Conference 7.286::atarist

    
    
    Hi Bernd,
    
    
    Dont re-invent that round object, read NOTE 428.*, IT GETS INTERESTING
    around .9 onwards, also 90.8. 
    
    Cheers
    Pat K.

    Thanks Pat! 
    I'm sorry for posting an already solved problem in the conference. It
    wouldn't have happened if I used the search utility more often...
    I wouldn't mind Mr.Moderator deleting this note.
    Bernd

I thought TOS 1.4 solved this in regards media change on some "generic"
mechanisms.  That is what the USENET is saying.

Chad

    Berndt,
    
    If you compare the times of your base note versus my 90.8 reply, you
    must have entered yours whilst I was still typing!
    
    A note elsewhere in 90.* tells you why the diode trick works.
    
    Regards, Brian
    

    
    Brian,
    
    I read your reply shortly after and was happy to have a solution on
    hand. Yesterday evening I soldered the diode in, and for the first
    some tries it worked great (with ESC and with manually open the drive
    by mouse). Then again the same problem and no change since then...
    I quess I have to jumper other pins with the diode in this TEAC-drive
    (I'm not sure but I think it's a TEAC FD35 or so), I have to look it up
    this evening. Anyway I got no specs for it because it was on sale for 
    about $30. As well there are a lot of jumpers in the drive beside the
    one "Drive Select" which I could identify. Maybe one of those could
    solve the problem also.
    
    Bernd
         

    
    After two nights now of extensive use involving lots of floppy changes
    (transferring the contents of my single-sided disks to double-sided via
    RAM-disk) I can report finding no problems with media changing.
    
    I was fortunate in that the various configuration selectors on my drive
    seem to have been in the right positions - there was no documentation
    with it!
    
    You mentioned it worked correctly initially then stopped. Have you
    checked all the obvious, such as the diode has disconnected itself, poorly
    seated connectors etc.?
    
    btw: Note 90 does mention alternative signals that might need to be used.
    
    Brian

The last night was without any sleep because a friend of mine and I 
oscilloscoped the system for hours. Maybe some points are of interest
for some of the people in the conference.
First I have to further explain my hardware configuration:
The base machine I'm using is a 1040 ST as mentioned before in this 
note. The mother-PCB when I bought the computer was as follows:

           internal Ribbon Conn.               14-Pin Floppy Conn.
           _____________________               ___________________
           | o o o o o o o....o|               |o o o o o o o o o|
           ---------------------               -------------------
             DS0 10| |12 DS1                     DS0 5| |6 DS1
                   |  (not connected)                 | |
                   |                                  |_| jumper on PCB
                   |                                    | set by Atari
Sound Chip         |                                    |
DS0: ______________|                                    |
DS1: ___________________________________________________|

As I had enough room in my PC-case I could place two 3 1/2'' drives into
the box. Later I wanted to connect a to-be-bought 5 1/4'' drive for 
MS-DOS to the system also, so I decided to connect the second small drive
to the ribboncable too. To go with two drives on the same cable and still
leave the 14-pin connector usable for a third drive, I ended up with 
following solution: I extended the ribboncable and placed a second Floppy-
connector to it, below the rest

            internal Ribbon Conn.              14-Pin Floppy Conn.
            ____________________               ___________________
            | o o o o o o o...o|               |o o o o o o o o o|
            --------------------               -------------------
              DS0 10| |12 DS1                    DS0 5| |6 DS1
                    | |                               | |
                    | |  small cable                  |-| jumper
                    | |----------|                      |
 Sound Chip         |            |                      |   
DS0:----------------|            |Drive B intern        |
DS1:---------------------o---o---|                      |
                             o--------------------------|
                        Switch    Drive B extern

For that I had to cut one line on the PCB. Everything works fine, even with
a third drive which is left jumpered as drive "0". The second small drive
inside the case now has to be jumpered as drive "1". To achive clear signals
and not to overload the floppy controller I swapped the 1k pullup resistor 
arrays in the drives against 6.8k arrays. 
So far so good, but... I have those two TEAC FD35 drives and TOS 1.4! I tried
the solution I was pointed to in the other notes concerning drives in this
conference to get a media-change signal, no way! Not on Pin 2 nor on Pin 34,
no jumper setting and no diode solved the problem. So we disassembled the
computer last night and found out the following facts.(on the pins 2 and 34
were no changing signals while swapping diskettes in the TEAC, the only
signal changing was Write Protect). Here are the signals detected on the 
Write Protect line:

                         Original Atari Drive:
                         ---------------------
                    Write Protected             Write Enabled
---------------------------------------------------------------------------
not selected        High with Low Pulses        High with Low Pulses

selected            Low                         High

after removing the diskette, the signals go back to High with Low Pulses!


                         TEAC FD35 Drive:
                         ----------------
                     Write Protected            Write Enabled
---------------------------------------------------------------------------
not selected         High with Low Pulses       High

selected             Low                        High

after removing the diskette, last status is kept on the lines, no change!!

There's no mechanism to give me a pulse when changing the diskettes in the
TEAC FD35, I didn't find any on the drive's PCB. Maybe that's the reason why
they were so cheap with around $30 or $40 (79.-DM).

If there's anybody who has schematics of that drive or thinks I forgot some-
thing while examining, please let me know! I've got to get them running
somewhen.

(P.S. my friend bought the same drive and has no problems using it with the
old TOS, though he couldn't remember if he only used write protected diskettes
while checking the drive. He finds out tonight)

Bernd 

I forgot if you found any other unlabeled jumpers on the drive.  My Toshiba 
drive has a jumper that sets up whether media change pin "works" or not.

Chad

    
    There was one close to the connector, labeled S1/S2. When I measured it
    with an Ohmmeter I found out that this jumper was designed to activate
    Pin 2 or Pin 34. The others were MO (closed), MS (open), FG (closed)
    and the Drive Select jumpers. Another row with six pins labeled 1 to 6
    weren't jumpered at all (I presume they are Testpoints).
    Anyway, Pin 2 and Pin 34 had no effect on changing diskettes, no matter
    how S1/S2 were jumpered, so the diode had no effect on the Write
    Protect line.
    
    Bernd
    

    
    We meanwhile found one point on the PCB which has 5V as long as there's
    no diskette in the drive and Low as long as a diskette is in. We
    used this 5V to drive a DIL reed relais and switch the Ready signal
    of Pin 34 via the diode to Pin 28. Everything works fine now! Thanks
    for all the suggestions I got by mail and in this notes conference!
    
    Bernd
    
    

You mean, you just poked around with a voltmeter until you found a
useful signal?

I admire that approach.  (I used that approach one to find a clock on
the Commodore PET that had an edge in the right place to refresh my RAM
expansion.  A time-honored technique.)

    
    Actually it was a scope, but a friend and I had to do something to make
    that thing work. I had no datasheet or schematic for the drives, in fact
    nothing than the drives themself.
    (There are several users in my area who bought that drive in sale
    from a local electronics mart, so we couldn't give up...)
    
    Bernd                                                    
    

	I read all the notes on using generic drives with the ST and
	didn't expect any problems but . . .

	I'm adding a Toshiba ND-352TH-A also known as FDD4216G0D as a
	second floppy.  If I leave the media change diode out, everything
	works fine, except of course media change is not detected.  With
	the diode, once the disk is removed once it always appears write
	protected.

	So I looked at the signals (only have a logic probe at home - but
	it's a good one :-)  ).  The "no media change" condition is 
	indicated by a "high" on pin 34 (jumper selected for this function)
	all the time.  "Media change" is indicated by that pin toggling
	with no drive select or constant low when the drive is selected.
	It seems that the only way I can get "no media change" is to
	power up the new drive with a disk inserted AND then reboot the ST.
	Once the disk is removed once, "media change" is always indicated.

	So, my questions are:
		1) What is the media change signal really supposed to do?
	    	   I guess that really is, what does Atari expect it to do?
		2) What is supposed to reset it from the "media change"
		   condition?
		3) Is it possible to use this drive with the ST?

	thanks
	chris

	-- I guess I'll also post this to USENET.  I'll summarize any
	   replies here.

I replaced my SS drive in my 520STfm 1meg machine with a Toshiba DSDD drive.
That part number you gave sounds sort of familiar.  What I did was stick a cheap
slider switch in the diode path.  When closed, media change works but often the
new disk apears write protected -- open th switch and this "write protectedness"
goes away.  Just remember to switch the switch closed again before changing
disks or the mchine won't realize you've changed disks and you can wipe
out the new disk (speaking from experience...).

Chad

	Well, I now know a lot more about how media change works on 
	Toshiba drives (at least the one I have) than I ever wanted to,
	but the problem is fixed.  As these things often are, its really
	quite simple.

	The disk change signal provided by the drive is essentially useless.
	It goes active when the disk is removed and isn't reset until
	apparently the first time the head is moved.  (It is gated by
	drive select of course)  So the diode fix doesn't work.  If you
	remove a disk and reinstall it, then try to write to it, the ST
	reads the directory off track 0 but doesn't move the heads.  The
	DC signal stays active and, since you cleverly installed the diode,
	the disk appears write protected to the ST.

	Of course all the circuitry controlling this is in a custom chip.
	I had a two gate circuit that was going to fix this but I got a
	much better suggestion off the Usenet.  If you make the drive think
	that it contains a write protected disk when no disk is inserted,
	all is well.  This drive and apparently others don't assert write
	protect when the disk is removed.

	There are two small mechanical switches (near the LED, Chad, if you
	want to change yours), one detects write protect, the other disk
	insertion.  If the etch is cut going to the disk switch everything
	works fine.  The drive asserts write protect when the disk is
	removed and no diode or switch is needed.  (The guy on the Usenet
	had a Chinon drive - he had to short the switch.)

	chris