I picked up an HP7260A table top card reader, with the aim of
restoring it to working condition. The major problem has been that
the "pick" roller that moves the card out of the input hopper and
through the first gate had melted into a pile of goo, ending up in a
Yes, HP rollers seem prone to that ;-(. I have an HP9820 calculaotr where
the internal printer platten roller had turned to goo and dripped onto
the printer driver board, and even onto the main backplane at the base of
the machine. Cleaning it up took nearly as long as making a replacement
Getting back to your machine, is this a normal cylindrical roller, or
does it have a flat on it or something?
Thanks to recomendations on the MoHPC froum, I'ce found that 3M 'Cold
Shrink' is a suitable material to make cylindrical rollers from. Put it
on in layers (don't pull too much of the 'core' out). It grips well.
That's what I used for the platten I mentioned.
Before I was alerted to this product, I had conisdered casting my own
roller using one of the 2-pack elastomers, like Devcon 90. The basic
procedure is to make a mould, coat that with the release agent, coat the
spindle with the primer, then mix the 2 parts of the elastomer and pour
it in. The main problem ia that you're supposed to mix the entire
elastomer kit in one go, once mixed you have 30 minutes to pour it, and
you have to wait overnight before taking the part out the mould. Thus,
yo have another go (something that I was sure to need to do), you'd have
to use a new kit, and that gets expensive fast (I found that an elastomer
kit + primer + release agent was going to cost over \pounds 100.00, the
kit alone was over \pounds 30.00...)
But of course you can make rollers of just about any shape that way,
including oens with flats, steps, etc.
I would have been more likely to give that a go if I'd had someody local
who'd used said product (even if not for repairing classic computers :-))
who could have told me 'No, pour it more slowly' or 'You'll never get
that mould to fill evenly' or whatever.
puddle in the output hopper. I only had to sacrifice
four blank cards
(hand fed) before getting one to pass thru the rest of the transport
almost entirely unblemished(*). So far the control/sequencing
circuitry seems to be working, but I haven't checked yet if any data
is comming out the serial port.
Thanks to the HP Computer Museum (
www.hpmuseum.org), I found the
http://www.hpmuseum.net? The site you mentioned is for handheld calcualtors
mostly.
-tony