 ohm resistor corresponds to a current draw of 175mA, so I'd say that the MSP430 is dead, assuming that it is the only device powered from this supply. You could isolate the capacitor on the 3.3V output, just to be certain that it is not the culprit.

Edit: A similar voltage drop across the 10 ohm resistor at Vout would account for the 0.9V measured voltage (2.6V - 1.75V = 0.85V).

Edit #2: Measure the resistance between the 0.9V point and ground.

_________________
A backup a day keeps DR away.


Last edited by fzabkar on August 7th, 2016, 16:08, edited 1 time in total.

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 113 posts ]  Go to page 1, 2, 3, 4, 5, 6  Next

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 21 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