USB drive toasted by Dell workstation - can it be recovered?

[wibit]

I would measure the voltage across each of the capacitors. You should find +3.3V, +1.8V and +5.0V. The voltage across the 3.3 ohm resistor would be a useful indicator also. If either the 3.3V or 1.8V supply is missing, then test for a shorted capacitor.

I will do it and follow up with what I find. It may take me a day or two. Thank you for the recommendation.

[arvika]

Typical controller problem, as I said before. Need chip off recovery. Be careful while you make a measurment, if you make some short it could erase the nand.

[wibit]

Didn't find any shorted components. Also didn't find 1.8V yet, but everything else is there. Given a supply of 5.08 VDC --

voltage drop across R1 (3.3 ohm) = 130 mV (for a current draw of about 40 mA)
across C3, C7, C8, C11 = 3.39 V
across C4 = 1.23 V
across C5 = 4.90 V

LED1 (D1, next to R9, above C4) lights up fine.

If I can't find the datasheet for this chip (GIL, not GHL), I'm just going to probe each pin on the controller until I find 1.8V somewhere.

[HaQue]

I don't think you need 1.8
bga152 ive done 3.3 was fine

[fzabkar]

The 40mA current draw is inconsistent with the over-current condition reported in the error log. Perhaps they are unrelated?

Normal practice is to bypass every supply with at least one filter capacitor. Since you have accounted for all the capacitors, it is unlikely that you will find an additional 1.8V supply. Perhaps the 1.2V reading (controller Vcore ?) is normal.

I would map out the ground and supply pins of the TSOP package. You would do this by testing for continuity between each pad and the 3.3V and 1.2V capacitors. Modern NAND flash typically has separate Vcc and VccQ rails. These can be at the same voltage (eg 3.3/3.3 or 1.8/1. or at different voltages (eg Vcc = 3.3V, VccQ = 1.8V). Perhaps there are some NANDs that use 1.2V for VccQ ???

Just out of curiosity, I would also test for continuity between the supplies and each of the unpopulated resistors (R5, R4, R15, R12) near C8. These resistors could actually be configuration links.

[wibit]

Some versions of this chip don't appear to use 1.8V just 3.3 and 1.2. For example the GPL does, but the GHL does not from what I see here. Still haven't found pinouts for the GIL.
http://www.usbdev.ru/glossary/galcor/au6998sn/

I agree the over-current condition reported in the system boot log (not error log) may have nothing to do with this problem. That line item is sometimes reported in error by certain Linux distributions. It could also be due to something other USB device--I haven't dug into all that on the workstation side for lack of time.

The bottom line is that this USB flash drive worked solidly for two years across four other PCs. When I plugged it into the new Dell workstation, that was the last time I saw my data. Tried a second UFD, same outcome--data and formatting wiped out.

Static discharge or some other kind of transient spike at the USB port?

[fzabkar]

ISTM that the GPL pinout doesn't have any really significant differences over the GHL pinout apart from the supply pins and 4 additional CE pins. The GIL pinout would be very interesting.

I don't know about static discharge, but I have sometimes seen or felt minor sparks or shocks when I've connected cables between devices on different power grounds, or devices with missing power grounds. In the latter case the "ground" terminal usually floats to half the level of the AC mains supply. I'm not sure how this would affect a USB stick, though. You could eliminate static discharge or power grounding as the cause if you were to connect the flash drive before powering up the computer, but that would mean risking another sacrificial drive.