RL02 Disk Cartridge Problems - why that particular pack is junk

From: Johnny Billquist <bqt_at_update.uu.se>
Date: Tue Dec 10 19:27:00 2002

Jumping in a bit late...

On Tue, 10 Dec 2002 ard_at_p850ug1.demon.co.uk (Tony Duell) wrote:

[description of RL02 physical format deleted...]

> > There is also a "guard track" on each side of the data track,whose
> > purpose is to cause the head to go back to the center of the data track
> > should it drift.
>
> Is there? I was under the impression there were guard bands on the very
> inside and very outside of the disk (used when the heads are loading,
> etc). Bot not between each pair of data tracks.

I think you are correct, and that the original poster misunderstood the
manual. (Actually, I'm pretty sure you are correct :-)

> There are, however, offset (from the track centre line -- in fact I think
> they're halfway between 2 adjacent tracks) signals in the header area.
>
>
> Something like :
>
> ===== ====
> ---DATA---- -----Next data-----
> ==== ====
> -- Data----- ----Yet more data--
> ===== ====
>
> Where the ==== are the servo signals.

Huh? No. Actually, the data track itself is used as the servo signal. The
head centers in on the track by centering on where the amplitude is
strongest.

> > When a drive is told to seek to a particular sector, it reads the
> > factory written sector header to determine if it is the correct sector.
> > If the drive can not find this info quickly enough, it will fault.
>
> Sorry, but no. The drive doesn't check the headers at all. I've been
> through the schematics and there's nothing that will look at digital
> data on the disk. It will fault if it can't find the servo signals, though.

Actually, the *drive* just moves a number of tracks back and
forth. (And seek is just to a relative track). When you progam a driver to
seek to a track, you check the current track, calculate the track delta,
and request that of the drive. When the drive report ready you once again
check the track to see which one you actually are on. Hopefully it is the
right, but if not, do another seek.

However, when you *read* a sector, the *controller* checks that you read
the right sector by checking every sector header that the head passes
over, and when the right one comes, the data is transferred (the same
applies for write). If the right sector don't show up, it will fail, but I
don't think it will fault.

> > It seems to me that to correctly format one of these disks after a bulk
> > erasure, you would have to be able to do the following:
> > 1. Correctly position the head - probably the easiest part
>
> I would think that generating the right pattern of bits for the header
> was the easy part. That's just (relatively slow) digital electronics.

Correctly position the heads when you don't have a servo track, and the
heads are actuated by voice coils, means you cannot position the heads at
all.
So that is a big hurdle. Find another type of drive for formatting, that
is item #1.

> > 2. Write the correct sector header, including addressing

Which no RL01/RL02 controller can do. There is no function to write sector
headers. So, in addition to finding another drive to be able to position
the heads, you need another controller, to be able to write the data.

> > 3. Write the guard tracks ( I bet this is the most difficult)
>
> If you mean the 'servo bursts', well, you need to be able to accurately
> position the heads on half tack spacing. Actually writing the signals is
> easy.

Yes, once the above conditions have been fulfilled, actually writing is
the easy part.

> > 5. Create the "Bad Block" file at the end of the disk.
> >
> > So, looking at the above, I can see where it would very difficult to
> > format one of these guys using a stock RL01/RL02 drive, even modified.
>
> IMHO the really hard part is going to be positioning the head accurately
> at half track spacings...

You don't have any half-track spacings. However, you also need to write
the guard bands, which are outside the normal data area.
But if you happen to have a drive you can move the heads about with
absolute precision (well, pretty good precision anyway), and being able to
write headers, I doubt writing the guard bands would be that hard. You
just have to know how the guard bands look physically. The drive recognoze
them, and immediately move back to the data area if they show up.

        Johnny

Johnny Billquist || "I'm on a bus
                                  || on a psychedelic trip
email: bqt_at_update.uu.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
Received on Tue Dec 10 2002 - 19:27:00 GMT