All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 9 posts ] 
Author Message
 Post subject: Crucial MX300 not recognized in BIOS
PostPosted: December 28th, 2017, 0:28 
Offline

Joined: December 27th, 2017, 23:53
Posts: 4
Location: United States
I have a 1TB MX300 that is not recognized in BIOS after pressing power button to turn off the laptop. I was using the drive to store all my backups as well as my main drive while I reformatted the backup drives. Surely, the machine would never just lock up and the hard drive fail during those two hours, right? Wrong.

I have already tried the recommended steps by Crucial to plug it in with no SATA connection for some time, several times.

The data is not worth the expense of professional recovery, and if it was, I would still prefer to spend the money on recovery tools or a nice oscilloscope. So on to troubleshooting:

The machine recognizes something is there, because Linux recovery boot disks will try to initialize it. MHDD simply says "drive not ready," and it looks like it's dependent on BIOS initialization even though it shows the ports. It also shows a PC3000 port even though I don't own one of those.

No data seems available online about the Marvell chip the MX300 is using, and the screenprinting on the board gives no clues. There are two pairs of ports that look like JTAG, but the datasheets for the chips near them indicate SpyBiWire interface, so I'm guessing they aren't even the port I would need to connect to. There's an unsoldered header for a 10 pin port with large pads behind it. This would make sense for an ARM 10-pin or 14-pin JTAG port. There's a small row of 4 pins right up next to the flash chips. Again, I'm guessing this is not the interface to the correct chip.

Is there some freely available software that allows me to send ATA commands directly to a port without relying on the BIOS?

Does anyone know of a good place to find the pinouts for the board, and/or expected voltages at test points? I do have a hot air station if it's just a regulator issue. Regulator issues are probably what I'll try to eliminate next for all the parts I can find datasheets on.

I'd also take suggestions for paid software that is affordable to a normal person, I just don't have several thousand dollars laying about. Thanks in advance for any suggestions.


Top
 Profile  
 
 Post subject: Re: Crucial MX300 not recognized in BIOS
PostPosted: December 28th, 2017, 0:47 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15461
Location: Australia
There are several threads which identify the voltage test points for various Crucial SSDs or their equivalents. If you upload detailed photos (or CCD scans) of both sides of the PCB, I could help you with these.

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

_________________
A backup a day keeps DR away.


Top
 Profile  
 
 Post subject: Re: Crucial MX300 not recognized in BIOS
PostPosted: December 28th, 2017, 2:47 
Offline

Joined: December 27th, 2017, 23:53
Posts: 4
Location: United States
Attached are images. You'll see the available test points aren't really similar to the crucial boards shown online. I have a smaller MX that is also unlabeled. It works but has a different layout entirely.

IR images were not instructive. It heated up exactly where I expected, so I didn't include them.

Headers visible on back of the board. I'll get poking at them tomorrow.


Attachments:
File comment: Closeup of remaining chips on back
IMG_20171228_003252.jpg
IMG_20171228_003252.jpg [ 3.37 MiB | Viewed 10693 times ]
File comment: Closeup chips right of both controllers
IMG_20171228_002730.jpg
IMG_20171228_002730.jpg [ 4.2 MiB | Viewed 10693 times ]
File comment: Closeup chips up, left of Marvell controller
IMG_20171228_003209.jpg
IMG_20171228_003209.jpg [ 3.62 MiB | Viewed 10693 times ]
File comment: Closeup main controllers
IMG_20171228_002454.jpg
IMG_20171228_002454.jpg [ 2.76 MiB | Viewed 10693 times ]
File comment: Front of the board
IMG_20171227_235824.jpg
IMG_20171227_235824.jpg [ 4.28 MiB | Viewed 10693 times ]
File comment: Back of the board
IMG_20171227_235858.jpg
IMG_20171227_235858.jpg [ 4.55 MiB | Viewed 10693 times ]
Top
 Profile  
 
 Post subject: Re: Crucial MX300 not recognized in BIOS
PostPosted: December 28th, 2017, 4:45 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15461
Location: Australia
I hope I have identified these ICs correctly:

TLV62084, Texas Instruments, marking SLO, 2A High Efficient Step Down Converter, 2x2mm WSON, 8-pin, 2.7 - 6V Input:
http://www.ti.com/lit/gpn/TLV62080

TLV70028, Texas Instruments, marking NL, 2.8V, 200mA, Low IQ, LDO Regulator, WSON, 6-pin:
http://www.ti.com/lit/ds/symlink/tlv700.pdf

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

MSP430, Texas Instruments, 16-Bit Ultra-Low-Power Microcontroller, 32-pin QFP:
http://www.ti.com/lit/ds/symlink/msp430f1122.pdf (not identical chip)

N25Q032A, Micron, 32Mbit, SPI serial flash memory, 8-pin, 2.7 - 3.6V, quad IO:
https://www.micron.com/~/media/document ... v_65nm.pdf

The 4-pin IC marked "2H=F" is probably a Richtek device. I suspect it may be a power supply supervisor.

The MX100 SSD in the following thread sustained a failure of the MSP430 IC. The OP provided much useful feedback.

viewtopic.php?f=10&t=33908

I would start by measuring the voltages at the inductors at the outputs of each of the buck regulators. Then check the LDO output and the voltage on the backup capacitor array.

_________________
A backup a day keeps DR away.


Top
 Profile  
 
 Post subject: Re: Crucial MX300 not recognized in BIOS
PostPosted: December 28th, 2017, 12:34 
Offline

Joined: December 27th, 2017, 23:53
Posts: 4
Location: United States
First of all, thank you! I didn't expect you to go to all that effort, but it will certainly save me some time.

I looked through the link you provided, and I'm guessing my failure mode will not be physical like his, where his PSU failed. Since non-detection after power off is a common problem with these drives which is usually fixed by power cycling, I'm still suspecting this is a software problem.

I'll try to eliminate regulator failure first, hopefully after work tonight, but I think I will eventually need to get a JTAG connection to that Marvell chip in order to give it some sort of reset command, if I can't get software that can get it responding normally through the SATA port.

Related to the other guy's problem: I may be able to use a logic analyzer on the EN pins to determine the initialization order and timing of the power circuits. Would that be useful information for others in the future, or too far beyond what people normally need?


Top
 Profile  
 
 Post subject: Re: Crucial MX300 not recognized in BIOS
PostPosted: December 28th, 2017, 15:22 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15461
Location: Australia
ZHN wrote:
I looked through the link you provided, and I'm guessing my failure mode will not be physical like his, where his PSU failed. Since non-detection after power off is a common problem with these drives which is usually fixed by power cycling, I'm still suspecting this is a software problem.

I'm not involved in data recovery, but AIUI the drive may be taking some time to repair and reconstruct its translator. The "fixes" suggested by the SSD manufacturers typically require that the drive be left powered on for several minutes, after which a system reset is applied.

ZHN wrote:
I'll try to eliminate regulator failure first, hopefully after work tonight, but I think I will eventually need to get a JTAG connection to that Marvell chip in order to give it some sort of reset command, if I can't get software that can get it responding normally through the SATA port.

These SSDs typically have a UART port. You may be able to see some diagnostic output.

Examining a firmware update may prove useful. Sometimes you will see diagnostic commands (for a debug monitor) in plain text.

You can sometimes place the SSD in "safe mode" by shorting appropriate test points.

ZHN wrote:
Related to the other guy's problem: I may be able to use a logic analyzer on the EN pins to determine the initialization order and timing of the power circuits. Would that be useful information for others in the future, or too far beyond what people normally need?

It probably won't be of use to the typical "chip-off" DR shop, but I would find it of academic interest.

_________________
A backup a day keeps DR away.


Top
 Profile  
 
 Post subject: Re: Crucial MX300 not recognized in BIOS
PostPosted: December 30th, 2017, 17:04 
Offline

Joined: December 27th, 2017, 23:53
Posts: 4
Location: United States
A quick check of voltage levels around the board didn't reveal anything amiss. I've not had time to check every chip yet, but I really don't expect to find anything damaged because I have not done anything that should cause damage.

Yes, there does appear to be a 2.75v serial port with a floating RX up near the flash chips, however I don't have any idea what commands it would accept. Other manufacturers who do document their serial ports seem to have a limited mode enabled all the time and an engineering mode that you can enter with the right command. Since this isn't exposed to the consumer, I don't expect ASCII "?" will give me a command list. Pin order is GND, ?, RX, ?, where one of the ? is TX and one is +2.75. My guess is GND, TX, RX, +2.75, but it will be easy enough to figure out for sure once I have a header on it.

Some people on the Crucial forums mentioned that if left plugged in long enough, the controller seems to leave the loop it gets stuck in and respond when it's plugged in hot. Without a desktop motherboard, I don't have an appropriate port to try that. I've had power applied for about 26 hours now, it's on, and nothing is getting warm except the main controller, which is not even reaching 30C. I'm going to build a SATA data cable with a male end scavenged from a dead motherboard and attempt to plug it in hot to see if it will respond before I try anything else.

If I can pull the data, I'll still play around with the drive as time allows later and update with my findings. One of their techs mentioned that there is nothing stopping me from disassembling the firmware. I didn't say anything to indicate I would have the interest or ability to do that (the answer is maybe, but it would take me a long time). I'm taking this as indication there are useful undocumented ATA commands to find in there, because I don't know why else they would mention it as an option.

Once the data is pulled, it might be quicker to write a fuzzing program to see what commands create a response. I can't try yet in case I encounter some kind of "wipe" or "lock" command besides the ATA-spec ones. I've already learned that lesson the hard way on a PLC owned by my workplace when I discovered an undocumented "wipe all bootloaders" command and had to disassemble the device to restore it.


Top
 Profile  
 
 Post subject: Re: Crucial MX300 not recognized in BIOS
PostPosted: December 30th, 2017, 17:19 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15461
Location: Australia
Just in case it helps someone else ...

DMP2035U, Diodes Inc, marking MP3, SOT23, P-ch Enhancement Mode MOSFET, -20V, -4.9A:
https://www.diodes.com/assets/Datasheets/DMP2035U.pdf

Analysis of Micron RealSSD C400 SED SSD Firmware Update:
http://www.hddoracle.com/viewtopic.php?f=59&t=1617

_________________
A backup a day keeps DR away.


Top
 Profile  
 
 Post subject: Re: Crucial MX300 not recognized in BIOS
PostPosted: December 30th, 2017, 17:31 
Offline
User avatar

Joined: September 8th, 2009, 18:21
Posts: 15461
Location: Australia
ZHN wrote:
Yes, there does appear to be a 2.75v serial port with a floating RX up near the flash chips, ... Pin order is GND, ?, RX, ?, where one of the ? is TX and one is +2.75. My guess is GND, TX, RX, +2.75, but it will be easy enough to figure out for sure once I have a header on it.

The 2.75V supply may be coming from the "NL" IC. A continuity test would verify this. Otherwise you could load the pins with a resistor to ground and look for a voltage drop.

_________________
A backup a day keeps DR away.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 9 posts ] 

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: pepe and 47 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:  
cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group