• Please review our updated Terms and Rules here

PDP 11/45, Part 2

Marty said:
Hi All;
I have checked out all four of the Core Stacks themselves and they all check out, Not only did I run them with Don's ZZ program, but I also boot from the TU58 emulator as well..

Don, your three programs that I have all Load and Run just fine..

RSX11M, of the two Files that I got from You, not so good.. The file xxdpd2d, Boots and Runs just fine.. The file xxdpddd Loads, but will not Run.. It stops at Address '000010.. Just like before.. I have not yet tried Reloading the program and seeing if it will run..

THANK YOU Marty
Click to expand...

Marty: What heck are you going to do with rest of your life now? :D
 
Hi All;
Agent Orange, I have just check the G235 Boards and two are Good, and two are Bad..
One Bad Board fails after 100 passes and the other fails after 10 passes..
I still need to see what I have in G114 Boards..
Then depending on what I find, I still need to get the Memory Management boards to Work, get a Real Tape Drive to work with the System and see if it will.. Get a Disk Drive to work with the System..
And last but not least get an Operating system to work on this machine..
That should keep me Busy at least this afternoon !!!
THANK YOU Marty
 
Hi All;
One of the G114 Boards is a (right from power up) No Deposit (No return) Board, I have checked all the fuses on the Board..

THANK YOU Marty
 
Marty said:
IT WORKED I HAVE SUCCESSFULLY BOOTED FROM THE TU58 EMULATOR !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Click to expand...

Congratulations Marty,if there was an award for perseverance then you should get it , NICE ONE.
Dave H
 
Hi All;
It looks like I have made it with the Core Boards, I think I have two complete sets of Boards.. The only thing that has yet to be checked it the other three Timing Boards.. Hopefully I can get at least one more good Board out of the ones that are left..
THANK YOU Marty
 
Hi All;
Only one Timing board is any Good, so 16K for now is my limit..
But, I have one thing I can try.. I can Pull the MF11-UP and put in two DD11's in the same space..
And see IF what I have in Memory Boards will work in the 11/45, I have both Core and IC Ram styles..
But, I think most of them are for the Later machines, so I don't know if they will work in this machine, timing issue..
They are all UniBus ..
THANK YOU Marty
 
Marty said:
Hi All;
Only one Timing board is any Good, so 16K for now is my limit..
But, I have one thing I can try.. I can Pull the MF11-UP and put in two DD11's in the same space..
And see IF what I have in Memory Boards will work in the 11/45, I have both Core and IC Ram styles..
But, I think most of them are for the Later machines, so I don't know if they will work in this machine, timing issue..
They are all UniBus ..
THANK YOU Marty
Click to expand...

UNIBUS memory is asynchronous, so timing does not matter. They will work.

HOWEVER, the bigger issue is slot compatibility. Semiconductor memory boards all require a MUD (modified unibus) slot. Not sure which core memories you have, whether they are multi-board or single, and what kind of dedicated slots they may need. So BE CAREFUL before you just start plugging things in and applying power!

If you have any UNIBUS expansion backplanes (like DD11-DK or -CK) they are MUD in the inner slots, so semiconductor memories would work there.

Don
 
Hi All;
Yes, Don, """ If you have any UNIBUS expansion backplanes (like DD11-DK or -CK) they are MUD in the inner slots, so semiconductor memories would work there. """
I know, I have DD11-Ck's, I learned alot about DD11's from the PDP 11/40 issues.. I learned to check pins against kind of slot it is, So I know that the Semiconductor memories will Work, in those slots..
SomeWhere someone had said in the Past in this forum that the Later Memories might not work in the Earlier Machines.. And the Rest of the Core Boards are Single Slot Boards, some though are Double thickness.. So they take up two slots, but plug into only one..
But, Thank You for Caring enough to mention this, that others might know..
Can You Imagine, before I have been using Core at roughly lets say 500ns, and this machine Runs like Greased Lightening..
What will it do with 150ns Semiconductor Ram Ic's..
THANK YOU Marty
 
Last edited:
Marty said:
Can You Imagine, before I have been using Core at roughly lets say 500ns, and this machine Runs like Greased Lightening..
What will it do with 150ns Semiconductor Ram Ic's..
THANK YOU Marty
Click to expand...

Yes, I can, and it won't be anywhere near as noticeable as you think. Depending on what generation MOS memory boards you have the access time can be anywhere from 150ns to 250ns, cycle time is about twice that. UNIBUS adds significant overhead timing because it is asynchronous and has fairly large embedded delays (50ns to 100ns) that are required to deskew the data and control signal timing.

In a double blind test I doubt you could tell the difference.

NOW if you want a really fast screaming 11/45 find the bipolar memory boards that were designed for it that turned it into an 11/55. THAT would be a screaming machine (for its time anyway). Basically the whole memory was cache, and this system went into many of the early flight simulator systems.

See: http://www.columbia.edu/cu/computinghistory/pdp11.html and http://www.pdp-11.nl/pdp11-55/pdp11-55startpage.html

Don
 
Last edited:
Hi All;
Don, I would Love to find a Board or two for that.. I have been looking and doing some tracing out of the connections for what all is used in that area.. I have not found any schematics for these Boards.. The M8110 and M8111 and G401.. Just the Maintenance Manual.. Would there be a military version of these manuals..
THANK YOU Marty
 
If you like that idea... Why not make your own fast memory using "modern" S-RAM parts? Heck - you've wire wrapped just about everything else - should be a breeze for you. You could fill the machine.
 
Hi All;
That is Exactly what I have for Quite some Time been thinking..
Which Is Why I have been looking for Schematics for these Boards..
I have planned on using 6116's or 6264's Cmos SRams.. I know How to most of the Circuit would be implimented.. At First it would be 6116's with jumpers to change the socket to accept 6264's at a later Time.. I have Done this on other Boards that I have made..
I just need to find out how the processor into and out of timing work and the protocal for "FastBus"..
Which Is why I have been putting Down on Paper, what connects to whom, in the affected slots..
THANK YOU Marty
 
Last edited:
Hi All;
I THINK I Found what was Wrong on the PDP 11/40..
The Backplane an DD11-C, the wires that were Labeled +12 and -12 got switched.. The Labels were on the Wrong wires..
I didn't Blow up the PDP 11/45, But it smoked a Cap on the Monolithic Memory Board, 303-158, MSC 3501/3603..
I at this present time Don't know If there was Damage to the memory Board, other than the Cap..
I rechecked the corrected connections with a Meter, and Both were correct this time..
I plugged in am M 7891 and Heard a snap!! I then turned off everything and unplugged the M9202.. Something Must be wired incorrectly on this particular DD11-C Backplane.. Since, I know what line Blew on the 11/40 I can trace that back on the PDP 11/45, and look for Damage there, and then OHM out that line on the DD11-C and find out possibly where the incorrect short or wrong wire is..
The PDP 11/45 seems to be OK, But I am taking no chances..
THANK YOU Marty
 
Last edited:
I was about to post some thoughts about the M8110 MS11 memory system after doing a dance through the Tech Manual...

I'm kind of at a loss for words at the moment.

Snap Crackle Pop is not my favorite adventure.
 
Hi All;
RSX11M, That is the same Manual that I have already downloaded.. Go ahead and tell me what You had in Mind...
So far, Here is what I have found, Some one (I would Guess) Has most likely rewired the DD11-C for something other than its original wiring.. So far I have found the +12 line is Shorted to at least the following lines, and there could be others, this is what shows up at first glance, with the meter..
the (AU1) NPG H, (BU1) SSYN L, and on the SPC slot (FD2) F01 N1.. I Don't know How many other things are NOT correct, on this DD11-C Backplane..

Well, Now I Really am at a Loss, I took Out this Backplane, and its HardWired, it a DD11-CK.. Its Not Hand Wired, as I had thought before..

Well, the Only thing I can do for Now, is Put the MF11-UP Backplane back in and see IF or What got Damaged or Doesn't work on the PDP 11/45..
THANK YOU Marty
 
Last edited:
Hi All;
I found my now Double Mistake !!!!!
The two 12 volt lines were originally correct, (I think)..
What I had Done, not having the Best of lighting, right behind the Front Panel, was Put the BackPlane in Backwards, "A" is "F" and "F" is "A" and so forth.. I still need to see what all the Damage would be having done this.. HopeFully it was to the Memory Board and not to the CPU..
The SSYN, as well as NPG H, is most likely what got hit.. So I need to find which Board it is on and see if it is Bad or OK !!!!
And they Both are Serviced by the same IC, E55, 8881..
And Now I know its Bad, I looked at the Board in the Light and pin 4 which is one of the ones that got Hit, has Leakage out from the side of it.. So Ic 55 is definitely Bad..
I will need to scope back a stage or two and look for more Damaged Ic's..
THANK YOU Marty
 
Last edited:
Hi All;
Its definately "Down", so I will first need to Trace out all of the Places that used Ic 55, and go backward from there..
And put Back in the MF11-UP BackPlane and get it all hooked back up That way I know at least that part is OK, since it wasn't in the machine when this happened.. Why Can't I leave well enough along.. 16K would have been OK for awhile..
THANK YOU Marty
 
Back when I worked for DEC in the late 70s we had a coop student for a while. We had a small 11/05 single rack system for lab use that had RK05s for disk storage.

The system started behaving intermittently, so we got in the habit of pulling the RK11 controller cards and re-seating them. Problem would go away for a few weeks or so, but inevitably would come back.

One day we told the coop to pull the boards, run an eraser up and down the fingers to clean them, and then reassemble and check it out.

So he pulled ALL the RK11 controller boards at once, cleaned them off very diligently, and then reassembled, putting all the boards 180deg reversed in the dedicated RK11 backplane slots (four quad boards IIRC).

Turns out, if you push hard enough, you can fully seat a UNIBUS connector style board in the DEC backplane reversed 180 degrees ...

Then he powered up the system. BLAM and sparks!

He kind of never lived that experience down. Of course we wouldn't let him either.

So then we just put everything back the way it should have been, and called in-house DEC field service saying our system was broken.

They said wow we never have seen this happen before. And then they replaced all the RK11 controller cards.

System working fine again.

Don
 
Last edited:
Hi All;
Don, Great Story.. I still looking for my DEC Field Service Person to show up..
No wonder You know so much about DEC systems, what did You do at DEC ??
Hopefully, since a lot of the machine works, I can find the problem(s) and fix the Trouble.. Who knows maybe I'LL find something else that was never fixed.. Thanks for the Encouragement..
Some years ago when I was working on a system that had Tape Drives and we had to test each board with a Tape Drive, beside the Disk controller part, and the formatter Board would "act up" I had to do the Erase thing to make it work..
THANK YOU Marty
 
You must log in or register to reply here.
Back
Top