Restoration of Panasonic DMR-EX78

[Australopithecus]

This is what was done to restore a Panasonic DMR-EX78 to full functionality when the RAM drive failed.
Obtained a second-hand machine with empty HDD and working RAM drive, but wanted to retain the 200+ recordings on the HDD on the original machine.
Note that if the working RAM drive had just been swapped to the original machine the procedure below would probably not have been required at all!

• The RAM disc drive would not read any discs and showed a persistent ‘disk in’ symbol on the display, accompanied by clunking from the drive when switching on and when a disc was inserted. This indicates failure of the drive so a second-hand replacement ex78 was obtained.
• The HDD of the replacement unit was empty and the HDD of the original unit had a lot of programmes stored that we wanted to retain access to.
• First thought was to just transfer the HDD to the new machine but this resulted in an on-screen message requiring the HDD to be formatted. So this was not the solution that was needed.
• Tried the empty HDD from the new machine in the old machine; this also resulted in an on-screen message requiring the HDD to be formatted. Said “yes” to this since the HDD was blank, and it appeared to work.
• Put the original HDD back in the original machine and it didn’t work — said it needed to be formatted! Horror! Had we lost everything?
• Consulted our son who knows about these things and he had seen somewhere on line that the process of writing anything to the HDD also writes a few sectors from the start of the HDD to non-volatile memory in the machine. We suspect that this is on the small board mounted vertically next to the HDD. Edit: No, it is on the main board IC7404, thanks to Gavtech for this info.
• Therefore when we formatted the empty HDD of the replacement machine on our old EX78, it overwrote the data in memory from the original HDD and when we put the original HDD back it didn’t match up and the machine rejected it.
• Before going any further, we took an image of the original HDD using an IDE to USB bridge, (a hard drive docking station) just to be safe. It turned out not to be necessary to restore from this image.
• Using Linux, inspecting the start of the disk image of the original HDD in a hex editor [link to “tweak”: Tweak: an efficient hex editor] showed that the pattern of data changed at address 0x3000 (i.e. 24 512-byte blocks in). Looking at the start of the newly-formatted replacement machine HDD also showed a change in the pattern at address 0x3000. We guessed that the data leading up to this point might be what’s stored in the NVM.
• Using the “dd” utility under Linux, we copied the first 24 512-byte blocks from the new HDD to the old one. (dd if=”/dev/new-disk” of=”tempfile” bs=512 count=24 and after swapping, dd if=”tempfile” of=”/dev/old-disk” bs=512 count=24)
• Putting the old HDD back into the old machine now worked; there was no prompt to format it and all the old programmes were present.
• The working RAM disc drive was then transferred to the original machine (which is what we should have done in the first place) and now every function worked as normal. It was noted that there was no clock display when it was switched off but this was restored by switching powersave to ‘Off’.
• Might at some stage replace the laser sled in the failed RAM drive.

Hope this helps someone at some time.

 

[Gavtech]

Good work.

FYI the small board mounted vertically next to the HDD is the so called 'Backend Board'. It is related to tuning and is different for each territory.

The eeprom is on the main board: IC7404.

 

[Australopithecus]

Thanks for that additional info about the location of the eeprom