Lisa Widget Optics Adjustment Help

[switch998]

Hi all,

I need a little help with the adjustment procedure for the optics system with NeoWidex on the Lisa Widget 10MB hard disk. I have two units here that are in need of an adjustment that I just can't seem to get right. One was a working drive that I sacrificed for a sanity check. I tweaked just one pot back and forth to the original position, and now the drive does not work.

Moving the actuator back and forth by hand results in a very unsteady image on the scope, so I suspect that's why my alignments are invalid. I want to know how to initiate the procedures on this page (http://john.ccac.rwth-aachen.de:8000/patrick/UsbWidExamples.htm#session4) using NeoWidex. as I do not have a UsbWidEx. Specifically, the alternate seek command (a) and the forward seek command (f). I have been using the keyboard to manually seek, watch the scope, adjust, then go back, which is tedious. Are these commands implemented in NeoWidex?

 

[stepleton]

I'm afraid not --- sorry about that! Do you have an environment where you could build NeoWidEx? It may be possible to hack in the functionality you need, short of actually implementing these exercises.

 

[switch998]

That did cross my mind, and while I could probably find a way to build it, writing anything in assembly is wildly beyond my skill set.

I suppose the bigger part of my question is what am I missing with the adjustment process? All the points of the circle are at 5v, LED current (TP5) is ~2v (which is fine, I guess, since they were working drives). The heads on both drives just wind up endlessly failing to position and smacking the end-stops. Maybe TP5 being at only 2 volts is a problem, or do I need to low-level after an adjustment?

 

[stepleton]

It's been a long time since I've attempted an adjustment, so I'm afraid I can't offer much help right now.

I don't know for sure, but I would expect that the coarse servoing that seeks the head between tracks only depends on the optical sensor --- if that's right, the signal from the heads isn't a factor, and low-level formatting the drive would not change anything. The microcontroller source code for the Widget controller and the servo are available, and it would be possible to refer to that to know for sure whether the heads are involved, if the information isn't in the ERS document already.

I think having the right tool for controlling the Lisa during the adjustment process would be helpful. Hacking NeoWidEx might not be as difficult as you imagine: you might get away with replacing one of the commands with copy-pasted code that seeks the head back and forth between two distant tracks. It's been an even longer time since I wrote NeoWidEx, so I'd have to refamiliarise myself there, too. But we might be able to work it out together.

Another option would be to bring your question to http://lisalist2.com, where Patrick Schäfer (if available) could answer your questions authoritatively.