Try this

remove the partition
then create an extended partition
now create two logical drives, each about half the hard drive
try to format both.

I'm guessing your hard drive has some bad sectors in a lousy location.

I think the same. 🙂👍

You can try running S0KILL to safely erase all data, too.
Re: IDE to Compact Flash as MS-DOS boot drive.

Then proceed creating all partitions from new.

Bad sectors.. I'm speaking under correction, but I think something is off if that error message appears.

Not because of DOS, but rather because HDDs that "modern" do usually silently use spare replacement sectores (reserved sectors).

That means that either some serious defect exists or that the HDD ran out of spare sectors.
Which isn't good, either.

On MFM/RLL era HDDs, bad sectors were normal and excluded during low-level format.
A sticker/label with a table was stuck on the HDD that provided information.
Also, DOS itself did exclude bad sectors during soft format (aka high-level format).. Not sure how reliable that was, though.🤷‍♂️

Edit: Please don't throw out that HDD, though.
Some HDDs (especially Conner models) that old or young do have a debug port on the PCB.
It's a serial connection, like with modern DSL router.
By using a terminal program, you can see what the HDD firmware is reporting.
PS: The voltages might be +-5v, not +-12v as used by RS-232 ports on PC.

weedeewee wrote on 2023-03-21, 16:28:

Try this […]
Show full quote

Try this

remove the partition
then create an extended partition
now create two logical drives, each about half the hard drive
try to format both.

I'm guessing your hard drive has some bad sectors in a lousy location.

Ok I will try but I never heard any read retry noises associated with bad sectors
everytime the drive did something like the NDD surface scan you could hear the heads normally running along its course

I'd grab a copy of Seagate's SeaTools for DOS; IIRC, they'll actually work with any IDE drive.

https://www.seagate.com/support/downloads/sea … legacy-support/

it may be a hardware issue , like ram... pull the drive run a full test externally on modern pc , if all is fine the drive is ok. (check for corrosion on PCB , if corrosion , clean with contact cleaner , remount pcb). Fdisk is no good , try other tools, (from hirens) or something else. better yet format partition and load win files from modern pc, then plug in old. Properly set-up hdd CHS specs in bios.

I usually use QuickTech Pro - PC Hardware Diagnostic tools , it has a good good hdd tester and formatter. (if the hdd is seen by pc)

ediflorianUS wrote on 2023-03-22, 10:18:

it may be a hardware issue , like ram... pull the drive run a full test externally on modern pc

Might be a challenge to make the drive work in a new PC.

Hard drives built before 1995 rarely have working LBA addressing and require CHS addressing to function. (Cylinders/Heads/Sectors)
Most USB adapters and PCIexpress controllers built after 2005 insist on using LBA addressing.

So if you attach that old drive to your new computer and it starts clacking like a rattle and seems unable to do anything, that's just the result of trying to use LBA addressing on a CHS only drive.

I agree douglar ! Needs a 486, Pentium, P.Pro or Slot 1 board running DOS to really fdisk and format that old of a drive IMHO and make sure to manually enter the CHS values of the "translated" of the CHS
good chance the 486 does have an issue...

I was able to use a classic ~80 MB drive on an AMD Athlon 64 X2 PC in 2008 or so.
The BIOS auto-detected the HDD with the right values if attached to the on-board IDE connector..

But that's really hit and miss. Depends on the drive's firmware, the PC's BIOS etc.
Real CHS (P-CHS, physical CHS) isn't being used since the 1980s, anymore. It's rather L-CHS (Logical CHS).
Some vintage HDDs also support LARGE model (E-CHS), the predecessor to LBA.

Yeah, it's really hard to predict how compatibility turns out in practice. 😅
HDDs since 1990s or so do cheat with the CHS values, anyway, to please both IDE and PC BIOS specs.
If it wasn't for DOS/BIOS, the very same ancient HDD controller electronics could handles gigabytes of storage just fine.

Edit: IMHO, the CHS or LBA translation matrices used are slightly similar to how x86 assembly worked in real mode: segment+offset.
I.e., it's possible that two different drive geometries *may* work in accessing the drive's data.
This possibility might also be the cause for confusion between PC-HDD, however.

The changes that were made with ATA-2 specification also harm compatibility.
Certain register meanings have changed, or registers react in a different way.

douglar wrote on 2023-03-22, 19:29:

Might be a challenge to make the drive work in a new PC. […]
Show full quote

ediflorianUS wrote on 2023-03-22, 10:18:

it may be a hardware issue , like ram... pull the drive run a full test externally on modern pc

Might be a challenge to make the drive work in a new PC.

Hard drives built before 1995 rarely have working LBA addressing and require CHS addressing to function. (Cylinders/Heads/Sectors)
Most USB adapters and PCIexpress controllers built after 2005 insist on using LBA addressing.

So if you attach that old drive to your new computer and it starts clacking like a rattle and seems unable to do anything, that's just the result of trying to use LBA addressing on a CHS only drive.

My connor drive ide 20 mb from Toshiba worked fine with external enclosure to backup everything .

ediflorianUS wrote on 2023-03-24, 08:45:

My connor drive ide 20 mb from Toshiba worked fine with external enclosure to backup everything .

Whats the make and model of enclosure are you using?

douglar wrote on 2023-03-24, 09:59:

ediflorianUS wrote on 2023-03-24, 08:45:

My connor drive ide 20 mb from Toshiba worked fine with external enclosure to backup everything .

Whats the make and model of enclosure are you using?

a Q-Tec 750H USB2 HDD CASE 3.5" item no.13290 rated 200-240VAC/VDC,2.0A/12V

(used the same enclosure to load/install win98 on my 486 WD43 hdd).

Nice adapter. Looks like it is old enough (~2005) to still support CHS addressing. That’s a keeper!

douglar wrote on 2023-03-24, 14:35:

Nice adapter. Looks like it is old enough (~2005) to still support CHS addressing. That’s a keeper!

~2001 https://datasheet.datasheetarchive.com/origin … DSA00381869.pdf

Re: IDE to USB adapter (again) Afaik Cypress EZ-USB FX2 USB 2.0 Chip reference IDE-to-USB design supports CHS and PIO modes
CY4611B – FX2LP USB to ATA/CF Reference Design https://media.digikey.com/pdf/Data%20Sheets/C … DFs/CY4611B.pdf
can be build using $5 "CY7C68013A-56 Develope Board Eeprom FX2LP Module Logic Analyzer Usb 2.0"
https://www-user.tu-chemnitz.de/~heha/basteln … x2/index.en.htm

looking at ide.c source code:
[code]extern char NumCylindersMSB; // global #Cyl MSB
extern char NumCylindersLSB; // global #Cyl LSB
extern char NumHeads; // global #Heads
extern char NumSectPerTrack; // global #SectorsPerTrack[/code]

but now that I started reading the firmware itself all I can see is LBA addressing 🙁

tnx, yes it's great had no issues.(cc.2003 or 04) (no-matter what ide drive I tryed from 20mb to 500gb).

ok a little update on the 170 mb IBM hard disk
To be honest the hard disk is 160mb ... because back in the good old days 1 MB = 1024 KB

Anyway... I managed to find some old MS-DOS disks that I had not put into a disk drive in 35 years ! And run the setup
the setup started and it told me the hard disk was not formatted and if I wanted to format... I formatted it and it just went and installed ms-dos 5.0

so now the HDD is running fines, I copied some files and folders and programs run from it...

So as a note to people... dont assume anything when dealing with old hardware... it is a REAL possibility you dont have enough information/knowledge when presented with an issue
I could very well have assumed the disk was toast and throw it or open it or just completely destroy it in my ignorance

but the HDD was not doing any unpleasant sounds and was responding to commands so I had a big feeling it was just some logical/compatibility/software issue

thanx to everyone that responded

here are a few pictures of the ancient bugger

I managed to fix load's of thoast drives (at least to the point you can MS-DOS-use them.) I even hate trowing away the one's I could not fix , I just set-tem asside. I even start Ram selection and would only trow the one's I cannot fix , use or won't boot to MS-DOS (1mb required).

I learned the hard way that if you use SSD, and they fail , data is lost , you are lucky if you can fix the ssd.... (so I hate them). yes they are fast , but have internal firmware(os) much worse than HDD's

and I leanerd single and dual platter drives are most robust , and sometimes even with 1 bad sector they can last as much as you need them .

I do like upgrading old pc's but , as I have files and check for backup before upgrade , I had no issues , and was limited to about 4 gb on vintage (pre P1 systems) Never managed to fill them. (you can use up to 120gb in fat32 in dos) with not much issue ,
but what's the point?

Ocasionally I do came across on drives that spin , but you can only format them if you wish to use , or worse , format in uncomon format's (like a seagate 500gb I never managed to do anything except reset-format then partition with exFat , no fat 32 or ntfs would ever run on it ) but drive run fine , useing as backup for music as secondary drive , copy-ing with cmd copy-x command... under xp64bit

Most of the times PCB is faulty , dirty or corroded , quick clean fixes many issues otherwhyse the system would spew-out as errrors.
Once I had to resolder 1 pin on IDE drive that gave multiple errors, ide conector was savered (1pin) ... never had issues after fix.
Or my Toshiba T3100e , had a missaligned by 2 pin IDE cable , and gave error , fix that , worked fine , than put asside the connor drive and upgraded to Seagate 4.3 , (the WD4.3 would not be verry compatible , could not partition it on the T.... , worked fine on the 486 project - my original 486 drive is K.i.a.... need to fix that , 320mb....).