I have an instrument that runs true DOS on a 486 motherboard
Data storage is on the internal IBM-H3256-A3 hard disk
I'm looking for a method to download/transfer data from the hard disk in a more efficient fashion than copying files through DOS on to the floppy drive and then transferring to laptop via floppy-USB unit (lots of files and many floppy disks are required).
Access to the hard disk and floppy is not straight forward so I was hoping to install a permanent solution
I was wondering if it was possible to replace the floppy with a flash drive or external hard disk?
Be able to access the hard disk as an external hard disk when the unit is powered down?
I don't really know......this isn't my forte but I'm sure (hoping) there must be a sensible solution?
It appears this HDD is an old "standard" IDE hard drive. Thus you should be able to disconnect and remove it from its current computer and mount it in any more modern one that has an IDE port, and connect to that. Pay attention to the jumper setting for Master or Slave. The data here:
show that there is a jumper block of 2 x 7 pins for setting this, and I'm guessing it already has a jumper on the 1st pin pair to make it a Master device. So where is that? Look at the bottom photo here:
of the unit turned over. It looks like the jumper block is NOT on the back edge, but instead is pointing in and under the unit, behind the 4-pin Molex power connector. The one in the photo has a grey jumper on the 1st pin pair.
If you are not familiar with Master and Slave jumper settings and how / why to set them, post here for further info.
The HDD should have on it a label (see middle photo of that set) that has the three old parameters: CYLinders, HEADS, and SECtors per Track, that MIGHT be needed to get your newer machine to use the drive. Normally in BIOS Setup if you leave the drive detection system for that IDE port on AUTO it can figure it all out itself, but sometimes you need to enter these parameters manually. While you're in BIOS Setup, check the Boot Priority Sequence and make SURE this IDE unit is NOT anywhere in the sequence.
Once it is hooked up, boot into Windows and see if you can READ anything from the drive - do NOT try to write anything to it! The first clue is whether it shows up in My Computer (in Windows), and then whether or not you can see its files there. I expect you should be able to. Then all you need to do is COPY all the files you like to another HDD in the new machine.
Do all the copying you can, then shut down and remove the drive from the new machine and replace it in the original instrument. Do NOT try to access or use any of the original files beyond copying. Try working ONLY with the copies you make to the HDD in the new machine.
I highly recommend you do NOT try to copy and use any .COM or .EXE files - those certainly are designed for use under an old DOS and may never work on any modern CPU. But data files might be accessible with some modern applications. Just do NOT try to access the original files on the old drive.
Unfortunately, I doubt you could just connect an external HDD unit to the old 486 mobo and make copies to it. The problem is likely both the mobo and the old DOS. The mobo may not have any USB port on it (most external HDD's use at least USB2, which is also backwards-compatible with USB1). But even if it does, I don't remember any DOS that had USB drivers. I think USB support first showed up in WIN98 or Win 98SE. And I'm sure neither mobo nor DOS had any support for Firewire, and certainly not for eSATA.
I am not (as you suspected!) familiar with Master and Slave jumper settings but I will see if I can find further information before making a further post
I do know the about the heads, sectors, etc because I need to manually enter these in the CMOS set-up for each instrument
I'm hoping I will be able to split and hard-wire the HDD ribbon cable to an external plug for connection somehow (usb to IDE device??) to an external pc - assuming I can get your suggestion working. The jumper settings could be an interesting problem though - since the HDD is sat within the rugged casing of the instrument, which I'd like to leave alone in the long run. Thus power down the instrument (pc) and connect the hard drive to a modern pc and extract the required data. Sever the connection and power up the 486 unit again
(had to Google 'mobo' - a futher education for me!)
It may be easier to use the RS232 or the parallel port (both require a cable available from Laplink that can probably provide a complete solution). Google "rs232 file transfer" for various RS232 solutions.
There are more serial than parallel port solutions, but it's slower. The best solution would be an ISA NIC as it would allow transfers at 100mbps (the hard disk would be the bottleneck), but it's more complex to setup.
It sounds as if you do not want to remove the HDD from the old instrument. In fact, it even sounds like you may not be able to disconnect the power and data cables from the back of the HDD. Is that correct? And I can understand that you probably would need to remove the HDD from the case in order to reach the jumper block, and don't want to do that.
Your best option may well be a serial or parallel connection between two machines using Laplink cable and software, as GhislainG has suggested. Bear in mind two things when planning this:
1. The software you get to do this will need to have a version that runs under your old DOS, whichever version the instrument runs on.
2. On the modern computer that accepts the data being copied, my guess is that it will be running some version of Windows as its OS. Make sure the software you get also has a version that does run under your more modern OS on the newer machine. There are some odd quirky tricks to getting old DOS software to use serial or parallel ports under a Windows OS, so it's much better to have Windows-compatible software on that new machine.
I REALLY recommend you do NOT try to splice cables from one machine to another to move data. The details of signals in the various data transfer protocols are so very different it takes actual intelligent signal processing to convert from one form to another. You cannot take the electrical signals on an IDE data ribbon cable and simply plug them into a USB port!
The other option for you IF you can disconnect the power and data ribbon cables at the old IBM HDD in the instrument MAY be to follow your other idea. If you can do that disconnection AND you can get the new machine close enough, you might run cables from the PSU of the new machine (for the 4-pin Molex power connection) and from the new machine's mobo IDE port (for the ribbon data cable) to the IBM drive in the old instrument. IF you try this, I REALLY recommend you turn off the instrument AND disconnect all its power cords, etc. so it is completely isolated from the world except for its connections to the new machine.
To do this you may need to hunt down (computer parts shop) an extra-long IDE ribbon cable and maybe an extension or something for the 4-pin Molex power connection.
If you are pursuing this latter line, here are some notes:
1. The ribbon cable. These all have 40-hole connectors on the ends (well, most have one hole blanked off so you can only plug in one way), but modern cables have 80 wires in the ribbon. The originals had 40. The 80-wire cable allows cleaner and faster data transfers, but only if the IDE port on the mobo AND the HDD itself are designed for that. In your case, that old IBM drive probably never could go "high speed", so even if you can only get a 40-wire ribbon cable that will do the job.
2. ANY IDE port can support up to 2 devices on it, so each needs a unique identifier. That is what the jumpers for Master and Slave are all about. Any IDE port MUST have a Master device on it to work; it MAY also have a Slave. To accommodate this, most IDE ribbon cables come with 3 connectors. However, some are missing the middle one. On modern cables (80 wires) there is a Blue connector on one end that goes to the mobo port. There is a Black connector on the other end that goes to the device with its jumper set to be the port Master. In the cable's middle is a Grey connector that plugs into the Slave device IF you have one. Some older cables do not have these color codes, but each connector WILL have bumps on one edge and probably a blanked-off hole so that you can only plug it into a mating connector one way.
3. The old IBM HDD. I would bet it is the only HDD in the old instrument, so it MUST have its jumper set to Master already. Don't even try to change that. When you connect it to the new machine, just make sure you do not also connect a second HDD to the middle connector on the ribbon drive. That way you'll have the new machine's IDE port connected to only one HDD and it will be the port Master, all OK.
4. I'm glad you are familiar with CYL / HEAD / SECT settings on older machines. You should know, however, that newer machines have BIOS's that are pretty smart about figuring these things out for themselves when an IDE drive is connected to a port. So you MAY not need to set the new machine - it might just work fine if you leave the port set to AUTO. HOWEVER, if you have any trouble getting that to work, you can fall back to setting the IDE port's parameters manually, as you have done on older machines.