Replace roller rubber on HP 9825 tape drive

Joe R. rigdonj at
Sat Jun 18 08:02:23 CDT 2005

At 10:35 PM 6/17/05 -0700, you wrote:
>On Sat, 18 Jun 2005, Tony Duell wrote:
>> > Presumably the roller is secured by a circlip and you can remove it for
>> No way...
>> It's fixed to the motor spindle, and carries a tacho disk (with clear
>> and black lines) at the lower end. This is hidden inside the plastic
>> chassis of the tape drive, between a light bulb and a phototransistor.
>Ok, I didn't realize all that was going on.
>> I've not had my 9825 apart for some time, but I can remember that you
>> take out the complete tape drive unit, and then remove the screws holding
>> the PCB and mechanism to the mouting bracket. The motor assembly is held
>> in place by screws from the 'bottom; (non-cartridge-side) of the drive,
>> but I can't remember if you have to desolder any wires to get it to come
>> free enough. Then you can remove the setscrew(s) that hold the roller to
>> the motor and pull it off the spindle. Problem is that the setscrews burr
>> the spindle, and often the roller doesn't want to shift...
>My experience differs, but then I may not have been paying adequate
>I had to remove the three screws holding the PCB to the aluminum bracket
>so that I could get to the one screw (of three) that attached the tape
>unit to the rest of the chassis.  Then I could remove the tape unit (after
>unplugging the cable).  The aluminum bracket has two screws attaching it
>to the body of the tape unit.  Once those are off, you can extract the
>aluminum bracket and get at the roller and read/write head.  I couldn't
>figure out how to remove the motor so I left it in place.
>I couldn't find any of the recommended materials at the local hardware
>store (never even heard of Tygon) but I did find latex tubing, and I
>figured that would work just as well as anything.  I cleaned off most of
>the old roller material and decided to leave in place what I couldn't
>scrape off to work as a sort of adhesive (that stuff is sticky).  The
>latex tubing had a 1/4" inside diameter and a 3/8" outside diameter.  The
>diameter of the spindle head is 3/8", so the tight fit gave it added
>traction.  The thickness of the latex tubing is of course 1/8", which is
>about double the thickness of the original roller.  But now the spindle
>was too thick and was wedged against the read/write head.  So I got some
>sandpaper and attached a 9V supply to the motor and lathed it down about
>1/32".  Latex is difficult to cut straight so I also smoothed out the top
>and bottom of the ring.  Latex boogers were all over the drive mechanism
>after that so I blew it out pretty good with compressed air.  It now
>looked pretty nice; a little rough--a finer grit sandpaper was probably in
>order--but passable.
>I re-assembled everything and tested it.  It had trouble at first: the
>drive would start up but then I'd get error 43 (unexpected BOT or EOT or
>tape error). 

   Check your tape! I'll bet that you'll find a spot where the media has
come of the plastic tape and now leaves a clear spot. The EOT sensor sees
the clear spot and thinks that it's the EOT hole.  

 I kept re-trying and eventually it seemed to work, but then
>I pulled the tape and examined it and it seemed like the tension loop
>inside had come undone.  I opened the tape casing and, being an idiot,
>pulled it apart from the wrong side, and all the internal spindles fell
>out.  So now I have a thoroughly scrambled tape that I now need to figure
>out how to re-assemble 

   Good Luck! I've reassembled the tapes but I could never get them to
tighten up the slack tape that was inside. The tapes are suppsoed to be
selftightening but they won't tighten up for me.



(I was supposed to be getting the data off this
>Anyway, more work is in order.
>Sellam Ismail                                        Vintage Computer
>International Man of Intrigue and Danger
>[ Old computing resources for business || Buy/Sell/Trade Vintage Computers
>[ and academia at  || at  ]

More information about the cctalk mailing list