Try on different computer / different IDE cable

_________________
You don't have to backup all of your files, just the ones you want to keep.

3 * 80GB drives = 240GB of storage space. For $50, you can have a brand new 250GB hard drive, for a little bit more, you could have a 500GB drive. So, if the data on the drives have no value, just replace them. If the data is valuable, I'd recommend shipping to a data recovery professional for a free assessment.

_________________
Luke
Recovery Force Data Recovery

Did you try a different IDE port? Is it the only device on the cable?

If it turns out to be a drive issue rather than the motherboard or cable, and if you want to do a little testing, measure the voltages at each of the data pins (DD0 - DD15) when the drive is idle. If one differs from the others, then there may be a problem with one of the interface resistors.

http://pinouts.ru/HD/AtaInternal_pinout.shtml

If hardware is O.K then it can be ide port or bus / cable problem

Well, I'm sure it's a problem with disc - not with MB or cable. Disk is not visible on 2 different PCs. Jumper setting doesn't change anything. Disk is only device on the cable.

Buying a new disk is not a problem, but I would like to recover some files from old one - family photos, contacts from communicator etc.

The same problem was about week ago, but after hour of testing different tools (MHDD, Avira Rescue CD, Xubuntu) disc suddenly "back from grave". Now I'm trying to do the same, but this time I'm not so lucky so far.

More info from Victoria:

Using API:

Screen:http://www.cws.i-slownik.pl/data/up/viki_api.jpg

Using PIO:
Disc reacts on jumper settings (Victoria finds it as slave when pull out the jumper)
When try [Power]:

Screen:http://www.cws.i-slownik.pl/data/up/viki_pio.jpg

But there's a light of hope - when I press "RHPA", it shows:

LBA suddenly working? Before that I've pressed "Unlock" with different settings.

Still fighting...

I don't see "modyfy/edit" button, so I post again...

Surface scan in Victoria doesn't work, but...
Victoria reads native LBA. When copy this value (156301488) to scan "End LBA", surface scan works! So there is hope. How can I copy all data from disc (image 1:1)? Many programs can do that, but in this case I need to set manually LBA begin and LBA end (ignore LBA value detected by copy program). Any ideas? Linux "dd" command will do it?

The Identify Device command is 0xEC while the Read Native Max Address command is 0xF8. The former command produces nonsense, but the latter executes successfully. There are differences in two bits of each byte.

0xF8 = 0b1111 1000
0xEC = 0b1110 1100

According to the product manual ...

http://www.seagate.com/support/disc/man ... 89997c.pdf

... the number of LBAs for model ST3802110A is 156,301,488 (0x0950f8b0).

This is the same value reported by command 0xF8.

0x0950 = 0b0000 1001 0101 0000
0xF8B0 = 0b1111 1000 1011 0000

ISTM that the commands are executed correctly when bit 2=0, or bit 4 of the least significant byte is a 1.

According to the ATA8 spec ...

Working Draft AT Attachment 8 - ATA/ATAPI Command Set (ATA8-ACS):
http://www.t13.org/Documents/UploadedDo ... A8-ACS.pdf

... the model number, serial number, and firmware version should consist of 20, 10, and 4 words, respectively. This suggests that " ?E" or "?E " triplets are actually 4 bytes long, with one nonprintable byte.

An examination of Victoria's report gives the following:

32533 = 0x7F15 = 0b0111 1111 0001 0101
16197 = 0x3F45 = 0b0011 1111 0100 0101
65066 = 0xFE2A = 0b1111 1110 0010 1010

ISTM that bit 4 or bit 12 of each word may be hot, but I'm not certain because the endianness is confusing.

If bit 4 is hot, then an EC command would be misinterpreted as an FC. According to the ATA8 spec (Table B-2), FC is Vendor Specific. Hopefully it doesn't have an undesirable outcome.

It would help to see the entire 256-word data block returned by the Identify Device command. This would confirm beyond doubt the existence of a stuck bit. I don't know if Victoria can do this, but Seagate's SeaTools for Windows automatically captures these data in a *.ATA file, where the asterisk is the drive's serial number. Otherwise CrystalDiskInfo will do this as well, in text format.

http://www.seagate.com/support/seatools/
http://crystalmark.info/software/Crysta ... dex-e.html

The latter is the better option, as SeaTools has a CRLF bug which sometimes affects the Identify Device output.

This is a very typical problem when there is a fault with the cable, however, typically the signal pins on the PCB have dry cracks or a component on signal line is duff

_________________
All went well until I plugged the drive in.

just read all of it.

It's worth checking these pins>

Data bus bit 0 « DD0
Data bus bit 1 « DD1
Data bus bit 2 « DD2
Data bus bit 3 « DD3
Data bus bit 4 « DD4
Data bus bit 5 « DD5
Data bus bit 6 « DD6
Data bus bit 7 « DD7
Data bus bit 8 « DD8
Data bus bit 9 « DD9
Data bus bit 10 « DD10
Data bus bit 11 « DD11
Data bus bit 12 « DD12
Data bus bit 13 « DD13
Data bus bit 14 « DD14
Data bus bit 15 « DD15


http://www.t10.org/t13/project/d2008r7b-ATA-3.pdf Page 9 Table 5

_________________
All went well until I plugged the drive in.

I had already suggested the same thing (maybe you don't read my posts), but I think the OP expects to fix the problem from behind a keyboard.

Quote:
measure the voltages at each of the data pins (DD0 - DD15) when the drive is idle. oppss! ;o)

_________________
All went well until I plugged the drive in.