30 Sep
2014
30 Sep
'14
5:05 p.m.
From: Antonio Carlini
> Does anyone know of a list of quad QBUS cards that
work (or do not
> work) in Q/CD backplanes
There are lots and lots of Q-bus cards so it would be
quite a long list.
Indeed; hence the "or do not work" in my original message - I realized that
that list was likely more manageable.
(Although one possible bug with the 'does not work' list is that one could get
a false positive failure...)
From: David Riley
Conversely, cards designed for the CD interconnect
(RLV11 board sets,
PMI memory, PMI CPUs) should NEVER, EVER be plugged into a Q/Q
backplane.
Off-topic somewhat, but I wonder why DEC put PMI signals, etc on power pins
(I assume that's what zaps a PMI/etc card plugged into a Q/Q slot). Surely
there are enough pins which are bus lines, etc, which one could use? Yes, the
machine probably wouldn't work if one plugged such a card into a Q/Q slot,
but you wouldn't zap boards...
> CB2 is tied to DB2, and the two feed (through a
jumper which is
> normally removed) special power (-5V) for a particular kind of EPROM
-5v is an optional rail on Qbus, anyway.
Actually, I think AB2/BB2 are 'normally' -12V - although not all power
supplies provide the -12 - e.g. the BA11-N only does +5V and +12V.
In general, as long as there are removable grant
jumpers for the CD
lanes
But not all quad cards have removable grant jumpers - e.g. the BDV11 doesn't.
(Not sure why it even _has_ grant jumpers, given that one would usually make
it the last card, especially since it has pull-ups - but I guess it's in case
one doesn't.)
Noel
30 Sep
30 Sep
5:44 p.m.
On Sep 30, 2014, at 20:05, jnc at mercury.lcs.mit.edu (Noel Chiappa) wrote:
Maybe that's what it was thinking of, along with +5v-battery.
My H9270 certainly gets along running off of a PC
power supply's drive cables just fine.
You can install it without the terminators so that, for
example, you could plug it into the first of several boxes
on an extended bus; the control switches are pretty
handy in that situation. But it's also a fairly early Qbus
card.
- Dave
Off-topic somewhat, but I wonder why DEC put PMI
signals, etc on power pins
(I assume that's what zaps a PMI/etc card plugged into a Q/Q slot). Surely
there are enough pins which are bus lines, etc, which one could use? Yes, the
machine probably wouldn't work if one plugged such a card into a Q/Q slot,
but you wouldn't zap boards...
Indeed, that's actually something I'm planning on figuring
out; my current quandary is mapping Unibus pins against
Qbus pins in order to figure out why one might cook the
other (main answer I see so far: Qbus has a standard +12v
power pin on what is normally a data pin for Unibus, but
it's nothing that can't be worked around with modern
electronics). Next step will be figuring out what's going on
with common CD bus pins that cause the same problem; it
may actually be the same thing, but I haven't looked closely
enough at a full pinout of the CD bus to get an idea yet.
> CB2 is
tied to DB2, and the two feed (through a jumper which is
> normally removed) special power (-5V) for a particular kind of EPROM
-5v is an optional rail on Qbus, anyway.
Actually, I think AB2/BB2 are 'normally' -12V - although not all power
supplies provide the -12 - e.g. the BA11-N only does +5V and +12V. In general, as
long as there are removable grant jumpers for the CD
lanes
But not all quad cards have removable grant jumpers - e.g. the BDV11 doesn't.
(Not sure why it even _has_ grant jumpers, given that one would usually make
it the last card, especially since it has pull-ups - but I guess it's in case
one doesn't.) 
7:08 p.m.
On Sep 30, 2014, at 17:44 , David Riley <fraveydank at gmail.com> wrote:
Indeed, that's actually something I'm planning
on figuring
out; my current quandary is mapping Unibus pins against
Qbus pins in order to figure out why one might cook the
other (main answer I see so far: Qbus has a standard +12v
power pin on what is normally a data pin for Unibus, but
it's nothing that can't be worked around with modern
electronics). Next step will be figuring out what's going on
with common CD bus pins that cause the same problem; it
may actually be the same thing, but I haven't looked closely
enough at a full pinout of the CD bus to get an idea yet.
I've composed a spreadsheet or two which compares these pinouts, with the thought of
seeing how hard it would be to make an FPGA-based double-wide card that might be usable in
multiple DEC busses with suitable programming. I'll export to PDF later this
evening...
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
7:33 p.m.
On Sep 30, 2014, at 10:08 PM, Mark J. Blair <nf6x at nf6x.net> wrote:
On Sep 30, 2014, at 17:44 , David Riley <fraveydank at gmail.com> wrote:
You and I are on the same page. I've done it for Qbus and
Unibus SPC (though as I mentioned, I have not covered the
CD lanes, due to a lack of time and somewhat sparse details).
I'd love to have a chat (probably off list) about what can
be done as far as bus drivers/receivers and how to protect
the pins that conflict; I have some concrete ideas that may
work. And it looks like I'll be getting an 11/44 soon, so
I'll actually be able to test Unibus cards...
- Dave
Indeed, that's actually something I'm
planning on figuring
out; my current quandary is mapping Unibus pins against
Qbus pins in order to figure out why one might cook the
other (main answer I see so far: Qbus has a standard +12v
power pin on what is normally a data pin for Unibus, but
it's nothing that can't be worked around with modern
electronics). Next step will be figuring out what's going on
with common CD bus pins that cause the same problem; it
may actually be the same thing, but I haven't looked closely
enough at a full pinout of the CD bus to get an idea yet.
I've composed a spreadsheet or two which compares these pinouts, with the thought of
seeing how hard it would be to make an FPGA-based double-wide card that might be usable in
multiple DEC busses with suitable programming. I'll export to PDF later this
evening... 
8:51 p.m.
On Sep 30, 2014, at 8:33 PM, David Riley
<fraveydank at gmail.com> wrote:
> On Sep 30, 2014, at 10:08 PM, Mark J. Blair <nf6x at nf6x.net> wrote:
>
>
>> On Sep 30, 2014, at 17:44 , David Riley <fraveydank at gmail.com> wrote:
>> Indeed, that's actually something I'm planning on figuring
>> out; my current quandary is mapping Unibus pins against
>> Qbus pins in order to figure out why one might cook the
>> other
This sounds like you are trying to figure out why a qbus board plugged into a unibus
backplane might blow?
Or vice versa?
Please explain, I may have missed some of this thread.
Thanks.
- Dave
9:27 p.m.
On Sep 30, 2014, at 11:51 PM, Roe Peterson <roeapeterson at gmail.com> wrote:
Both, really. In the interest of developing a "universal"
Qbus/Unibus paddle board. If anyone has more concrete reasons
why one might cook the other than the 12v pins on Qbus and the
+15/+8v line on Unibus (not necessarily connected on a Qbus
card), I'd love to hear.
There are advantages to making a single board that can talk
to both without catching fire, especially if you don't have
to move any jumpers to do so.
- Dave
>> On Sep
30, 2014, at 17:44 , David Riley <fraveydank at gmail.com> wrote:
>> Indeed, that's actually something I'm planning on figuring
>> out; my current quandary is mapping Unibus pins against
>> Qbus pins in order to figure out why one might cook the
>> other
This sounds like you are trying to figure out why a qbus board plugged into a unibus
backplane might blow?
Or vice versa?
9 p.m.
I also haven't done the QBUS C-D rows yet, do I'm adding them now. My spreadsheet
also includes Omnibus. Because, why not! :)
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
9:33 p.m.
Well, it turns out that I haven't added the C-D rows to the table because I don't
have any good documentation about them yet. Here's what I have so far:
https://dl.dropboxusercontent.com/u/29876211/DualWidthBusComparison-v1.pdf
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
3502
days inactive
3502
days old
7 comments
4 participants
participants (4)
-
fraveydank＠gmail.com -
jnc＠mercury.lcs.mit.edu -
nf6x＠nf6x.net -
roeapeterson＠gmail.com