• Please review our updated Terms and Rules here

TRS-80 Model II disk controller with alternate drives

NeXT

NeXT

Veteran Member
Joined
Oct 22, 2008
Messages
8,208
Location
Kamloops, BC, Canada
I have a 8709474 Rev PP2 hard disk controller in my TRS-80 model II and I'm lacking the appropriate external hard disk. I'm curious if I am able to substitute for another drive like the ST-225, TM-602, Miniscribe 8425, or whatever else I have in my stockpile of MFM drives. If not, what is the cause? is the drive type hard-coded?
Be cool to have a drive tucked away inside the system itself.
 
Please see Frank Durda's excellent HD reference at: http://nemesis.lonestar.org/computers/tandy/hardware/storage/mfm.html

If you ID your card based on his description of the controllers I can better answer your question. Basically, if you have what he calls controller #4 then any MFM drive can be made to work. The HD information is not hardcoded in any of them. You have to enter all parameters. I think the configuration is stored on 0/0/0 or some other very early reserved position. At least that is what I remember. I know for Xenix I've been able to use any drive.

If you have any of the other cards 2 or 3 you will also need an interface board similar to the WD 1005-HDO to convert the tandy interface to the correct MFM signals.

If you have card 1 I think you're out of luck.
 
The presence of a WD1010 and two disk channels indicates I have a #4 controller. There's blue BERG connectors for DATA and CONTROL. There is also two pin headers for DATA and CONTROL with the two CONTROL connectors electrically linked and the two DATA connectors not being linked indicating two separate drive channels. From what I read this controller is ST-412 compatible so in theory I can just hook up an ST-225 and it will work fine with it, right?

Funny enough, I do also have a #1 card laying around. Not much use though as it seems it's for the 8" drives only.
 
Last edited:
Well, for Xenix you can then use any drive. If the drive has more than 8 heads you have to mask one of the pins (pin 2 on the data cable to the drive). For other OS's I don't know what support there is. LS-DOS is pretty generic, but it may not understand the board. That was a pretty late board, only used in the 16B and 6000 line officially. I don't know if CP/M or TRS-DOS will understand it either. Give it a try.
 
Well the ST-225 has four heads so I should be good there. I have disk images of Xenix stashed away that I'll try once I finish a 5.25" mod but all I really have immediately on hand is TRSDOS 2 and CP/M 2.25B.
 
OK, for Xenix there are some other issues. What 68000 boardset do you have? If it is the older "dog eared" card, I think you can only use versions prior to 3.0. If you have the newer types of cards you then have to verify that the PAL in U36 is labled 3646D4. If it is, you're good up to the last version (3.2). If it has a different checksum I think you can only go to 3.1.
 
Last edited:
kb2syd said:
Well, for Xenix you can then use any drive. If the drive has more than 8 heads you have to mask one of the pins (pin 2 on the data cable to the drive). For other OS's I don't know what support there is. LS-DOS is pretty generic, but it may not understand the board. That was a pretty late board, only used in the 16B and 6000 line officially. I don't know if CP/M or TRS-DOS will understand it either. Give it a try.
Click to expand...

Hey Kelly,

This piqued my interest. I have drives with more than 8 heads I've thought about using with Xenix. I pulled up the pinouts for the ST-506 interface. Any chance you mean pin 2 on the control cable? And by mask, do you mean to break the connection?

Kevin
 
U36 does not have a sticker on it. It isn't even a PAL. U34 on the other hand has one marked 34063F.
So I found no traces of an HDD format tool on my TRS-DOS disk and oddly the hdinit utility on both copies of my CP/M disk are corrupt in the exact same location if I hex dump them so I can't use them either to mess with the drive.
The ST-225 will at least light up the ACTIVITY light when the system is turned on but it never really does anything.
 
Kevin Adams said:
Hey Kelly,

This piqued my interest. I have drives with more than 8 heads I've thought about using with Xenix. I pulled up the pinouts for the ST-506 interface. Any chance you mean pin 2 on the control cable? And by mask, do you mean to break the connection?

Kevin
Click to expand...
You are correct. My bad. It is pin 2 on the Control cable. Note that you can only format up to 8 heads. I think the max for that geometry is around 70 MB. And by mask I mean break off, or put a piece of tape over the card edge connector on the drive.
 
NeXT said:
U36 does not have a sticker on it. It isn't even a PAL. U34 on the other hand has one marked 34063F.
So I found no traces of an HDD format tool on my TRS-DOS disk and oddly the hdinit utility on both copies of my CP/M disk are corrupt in the exact same location if I hex dump them so I can't use them either to mess with the drive.
The ST-225 will at least light up the ACTIVITY light when the system is turned on but it never really does anything.
Click to expand...

On all of my newer 68000 boards there is a PAL at U36. I could not find one of the older "dog eared" versions. The old boards are physically larger than the other boards in the model II, and have a cutout for that retaining bar. What chip do you have in U34? If not a PAL it will not run the later versions of Xenix.

Here is some more info:
http://mobile.google.com/group/comp.sys.tandy/msg/2882cef96dfab46d?dmode=source&output=gplain
Code: 
----------------------------------------------------------------------------

DATE:		July 11, 1985

REVISION DATE:	July 11, 1985

BULLETIN NO.: 	12/16B:44

PRODUCT:	26-6004/5/6 Model 16B

SUBASSEMBLY:	AX-9416 6 MHz (short) 68000 CPU board (All Revisions)

----------------------------------------------------------------------------

PURPOSE:	Describe modifications to allow XENIX 3.0 to run on a
		6MHz CPU board.

----------------------------------------------------------------------------

DISCUSSION:	The following describes the procedures for modifying 6 MHz,
		or "slow" 68000 CPU boards to allow the running of XENIX
		System 3.  It is important to note that this modification
		applies only to the "short" 68000 CPU's only.  

		The "long" 68000 CPU boards cannot be modified to run
		XENIX System 3.  [Although it was technically possible. - Ed]
		
		The terms "long" and "short" 68000 CPU board refers to the
		width of the board itself.  These boards can also be identified
		by the component location number silk screened on the board.
		For "long" boards, the 68000 processor is located at U22.
		On the "short" boards it is at U19.

----------------------------------------------------------------------------

PROCEDURE:

NOTE: The parts required for this upgrade will be supplied in the upgrade kit.

1.	Install the new PAL in U36.

	a.)	Remove the old PAL in the U36 position on the 6MHz 68000 CPU
		board.
	b.)	On the new PAL (U36), bend pin 5 up so as not to come in 
		contact with the socket when installed.
	c.)	Install new PAL U36  (device number 3646D4, or 46D4)
	d.)	Jump U36 pin 5 to U25 pin 9.

2.	BERR modification.

	a.)	On the SOLER SIDE, cut trace at U21 pin 9.
	b.)	Jump U9 pin 4 to U33 pin 13.

3.	Install the new U48 PAL.

	a.)	Remove old U48 PAL (device type 16R6)
	b.)	Install new U48 PAL (decice type 16R4)  This new PAL
		will have check sum 483999 stamped on top.

4.	WAIT modification.

	a.)	Add 330pf cap from the ground end of C43 to the feed-thru
		near the 5 volt end of C37 (this feed-thru is hooked to
		TP-28).
	b.)	On the SOLDER SIDE of the board, add a ground strap jumper
		(22-24 gauge stranded wire) from the ground end of C29 to the
		ground end of C28.

5.	Configure the refresh jumper to E1-E2 (E2 is the middle pin).

6.	Ensure compliance with ALL other Technical Bulletins, paying
	particular attention to the following: 

		12/16B:10
		12/16B:23
		12/16B:24
		12/16B:39
		12/16B:40

7.	Reassemble unit and test with 68000 diagnostics and XENIX 3.0
	[or later].


REPLACEMENT PARTS LIST
New PAL U36		MX-2125	26-6021	(REPAIR PART ONLY)
New PAL U48		MX-2124	26-6021
330pf capacitor		CF-1830 26-9999C
Ground Wire		Use 22-24 gauge insulated stranded wire.

----------------------------------------------------------------------------
 
Last edited:
I completed the addition of the hard drive into the system.

100_2767.jpg

100_2769.jpg

100_2772.jpg

100_2778.jpg


A bit ghetto but it seems to be working well and not interfere with the CRT, probably thanks to the use of the sheet metal for shielding. Power is supplied through the guts of a wallwart for a USB CD drive. The rest of the wallwart was used as an insulated support that sits above the main PSU.
You can also barely see my M68000 boardset. Yes it is the larger, older version.
 
Looks nice. Does it boot anything? I think a lot of the older operating systems may not understand the controller you have, but I would like to be proven wrong.
 
Unfortunately in both copies of CP/M I own, HDINIT.COM is corrupt in the exact same location. I assume that either the drive that made the disks was funky or the image the disk was made from was partially corrupt.

100_2779.jpg

100_2781.jpg


I also noticed you were looking for the same TRS-DOS disk tools I am looking for about six years ago. Did you ever find them? It could be cool if I could partition the drive so it could be used for CP/M, TRS-DOS and Xenix.
Otherwise to answer your question, no the drive does not yet boot anything.
 
No. I was never able to find an unadulterated copy of that disk. Unfortunately the individual passed away before we got too far in the process.

I'm working on a disk imaging project. I'll see if I have Tandy's CP/M for the II and send you a copy. Did you try the CP/M from Dave Dunfield's archive?
Tandy Xenix did not like partitions. It used whatever disk space you gave it and that was it. In fact, the boot laoder, disk geometry and bad sector map was always on absolute 0/0/0 or something very rigid like that.

If you just 'cat /dev/hd0' it would return the geometry and the bad space map.
 
Did you try the CP/M from Dave Dunfield's archive?
Click to expand...
No I have in fact not yet because I have yet to figure out how to dump the images to 5.25" floppy instead of its expected 8". I'm using a set of floppies I was give off another forum member and were the only bootable disks I owned for the system up until very recently.
 
NeXT said:
No I have in fact not yet because I have yet to figure out how to dump the images to 5.25" floppy instead of its expected 8". I'm using a set of floppies I was give off another forum member and were the only bootable disks I owned for the system up until very recently.
Click to expand...
Get a computer with 5.25" drives and run testfdc to see if your computer can handle the FM mode expected. To do so it must pass Single Density and Double Density at 150 and 300 kbps.

If it can then you can just use ImageDisk to write your image to the 5.25" drive. Set Sides to 1, Single Step with 77 tracks then write the image.

Does that help?
 
BuMP.
Finally got hold of an OS that I could install onto the disk. Immediately found a problem.
I take a known working ST-225 and set it to DS0 the controller will light the BUSY light on the drive and it will never go out.
Furthermore, the format utility (DISKUTIL on the Xenix boot floppy) detects that a drive is attached on channel 0 and gets as far as starting the format before immediately failing and telling me the whole disk is bad.
CRW_7790.jpg

I can confirm the utility can detect if a drive is or is not connected.
If I set to '225 to DS1 the BUSY light does not come on until I tell DISKUTIL that the drive I want to work with is on Channel 1. Again it fails with the same error. I also tried with an ST-251, Kalok KL 330 and Kyocera KC-20B and they all give the exact same symptoms. Not sure if it's cabling or one of the two jumpers on the controller were fiddled with.

CRW_7791.jpg
 
Last edited:
BuMP.
To double check that the drive survived mving I ran the XT-GEN format utility that is in the ROM of a Western Digital WDXT-GEN controller. No errors reported and the format was successful.
Try formatting with the xenix utility again. Same error.
 
You seem to have the old 68000 board set. It does not officially support this controller. I think the clock rate is incorrect.

The controller itself may also be out of spec. I have one that gives the same error all the time.
 
....Oh.

Dammit.
Well I guess that is the end of this project. It isn't affordable to source the boards, regardless of revisions.
 
You must log in or register to reply here.
Back
Top