IS501 sensor -- my solution

Some days ago I was asking about a source for an IS50 OPIC optical sensor used in the shaft encoder of my Olivetti Sparkjet printer. And it appears this device is unobtainium In an earlier message tonigth I mentioned I had recently bought a cheap PS/2 mouse, and that it is now very dead. These 2 events are connected... The reason I boguth a cheap mouse is that it was optomechancial, rather than optical. And the reason it was PS/2 is that that was the cheapest one I could find. The salesdroid in Maplin thought I was craxy when I daid I didn't care what the interface was. On the other hamd, trying to ecplain classic computing to him was going to take too long, so I simply said that PS/2 was what I wanted... And of coruse the reason I wanted an optomechancial mouse is that it contains a pair of dual phototranssitor sensors to sense the motion of the encoder wheels. Of ocurse I took the mouse apart. It was very cheaply made, with the slotted disks and spindles moulded in one piece, then clipped into 'bearings' moudlded in the base of the mouse. Ouch. But I didn't care about that. I soon desoldered the sensorts which appear to be dual phototransistors with a common collector connection. And the pitch of the slots in the mouse's encoder disks was very close to that in the disk in my Sparkjet printer. Next job was to carefully remove and dismantle the encoder in the Sparkjet. I desoldered the OPIC, which shed a couple of pins in the process, but I cared more about the PCB. I also removed the dual resistor assockated with the OPIC> I then fitted one of the mouse sensors in place of it, common collector to the +5V tack. And with a little cut and muper, I routed the 2 emitter connections to pads that had held the dual resistor. I carefully ressembled the sensor and fitted 10k resisotrs in placeof the dual resistor assembly. That would do as an emitter load for testing. Turning the spindle produced a change in voltage, but it was close to +5V all the time. Dropping the resistors would help, in the end I foudn that 470 ohms (yes, that low) was ideal. I added a 74LS14 schmitt trigger IC on a bit of stripboard to clean up the signals, conneted the outputs to the pins on the connector to the printer's main board and gave it a try. Using the logicDart I could get a display of the wavefortms, and by adjusting the encoder PCB position I got a pretty good pair of quadrature signals. Doing the self-test on the printer got the carriage jiggling about is it should do (the printer was dismantled so that there was no platten, HV generartor, carriage rails, stc, so I couldn't see if it was preinting anything, but I coudlsee the montion of the carriage drive). Alas evey so often it would miscount and the carriage would jump in position. Much gabbing of signals with the LogicDart later, I spotted the odd glitch in one of the output waveforms. No idea what was causing it, but a 1nF camacitor in parallel with the 470R resistor on each phototransitor cleared it up. And cleaered up the posiiton problem Now 'all' I have to do is align the carriage rails properly. Iv'e refitted the rest of the printer mechansim, and it does print, but it fades out after few lines, to recover only wne I clean the end of the ink cartridge. I suspect the spark is flying in slightly the wrong direction and putting carbon ('ink') on the end of the cartridge, causing HV leakage. So, the replacement for the IS501 consisted of the sensor from a cheap mouse and a 74LS14, a couple of R's and a couple of C's from my junk box... -tony