> On Jun 17, 2015, at 1:50 PM, Johnny Billquist <b...@update.uu.se> wrote:
>
> On 2015-06-17 19:40, tony duell wrote:
>> [Writing alignment disks]
>>> As far as I know, in special machines mounted on slabs on stone
>>> weighting tons, standing on dampeners, so that you had absolutely
>>> vibration free environment, and then a very precisely controlled head
>>> control system that could write the tracks at the exact place they
>>
>> I have an idea that some of these units used an optical interferometer to
>> determine the head position
>
> Quite possible. But it also requires the movement control being different
> from a standard drive, in order to drive at the precision, as well as the
> feedback from the inferometer.
Interferometer would make sense, at least for drives of that era. I think
modern drives have track spacings small enough that a visible light
interferometer may not be sufficient any longer.
> ...
>> Incidentally, I once saw a procedure (maybe HP) for rewriting the servo
>> surface of
>> a fixed/removeable drive in the field. It used special electronics, but not
>> any special
>> mechanics. It went like this :
>
> [...]
> Well, a drive like the RK05 can also be reformatted in the field. So it all
> depends on the drive…
True, but an RK05 doesn’t have servo data on the platter; positioning is done
by reference to an optical widget in the drive. So it depends on mechanical
reproducibility being significantly better than the track spacing. Higher
density drives use on-pack servo to avoid that constraint. And embedded servo
avoids an additional constraint: accurate positioning of one head relative to
another.
paul
