Help with OCZ ARC100 120GB ssd water damaged

[n3dv3d]

Hi,

I have the SSD that it has been water damaged. After an accurate cleaning I tried to connect it to pc3000 but I have clear green status LEDs so I suppose there is an electrical problem.

After trying a few measurements, there doesn't seem to be any short circuits. Does anyone have any suggestions? What components could I check?

Thanks in advance for your help

[fzabkar]

Measure the voltages at the inductors.

[n3dv3d]

Hi,

thanks for your reply.

Below a photo with some measurements:

Attachments

[Lardman]

Things look in general, low by about 0.2v , but I doubt that's your issue. What the current draw like ?

[n3dv3d]

About 0,3A

[Lardman]

Check the caps around the nand for voltage

[fzabkar]

R72 (10R0) has 5V on one side and 1.6V on the other. This means that it would be dissipating 1W. Also, the resistance between ground and the 1.6V side of the resistor would be 5 ohms. Clearly something is amiss here. I would measure the resistance of R72, and check for 5 ohms to ground.

[fzabkar]

This photo has a "000" link at R72:

https://images.bit-tech.net/content_images/2013/09/ocz-vertex-450-256gb-review/vertex450-7b.jpg

This makes U15 seem pointless. :-?

U15 appears to be a P-channel MOSFET, possibly configured as a high-side load switch for soft starting:

https://www.diodes.com/assets/Datasheets/DMP2035U.pdf

[fzabkar]

This is my theory, but ICBW.

I believe that Q24, U15 and R72 constitute a soft-start circuit. Initially after power-on, the SSD receives its 5V supply via R72. This 10-ohm resistor provides inrush current limiting.

The 3 DC-DC converters (U17, U19, U20) then power up partially, providing some power to the flash controller's IO section. A signal from this section then switches on the base of Q24, which then switches on U15, which in turn shunts R72. The 3 converters then receive a full 5V supply via U15 and then begin regulating at their specified voltages.

In the present case Q24 is not being switched on, so this would suggest that the flash controller may be dead.

If you wish to bypass the soft start, replace R72 with a shorting link, or flow a blob of solder over it. Of course, test for shorts before considering this approach.

[fzabkar]

I believe that Q24, U15 and R72 constitute a soft-start circuit.

Sorry, all references to Q24 should instead be Q4.

[n3dv3d]

Hi,

first of all thanks for your reply.

Check the caps around the nand for voltage

The caps seem to be good

This is my theory, but ICBW.

I believe that Q24, U15 and R72 constitute a soft-start circuit. Initially after power-on, the SSD receives its 5V supply via R72. This 10-ohm resistor provides inrush current limiting.

The 3 DC-DC converters (U17, U19, U20) then power up partially, providing some power to the flash controller's IO section. A signal from this section then switches on the base of Q24, which then switches on U15, which in turn shunts R72. The 3 converters then receive a full 5V supply via U15 and then begin regulating at their specified voltages.

In the present case Q24 is not being switched on, so this would suggest that the flash controller may be dead.

If you wish to bypass the soft start, replace R72 with a shorting link, or flow a blob of solder over it. Of course, test for shorts before considering this approach.

I tried to bypass the soft start as your suggestion and this is the situation:

Attachments

[fzabkar]

L3 and L4 don't seem right. At least one of them must be Vcore. What resistance do you measure between ground and each of L3 and L4?

[n3dv3d]

L4 61ohm
L3 1,6Kohm

[fzabkar]

Check the EN1, EN2 and EN3 pins. They should all be high.

https://www.ti.com/lit/ds/symlink/tps652510.pdf

[n3dv3d]

EN1 -> 2,5V
EN2 -> 0,09V
EN3 -> 2,1V

[fzabkar]

It appears that buck regulator #2 is not being turned on. One of the other regulators (L3) is enabled but has no output or an output which is too low, so it is either faulty or overloaded.

This is my guess for the power distribution:

Code:

                      DC-DC converters       Main ICs
                      .--------------.    .-------------.
            R72       |              |    |             |
  +5V       ___   Vin |     U20      |    |    NAND     |
    o--+---|___|--+-->|              |    |             |
       |    10R   |   |     U19      |--->|    SDRAM    |
       |          |   |              |    |             |
       +----+^+---'   |     U17      |    |  Controller |
       |    |||       |              |    |             |
      .-.   === U15   |              |    |             |
      | |     |       |              |    |             |
      | |     |       |              |    | PSU         |
      '-'     |       |   ENable     |    | control     |
       |      |       '--------------'    '-------------'
       '------+             ^                | |
              |             |                | |
      ___   |/              '----------------' |
  .--|___|--|   Q4                             |
  |         |>                                 |
  |           |                                |
  |           |                                |
  |          ===                               |
  |          GND        <-- PSU-ON             |
  '--------------------------------------------'

AISI, the SSD is stuck in soft-start mode. All the current for the SSD passes through R72. Applying Ohm's Law, we calculate the current draw as 0.34A.

I = V / R = (5V - 1.6V) / 10 ohms = 0.34 amp

To switch from soft-start mode to full-on run mode, a PSU-ON signal must come from somewhere. Also, two of the DC-DC converters need an ENable signal, possibly from the same source. I expect that this source is a small "PSU control" section within the flash controller. U19 is a 2.5V LDO linear regulator, so perhaps it provides the enable signal for buck regulator #1.

I think I would need to determine the sources of the ENable signals and the PSU-ON signal before I could begin to properly understand this circuit. I confess I am stuck at this point.

Attachments

[n3dv3d]

Hi,

thank you very much for your analysis.
I tried to take some measure from a similar part ( OCZ VTR1-25SAT3 ) and seems there are significant discrepancies on capacitor and resistors near the pin 1 of TPS652510:

C133 -> 0,9V corresponding cap on VTR1 -> 6,5V
R71 -> 0,6V corresponding on VTR1 -> 1,2V
C12 -> 0,9 corresponding on VTR1 -> 2,1V

These other measure could help you ? Can you suggest other measures I should take?

THanks a lot

[fzabkar]

I would replace U17, but I expect there will be other problems.

[fzabkar]

I just noticed that, after you bypassed R72, the PSU-ON signal is now active. If this signal is coming from the flash controller, then perhaps it is alive.

[n3dv3d]

I just noticed that, after you bypassed R72, the PSU-ON signal is now active. If this signal is coming from the flash controller, then perhaps it is alive.

Considering this, could it work by replacing the U17?

Thanks a lot