Switch to full style
Data recovery and disk repair questions and discussions related to old-fashioned SATA, SAS, SCSI, IDE, MFM hard drives - any type of storage device that has moving parts
Post a reply

st1000lx015

January 7th, 2018, 22:59

Yep.....
Combined HD & flash storage, maybe 3-4 months old, good location, good cooling , not banged about.

Was working fine, no SMART errors a week ago test ,switched off computer came back 1 hr later turned it on and the drive is completely dead.

no spinup, no sound, swapped it to another machine , same result.

Quick look, nothing appears burned out, nothing hot

Any known issues with these drives?

Re: st1000lx015

January 8th, 2018, 2:45

code_slave wrote:Any known issues with these drives?

There are some, but qualified diagnostics is required.

Re: st1000lx015

January 8th, 2018, 4:48

If it's not spinning it could be a Power issue on the pcb or it could be sticktion...if you put Power on it, is there a faint beeping sound or buzzing or is it totaly dead?

Re: st1000lx015

January 8th, 2018, 4:54

Bad NAND flash?

To exclude a possible PCB fault, I could help with voltage test points, but I would need a detailed photo of the component side of the PCB.

Re: st1000lx015

January 9th, 2018, 6:14

Ok after 1 day of trying to find out WHY my cat chewed.....TTY board did not work.

The Drive, there is absolutely no sound what so ever, does not even get warm after 1 hr of power on.
Checking the 3 pin motor connector, there is no power or waveform on any of the connections, when metered or scoped

Serial connection I see this:

Boot 0x80M
DensityList:0x0000
LED:0x000000BD FAddr:0x00009C4C

LED:0x000000BD FAddr:0x00009C4C

LED:0x000000BD FAddr:0x00009C4C

LED:0x000000BD FAddr:0x00009C4C

..... continue as above

There is no response to ctrl-z ,ctrl-a etc.

Basically i just want to get it back on line so i can read the files.

Re: st1000lx015

January 9th, 2018, 7:05

I think it is Bad NAND flash. It is possible to recover the data. You should visit someone who know what to do.

Re: st1000lx015

January 9th, 2018, 19:02

Yep there is the kahuna fix which seems to apply to prior drives with flash.

I was thinking about a PCB replacement.
But I don't think there is a EEprom chip fitted. (I have a way to pull content from EEproms, without desoldering),
but this PCB does not appear to have a standard one,
(I may have a EEprom potential candidate...just looking at the PCB traces,but it means a new pogo pin Jig side connection not top hat)

It was just on the off chance one of you guys had run into the same shit and could save me some time.

I was also thinking about JTAG.
as regards the "locked" drive, i see that these drives can still have the user update the internal seagate firmware, some time ago i had thoughts of this as a side channel

Re: st1000lx015

January 9th, 2018, 19:06

Can you upload a detailed photo of the PCB?

Re: st1000lx015

January 9th, 2018, 20:19

PCB.

I suspect the EEprom is near the connector for the terminal, but with 6 pins instead of the usual 8

I'm also going to disable the CE line for the nand, this might cause the PCB to think the nand is not fitted , since it cannot read the chip ID nd this might bypass the code and put it into an init mode
Attachments
pcb.jpg

Re: st1000lx015

January 10th, 2018, 1:27

viewtopic.php?f=1&t=35857
viewtopic.php?f=1&t=34804

Re: st1000lx015

January 10th, 2018, 5:47

Jez.....

looks same as
viewtopic.php?f=1&t=35857

that pcb is serial eprom 25S81

MyPCB is 3A6K, is it the same serial eprom part or is there another supplier ?

ok looks like pogo pin time and a readup of the datasheets to see if I have a suitable emulator in my stable

and a big thanks for the amount of time you guys have saved me...fzabkar

Also the component marked "A1" is a TVS diode , not a fuse, it is clearly banded at one end (same as my pcb) and reads as a diode on a multitester.

Re: st1000lx015

January 10th, 2018, 15:01

3A6K is W25Q80BWBYIG.

    3A = W25Q80BWBYIG
    6 = year (2016)
    K = work week

W25Q80BWBYIG, Winbond, marking 3Axx, 8M-bit, 1.8V, serial flash memory, package WLBGA-8:
https://www.tvsat.com.pl/pdf/W/w25q80bw_win.pdf
http://www.srdzkj.com/pdf-1/W25Q80BW.pdf

Re: st1000lx015

January 10th, 2018, 22:54

Thanks again.....

I cannot believe that Seagate is using the likes of "winbond" as a parts supplier, their reliability is not upto par, we used to buy part from them.
The epoxy they used was hygroscopic & the spec of the parts changed depending on the humidity.....

I'm off to get a replacement drive lets see how it goes....

Re: st1000lx015

January 11th, 2018, 2:59

Are you able to find a firmware update at Seagate's Download Finder? It won't help you fix your problem, but the firmware modules may have some useful information. For example, it may tell us if the ROM architecture differs from regular F3 ROMs.

If you are so inclined, we could measure the onboard voltages for the benefit of others. I suspect that the 6-pin IC may be an LDO regulator. Perhaps disabling the NAND supply would achieve the same end as disabling CE?

Re: st1000lx015

January 11th, 2018, 4:01

What is the use of typing 100500 messages which are not in the least concern with the solution? If TS really needs data its better to ask who really CAN handle this drive. In fact, there are VERY few people around who can.

Re: st1000lx015

January 11th, 2018, 4:29

@"Masterclass", if you have something useful to add, then please do.

Re: st1000lx015

January 12th, 2018, 0:46

fzabkar wrote:if you have something useful to add, then please do.

I did! In fact, my advice is much more useful than posting pseudo-useful pictures and "datasheets". Obviously, the drive has NAND troubles. Its features go deep to code and its really useless to check voltage, clean with duster or replacing DVI-VGA monitor cable. There is no any fw update for these drives and even if there were - it would be also useless.

Re: st1000lx015

January 12th, 2018, 2:54

Other Seagate SSHDs use an ST1L05CPU33R 3.3V LDO regulator to power the NAND flash. Its pinout looks very similar. Perhaps the NAND could be disabled by driving the regulator's EN pin low.

ST1L05CPU33R, ST Microelectronics, LDO regulator, 3.3V, 1.3A, marking 1L05C33, DFN6:
http://www.st.com/web/en/resource/techn ... 186925.pdf
Attachments
3V3_LDO_reg.jpg
3V3_LDO_reg.jpg (75.9 KiB) Viewed 10085 times

Re: st1000lx015

January 12th, 2018, 5:55

fzabkar wrote:Perhaps the NAND could be disabled by driving the regulator's EN pin low.

Are you joking?
Take Kahuna drives for example (drives which are much more simple than Rosewood ones). If you know the field well, you might remember that the solution is far from hardware modifications. Why do you advice to do the things which 100% will not help? To make people spoil their drives?

Re: st1000lx015

January 12th, 2018, 8:43

Masterclass wrote:Why do you advice to do the things which 100% will not help? To make people spoil their drives?

How will disabling the NAND "spoil" the drive?
Post a reply