USB-IDE Bridge Fail - old Freecom External : HW encryption?

[fzabkar]

I'm always hesitant to recommend shorting, but that's what Ace Labs do, so if it's OK for them, then who am I to argue. You would only need to short those two points for a couple of seconds while you are applying power to the board. The short needs to be in place before and after. I would prefer to desolder the Data pin, but that has its own risks if you're not adept at soldering.

If it does turn out that the board is unsalvageable (eg a bad bridge IC), then you might like to purchase any PL-2507 based enclosure that has an external EEPROM, and then transplant your own EEPROM. Note that the "C" version of the PL-2507 definitely supports 256Kbit EEPROMs, but I don't know whether earlier versions do. In fact the datasheet that I found states that the maximum supported size for the EEPROM is 256 bytes, so it may in fact refer to an earlier version of the chip.

[ConnorHdd]

Looks like no dice on the eeprom short :

Where do we go from here?

Thanks for your continued interest and advice.

[fzabkar]

I don't use USBDeview, so I'm not certain that I fully understand its output.

The device with VID/PID = 07AB/FCCD is a "FREECOM Classic HD 160GB". It has a "Last Plug / Unplug Date" of Jan 25, 2014. Therefore I'm wondering whether the device list is just a registry dump rather than a report of the currently connected devices. Instead I use Microsoft's UVCView which reports the currently connected USB hardware.

In any case it does appear that no USB device has been detected since January 31, so the Prolific bridge does appear to be brain dead. The next thing to test would be the 12MHz crystal. I have an oscilloscope, but I suspect that you might not.

[ConnorHdd]

I don't use USBDeview, so I'm not certain that I fully understand its output.

The device with VID/PID = 07AB/FCCD is a "FREECOM Classic HD 160GB". It has a "Last Plug / Unplug Date" of Jan 25, 2014. Therefore I'm wondering whether the device list is just a registry dump rather than a report of the currently connected devices. Instead I use Microsoft's UVCView which reports the currently connected USB hardware.

My understanding is that the green entries are current. Any others are just historical.

In any case it does appear that no USB device has been detected since January 31, so the Prolific bridge does appear to be brain dead. The next thing to test would be the 12MHz crystal. I have an oscilloscope, but I suspect that you might not.

I have a 1940s oscilloscope - valves and all - but I suspect the caps would need replacing as a min, so yeah, you suspect right. I guess I'm right in saying that w/o CLK edges the PL-2507 will look as dead as a dead PL-2507, no?

Also, I had no IDE drive connected during these eeprom short tests: just thought I might mention this on the off chance you believe it might make a difference - can't see it myself, by what do I know?

[fzabkar]

Some bridge firmware won't enumerate itself until the drive comes ready. Other bridges don't care whether a drive is present. In fact I have seen reference circuits that have a jumper to select how the bridge behaves under such circumstances. That said, since the bridge doesn't switch on the dual MOSFETs that supply power to the drive, then this would tend to confirm that the bridge really is dead.

In answer to your other question, yes, a dead crystal means that the bridge will appear to be dead. In fact when you are troubleshooting a microprocessor based circuit, the first things you should check are the power supplies and the crystal/oscillator. I have both a CRO and a frequency counter, but I don't know how you would achieve the same end without either.

You say you have an old scope. Do you have a soldering iron, too? Could you try replacing the crystal?

Edit: Can you hang on for another week? I may be able to retrieve my old PL-2507 enclosure by then.

[ConnorHdd]

I have a couple of soldering irons, one a massive job best reserved for mains sub-station work ( ), the other a butane, pen-like affair. Nothing of the standard / scale needed for pcbs (I'd guess).

I can indeed hold off on this - there's no great urgency, but I wouldn't care to put you to any more trouble, unless that is, you are happy to scatter your 'pearls' of wisdom in the hope of elevating a 'swine' like me from the mud of ignorance!

In the meantime I'll repeat the eeprom short test with drive attached.

I also have my eye on a drive on ebay, supposedly possessing an identical pcb. As you probably noticed from the post in which I quoted a freecom rep, it would seem that the PID and VID are are the main issues I face : he said nothing about any holding of encryption keys on the pcb, even though I asked specifically about these in the email I sent to him some time ago. So maybe a swap of the enclosure's HDD for mine is all that is needed to release the data.

Either way, I'm happy to pursue a solution here as it's giving me the motivation to learn and revise stuff I've neglected for far too long. Whether the pcb can be fixed or not, I'm very grateful for the time you've spared and knowledge you've generously shared.

[ConnorHdd]

Just to say I've bid on and won an external drive that, apart from bare drive capacity, seems to be a carbon copy of the unit that has failed. Freecom have led me to believe that the pcb in this unit will be capable of of allowing me to decrypt the data on the drive I have. I'll let you know how things turn out when the item arrives, but my understanding is this will not be until Friday. I figure if this does the job for £10.50 delivered, it will be a result.

[ConnorHdd]

Just to add that Freecom have come back with this:

We confirm the password should be stored in an encrypted file not on the PCB.

So it looks like a case of just waiting for the drive I bought off ebay to arrive.

I have to say I'm pretty impressed with Freecom, given the drive is pretty ancient. I was half expecting them to say they had no information on (meaning interest in) such an old piece of equipment. Of course, like any company, they don't come close to the generosity of spirit exhibited here, but you guys are something else....

[HaQue]

yes great to see they shared the info, many companies wouldn't. There is hope after all. +100 rep points for Freecom

[ConnorHdd]

Moreover the info was right: my 2nd-hand, ebay-bought, Freecom Classic HD arrived today, and after a quick swapping out of the drives, spun up like a good 'un. The Freecom Personal Media Suite recognized the drive as one of its own and promptly decrypted all files dragged from its "Secure & Compress" folder: just like old times!

Now I have a working bridge, I'm wondering whether this will help in diagnosing the fault in the busted one.

[pcimage]

Great work!

Nice result