Replace roller rubber on HP 9825 tape drive

[Tony Duell]

> 
> > Hi Tony
> >  I'm proposing an experiment for you since I know that
> > you are the type that would actually try it.
> 
> I am going to try this, although probably not tonight...
> 
> However, even if it does bahave as you suggest, it still doesn't explain 
> _why_... I will have to think about this some more...

Sorry to follow up my own message, but I spent last night thinking about 
it, and I think I now understand it.

The things that are important are (a) the belt is elastic and therefore 
stretches, and (b) the drage from the spools means that the tension in 
the belt is not the same all the way round. There is more tension in the 
belt on the takeup side than on the supply side. 

Supppos that a certain length of belt has gone past the supply spool (and
thus wound off that length of tape). As it goes round the takeup spool the
belt will be stretched a little more, so it will move the surface of that 
spool by more than the amount that the surface of the supply spool was 
moved. It will try to wind more tape onto the takeup spool, thus 
tensioning the tape.

For obvious reasons this effect is small. It helps to have the tape 
fairly well tensioned before using the cartridge. If you're reassembling 
a defective one, I'd tension the tape as much as possible by hand.

If you could make a cartridge with the belt round the outside, and with 
the drive roller possitioned far enough forward to let the bead get in 
contact with the tape (inside the loop formed by the belt), it wouldn't 
work. There'd be more movement on the supply side, the tape would get 
ever more slack.

-tony