• Please review our updated Terms and Rules here

Pdp 11/40

Hi KM11;
I think what is confusing you , when I say I had to fix a broken wire is -- that I (for Now) Do NOT have a unibus Jumper Card to connect my main processor backplane to my secondary backplane that has my memory board and my M7800 card.. as well as the terminator card.. One of the wires broke, that I have connecting the two together..What I have is in slot 9 of the processor backplane is a W9301 (wire-Wrap Board) Row A and B are wire-wrapped to an MDB card in slot 1 of the DD11-c, for unibus connection.. Row D of the W9301 has the Grant Continuity wire-wrapped in.. Slot 1 of the DD11-c in rows C-F have the M7800 in it.. Slot 2 has M7891 memory Board (half populated), slot 3 has a M7867 DUP-11 in it, (Just for something to be there), slot 4 has the M9312 Bootstrap/Terminator in it, and the other M7800 , which is addressed and vectored differently than the other one.. In the main Processor Backplane I have Nothing in slot 1, Eis in slot 2, U word in slot 3, data paths in slot 4, Ir decode in slot 5, status in slot 6, timing in slot 7 and memory management in slot 8.. When I first turn it on all lights are on, except for the top two (16 and17) in the address .. Run is on, processor is on and Bus is on.. I push down (HALT) all data lights go out, 026765 in address light gos on, run, processor, bus and console light come on.. user and virtual are off. Push Load Address (Switch) and no Data lights are on.. Address lights on are 063520 are on.. when I press Examine Data lights 022660 come on, press examine again and i get in address lights 001522 and the Data lights are 022660.. Does this Help ?? I know its long..
THANK YOU marty
 
Hi KM11;
Whenever I press the Load Address switch I always get a variation of the 1520 displayed in the Address lights.. When I say variation, IF I add or remove a card, the upper address lights change, but the lower ones stay the same.. I even tried it with NO DD11-c connected to it and the terminator in processor slot 9 and the M7800 there as well.. Still the same thing.. I would expect it to have 000000.. I can press examine and the address bits increment, I can Switch in any other bit in the Front Panel and press Load Address and it shows up correctly, But I cannot alter the 000520 with the switches..
THANK YOU Marty
 
Hi All;
Now that I am not as confused as I was earlier, Would it help to move all Of the Jumpers for the KT11 Memory Management Board, so that IT is not there, as well as for the EIS Board.. Take them Both out and see if that makes any difference on the Front Panel lights ?? I would make a list of the changes, for myself, even if no one else would need it..
THANK YOU Marty
 
Hi All;
I have taken out the Eis Board and the Memory Management Board.. It does the same thing.. I also found out that depositing to memory works, which it does, but the Data is 000520.. But at least I know that the Deposit works.. It looks like I need to put the Data Paths Board on an extender.. Which right now I don't want to do, as there is no card guides holding the Boards straight in.. And with the added weight of the Board on top of the extender and the pull of the cable from the Front Panel (Console).. It could be too much and cause it to Break off the Fingers.. Not a good thing.. My best guess would be the 3101's or the 74174's or the 8881's all on the Data path Board.. I have plenty of these, except for the 8881's.. Since I am not good any more at removing parts now, Years ago I was able and had the equipment to do so.. I would need to cut them out and then remove the cut legs, but I would rather not do that to a good part, without evidence that it is bad.. Please reply with your best advice, so I can continue finding what is wrong with this system.. I guess I could solder wires on the legs of some of the Ic's (74174 and 8881) and them without the extender, I could look at the signals with my scope..
THANK YOU Marty
 
Hi All;
The back end seems to be OK.. Did the wire thing.. So I will start from the front end. I am going to follow bit 04, since it is one of the problem bits, all the way thru from the Switch on thru and see If I can find anything that way.. I am done for the night, as I have found that when I am tired, that is when I make mistakes.. So I will restart on Monday, starting with the switch and checking each IC along its path.. I see from the print set, that it goes from a mux to Reg B to another mux and then thru the ALU to reg D and out.. Also need to check the 7404's and the BA Mux.. Lots to check, on page 3..
THANK YOU Marty
 
Marty:

From your description, I'm picturing this sort of "Frankenstein" PDP-11 assembled from different parts of different vintage that actually almost works! (No offense intended and I salute your ingenuity.) Seriously, it sounds like most of the basics may be functional and you probably stand a good chance of sorting the rest out. It’s sometimes surprising at how equipment of that era has stood the test of time.
 
Hi KM11;
What I have is all of the processor boards, along with the Memory Management and Eis options.. For now I am missing the Stack option M7237 and the Clock option, M787 Boards.. But, these and a plain Terminator Board, and a Unibus Jumper Board are at my friends place, and I hope to get them later this week.. The only Board that may be from something else is the M9312 and the Memory Board.. I agree, since all of the Boards from Both the PDP 11/45 and the PDP 11/40, were thrown all in the same boxes, that not everything for the 11/40 is correct, for a 11/40.. But, as far as I can tell, I have the correct Boards.. After thinking about it over the weekend, I am going to make a Partial Maintance Board, Switches only.. Which will allow me to Single Step the Clock and with a modification to the circuit, have a slow clock.. I hope that will allow me to find more easily the problem Ic's.. I can follow a good Bit from switch setting to turning on of the Led, (After doing a Load Address) and when I do it just one bit higher than my offending bit, It most likely will be all the same Ic's.. So, I can compare a good signal to a Bad signal, side by side.. And, hopefully trace where it is having a problem..
THANK YOU Marty
 
Hi All;
I found out something.. One my friend came by with some of the other boards, only need one.. Stack Limit Board.. Anyway, I found that if I put all the switches to '1' and them do a load address, and then do an Examine, it clears the switch register and it has all Zero's.. I am now going to set it up with memory and see IF I can Deposit anything into Address '0000000'.. And not just '000520'.. Also, I have (now) an M9202 and an M920 to connect my Unibus DD11-C to the main system backplane.. My question is which one should I use.. The M9202 Says Unibus Inverter while the M920 says Internal Bus Connector.. I would like to use the M92902, But, IF it inverts, I am not sure that ,, that is correct.. Or is that just a mis-label on their part, or a mis-translation on my part ???
THANK YOU Marty
THANK YOU Marty
 
Hi All;
I FOUND the PROBLEM !!!!! It ended up being a bad Cable, which would work some of the time, but not all of the time. Like when I would move things around.. I have been able to now Deposit and Examine memory.. Next I am going to get a small program I saw in the internet to make the led's go back and forth on and off.. That will at least tell me that it is able to execute programs or is not working..
THANK YOU Marty
 
Hi All;
It will run the program for a short time , before going back to consloe mode.. But, I can restart it and it will run again, never stopping at the same address.. The Led's do the right things.. So I can at least say it seems to execute a program.. I am NOT sure Why it stops in the middle of the program, though.. Next I need to see if I can get the M7800 Board to work..
THANK YOU Marty
 
Hi John;
Is the PC what the Leds are showing when it stops ?? If, so then it is stopping at the next location, after the end of the program.. How or what would I need to Check for false interrupts ??
THANK YOU Marty
 
The PC is the "program counter," register 7, which holds the location of the next instruction. I believe the LEDs show the last PC address on a halt, but I'm not familiar with front-panel operations.

You say it's pointing at the address after the end of the program, but didn't you say it was stopping at random points within it before that? Huh. Does your program loop, or does it run through and then halt?
 
Hi John;
Sorry about the confusion.. What I guess I am trying to say is that, it mostly stops at the same memory location.. But the Led that is turned on, is a different one each time, depending on how long the program has run.. The program does loop.. I know what a PC is, what I didn't know was IF that was, what was showing up on the Leds after a stop.. Here is a listing of the program..
1000 5000
1002 5200
1004 6100
1006 0005
1008 0775
start the program at 1000
THANK YOU Marty
 
Huh. Is that in hexadecimal, or octal? I'm having a hard time working out what that's supposed to do. Do you have a listing you were working from?
 
Hi John;
No, I do not have a listing, as it came off of the internet -- youtube PDP 11/40.. It is octal, except for the last address, which I said was 1008 and should have been 1010 octal..
THANK YOU Marty
 
Hi John;
I just took a quick look at Lawrance book on programming.. I am going to Print it out and start looking and reading it.. It looks like it does the following as far as I have figured out..
1000 5000 clear register 00
1002 5200 increment register 00
1004 6100 Rotate left register 00
1006 0005 (I haven't found that yet)
1010 0775 (I think) a jump back to 1002
THANK YOU Marty
 
Okay, so that's:
Code: 
1000 CLR R0
1002 INC R0
1004 ROL R0
1006 RESET
1010 BR 1004
As expected, it's supposed to loop indefinitely...I don't know enough about bus stuff to say why it would stop at the RESET instruction, but it must really be going off the rails if it gets past an unconditional branch to the HALT you placed after the program...
 
You must log in or register to reply here.
Back
Top