Chilli

Jaclaz, how can you tell? I did not find any match in this forum neither in google for "process effect list". Furthermore Jamesblond stated he has a LBA0 error. Is it necesarry to spin down in this case at all? At least Gradius description for LBA0 starts immediately with partitioning. Until now my understanding was that spind down is necessary to break the infinite loop of bsy drives (however since mine always failed busy, I have no expericene with LBA0). I had some private conversation with JamesBlond in german language and it seems to me that his dirve has a problem whenever it really needs to access the platters. This might be a physical problem either in mechanics or electronic hardware, or wrong calibration values if the data recovery people made some changes for fun. If one of the latter is correct the role of the data recovery company is very doubtfull (considering the quote versus actual costs for recovery of a drive with physically defect mechanics). Regards Chilli

Thanks for the link, I'll look at it when I'm at home. As I said I gave up searching. So maybe I can sell the drive for some bugs or use it as back up drive if I can update the firmware. I don't want to argue with you about the difference of a minimum timespan and an average, it will lead to nowhere ;-). I think I stumbled somewhere about an explanation that there are more than one failure mode of the harddrive caused by the SW (eg. there was a Segate ICT or EOL test pattern dependend issue and some other stuff). Based on my usage (the PC is not booted daily) and the short (almost every time decreasing) MTTF of my drive I would say the normal counter overflow theory is no explanation for the failure of my drive. The drive is what some called "locked in", that means I have to power up with heads & motor attached and than quickliy unattach the heads after first reponse from the drive on the serial line. -Maybe this makes a difference. Still instead of using some plastic cards, I first loosen all PCB screws (to avoid bending/stress on PCB) and than quickly unscrew the screw above the head connector at the right time. But I also won't argue if it is more likely to damage the PCB with a plastic card or to damage the heads because of paritial head connector contact during too slow screwing ;-). -Can just say my drive survived the proceduer, and also survived 5V TTL (thanks to the serial reaistors in the serial line on HDD PCB).. Regards Chilli

I can not agree with this. Just unbricked my drive last week for the 4th time. I undbricked it the last time two months ago. However, since I did not solve the Fujitsu Siemens firmware issue (and to be honest I'm not interrested to look further in this topic anymore), I have a replacement drive at home since Tuesday, that just waits until I figured out how to solve the 4kByte sector alignement issue, before I send my seagate drive into rent. And, ofcourse, the replacement drive is no seagate drive. Regards Chilli

Sucessfully recovered my Segate from locked state for the third time. Eventually i will spend some more hours to find out how to overwrite Fujitsu Siemens firmware, but very likely I will just buy another harddisk, mirrow the data, activate windows..., thrash the seagate and never buy seagate again... Regards Chilli

I had some IT guys that stored my drive in a fridge in the hope to restore some data. Infact they could not restore a single byte. I got the "piece of trash" home and found out that somehow the partition table was corrupted. I could restore almost all the data (ofcourse I invested quit some time to find the right tools). The difficult thing on IT guys is that you need to know how really knows something and how is just talking. -Well i guess that's not only true for IT guys . Regards Chilli

@dragon65: The whole procedure described here is specific for the seagate firmware bug. Maybe others in this forum can help you. But the work around of this thread will not work for you and also, if you look at the topic of the thread, this is not the correct place for your question . Nevertheless: Good luck.

I think that the black cable is not necessary in most cases. 1) If you have potential separation in your serial adapter cable (that's what good adapters do...) it's sufficient to connect the GND on the HD side of the serial adapter cable to the serial GND at the HD serial connetor (no black cable necessary). 2) If you do not have potential seperation and connect the serial cable GND to the HD serial GND (on the HD serial connector) you "short cut" it to the computer GND anyway (no black cable necessary). ad 2) I did not check if serial GND on HD has a protection resistor like Rx and Tx, but if not, a break of the power GND would drive the whole HDD current over the serial GND in that case (what it is not intended for). So black cable or missing potential separation of serial adapter might even have draw backs. The only thing you must not do is use 20year old power supplies that do not have potential separation from the power line... In this case, if you apply the black cable, the smoke on the black cable or the popping noise of the electrolytic capacitors in the power supply might prevent you from destroying your HD ;-). Regards Chilli

Hello all, A just wanted to drop a note that since I could not update the firmware of my drive (fujitsu siemens OEM) after revocery in January, the drive failed again with boot of death within 4-5 months. I just folloew the same procedure as last time to recover it again. Now, that I have sufficient resources backup the whole drive I might try potential destructive fimrware reprogramming methods, if I find tim to play around. @PC995 If you do not write a diploma work, what is the question good for? If seatool works, don't mess around with serial communication. If your drive is looked unlock it with serial communication and make whatever you want in addition with official tools via SATA. The serial connection is for serive purpose. If you do not exactly know what you are doing (and you question impliees this) you can screw up your drive easily. Regards Chilli

At least they could. At data recovery service it is not unusual if you except the electronics to be broken to exchange it with a working one. That would not work on our desktop because the new electronic does not have the correct drive parameters. But a skilled service guy has probably a patented "One - Button - Drive Calibration" software ;-). BTW: Seatools for windows can finish non of the tests on my laptop hdd. This is rather a sign of bad tool and not of bad drive... (seatool does not give a usefull error message either). Regards Chilli

Last time I needed low level format must have been about 15years ago... So I would not be surprised if the command is obsolete (or legacy ;-) ). Regards Chilli

Very strange. You might have another problem too, that boots log entry count. I flew over the diagnostic commands and astonishingly (is this word english?) I did not find a command for low level format. I did not find a way to erase the data via command inteface. Regards Chilli

NOP. Don't have much time either. Seems that interresting stuff is "hidden" in bootblock for each, seagate flash tool and FSC flash utility. Don't have an utility installed yet to view boot block content. Btw. I took a look what flash utility did on the USB stick: it made a 100MB Ext3 partition and used 14 MB of it (excluding boot blcok). So I don't know what FSC ment with 1,7GB free space on stick... It's funny enough that you have to download a utility with about 340MB to get 14MB on the stick... However with static library you're always on safe side ;-). Regards Chilli

@bill4d:As Jaclaz is also writing between the lines: If the drive is bricked as soon as you hook it into you computer, than it was not sucessfully unbricked before. We us the term unbricked for drives that are accessible by the computer again. As long as is it not accessible by the computer it is as much as a brick. Whether you can communicate per seral interface or not has nothing to do with unbricked or not. Regards Chilli

"Card trick" on the head connector and short cutting ist the same. The short cut also only "wipes out" the heads. For many drives the cardboard trick for either the head connector or the motor connector works. But there are some reffered to as "locked in" that do not communicat anymore as soon as you power them with a cardboard on either of the connectors. For these there was the short cut invented, which has to be applied at a certain time. However if you manage to disonnect the head connector at this certain time it does the same. Regards Chilli

For my ES:2 I disconnected the heads at the moment you described. I could accomplish this by loosing all screws around the head connector (at unpoweerd disc), tightending the head connector screw a bit, so that it gets contact again, making procedure like you described and instead of short cutting just quick opening the connector screw. After spin down you can tighten the screw again. Maybe better for people that have no solder equipemnt or another PCB revision. Also the motor connector can be "disconnected" by unscrewing, if needed. Regards Stephan

Just tried the Fujitsu (Siemens) update utility for my FSC OEM drive. The SW refomates an USB Stick to EXT2 (so just saying 1,7GB free space needed is understatet because the whole stick content is deleted), and installs a CentOS boot image on the stick. The flash uitlity itself says "for internal use only" and "Beta 0.2". The funny thing is: it seems that the flash utility finds the drive and says: "No update required" !!! Of course I can not tell if this is intended or a bug. A have also to say that I do not have a FSC computer and if the SW says "for internal use only" it might be that it works only on certain HW. However CentOS itself dedected hardware without a flaw (as far as I could read the boot messags). Since the update tool seems to be from end of 2009, the new firmware FSC7 should contain the fix for boot of death... Looks like I need to find another solution ;-). regards Chilli

I dont know where custom vendor firmware differs from standard Seagate firmware. There is a very little chance that there is a different EEPROM mapping for different firmware versions. However since ICT and EOL calibration data will be in the same area for all models the firmware should be able to rebuild EEPROM data if the checksum fails. (Note I don't knwo where all the data is stored I just call it EEPROM. Could be flash or somewhere else...). So my first step is trying to flash FSC software because the risk and my working time will be minimum. Regards Stephan

Was just about to write to Seagate support but did not submit because I did not have seial number at the hand. But in the meanwhile I stumbeled about an update for the FSC6 SW (FSC7) http://support.ts.fujitsu.com/Download/Sho...FD-453D2011317F The strange thing is why their update SW needs 1,7GByte(!) when unpacked on the stick... Hope this will not come up with "This is no FSC system"... If it where so, there is an 50% chance that I can pick up the binary from the FSC image and palm it off to the Seagate flahs utility. But there's a 99% chance tht I don't have time for it ;-). Hope I have time for trying the FSC way of flashing this evening. Regards Chilli

Has anybody flahed a OEM drive with Seagate firmware yet? Is there a tool to read out SW? Since you can update firmware at all, configuration data for seagate firmware seems always to be at the same addresses. I would guess that the same applies also for OEM firmaware. But a risk is still there. I'm afraid this topic has to wait for the next weekend (Spent much too much time on in this one). Greeetings Chilli

So you never see a prompt, not even (all connectors connected) when sending countinously CTRL z at startup? If you use Hyperterminal, Ctrl z should be fine. If you are using another terminal, can you send ASCII (026) dec? Not that some terminal emulation does not allow Ctrl z... As far as I see there only three possibilities: 1) Your drive can not "hear" you 2) You do not physikally send Ctrl z 3) Your drive has a problem that has nothing to do with the BSY bug even if symptoms are similar. What you can try: Use another terminal programm or play with terminal adjustments (emultaion adjusttmens - baud rate etc. must be OK if you can receive Use another cable for serial level conversion (or check yours with Osci/level converter to insure it sends what you want to send) Send the drive to me ;-)

@sunlit Than pls go back just a few posts to my stepwise instruction,. You might have the same problem I had. Regards Chilli

No, if the loopback tests work and you get the messages the thingy is working. (unless for some reason the interfacce TTL levels is not adequate, i.e. the Rx "understands" the data sent by the PCB, but the "Tx" is not understood by the PCB. ) Yes and no. In this casewhere you already do receive it should also work the other way. If the power you use for the HDD and the PC with the USB cable is not the same you propably do not have a common ground. I connected my Srerial TTL cable to the HDD GND (the jumper pin right next to the Rx TX pins of the HDD). Ofcourse you can olny do that if you have potential free power supplies (which is standard) otherwise *BOOM* Regards Chilli

No luck with FSC FSC gave it's HDD business to Toshiba and Toshiba has just OEM business (no customer support). I'll keep searching, but things don't look good. Maybe need to crack flash software ;-). Regards Chilli

At the previous post I ment Strg z (low case) or ASCII(026 dec). Now I have sucessfully recovered the HDD but the flashtool refuses to flash the drive (see attachment). Maybe because it's an OEM drive? I don't really know from which OEM vendor or model this is. It might be a coincitance that FSC is short for Fujitsu Siemsn Computers, but might also be that this comes from a charge detemined for FSC. I guess there is a trick for reflashing OEM HDD? Regards Chilli flashtool_output.bmp

Done! Without jumper And without cardbord! ;-). *) I figured out that if you losen the screws, the PCB looses contact to the connector. *) For each connector there is a screw to to press the PCB against the connector. --> If you first losen all screws that affect one connector, than you can open and close the contact to the connector with screwing the connector screw in or out. Attention: I was wrong: It's not the motor that has to be blanked out, it's the heads (at least at my model). (was pointed to by: "After the closure of these contacts, the head within 5 ~ 7 seconds klatsayut, then the engine itself stops") So procedure is: 1) optional: Loosen all screws and lift PCB so far that you can be sure PCB is not stuck at the metal body of the HDD anywhere and remount the PCB 2) Loosen all screws so far that the head connector is dissconnected. Keep in mind that the PCB must not be bent. 3) Tighten only the head connector screw so that it get's contact again 4) Supply power to the disc, wait until you get the LED: blablabla message 5) Keep sending Ctrl Z until you see the F3 T> promt 6) Quickly loosen head connector screw to disconnect heads (it's important to make this very quickly otherwise heads could be at least stressed) 7) If you managed this you have all the time of the world. Continue like described in the guides -> send /2 to change to level 2 and wait 15s and so on I used terminal V1.9b and made a Makro for each, Ctrl Z, /2, Z. Regards Chilli