Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Dilog SQ703A to DAT on 11/53

12 views
Skip to first unread message

shadoooo

unread,
Oct 12, 2009, 1:30:20 PM10/12/09
to
Hello.
I'm trying to use my Dilog SQ703A with a DAT device.
I don't have any manual for this card, so basically I followed the
internal menu to try to configure it.
I entered the console configuration utility and tried to map my SCSI
01 ID to the host 00 device in TK50 emulation.
The termination has been enabled on the DAT device, obviously
configured for SCSI ID 01.
The DAT device is a common Seagate CTD800R, so nothing special. It
works good with my VAX3100.
On the configurable values, I didn't understand what the IP/SA value
is, now is 160404.
My machine is a PDP11/53, what should be this value?
To simply try the SCSI communication, I tried to send a direct SCSI
command through the interface to the DAT.
For example I sent the TEST UNIT READY or the INQUIRY. In both cases I
have to wait for 20 seconds, after that I receive a message telling me
that the SCSI command was executed but nothing else... But I expected
a fairy fast response and some data, specially executing the INQUIRY.
Trying to use the drive as it was a real TK50, it doesn't works.
I don't understand where the problem is... Somebody experimented
something with this board?
Thanks

Rob Brown

unread,
Oct 13, 2009, 1:35:44 PM10/13/09
to
On Mon, 12 Oct 2009 at 10:30 -0700, shadoooo wrote:

> On the configurable values, I didn't understand what the IP/SA value
> is, now is 160404. My machine is a PDP11/53, what should be this
> value?

IP and SA are the bus addresses of the two registers on the SQ703A.
160404 is a *valid* address for this, but I can't say whether or not
it is the *right* valid address. There are a few possibilities here:

1. The board address is selected by jumpers on the board, so the
board knows what address it is configured for.

2. The board address is whatever you tell it at this configuration
step.

3. The 160404 number offered you is "default" or "random" and has no
basis in reality.

In any case, you need to set an address for this board that does not
conflict with another installed device. The fact that your machine is
an 11/53 is of less importance than the addresses of the other
peripherals installed.

The "fixed address" for the first TK50 is 174500. If you don't have a
TK50 or TU81 already installed, this is probably the address you
should use. The first floating address for a TK50 is 160404, although
this can be affected by what other devices are installed. Or, you
could just set it for any unused address, which will probably work but
will confuse any auto-configuration software you might use.

If you don't know what addresses are available, then your best bet is
to remove the SQ703A, and then probe various addresses to see what is
available. Once you find an available address, install the board and
configure to that address (in whatever order is required), and see how
it works.

For probing addresses use a command supplied by your operating system
(eg. OPEnregister in RSX), or use console ODT.

You will find a Floating Address Worksheet in appendix D of the
RSX-11M-Plus Sysgen manual. Try
<http://www.bitsavers.org/pdf/dec/pdp11/rsx11/RSX11Mplus_V4.x/2a/AA-H431H-TC_RSX-11M-PLUS_4.3_System_Generation_and_Installation_Guide_Jan90.pdf>.

hth

- Rob


--

Rob Brown b r o w n a t g m c l d o t c o m
G. Michaels Consulting Ltd. (780)438-9343 (voice)
Edmonton (780)437-3367 (FAX)
http://gmcl.com/

Rob Brown

unread,
Oct 13, 2009, 6:52:05 PM10/13/09
to
On Tue, 13 Oct 2009 at 11:35 -0600, Rob Brown wrote:

> If you don't know what addresses are available, then your best bet

> is ...

Of course your operating system may be able to tell you everything it
knows about. For example RSX-11M-Plus's CONfigure DISplay ATTributes
command.

shadoooo

unread,
Oct 17, 2009, 2:41:40 AM10/17/09
to
I changed the address, because it was configured for the secondo TK50
interface, while I need the first one.
Then I analyzed a little the circuits connected to the SCSI connector,
enough to understand that it's a HVD SCSI interface, not a SE.
But not too bad, I build an adaptor. So I could retry with inquiry
commands, and this time it worked, my DAT is reported as "ARCHIVE".
Unfortunately it seems not to work good yet, issuing a dd command in
ultrix causes errors.... Sometime it shows a series of "SV 0" messages
on the screen, sometime it goes to a kernel panic trap. There are
remaining configuration parameters like "dwell time", "burst size",
"IRQ priority", which I don't understand very well...
Maybe my DAT is not supported? Or maybe there's something wrong with
the TK50 emulation?


Rob Brown

unread,
Oct 20, 2009, 2:00:36 PM10/20/09
to
On Fri, 16 Oct 2009 at 23:41 -0700, shadoooo wrote:

> Then I analyzed a little the circuits connected to the SCSI
> connector, enough to understand that it's a HVD SCSI interface, not
> a SE.

At least some of the DiLog SCSI controllers were jumper selectable
between single ended and HVD. On the SU726A:

DIFFERENTIAL INTERFACE
JP15 - Differential sense
installed = SCSI Differential Sense Pin J1-21 normally
high. J1-21 grounded will reset the SCSI
interface Chip and disable SCSI Drivers.
removed = Disable SCSI Differential sense

SINGLE ENDED INTERFACE
JP18 - jumper 1-2 = Cable shield to logic ground
jumper 2-3 (or no jumper) = Cable shield not connected
to ground

INTERFACE CONTROL
JP21 - Jumper Removed = Single ended
Jumper 1-2 = Reserved
Jumper 2-3 = Differential

FWIW

I have no idea about the SQ703A.

Good luck. Have you tried bitsavers? And for that matter, the last
time I checked DiLog was still in business. http://www.dilog.com/

shadoooo

unread,
Oct 23, 2009, 7:07:22 PM10/23/09
to
>
> At least some of the DiLog SCSI controllers were jumper selectable
> between single ended and HVD.

Hm, not mine I think. This PCB clearly could be assembled with the
50pin in two different positions, one for HVD, the other for SE.
The only way could be an assembly modification, but the desoldering
operation could be very dangerous...

0 new messages