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 Historical content remains viewable.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

brew-your-own unibus card - bus transciever chips?

22 views
Skip to first unread message

Brad Parker

unread,
Jan 25, 2004, 9:14:10 PM1/25/04
to
Hi,

I've been asking people on other lists (pups, cctalk) about making my
own unibus cards.

The discussion has come down to bus transceiver chips. None of the
National chips which DEC used to use are in production anymore (sadly).

If anyone here has thoughts on this, or would like be cc'd as part of
the discussion, send me email and I'll add you to the cc' list.

Most of the issue surrounds finding drivers and receivers which have
very low bus current (<100uA), the right input levels (1.3-1.7v) and
drivers with the ability to sink 70ma but don't slew too fast.

Some RS-423/485 style drivers come close not are not perfect. Someone
suggested using FET's with RC's on the gates (not a bad idea). The
receive side is tough as most receivers are .8-2.0v... Maybe a CMOS
receiver with vcc=3.3v.

anyway, if you're interested in this or have ideas/info, send me email.

thanks!

(this all started when I asked if anyone else on the planet would like
a simple unibus IDE interface card :-) I plan to do this, but I want
to make a clean card which will work on any unibus, hence the
driver/receiver issue)

-brad

Al Kossow

unread,
Jan 25, 2004, 9:32:58 PM1/25/04
to
From article <401477F2...@parker.boston.ma.us>, by Brad Parker <br...@parker.boston.ma.us>:

> The discussion has come down to bus transceiver chips. None of the
> National chips which DEC used to use are in production anymore (sadly).
>

They only went out of production in 2002. I can't believe you won't be able
to find the small number that you'd need for your projects.

Have you tried surplus channels?


Geoffrey G. Rochat

unread,
Jan 26, 2004, 12:48:07 PM1/26/04
to
> > The discussion has come down to bus transceiver chips. None of the
> > National chips which DEC used to use are in production anymore (sadly).
> >
>
> They only went out of production in 2002. I can't believe you won't be
able
> to find the small number that you'd need for your projects.
>
> Have you tried surplus channels?

The National chip was the DS8641. The DS3662 was used for later designs,
and the DS8838 for earlier ones, Omnibus PDP-8s in particular.

Check out the AM26S10 and SN74ABTE16246, both in production, AFAIK, by Texas
Instruments. Although neither one is sanctified as an official Unibus or
QBus chip, and the pinouts don't even approach the National parts,
electrically the specs suggest that they may do fine. (Please keep in mind
that this "advice" may be worth no more than you paid for it.)


Toby Thain

unread,
Feb 5, 2004, 6:31:17 AM2/5/04
to
Brad Parker <br...@parker.boston.ma.us> wrote in message news:<401477F2...@parker.boston.ma.us>...

> Hi,
>
> I've been asking people on other lists (pups, cctalk) about making my
> own unibus cards. ...

>
> (this all started when I asked if anyone else on the planet would like
> a simple unibus IDE interface card :-) I plan to do this, but I want
> to make a clean card which will work on any unibus, hence the
> driver/receiver issue)

Brad,
I hope you will keep us all informed of your progress, and will
publish your eventual design; a UNIBUS card such as you are designing
would be a godsend to PDP preservationists such as myself...
T

>
> -brad

Bill Gunshannon

unread,
Feb 5, 2004, 9:16:37 AM2/5/04
to
In article <d6ce4a6c.04020...@posting.google.com>,

I can't speak for anyone else, but if they could be made fairly
inexpensively by getting a bunch of orders together I would really
like to have some. I would be interested in the same for QBus.
Assuming I could afford them, I would be looking at as much as a
half dozen of each!! I have a bunch of PDP-11's here just crying
for something better than my last few remaining RA disks. Especially
if they look like DU: to the system.

bill

--
Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
bi...@cs.scranton.edu | and a sheep voting on what's for dinner.
University of Scranton |
Scranton, Pennsylvania | #include <std.disclaimer.h>

glen herrmannsfeldt

unread,
Feb 5, 2004, 3:48:16 PM2/5/04
to
Brad Parker wrote:


> I've been asking people on other lists (pups, cctalk) about making my
> own unibus cards.
>
> The discussion has come down to bus transceiver chips. None of the
> National chips which DEC used to use are in production anymore (sadly).
>
> If anyone here has thoughts on this, or would like be cc'd as part of
> the discussion, send me email and I'll add you to the cc' list.

I was just wondering if this would work in a VAX 11/780?

That is, could the system disk be on such a unibus card?

-- glen

Mike Cesari

unread,
Feb 5, 2004, 5:01:40 PM2/5/04
to

Yes. UDA50's and various Emulex and CMD controllers come to mind.
RK07's, too, before VMS got big.

Mike
--
mcesari
(at)
comcast(dot)net

Toby Thain

unread,
Feb 11, 2004, 4:28:40 AM2/11/04
to
Brad Parker <br...@parker.boston.ma.us> wrote in message news:<401477F2...@parker.boston.ma.us>...
> Hi,
>
> I've been asking people on other lists (pups, cctalk) about making my
> own unibus cards.
> ...

> (this all started when I asked if anyone else on the planet would like
> a simple unibus IDE interface card :-) I plan to do this, but I want
> to make a clean card which will work on any unibus, hence the
> driver/receiver issue)

Brad

Have you seen the work by Dmitry Pryanishnikov cited in this posting?
http://mail-index.netbsd.org/port-vax/2000/01/03/0015.html

Like Brian, I would love to obtain one or two (or 6!) of these. I am
sure we could arrange quite a number of pre-orders among the PDP/VAX
community for anyone who was willing organise a short run. As Brian
also mentions, the remaining issues with Dmitry's project (apart from
DMA) seem mostly to involve writing drivers for a wider range of O/S
(but MSCP emulation would be nice).

\\
Q: What does this hardware do?
A: This hardware is an adapter (not a controller) that provides up to
4 logically-independent ATA channels for UNIBUS/Q-bus machine which
allows use up to 8 (theoretically) ATA (IDE) devices like HDD, CD ROM
and others. ... Adapter just maps ATA registers to UNIBUS/Q-bus I/O
space and generates standard ATA cycles and UNIBUS/Q-bus interrupt
sequences. This version of adapter DOES NOT provide DMA capability.
This only has effect on performance. ...

Q: On what machines has this adapter been tested?
A: We have tested Q-bus variant on LSI-11 (Q-bus with 16-bit address),
DVK-4 (22-bit Q-bus). ... UNIBUS variant has been tested on SM1420
(Soviet PDP-11 clone) and SM1700 (Soviet VAX-11/730 exact clone).

Q: What operating systems are supported by this package?
A: For now, RT-11 version 5.x, RSX-11M and RSX-11M-PLUS (should work
with any modern version of RSX, e.g., known to work with RSX-11M V4.1,
4.2, RSX-11M-PLUS V4.5). On VAX, support is limited to VAX/VMS v4.x.
...
//

Since Dmitry has given permission to redistribute this work, I am
mirroring the files at http://www.telegraphics.com.au/pdp11/dsu/

Only slightly OT is Chuck Dickman's QBus-IDE adapter - announced
http://minnie.tuhs.org/pipermail/pups/2002-January/000339.html &
described http://www.chd.dyndns.org/qbus_ide/ (inaccessible right
now(!!) - anyone have a legal mirror?)

Toby

>
> -brad

Brad Parker

unread,
Feb 15, 2004, 12:57:32 PM2/15/04
to
Toby Thain wrote:
>
> I hope you will keep us all informed of your progress, and will
> publish your eventual design; a UNIBUS card such as you are designing
> would be a godsend to PDP preservationists such as myself...

Don't worry, most people complain I keep them *too* informed :-)

I don't normally read usenet (anymore), but I plan to start reading
this. I'm in the middle of bringing, my 11/34a back to life (it's going
well so far) and I just got a 730. Once I replace the 730's TU58
rollers I should be in good shape, unibus wise.

I don't have any QBUS systems but, heh, I doubt that will last long.
I'm still working out the bus transciever issues but I should have a
solution soon and I plan to initially make a very simple UNIBUS IDE
controller. I may include a CF slot just for fun but I want to keep it
as simple as possible initially. Once that's debugged I plan to make
one with a xilinx chip and a CPU of some sort (perhaps a PIC) which I
cna use as an "emulation platform". My plan is to make the xilinx act
as a smart register set so I can experiment with emulating various
controllers. A 40MHz 18F PIC should be able to keep up :-)

While bringing up my RL02 the other day I started musing about making a
simple RL01/2 emulator. Probably the result of eating some bad chicken
:-) Still, it seems like it might be pretty straight forward. blah.
It's my day job which keeps getting in the way of the fun :-)

-brad


Toby Thain

unread,
Feb 21, 2004, 1:31:10 AM2/21/04
to
to...@telegraphics.com.au (Toby Thain) wrote in message news:<d6ce4a6c.04021...@posting.google.com>...
> ...

> (but MSCP emulation would be nice).

On the topic of MSCP, I notice that the Emulex UC04 controller manual
(February 1986, p6-2) says,

\\
A comprehensive description of MSCP may be ordered from DEC's Software
Distribution Center...
* UDA50 Programmer's Documentation Kit (QP905-GZ). This kit consists
of the following three software manuals:
- MSCP Basic Disk Function Manual (AA-L619A-TK)
- Storage System Diagnostic and Utilities Protocol (AA-L260A-TK)
- Storage System UNIBUS Port Description (AA-L621A-TK)
//

Anyone scanned these?

Toby

Toby Thain

unread,
Feb 21, 2004, 2:15:46 AM2/21/04
to
to...@telegraphics.com.au (Toby Thain) wrote in message news:<d6ce4a6c.04021...@posting.google.com>...
> ...
> Have you seen the work by Dmitry Pryanishnikov cited in this posting?
> http://mail-index.netbsd.org/port-vax/2000/01/03/0015.html
>

Dmitry's site (via http://vaxarchive.org/hw/) is ftp://digital.dp.ua/DEC/

>
> Since Dmitry has given permission to redistribute this work, I am
> mirroring the files at http://www.telegraphics.com.au/pdp11/dsu/
>

T

vax,3900

unread,
Feb 23, 2004, 11:41:41 PM2/23/04
to
Toby Thain wrote:

Yes, at least "MSCP Basic Disk Function Manual (AA-L619A-TK)" was scanned
and is downloadable from the internet. I have a copy in my HD that has the
name "l619atk.pdf". Googling the file name should be fruitful. However,
there are many sections "deliberately omitted" thus the namual is not very
useful. You even have no idea what those omitted sections are about.

>
> Toby

Squidster

unread,
Feb 24, 2004, 10:39:30 PM2/24/04
to
"vax,3900" <vax...@yahoo.com> wrote in message news:<c1ekm5$e8m$1...@charm.magnus.acs.ohio-state.edu>...
<snip>

> > Anyone scanned these?
> Yes, at least "MSCP Basic Disk Function Manual (AA-L619A-TK)" was scanned
> and is downloadable from the internet. I have a copy in my HD that has the
> name "l619atk.pdf". Googling the file name should be fruitful. However,
> there are many sections "deliberately omitted" thus the namual is not very
> useful. You even have no idea what those omitted sections are about.
>

Unfortunately www.pdp11.nl has dissapeared from the face of the Internet.
Could you please post your copy on any webserver and publish the URL?

Thanks.
/squidster

hea...@aracnet.com

unread,
Feb 25, 2004, 12:29:11 AM2/25/04
to
In comp.sys.dec Squidster <squi...@techie.com> wrote:
> Unfortunately www.pdp11.nl has dissapeared from the face of the Internet.

The owner of the system is currently in the US, while the system is not. It
will take him a while to get back to where it is located and get it back
online.

Zane

Bill Gunshannon

unread,
Feb 26, 2004, 8:48:15 PM2/26/04
to
In article <c1hbr...@enews4.newsguy.com>,

hea...@aracnet.com writes:
> In comp.sys.dec Squidster <squi...@techie.com> wrote:
>> Unfortunately www.pdp11.nl has dissapeared from the face of the Internet.
>
> The owner of the system is currently in the US, while the system is not.

Is that Fred? Who if he's reading this should consider this a reminder
that I still have a computer here for him. :-)

Christian Corti

unread,
Feb 27, 2004, 4:06:08 AM2/27/04
to
Bill Gunshannon <bi...@cs.uofs.edu> wrote:
> Is that Fred? Who if he's reading this should consider this a reminder
> that I still have a computer here for him. :-)

And that he promised to send me my 9 track tapes back after more than a
year and many emails. He said that he would be back in Europe end of
January. But I'm still waiting and waiting....

Christian

0 new messages