Conference hydra::amiga_v1

>   I'm trying to add an RX50 to my ami 500 and I've built the motor
>   latch and ID circuit described in previous notes .

This is a good start...

>   From what I've seen in this file it seems no one has actually added
>   an RX50 to an Amiga (please correct me if I'm wrong).

I think you are probably correct, though I recently met a person who
intends to when he returns home from training.

>   I've seen notes from people who have tried and they seem to have
>   the same problems as myself . IE "bad sector header" on a format
>   etc.

This may be due to the one sided nature of the drive. Yes the RX50
is a one sided drive while the default 5.25" drive in the mountlist
is 2 sided. Try making "Surfaces = 1" and that may make the drive
usable, if limited in space.

>   I think the reason for this is that the RX50 sends and recieves
>   MFM data to and from it's controller whereas the IBM style drives
>   that most people have added on succesfully have the MFM decode
>   circuitry built in .

The amiga does the MFM encoding/decoding internally. None of the
floppy drives I have seen decode anything.

>   Also , does any one know if the raw data from the amiga has crc
>   bits in the stream etc.

The amiga floppy (slow) file system has extra information in each
sector to help detect and correct disk problems.

    	-SES

    	Which note has the details of the driver for an RX50?
    

    The Circuit is in note 307.1
    
    If I have the circuit decoded right all it does is use the SEL1B signal 
    from the Amiga to clock the 7474 flip flop.  The MTRXD on the D input 
    resets the flop resulting in pin 6 (Q not) going high.  That anded with 
    the DRESB being high (Drive not being reset) causes the output on the 
    nand gate at pin 11 to go low turning on the drive motor.  It also 
    supplies RDY back to the Amiga.
    
    This is good since the normal Amiga drives are separate drives each one 
    requiring its own Hub Motor enable signal.  With the RX50 this is not 
    the case.  This is a dual drive with its own select logic built in.  It 
    only has one hub motor.  I think we are re-inventing the wheel here.
    
    
    Now we use RX50 terms:
    
    With the RX50 there are four select lines SEL0 thru SEL3.  This 
    combined with a jumper determine how the drives are selected.
    
    with J17 out SEL0 selecct drive A   pin 10
    with J17 out SEL1 selects drive B   pin 12
    with J17 in SEL2 selects drive A    pin 14
    with J17 in SEL3 selects drive B    pin 06
    
    
    Here are the RX50 pins per Rainbow Tech Manual page 6-31:
    
    2	      TKG43 L		Controls write current low after track 43 
    	      	   		is passed
    4	      Reserved
    6	      Drive Sel 3	Selects Drive side B if J17 installed
    8	      Index L		Indicates Index mark of selected side
    10	      Drive Sel 0	Selects Drive side A if J17 not installed
    12	      Drive Sel 1	Selects Drive side B if J17 not installed
    14	      Drive Sel 2	Selects Drive side A if J17 installed
    16	      Motor On L	Turns Spindle Motor on or off
    18	      Direction		Controls Head movement direction
    20	      Step L		Controls Head movement distance
    22	      Write Data L	Data to be stored on diskette
    24	      Write Gate L	Activates write circuits
    26	      Track 0 L		Track 0 head location LED
    28	      Write Protect L	Indicates selected diskette is write 
    	      	   		protected			
    30	      Read Data L	Data Retrieved from Diskette
    32	      Diskette Side 0	H = side 0   L = side 1 ( Always High for 
    	      	   		the RX50 also marked as reserved)
    34	      Ready L		Indicates selected drive side contains a 
    	      	   		diskette
    
    All odd pins are ground.
    
    Keep in mind I have not built this yet.  But I am just proposing it to 
    see what anyone else thinks.
    
    But how about this cable:
    
    Amiga pin 	   		RX50 Pin
    
    1	      /RDY 			34	Ready L
    2	      /DKRD			30	Read Data L
    3-7	      GND  			all odd pins
    8	      /MTRXD			16	Motor On L
    9	      /SEL2B			12	Drive Sel 1 L (B)
    10	      /DRESB			
    11	      /CHNG			
    12	      +5V
    13	      /SIDEB
    14	      /WRPRO			28	Write Protect L
    15	      /TKO 			26	Track 0 L
    16	      /DKWE			24	Write Gate L
    17	      /DKWD			22	Write Data L
    18	      /STEPB			20	Step L
    19	      DIRB 			18	Direction
    20	      /SEL3B
    21	      /SEL1B			10	Drive Sel 0 L (A)
    22	      /INDEX			8	Index L
    23	      +12 V
    	      	   			2	TKG43 L
    	      	   			4	Reserved
    	      	   			6	Drive SEL 3 L
    	      	   			14	Drive SEL 2 L
    	      	   			32	Reserved 
    
    	      
    The RX50 requires the following signals:
    
    I see this as a problem.  The RX50 requires the TKG43 L signal to 
    reduce write current on tracks 44 - 79.  This is supposed to come from 
    the controller.  Without it write current is not reduced.
    
    TRACK 0 L  Valid for the selected drive when the head is over track 0. 
    
    MOTOR ON L the motor reaches the rated speed in .5 secs.
    
    DIRECTION High = Step out (1)  Low = Step in (0).
    
    STEP L must be a minnumum of 1 usec.  The minimum width between 
    Step pulses must be 6 msec.
    
    Write Data L write pulses must begin a minimum of 500ns after write 
    gate is asserted.  A minnimum of 36 ms is required after the last step 
    pulse occurs.
    
    WRITE GATE L Enable writing and tunnel erasing of the diskette.  
    Ignored if write protected.
    
    INDEX L is valid 250 ms after MOTOR ON L is asserted.
    
    WRITE PROTECT L Asserted when diskette is write protected.
    
    READ DATA L is valid 250 ms after MOTOR ON L is asserted and 36 ms 
    after receiving the last Step pulse.  1.21 ms after write gate signal 
    is Unasserted or 30 ms after the drive side is selected.
    
    READY L indicates that a diskette is present in the selected drive.  It 
    is valid only for the selected drive.
    
    
    On the Amiga Side I do not see a use for the DRESB and CHNG signals.  
    The CHNG signal is not used on a 5 1/4 inch floppy.  You have to use 
    the Diskchange command to tell the system that a new floppy has been 
    inserted (Correct me if I am wrong).  Is the Drive Reset signal used on 
    a 5 1/4 inch floppy?
    
    I also do not know what affect not using reduced write current on the 
    floppy will have.
    
    The mount list should take care of the Non Standard format: 80 tracks 
    10 sectors per track and 1 head.
    
    But will it read and write A Rainbow MSDOS Floppy?
    
    
    Any feed back is welcome.
    
    Alan
    
                                

    I went ahead and built the cable and the RX50 drive select logic is not
    as smart as it looks.  If the Motor on signal goes low the motor spins. 
    It does not matter if the drive is selected or not.  So the latch
    circuit would be needed.  In either case I still could not format the
    drive.  It would fail trying to verify cyl 0.  Does anyone have the
    specs for the Amiga's write data signal.  Perhaps the RX50 spins at the
    wrong speed?
    
    Alan
    

I think RX50 requires it's own media (sold only by DEC) since it has very 
critical allignment requirements.

John