> -----Original Message-----
> From: cctalk-bounces at classiccmp.org
> [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Jay West
> Sent: Tuesday, March 28, 2006 9:33 AM
> To: General Discussion: On-Topic and Off-Topic Posts
> Subject: Re: 11/45 RTC
>
> Julian wrote...
> > Have you tried this with M7800 and M787 instead of M7228?
> I'd say if
> > not, try that, if yes, and if that works, then look over the DL11-W
> > manual for some sort of compatibility issue with the 11/45.
> This is
> > just speculation, but it's possible that you may have to use a M787
> > and disable the RTC on the DL11-W or something strange like that.
>
> I don't have a M787 card. This system arrived with two
> M7800's and a M7228.
> No KW11-L. It came with the M7228 (KW11-P). From what I can
> tell, the M7228 can provide several different time clocks,
> one of which is LTC based as an option if you run an
> additional wire, but not at the normal address for the LTC.
>
> I want to see first if the power supply is generating a LTC L
> signal, and then see where it goes. Right now I'm trying to
> just find the signal. It would obviously go to slot 1C
> somewhere for the M787. But I can't seem to figure out if it
> was by default also provided to a pin in the 3 unibus slots
> built in to the system unit in order for the DL11-W to take
> advantage of.
How about this: Dig in the KW11-P manual if its available and see if there
is any info on replacing a KW11-L with it. Maybe that will tell you where
that LTC L signal is.
Given that the system seems to pass a lot of diags, but fails the LTC test,
I'm focusing there. From looking in the programming manual for the DL11-W,
it appears that bit 7 never being set means that there is no LTC L signal
coming in to the DL11-W. It would appear that software clears this bit and
then waits for it to set (by the square wave from the AC at 50 or 60 hz or
so). According to the diag, bit 7 never sets, so I'm thinking I'm not
getting any LTC signal from the power supply. I need to find that square
wave signal and see if it's working and directed to the right place.
My problem is finding that signal. My 11/45 is a KB11-A, serial number <
2000. The maintenance manual shows the LTC coming from the power supply to
the power distribution board (top connector towards the cabinet door) shows
six wires, the bottom one (brown) going to slot 1 row C where the M787
option plugs in. My top connector does not have that many wires, only 3 or
so. My system never had (I don't believe) a M787 card. The system came with
two DL11-A (M7800), and a M7228 (KW11-P). But I don't think there was ever a
LTC in slot 1. So is it possible that LTC isn't even routed out of the power
supply? I'm wanting to use a single DL11-W with the onboard LTC. But doesn't
that need to get the LTC signal from the backplane? If it wasn't ever routed
to the system unit....
So I'm not sure where to look. I'll pour over the manuals some more.
Jay West
>
> Julian Wolfe wrote:
> >> A standard BA11-K box has 2xH744 (or later 2xH7441), 1xH745,
> >> 1x54-11086, and usually a spot for 1xH754 (for core
> memory, not req'd
> >> if no core).
> >
> > Oh CRAP. I built this machine from parts (albeit all
> marked 11/34),
> > and did NOT know that I needed an H754 for core. Would
> this couse the
> > deposit errors I'm experiencing?
> >
> >
> >
>
> Probably; your core memory is unpowered (H754 supplies +20V &
> -5V to core memory modules).
>
> AFAIK the ONLY use for the +20V/-5V H754 brick is for core memory.
>
> The MOS memory boards for the 11/34 box (MS11-L) use the
> +15V/-15V from the backplane and have onboard regulators to
> generate +12V/-5V for the DRAM devices.
>
Actually, what I meant was EXAM errors. I can DEP but not EXAM.
However, this is sort of amusing, because what I'm basically doing is
entering an address, and depositing a value into /dev/null.
This could very well mean there's nothing wrong with any of my equipment,
and once my MOS memory arrives, the problem might clear itself up.
Thanks again, Don.
> A standard BA11-K box has 2xH744 (or later 2xH7441), 1xH745,
> 1x54-11086, and usually a spot for 1xH754 (for core memory,
> not req'd if no core).
Oh CRAP. I built this machine from parts (albeit all marked 11/34), and did
NOT know that I needed an H754 for core. Would this couse the deposit
errors I'm experiencing?
Yes, the box is basically a 6502 microprocessor and support chips, and has no
mechanical connections to the Selectric.
Al, do you want scans of the manuals for Bitsavers?
> From: Richard <legalize at xmission.com>
>
> In article <4428D6CA.EFA31B9E at rain.org>,
> Marvin Johnston <marvin at rain.org> writes:
>
> > Since I've had a number of requests for the picture, it is online at:
> >
> > http://www.rain.org/~marvin/escon.jpg
>
> Based on the picture, it looks like it connects electrically to the
> Selectric and not mechanically. I'm assuming the entire enclosure
> looks like a box and doesn't have "typing finger" solenoids on the
> bottom?
> --
> -----Original Message-----
> From: cctalk-bounces at classiccmp.org
> [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Jay West
> Sent: Tuesday, March 28, 2006 11:17 AM
> To: General Discussion: On-Topic and Off-Topic Posts
> Subject: Re: 11/45 RTC
>
> Julian wrote...
> > How about this: Dig in the KW11-P manual if its available
> and see if
> > there is any info on replacing a KW11-L with it. Maybe
> that will tell
> > you where that LTC L signal is.
>
> Yes, but I'm missing something somewhere. I see the spot
> where the power supply LTC L signal hits the cpu backplane.
> It comes from the brown wire, #7 on P2 at the top front.
> There's a trace that goes down from this to slot 1, row C,
> pin R1. That's just peachy, and gives me a spot to test in
> order to make sure the power supply is providing a clock signal.
>
> But of course I want to make sure that signal is also present
> on the three unibus slots built into the cpu backplane, so
> that my DL11-W can use it. So I was looking for THAT pin.
> From eyeballing it and testing, it appears that slot 1 row C
> pin R1 (the LTC "source") hits all the unibus slots on the
> cpu backplane at row C pin D1. Ok, I'm fine there.
>
> But here's where the KW11-P manual gets confusing. I'm only
> bringing this up as a sanity check - It says that in order
> for the KW11-P to get the LTC (if you even want to use that
> option, not required), you have to install a jumper wire to
> the slot that the KW11-P is in to provide that signal. It
> says the KW11-P expects to get it on CE1, and specifically
> says to run a wire from CE1 to slot 1 row C pin R1 (on the
> 11/45). But why would they say that if the LTC signal is
> available on pin CD1 right there in the same slot already
> next to CE1?? Or better yet, why didn't they design the
> KW11-P to get LTC from CD1 instead of CE1? This is the part
> that makes me think I'm messing up somewhere.
Do you maybe have a different/special backplane alltogether? Is the
backplane designator sticker still on it?
>
> I guess I could find prints for the DL11-W and see if in fact
> it's looking to CD1 for LTC.
That sounds like a good idea too.
>
> Jay West
>
Since I've had a number of requests for the picture, it is online at:
http://www.rain.org/~marvin/escon.jpg
Included, but not shown, are two manuals. The first has the installation and
instruction manuals including schematics, while the second contains the
illustrations and diagrams for use with the Vol I instruction manual. There are
also a number of what look to be Selectric parts.
>From the information in the installtion manual:
"The ESCON equipment covered by these instructions fits all IBM office type
Selectrics. The length of the platen or if it is single or dual pitch, has no
affect on installation of this system. If your machine has a manual velocity
control, contact the factory for special instructions."
Another interesting quote from the instruction manual:
"the magnets used in the ESCON conversion kits have 55-ohm coils, except for the
shift magnet which has a 130-ohm coil."
There were no magnets in this package. I suspect if there was a Selectric
already prepared with the magnets, it got dumpstered before I got a chance to
check this stuff out.
> >I've got an Escon Microcomputer/Selectric Interface System that looks
> >like it was sold as a kit. It appears to be a unit to convert a
> >Selectric I or II typewriter so it can be used as a computer printer. Is
> >anyone familiar with this unit, and any ideas about what it is worth? It
> >belongs to the widow of my friend who died last month, and I am trying
> >to help separate the good stuff from everything else.
>
> Marv, does this thing work with a Selectric typewriter or a Selectric I/O
> writer? There's a big difference--the typewriter has none of the
> electrical components (solenoids, etc.) that the I/O writer uses to
> activate the mechanism. The golf-ball print mechanism is the same in both.
>
> Since Selectrics are still pretty easy to come by as typewriters, I suspect
> that this may have more interest if it were intended for typewriters (i.e.
> it included the actuators). On the other hand, the I/O writers, while once
> more commonplace, are fairly rare today and the unit would have
> substantially less appeal.
>
> Cheers,
> Chuck
Hello michael,
I am in NZ. and looking at buying an old ATC 510 which is for sale. I am a retired airline pilot and am helping a young guy next door to get his licences.
Is the 510 useful for basic instrument training, or would I be better off using a microsoft sim? What accessories have you got and where are you?
Many thanks
Dave McDonald
On Mar 27 2006, 19:00, Jay West wrote:
> I'm just shocked diags seem to receive such short shrift at DEC. I
guess it
> makes me feel quite fortunate that for HP 2100 & 21MX we have a
complete set
> of full source listings that exactly match diag documentation that
exactly
> match the code version on the media. I've refered to that code many
times.
> Hopefully someday a late full library of DEC diags will be unearthed
with
> matching source & docs.
Actually DEC took a lot of care over diagnostics. All systems were
sold with comprehensive relevant diagnostic media, normally accompanied
by the listings. In the early days the listings were on paper, but
later on microfiche. Furthermore, DEC maintained those diagnostics
carefully, and there were regular revisions and patches when necessary
which were sent out to customers who had maintainance contracts. I
don't know about the paper listings, but microfiche was certainly sent
out (inserting the updates used to be one of my jobs). In fact,
there's the problem -- when you do find listings, which have often been
separated from the machine and media they originally were meant to be
with, they may be for a different revision or patch level. The code
will be close, and the listings may be useful, but may not be quite the
same. Paper listings in particular will usually be for older versions
of the diagnostics, and many diags came along late enough that the
listings were only on microfiche.
A full set of PDP-11 microfiche diagnostics listings, such as might
have been held by a Field Service office, reseller, or a third party
maintainer, is one whole microfiche box, some 14" deep. There would
normally be another box beside that with the hardware documentation:
manuals, engineering drawings, parts lists, illustrated parts
breakdown, field and engineering change orders, etc.
--
Pete Peter Turnbull
Network Manager
University of York
I don't know that DEC ever distributed soft
copies of diagnostic listings on disk or tape (now finding that would be
the holy grail at this point).
-
I've never seen them either. If there is ever a HP vault trip, that
would
be something to look for.
Paper copies of the diags stopped when the maint fiche doc sets came out
This is probably what the diag handbook was based on.
If there are specific diags needed, I'll scan them from fiche. It is a
fairly long process to get clean copies, since the scanner is a
fiddle-fest
to set up at high magnifications.
