WD3200BEVT-22CTO - BAD ROM

[mhp666]

This disk has the ROM on the MARVELL chip, and it has a ROM that is not correct.
When I turn on the PCB it does not give any signal.

Is there any way to more or less easily reprogram the ROM without removing the MARVEL from the board?

This PCB carries the 88i6745N-TFJ1.

It is a test of learning.

[pikus_09]

Remove, R6 and R13 and program external flash chip. works on 100%

[mhp666]

I just found a topic that was already open before.
https://forum.hddguru.com/viewtopic.php?f=13&t=20324

another
https://www.malwaretech.com/2015/04/har ... art-1.html

I'm pretty interested if I can make some progress I'll publish it.

[mhp666]

Remove, R6 and R13 and program external flash chip. works on 100%

Thanks for answering.
I put a picture of the PCB.

Attachments

[pikus_09]

here is:

Attachments

[mhp666]

Remove, R6 and R13 and program external flash chip. works on 100%

I do not understand what you mean.
I remove the resistances and after that I do?

You could explain it a little more detailed.

Thank you.

[fzabkar]

Those resistors are configuration resistors. They program the relevant MCU GPIO pin to be either a 0 or a 1. The MCU's kernel code reads these GPIO pins at startup and then loads either the internal ROM code or the external flash memory. By removing these resistors, you are telling the MCU to boot from the external ROM.

BTW, doesn't MRT enable you to read and write the ROM via the terminal port?

[pikus_09]

Ehh, i mean, that if Main MCU have corrupt Mask rom. This are 2 options. First is play with Jtag and trying program mask rom, or Boot PCB from External Flash chip. And removing this 2 resistors R6 and R13 Main MCU can Start With External Flash Chip with correct ROM content.
If you remove this 2 resistors, this now program flash chip SO8 For example Pm25LD020 or W25X20 with correct rom file, and solder to PCB.
And run disk again.

[mhp666]

Thank you very much for your answers, I already have it clear.
I will leave it documented that it is very interesting.

BTW, doesn't MRT enable you to read and write the ROM via the terminal port?

I have tried but the PCB does not respond to anything.
I tried to select it as "Marvell COM port", and it does not obey either.
Any ideas?

I'll test the resistance and comment.
I will also get some jtag that has support for these chips.

[fzabkar]

I have tried but the PCB does not respond to anything.
I tried to select it as "Marvell COM port", and it does not obey either.
Any ideas?

Can you tell us the voltages at each end of those two resistors?

It should be possible to temporarily reconfigure the GPIO pins without removing these resistors. Then perhaps you will be able to communicate via the UART port.

[mhp666]

The 2 have the same value
2.85v and 3.33v

Tomorrow I will remove them and put the value without them.
Greetings.

Attachments

IMG_20180930_222038.jpg (95.2 KiB) Viewed 10337 times

[fzabkar]

You don't need to remove the resistors. Just short the GPIO_A and GPIO_B test points to ground before you apply power to the PCB, then release the shorts after the board powers up.

Attachments

R13_R6_config_resistors.jpg (65.91 KiB) Viewed 10329 times

[mhp666]

You don't need to remove the resistors. Just short the GPIO_A and GPIO_B test points to ground before you apply power to the PCB, then release the shorts after the board powers up.

With this, can I read the internal ROM or is it to program an external ROM?
I will try it and comment it.

[fzabkar]

You don't need to remove the resistors. Just short the GPIO_A and GPIO_B test points to ground before you apply power to the PCB, then release the shorts after the board powers up.

With this, can I read the internal ROM or is it to program an external ROM?

I believe that the MCU will see that R13 and R6 are missing, and then try to boot from an external ROM. Because there is no ROM at U12, the MCU should then wait for an upload via the UART port. Or so I think ...

[mhp666]

I believe that the MCU will see that R13 and R6 are missing, and then try to boot from an external ROM. Because there is no ROM at U12, the MCU should then wait for an upload via the UART port. Or so I think ...

Effectively, when placing GND values of 2.85v, it is as if we removed the resistances.
I have made a GND socket using a push button with the intention of loading an external ROM and then release the button and be able to record in the internal ROM. The problem is that once the MCU starts, it is configured for external or internal and it continues like this until the next power on, whether or not we change the values of the gpios in question.

Then after copying a ROM from another disk of the same model, restoring the HDD is very fast.

1. The external ROM is recorded. Power off, Power on.
2. The disc has access to read tracks.
3. We look for modules in the SA by tracks.
4. We rebuild the ROM from the SA.
5. We recorded the new ROM.
6. We turn off and on and the disk works perfectly.

Greetings and thank you for your help...