Re: anyone know where I can find a service manual for Disk II

Mike Willegal wrote:

> Michael J. Mahon wrote:
>
>> Mike Willegal wrote:
>>
>>> Hi Tom,
>>>
>>> Perfect, just need to find or make a good alignment diskette.
>>
>>
>>
>> Finding may be easier--it's an analog recording!

Hi Michael,

Are you sure about the analog recording thing? I was speculating that whole track was simply written with some repeating pattern like 1s and 0s. Last night I was looking at the signals coming from a regular formatted disk and it shows many of the same patterns that are seen in service manual, with a lot of added noise and jitter. I figure that the noise and jitter came from the sector headers and actual data not being a consistantly repeating pattern.

Actually, since 1's are represented by a flux change and 0's are
represented by no flux change, the highest frequency signal on the
media is a stream of 1's (250kHz).

You can use a known-good-alignment disk to check azimuth alignment
by simply maximizing read response on the test drive, but azimuth should
be adjusted after radial alignment (and then radial checked again). In
practice, azimuth alignment is hardly ever an issue unless a drive has
been physically abused.

The radial alignment test track is an analog recording made by a special
dual head with adjacent gaps. Two different sine wave frequencies are
recorded, each near 250kHz.

When the reading head is properly centered over the track, it reads
the sum of equal parts of each recorded signal, producing a symmetric
"lobed" pattern (looks like a suppressed carrier double-sideband
signal). If the head is off-center relative to the recorded "track",
the "lobes" will be of different amplitudes.

Since this scheme is based on phase coherent recording of two tracks
with the "boundary" located in the middle of the desired track position,
there is no way to write it with a single-head drive.

It is possible to approximately determine head radial alignment in other
ways, assuming that you have a known-good diskette to treat as the
"standard" to which all drives will be aligned.

Simply position to a middle track and read the known-good diskette,
noting the error rate (should be low). Then progressively step the
head by quarter (or half, for less precision) tracks away from the
nominal track position in each direction (one direction at a time)
and note the track read error rates.

If the drive is properly aligned, the error rates will be symmetrical
about the nominal track position. If there is assymetry, then the head
positioner should be adjusted and the test repeated until symmetry is
achieved.

Of course, the information on the track should be read to ensure that
it is indeed the intended track, to rule out gross positioning errors.

One could, in principle, make an "alignment disk" by writing (on a
known-good "standard" drive) a track with sevaral sectors displaced
by -3/4, -1/2, -1/4, 0, +1/4, +1/2, and +3/4 track. Then, reading
the track at the nominal position on the test drive should result in
good reads that are symmetrical around the 0-alignment sector.

I don't know for sure, but I expect that this approach was used in the
popular "diagnostic disks" by Verbatim, et al.

If you go this way, I think your code would be interesting to lots of
people--particularly the code to *write* the diagnostic disk, since
copying one is unlikely. ;-)

-michael

NadaPong: Network game demo for Apple II computers!
Home page: http://members.aol.com/MJMahon/

"The wastebasket is our most important design
tool--and it's seriously underused."
.

Relevant Pages

  • Re: aligning the heads on a 5.25" floppy disk drive
    ... To properly align a drive you need an alignment disk and a scope. ... head cleaning disk. ... On some drives you can pull the heads ...
    (comp.os.msdos.misc)
  • Re: Bad head in my reel to reel?
    ... the playback surprised me in regards to the noise floor levels. ... Later I decided to try to make a recording, ... follow the playback alignment procedure. ... that a recording head could somehow be bad just for one channel. ...
    (rec.audio.pro)
  • Re: File system compression, not at the block layer
    ... the problem is track alignment. ... you'd need seperate motors for each head. ... Haven't you noticed that drives with many ... If the skews are ...
    (Linux-Kernel)
  • Re: Partitioning hard drives for audio use?
    ... I did recover some of the files on the trashed partition. ... move it to another partition or drive when I'm done recording. ... My partitioned external drives were ... head seeking all over the surface of a 500GB drive for these files, ...
    (rec.audio.pro)
  • Re: Recorded TV playback
    ... hyperthreading shouldn't be an issue here (if it was an issue with MCE, ... your background task uses the "CPU" other then the one that you are using ... both in recording and playing. ... and not the USB/IEEE drives? ...
    (microsoft.public.windows.mediacenter)