Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!samsung!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!cbnewsl!moss!feg
From: feg@moss.ATT.COM (Forrest Gehrke,2C-119,7239,ATTBL)
Newsgroups: comp.sys.ibm.pc.hardware
Subject: Re: TECHNICAL ?'s ABOUT 1.44M FLOPPIES
Message-ID: <1991Feb18.202551.2088@cbnewsl.att.com>
Date: 18 Feb 91 20:25:51 GMT
References: <1991Feb12.215709.17456@coplex.uucp> <355@einstein.eds.com> <1991Feb17.152818.4901@cc.helsinki.fi>
Sender: @cbnewsl.att.com
Reply-To: feg@moss.ATT.COM (Forrest Gehrke)
Organization: AT&T Bell Laboratories
Lines: 98

In article <1991Feb17.152818.4901@cc.helsinki.fi> torvalds@cc.helsinki.fi writes:
>> In article <1991Feb12.215709.17456@coplex.uucp>, dannie@coplex.uucp (Dannie Gregoire) writes:
>>> I have been trying to find some information on the hardware differences
>>> between 720K and 1.44M floppies.  Specifically, what are the differences
>>> in the pin outs (and signals sent to the PC) between the drive's connectors.
>>> 
>>> What signal (if any) is sent to the PC when the drive detects that a high
>>> densisty has been inserted.
>
>Right, this info is straight from the EPSON SMD-340 3.5" floppy drive
>manual (not guaranteed, but should of course work for others as well :-)
>
>Pin	Name	what?
>
>2	HDI	High density in (low in 2MB mode (2MB unformatted=1.44MB))
>4	NC
>6	DS3	Drive select 3 (low -> selected)
>8	IDX*	Index
>10	DS0	Drive select 0
>12	DS1	Drive select 1
>14	DS2	Drive select 2
>16	MON	Motor on (low - on)
>18	DRTN	Direction (low - step inwards)
>20	STP	step (transition low->high steps motor, usuallu 3 ms)
>22	WTD	Write data
>24	WTG	Write gate (active low)
>26	TK00*	Track 0 (low means we are on track 0)
>28	WPT*	Write protect (low means write protected)
>30	RDD*	Read data
>32	SSL	Side select (note low means side 1, high means side 0)
>34	DCH*	Disk change
>
>pins 1,3,5 not connected, others grounded.
>* means signal from disk drive, otherwise to drive unit.
>
>NOTE! The HDI is not from drive to computer but the other way around.
>This can force the drive to use 2MB mode (1.44MB). This is in fact not
>used on most drives to my knowledge, as the drive senses the density
>from the disk covering. This can be enabled by a jumper on my drives.
>
>There are drives available that use pin 2 as HDO (High density out) to
>let the computer sense the density used, but I wouldn't know where to
>find them. Some can also use pin 34 to indicate RDY, ie that the drive
>is up to speed etc. The MON can also be ignored by some drives which use
>the select lines to activate the motor as well.
>
>This information is in no way guaranteed to be correct.
>
>		Linus Torvalds  torvalds@cc.helsinki.fi

To add to the above, as concerns Epson drives (I don't know
about others):

Just behind the 34 pin edge connector there are two rows
of 8 pins each which are jumpered in various ways as 
follows (counting pins with least significant from left
when looking at rear of drive):

        9   x   x   x   x   x   x   x   x  16

		1   x   x   x   x   x   x   x   x  8



	14 to 15  Drive select 0
	15 to 16  Drive select 1 (std. setting)
	6 to 7    Drive select 2
	7 to 8    Drive select 3
	4 to 5    Mode switched to internal sensor (std. setting)
	5 to 13   Mode switched by HDI input
	1 to 2    2M mode set by HDI "high"
	2 to 3    2M mode set by HDI "low"  (std. setting)
	9         Grounding
	10 & 11   Test terminals
	12        Test terminal

As shipped 2 to 3 and 4 to 5 are jumpered.  This results in 
the internal sensor turned on and formatting capacity 
determined by whether the floppy has one or two holes.

Changing this to jumpers across 5 & 13  and  1 & 2 will
result in the internal sensor being turned off.  If a
format is requested without command line sensors the
floppy will be formatted to 1.44 MB regardless of whether
the floppy has one or two holes.  Formatting to 720 KB 
has to be requested with parameters, henceforth.
Reading will be determined by whatever the diskette
has been formatted, again, regardless of number of holes.

I have formatted single sided double density diskettes I 
purchased in 1985 to 1.44MB without any difficulty.  I
don't think there is any difference in the media, at least
among the various diskette manufacturers I have tried.

This gets rid of the business of chopping holes in one
hole diskettes.

Forrest Gehrke feg@dodger.att.com