Image Map Image Map
Page 1 of 3 123 LastLast
Results 1 to 10 of 27

Thread: FDD Cable Twist - necessary?

  1. #1

    Default FDD Cable Twist - necessary?

    Hello All,

    In my continuing efforts to restore the Compaq Systempro I have run into an interesting issue.

    The Systempro seems to have two separate ports for connecting FDDs:

    1. A proprietary connector that delivers both power and signal to a 3.5" 1.44MB drive
    2. A standard 34 pin connector to be used for a second drive

    My system came with the 1.44MB drive installed and that works well enough. I was able to get my hand on a 1.2MB 1/3 height Canon FDD (Compaq FRU 112566-001) which has what appears to be a standard edge connector on it (see picture here). Compaq apparently used multiple vendors and the picture is from a Mitsubishi drive. Unfortunately, the cable was missing.

    Thinking that the two ports were independent, and a twist was only needed when both drives were on the same cable for "cable select of A: vs. B:", I made a straight through 34pin to edge connector cable. Connecting the cable allows the drive to be recognized and it even does the initial seek test on POST. However, I can not access the drive. Attempting to change drives least to a drive failure error with no attempt to seek. Furthermore, even though I set the drive as a 1.2MB drive in the NVRAM it keeps being changed back to 360KB (remember this is an EISA system so no BIOS/CMOS settings).

    I am now thinking that the two ports are not truly independent even though they are physically separate. Also looking at other Compaq FDD cables such as this they exhibit the twist even though they are clearly for a single drive connection. Plus, the behavior I am seeing is reminiscent of plugging the FDD cable backwards. The EISA cfg utility also has an option to flip A:/B: drives which I would think requires a twist in the cable.

    So could it be that I need to put in a twist on the cable that goes to the second drive? Since the default/fail safe setting for the system is to have the A: drive be the 1.44MB drive then that would make any other drive B: by default. However, in the std. manner the drive after the twist is considered to be A:. If this theory is right the motherboard flips the two drives in its default setting and reverses the ports (i.e. if we were using a standard two port floppy cable the drive in the middle would be A: and the one at the end would be B: ). This is obviously not the IBM standard but was it done?
    Current Wish List: 1. IBM 7531 Industrial Series PC 2. NEC MultiSync XL (JC-2001) Monitor 3. MicroSolutions MatchPoint AND/OR UniDOS card 4. Compaq 14" VGA CRT Monitor (the one that came with the SystemPro). 5. Stacker HW CoProcessor Board MCA BUS. If you have any of the above for sale please PM me. Thank you!

  2. #2

    Default

    Quote Originally Posted by Shadow Lord View Post
    I am now thinking that the two ports are not truly independent even though they are physically separate.
    I'd start by pulling out your multimeter with both drives disconnected and buzzing out the two ports to see if they are actually sharing the DS and MOTOR lines or not

  3. #3
    Join Date
    May 2009
    Location
    Connecticut
    Posts
    4,560
    Blog Entries
    1

    Default

    Have you checked if the HP website still has the manual for the specific Systempro? If the two floppy ports are connected, there should be a run of traces between the two.

    I thought the twist was needed for the A drive but no twist for the B drive but no telling what trickery Compaq did to connect the two. The system sounds like one done after Canion left and Compaq was doing all types of inconvenience to sell more proprietary parts.

  4. #4

    Default

    Quote Originally Posted by krebizfan View Post
    Have you checked if the HP website still has the manual for the specific Systempro? If the two floppy ports are connected, there should be a run of traces between the two.

    I thought the twist was needed for the A drive but no twist for the B drive but no telling what trickery Compaq did to connect the two. The system sounds like one done after Canion left and Compaq was doing all types of inconvenience to sell more proprietary parts.
    HP has purged most everything Compaq from its archives.

    I have a service manual for the Systempro (original). Unfortunately, on that system the MB had a std single 34pin FDD Connector.
    Current Wish List: 1. IBM 7531 Industrial Series PC 2. NEC MultiSync XL (JC-2001) Monitor 3. MicroSolutions MatchPoint AND/OR UniDOS card 4. Compaq 14" VGA CRT Monitor (the one that came with the SystemPro). 5. Stacker HW CoProcessor Board MCA BUS. If you have any of the above for sale please PM me. Thank you!

  5. #5
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    33,367
    Blog Entries
    18

    Default

    Compaq could be strange wrt the floppy cabling. For example, on my Deskpro ENP, a flat cable from the mezzanine defines the "A" drive (it comes from the factory this way). If you want a B: drive, you use a cable with a twist to get it. In other words, the cable would be <motherboard>->(flat)->Drive A->(twist)->drive B. There is no "reverse the A and B drives" in the BIOS setup that I can recall.

  6. #6
    Join Date
    Feb 2011
    Location
    NorthWest England (East Pondia)
    Posts
    2,341
    Blog Entries
    10

    Default

    Firstly generally the BIOS does only minimal testing when you specify a drive. If its going back to 360k there is some issue.
    Secondly the cables are highly unlikely to be independent, there will only be one FDC. What is important is how the Drive Select and Motor Start lines are wired.

    Before the IBM PC drives often had links to set options.
    So typically they had up to four drive select links DS0, DS1, DS2 and DS3.
    Also often there was only one motor start that was common to all drives so all the motors ran when motor start was activated

    IBM wanted to ship drives that didn't need plugging.
    So 5.25" drives have terminator resistors and they must go at the end of the cable, so the first drive needs to be last on the cable.
    There is a twist so that both drives can be strapped the same. It flips DS0<->DS1 and MS0<->MS1

    So on the last drive on the cable sees what is DS0 from the motherboard on DS1.
    So all drives are shipped as DS0.

    I think I would check your 1.2 drive to see if it has links. Especially the Motor Start.
    Then get a meter out and see how pins 10,12,14 and 16 are connected

    Oh and why should swapping drives need a twist? Sp long as they work as "A" and "B" ?

    https://old.pinouts.ru/HD/InternalDisk_pinout.shtml

    may elp
    Dave
    G4UGM

    Looking for Analog Computers, Drum Plotters, and Graphics Terminals

  7. #7

    Default

    Keep in mind that many 5" floppy drives do have drive select jumpers, e.g., DS0, DS1, DS2 and DS3.
    PM me if you're looking for 3" or 5" floppy disks. EMail For everything else, Take Another Step

  8. #8
    Join Date
    Mar 2013
    Location
    Chaffee, MO
    Posts
    1,337

    Default

    Dave,
    So all drives are shipped as DS0.
    I believe this is a typo as it really should be DS1 {DS0..DS3}

    My SFD-321B is strapped as DS1.

    In the testing that I have done the Motor ON Signals (Pins 10 & 16) both respond the same.
    At least on the Computer I was testing. It may not be true of all IBM Clones.

    Thanks.

    Larry

    floppy_crossover_cable_wiring.jpg
    Last edited by ldkraemer; December 16th, 2019 at 05:06 AM.

  9. #9
    Join Date
    Feb 2017
    Location
    Guildford, U.K.
    Posts
    294

    Default

    This appears to be a similar cable twist to that used from the outset in IBM PS/2s (e.g. Model 60 and Model 80). The drives being "jumpered" DS1 and the cable twist between the Drive B: and Drive A: connectors on the ribbon cable (either 40-pin w power, or 34-pin) swapping the Drive 0 & Drive 1 signals. This allowed any 3.5" diskette (or tape streamer) drive to be installed in either upper (A:) or lower (B:) position without changing anything else.

  10. #10
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    33,367
    Blog Entries
    18

    Default

    The Samsung SFB-321B as shipped is solder-jumpered (look near pin 2) to DS1. I occasionally have to re-jumper one of these to DS0 for the PLC crowd. I like these drives for their configurability.

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •