WD40NMZW PCB Issues

[ddrecovery]

I have had a couple of these USB drives in recently where we have swapped the ROM over to a matching SATA board. Not an issue in itself, but I have recently had 2 SATA boards just stop responding after doing some work in PC3000 such as blocking SA, turning off heads etc. The 5v and 12v status lights come on, but the PHY and BSY, DRD & DSC lights do not light up and the drive no longer spins. I have re-written the ROM data back to the chip in a reader thinking it might be corruption, but get the same result. These SATA boards are either very hard to find or very expensive and I want to get to the bottom of the issue if possible.

Anyone else had the same issues?

[rogfanther]

Are you removing the chips and using an external programmer ? Then have you checked soldering in the board , both on the flash chip and in the nearby components ?

[ddrecovery]

Yes we have been using an external programmer. All solder points around the board look fine. However I am not sure if bending the board while isolating the heads with this model of board might be causing something to crack as the connector is so big. But it does seem strange as it only happens when writing to the ROM.

I did notice a post from digisupport who said he was soldering a SATA connection to the USB board. Anyone know which contact pads are relevant with this board? Or are we just soldering data wires on the CPU side of the board?

[fzabkar]

I did notice a post from digisupport who said he was soldering a SATA connection to the USB board. Anyone know which contact pads are relevant with this board? Or are we just soldering data wires on the CPU side of the board?

Can you upload photos of both side sof both PCBs? I can help you test the voltages as well.

[rogfanther]

Which software do you use ? You could reprogram the rom in the board without needing to remove its chip

[jono-ats]

I had one of these die after programming the ROM through the SATA port.

Swapping ROMs didn't help.

I wonder if it is a NAND memory problem?

[ddrecovery]

Can you upload photos of both sides of both PCBs? I can help you test the voltages as well.

Sure thing.

Which software do you use ? You could reprogram the rom in the board without needing to remove its chip

I have PC-3000. AFAIK there is no way to download the ROM from direct from this type of USB board (unless anyone knows different?)

Attachments

[fzabkar]

I had one of these die after programming the ROM through the SATA port.

Swapping ROMs didn't help.

I wonder if it is a NAND memory problem?

Which chip would that be?

[jono-ats]

The big rectangular one.

[fzabkar]

Do you mean this big rectangle?

Winbond_W9751G6KB-25_big_rectangular_one.jpg (70.72 KiB) Viewed 14922 times

https://www.winbond.com/resource-files/da00-w9751g6kbg1.pdf

GENERAL DESCRIPTION. The W9751G6KB is a 512M bits DDR2 SDRAM, organized as 8,388,608 words × 4 banks × 16 bits.

[fzabkar]

In the unlikely event that both boards have failed, I would measure the voltages at the inductors and/or capacitors.

Attachments

[ddrecovery]

Thanks for the help.

Attachments

TVS_regs.jpg (91.31 KiB) Viewed 14897 times

[rogfanther]

What is the code of the flash chip ? ( that would be the small, 8 legged one near the screw hole, right ? )

@jono-ats, can you get a dump ( by external programmer ) from the chip in a killed board ? And a dump of one chip from a working board ? Maybe comparing both we could discover something.

Also, in a damaged board to the OP, have you tried resoldering those components around the flash chip ?

[ddrecovery]

Also, in a damaged board to the OP, have you tried resoldering those components around the flash chip ?

Yes I have. I have also tested most of the components and all test out fine. I did take a dump of the ROM after the first failure (attached). It is pretty corrupt so not sure if we can learn anything. But reinstalling the correct ROM to the chip does not resolve the issue.

Attachments

Corrupt ROM.7z

(290.96 KiB) Downloaded 455 times

[rogfanther]

Could you post a dump of the good rom that was written to that chip also ?

[rogfanther]

Oh, could you post pictures for the identification of the original flash chip in the USB board, and the original flash chip in one of these Sata boards, if you still have them ?

[fzabkar]

@rogfanther, here are better photos:

2060-800041-000 REV P1 USB-to-SATA conversion:
http://i.imgur.com/LdT6YM8.jpg
http://forum.acelaboratory.com/viewtopic.php?t=9615

[rogfanther]

Thanks Frank, would you have one from a SATA board with its original flash chip ?

[fzabkar]

This is the only photo (original resolution):

SPI_flash.jpg (68.24 KiB) Viewed 14872 times

[rogfanther]

I´m kinda thinking there could be two issues.

The one pointed by @ddrecovery, where after some desolder/reflash/solder cycles the board stops working, could be due to some effect of the heating in that corner region of the big chip. Either in its pins , or on those vias around the chip. Something that could cause intermitent contacts, so to generate some spurious commands to the chip and corrupting its data.

Correct me if I´m wrong, but the flash chip wouldn´t be written to during normal operation of the hdd, no ?

The other, pointed by @jono-ats, where programming the rom "killed " the board, could be something like that other model ( Pebble ?) where different families have different sized roms. Or even if the VSCs to write to that roms through SATA are not correct for this family.