ATMLH828 ROM chip on 7200.11 drive, no spin!!

[michael chiklis]

Customer gave me this drive today:

Barracuda 7200.11
ST3320613AS
FW: SD22
PCB: 100504364
PCB sticker: 100504353
ROM chip: ATMLH828 never seen this type of rom on 7200.11 drives

Initial symptom was NO SPIN, this is what customer told me.
He told me that he first gave the drive to a pro DR few weeks ago, they have shown him file list but he didn't accept their quote (too much expensive), so now the drive is in my hands.
I see that on the pcb they have unsoldered 12V TVS diode and signs of soldering on rom chip.
Drive spins when is attached donor pcb with not native rom ST25P40VP but of course heads do click due to not original adaptives.

I unsoldered the rom chip from original pcb and read the content:

ATMLH828 ROM.rar

(244.85 KiB) Downloaded 662 times

then i soldered original rom to donor pcb but drive doesn't spin (no sounds).
Then i wrote the original rom content into donor ST25P40VP rom chip, i soldered ST25P40VP on donor pcb but still doesn't spin.

I suspect that DR stolen original rom chip and put a fake one on because customer refused recovery so no one can recover data exept them, what do you think?
Someone can tell me if the rom i posted is good or not for this drive?

Original ATMLH828 rom (the one i found on original pcb) is 256 Kb
Donor ST25P40VP rom from my donor pcb is 512 Kb

Opinions?

then i soldered it to donor pcb (same codes of the original one) but the drive doesn't spin

[fzabkar]

The M25P40VP is a 4M-bit (512KB) device, as you have said.

M25P40VP, ST Microelectronics, 4 Mbit (512 K + Serial Flash memory:
http://pdf.datasheetcatalog.com/datashe ... s/7737.pdf
http://www.farnell.com/datasheets/40267.pdf

The ROM dump should contain the drive's serial number. In an ST3500320AS with SD1A firmware, this is located at offset 0x61644. It does appear that your dump is missing the second half of the ROM.

A quick search for "ATMLH828" doesn't turn up any datasheets amongst my usual sources. Perhaps the other numbers on the chip might help. In fact I'm wondering whether the "828" is a date code.

[michael chiklis]

Here some pics of the rom chip, not clear bit i did the best i could:

[fzabkar]

The "7H0553" looks like it might be a batch code. AISI, the "53" discounts it as a possible date code.

Obviously it's an Atmel part. I think that the "S4" marking probably identifies it. I suspect that the "4" might mean that it is a 4-Mbit part, in which case there could be something wrong with the way you are reading it. Have you managed to get the part's manufacturer and product ID via your programmer?

BTW, what are the date codes on the other chips and on the drive's label?

[fzabkar]

The part is an Atmel AT25F4096:
http://www.baite-group.com/uploadFiles/ ... 071201.pdf

AT25F4096, Atmel, SPI Serial Flash Memory, 4Mbit (524,288 x , 2.7 - 3.6V, marking "S4 3":
http://datasheet.octopart.com/AT25F4096 ... -39730.pdf

[fzabkar]

Sorry, the S4 part is an AT25FS040. That's the functional replacement for the obsolete AT25F4096.

http://www.mouser.com/ds/2/36/doc5107-27762.pdf

[michael chiklis]

That's right fzabkar, my eprom detects the chip as AT25FS040

[fzabkar]

That's right fzabkar, my eprom detects the chip as AT25FS040

Then either the chip is bad, or your chip programmer/reader is not reading it correctly.

[michael chiklis]

Problem solved!
chip programmer/reader was not reading it correctly.

The solution was to manually select the chip in the Willem software, autodetection doesn't detect properly this Atmel chip.
I don't know why, maybe software bug?
I wrote the full rom on chip ST25P40VP and soldered it on donor pcb.

full rom.rar

(431.04 KiB) Downloaded 536 times

Usually i just select "autodetect" in Willem (rom chip are usually ST or Winbond which are detect with correct size), this is the first time i see Atmel rom for 7200.11 drive which wasn't detect with correct size with automatic detection.
Now drive is spinning and calibration is correct.

IMPORTANT
The Atmel chip was not working on donor pcb, so i guess even the pcb doesn't detect it properly.
Maybe DR purposely used Atmel that hot air soldering swap wouldn't work?

[michael chiklis]

Maybe DR purposely used Atmel so that hot air soldering swap wouldn't work?

[fzabkar]

The Atmel chip was not working on donor pcb

The serial number at offset 0x61644 is "6SZ086E4". Does that match the label?

Do you have any terminal output?

Do you have any tools that can verify the integrity of the ROM? Can you verify the checksums of the code and adaptive sections of the ROM?

[michael chiklis]

Customer emailed to Seagate.
Seagate confirmed that Atmel chip is not original on this pcb, so this means that DR changed it.
They have answered also that this type of thing do happen only in Italy.

P.S.
DR asked to customer 890€ for just pcb issue.

[michael chiklis]

All data are recovered!!
It was just pcb faulty, DR should charge not more then 400Є for this job.
They have asked to customer 890Є because they are thieves

[fzabkar]

Who sold you the bad donor?

[michael chiklis]

Who sold you the bad donor?

I got donor pcb from china shop.
Donor pcb is good, also rom ST25P40VP on donor pcb was good (i written on it the rom content of my customer and worked fine).
The only bad thing was the Atmel rom chip that i found on original customer pcb, this because DR company changed it, probably original rom was ST or Winbond

Terminal Log.txt

(383.9 KiB) Downloaded 932 times