Image Map Image Map
Page 1 of 2 12 LastLast
Results 1 to 10 of 13

Thread: Fixing a RX02 drive

  1. #1

    Cool Fixing a RX02 drive

    A year back I got a very nice RX02 drive for desktop use. Essentially it was the reason
    to make a clone of the RX8E controller board. Unfortunately this drive refused to work.

    01 RX02 drive.jpg

    When I tried to boot the PDP8 the RX02 was constantly reading disk 0, disk 1, disk 0, disk 1, etc...
    In the meanwhile I got another rack mount RX02 which I got back to work in no time.
    Just some cleaning of the heads did the trick.

    But I wanted to have the tabletop model working again. So last Saturday I started with the drive.
    At first I swapped the boards to check where the problem is. The problem was in the lower circuit board.
    After changing it with the other board the drive worked normally. So I returned the board into the original
    drive and began with fault finding. And I have a known good drive next to it! So it should be easy then right?

    02 measuring.jpg

    It seemed that there was data coming out of the board, but it was garbage.
    So at first I checked the head amps. The output is very noisy of the amp.
    Even with shorted inputs. But after manually doing a head load the amplifiers are
    pushed almost into clipping area. So when a disk is being read, the noise disappears.

    03 check head amplifier.jpg 04 head amplifier noise.jpg

    To check if the differential head amplifier amps were working I put a very small signal on
    the input of the amplifiers. A nice square wave appeared at the output. So that was okay.
    But I did forget one thing... To remove the disk of the drive. The other RX02 refused to boot
    either of the disk. Yeahh guess what, I erased the disk with my signal injection
    Tried to read it back with IMD... Yes confirmed!

    05 wiped disk while signal injecting.jpg

    Oh well, just write the disk image back onto the disk and we can go further...

    But while checking the data further in the drive it really got corrupt in a weird way
    at a very high sort of bitrate... So I began to wonder if the PLL worked well.
    The PLL (Phase locked loop) is there to synchronise the drive clock with the data on the disk.
    The PLL sends an analog correction voltage to a VCO (Voltage Controlled Oscillator) to generate the drive clock.

    06 conclusion PLL problems.jpg

    Well the clock was not even near what it should be. It ran on 10 times the frequency which it should be.
    The cause was this little LM318 OPAMP. At the - input there was +2.3V and at the output there was +4V.
    (Normally when the - input is higher than the + input the output should be at least somewhat negative.
    So that was already wrong in the first place) But in this circuit both should be around zero when the PLL is locked.

    07 broken opamp LM318.jpg

    So I replaced that with a spare one which I had left from my VC8E project.

    08 replaced opamp.jpg 09 PLL opamp.jpg

    And it booted OS8 right away! So another RX02 back to live!
    But I have to admit, this wasn't an easy one to find...

    Regards, Roland
    Last edited by Roland Huisman; April 6th, 2020 at 03:22 PM.
    WTB: Case for Altair 8800 ...... Rolands Github projects

  2. #2
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    34,057
    Blog Entries
    18

    Default

    Those look like Calcomp 140 drives.

  3. #3

    Default

    Congratulations on getting it working! That actually sounds like a fun sleuthing project to me.

    --Jay

  4. #4

    Default

    Quote Originally Posted by Chuck(G) View Post
    Those look like Calcomp 140 drives.
    Hi Chuck,

    Yes, this RX01 on the web is actually saying Calcomp 140.
    https://www.vintagecomputer.net/digi...4238_front.jpg
    https://www.vintagecomputer.net/digi...p_14238_sn.jpg

    My Motorola EXORciser has also these kind of drives but in Black.
    But the electronics is completely different off course.

    This is a drive from my other RX02. I had it out to clean it. But it is clearly an updated version of the 140...
    IMG_20200407_121500.jpg IMG_20200407_121509.jpg IMG_20200407_121524.jpg IMG_20200407_121533.jpg



    Quote Originally Posted by Dare View Post
    Congratulations on getting it working! That actually sounds like a fun sleuthing project to me.
    Hi Jay,

    Yes indeed, fun to do! Especially when things work out in this way

    Regards, Roland
    WTB: Case for Altair 8800 ...... Rolands Github projects

  5. #5
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    34,057
    Blog Entries
    18

    Default

    I recall that the 140 has an "interesting" 40 pin connector wherein the DC (5 and 24 volt) are fed in. In my case, not using a keyed connector led to some "interesting" events. Not a standard Shugart 50-pin interface.

  6. #6
    Join Date
    Sep 2019
    Location
    Zurich, CH
    Posts
    150

    Default

    Quote Originally Posted by Chuck(G) View Post
    I recall that the 140 has an "interesting" 40 pin connector wherein the DC (5 and 24 volt) are fed in. In my case, not using a keyed connector led to some "interesting" events. Not a standard Shugart 50-pin interface.
    40-pin connector? That would suggest that is not the drive in the RX01/RX02. You can see the connector in some of the pictures. It's four bundles of cables ending up in four keyed 8-pin (I think) connectors. And the keying consist of one pin or another being plugged.

  7. #7
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    34,057
    Blog Entries
    18

    Default

    Probably the mech is Calcomp (I know of no other drive make that looks like that), but with Digital's electronics. Not uncommon back in the day--even Apple and Commodore did it.

  8. #8

    Default

    Quote Originally Posted by Chuck(G) View Post
    Probably the mech is Calcomp (I know of no other drive make that looks like that), but with Digital's electronics. Not uncommon back in the day--even Apple and Commodore did it.
    Yes I have seen many 'common' drives with customized electronics.
    The RX01 and RX02 drives do have their own electronics.
    Probably Calcomp was specialised in making good mechanics.
    The Digital RX180 has also Tandon 5.25" drives with non standard electronics...

    I think that it wasn't the goal of DEC to reinvent a floppy drive.
    Just use good stuff which is available.
    WTB: Case for Altair 8800 ...... Rolands Github projects

  9. #9
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    34,057
    Blog Entries
    18

    Default

    DEC used its own drive electronics for the RX02 drive (M7745 R/W board). Again, not an uncommon practice with peripherals in general. Witness the number of systems that used CDC/MPI hard disk technology.

  10. #10

    Default

    My other RX02 drive seems to have an alignment problem in the right hand drive.
    This drive refused to read almost all of my disks. I have cleaned it thoroughly but
    that didn't help. The felt pad was still there and pressed the disk nicely.
    I checked the wires and ohmic values of the R/W head. That looks okay.

    So I assumed that this might be an alignment problem. I have turned the track-0
    alignment screw for 1/4 to the left. It was a bit experimenting but now it reads
    about 75% of my disks again. But I want it to read 100% of them off course.

    Does anyone know where to find the alignment procedure for the RX02?
    And any idea if there is a way to produce an alignment disk in a known good drive?

    Regards, Roland
    WTB: Case for Altair 8800 ...... Rolands Github projects

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
  •