Date   : Mon, 24 Oct 2005 23:17:14 +0100
From   : Jules Richardson <julesrichardsonuk@...>
Subject: Re: Collected notes on floppy drive butchery?

Pete Turnbull wrote:
> On Oct 24 2005, 10:10, Richard Gellman wrote:
>>may say, floppy drives *do* have a write clock - this is generated
>>internally on the drive itself, and synchronised with the index hole.
> 
> What?  If you mean the one-pulse-per-rotation index pulse, yes there
> is a clock.  If you mean a separate clock signal that gets recorded
> with, or in some way matched with, the data, in order to syncronise the
> bits, then I'm afraid you are completely WRONG.  There is no clock in
> the drive.  There are clock bits in the data stream from the
> controller, generated by the controller.  There are also sync blocks
> between sectors, written there by the controller.  There is no write
> (or read) clock in the drive.  The clock used in the data recovery is
> the stream of clock bits recorded with the data.  Once again there is
> no clock signal created by the drive electronics.

Seconded - it's the controller which takes care of generating the bit 
stream (included embedding clock pulses, using whatever recording method 
- eg. FM, MFM) when writing to the drive. Ditto upon reading data.

My memory's having a off-day, but don't some/all *8 inch* drives have 
data seperator circuitry on board and are capable of presenting seperate 
clock and data signals when reading FM encoded floppies? I have a 
feeling they still need a combined data stream when writing however...

The relevant manuals are on the shelf back home and I don't feel 
motivated to download scanned copies :-)

cheers

Jules