• Please review our updated Terms and Rules here

Digital PDP 11/05 in BK11-K mounting box

billdeg

billdeg

Technician
Joined
Nov 18, 2003
Messages
3,885
Location
Landenberg, PA USA
I was sorting out some parts and found that I had enough to make a simple PDP 11/05 in a BK11-K mounting box with an H765 power supply. I'll at least assemble it and run some power tests to see what kind of work will be needed to build a basic system. This system would have a more compact footprint than the other 11/05 S (full height style) I have. Unfortunately the backplane has only 9 slots.

Question - There is a weird connector coming from J3 (middle) of the Power Distribution Board that is not mentioned in the documentation, there is nothing to connect it to on the H765. I think it was installed for testing purposes, you can see coming off the custom adapter there is a small two-wire connector that I believe connects to the chassis' front panel connector. Agree? I am going with this assumption.

thm_1105S_power-dstribution-board.JPG


http://vintagecomputer.net/browse_thread.cfm?id=622
 
Last edited:
UPDATE - See page 66/84 of the DEC-11-HBKEF-A-D BA11-K Mounting Box Manual, I needed to position the power supply hinged at an angle to see better where/if to attach. Done/thanks.
 
Last edited:
One other thing...I discovered that my other PDP 11/05 is a "NC" model, full height but not the 11/05 S model. I did not know that there was an NC also. The power supply is oriented differently, part of the reason I was confused at first, only having had experience with the NC. So now I know.
 
You have cards to populate that system? You’re not doing a serious Old School five or ten unless you got core!
 
Yes I do! I have not tested the core but I have tested the processor in my other 11/05. So I have confidence I can make some progress, once I fix the power issues. The issue at this point is that I have 4 H744's that are all faulty for one reason or another, I am going to see if I can cobble together/repair 2 working units. I don't want to cannibalize from other places. My spare regulators are in pretty sad shape though.

I have not given up on booting my 11/40 using RL02 drives, but I thought I'd try working on a different system (this 11/05 S) and see if the RL02s work on it instead.

I also plan to dumping the RAM from my other 11/05 (it's a "NC" model) with 8K core. I will transplant via serial port to the 11/40 to see if I can get BASIC running on the 11/40 that way. I have BASIC working in core on the source machine.
 
Last edited:
Progress...

DECPDP1105S_H765-open1.jpg

Here is the h765 wired but not attached to the 11/05 S chassis and backplane.


DECPDP1105S_front-panel-lights.jpg


I set up a basic system and so far I can at least load RAM addresses and deposit. Some bits can get stuck. In this picture the 8th bit is stuck. I can "shake a stuck bit free" by playing with the system, overwriting memory, but I need to get to the point where this is not necessary. More testing is needed but now I have a nice little test system with a good power supply that I can use to test PDP UNIBUS cards.
 
Last edited:
Hi All;

Bill, more Pictures.. Please, but, I like Your progress..
I think that after, either, I need a break from the PDP 8i clone I am building or cannot get it going..
I would like to build one of these from TTL, and a single Board or a few Boards, like I am trying to do with the PDP 8i, only this would be an 11 machine..

THANK YOU Marty
 
Here is my site thread
http://vintagecomputer.net/browse_thread.cfm?id=622

Here is a direct link to photos and manual for the 11/05 S with BA 11-K
http://vintagecomputer.net/digital/PDP11-05_BA11-K/

I need to read up on how to attach the front panel to the chassis, and to see if I am missing some parts. The +20 is a little low, I need to adjust the regulator. I need to stabilize the RAM, test the CPU, perhaps attach a teletype and load some programs. There is still more I need to do for sure.

b
 
Last edited:
DC LO AC LO look good. (adjusted the +20 up a hair on the h754)

I transplanted the CPU cards into another system to test and they work also. Aside from the console itself, it must be a RAM problem. Maybe a front panel problem I can test that asap.

The system has 8K Core
G235 (xy drive module)
H217D (memory stack)
G114 (sense / inhibit)
M8293 (memory control module)

I posted a question to MARCH and CCTALK about whether I can swap H217D with an H217C in this box and configuration.

UPDATE - found the MF11-U/UP core memory maintenance manual, this explains how to deal with RAM configuration and the backplane.

I tested the G114, seems to be OK.
 
Last edited:
Mission accomplished. I got all three downstairs, the full height 11/40 and industrial 11, plus the TU10 9 track tape unit. I completely tore them down over the period of two months. One piece at a time. For the final move to take the empty enclosure boxes/racks out I got help from my brother and a nearby friend. Now that I have done this a few times it's not so bad to do. The actual cabinets when completely bare are easy to move.

It will be hard to hoist the TU10 drive back into the cabinet, but I think I will take it apart until it's light enough to lift up 5 feet. As it is now I'd need 4 people to do it each taking a corner.

I have started testing the voltage regulators and the h742a and h7420a power supplies used by the 11/40's (main unit and expansion chassis). I have already been through the process of restoring an h742a so I should be able to get hopefully at least two of the four going quickly.

Returning to the main thread topic, I will have to add the RAM parity jumper to the backplane if I want to use for 11/40's H217C core RAM on the BK11-K because it's wired to be a pdp 11/05 S, which uses non-parity RAM (h217D). As I stated earlier it would be nice to be able to test RAM using the BK11-K because I know everything else more or less works. The only no parity RAM I have available is not inhibiting correctly address line 2 (or whatever you call light 2). When I deposit into line 2 (i.e. 000 100) nothing shows when I examine. I can load address 000 100 (the light works) and I have tested the front panel (lights work). There are four cards used by non parity RAM. OF these, the G235 I think does the write inhibit to the h217D's core stack. Hopefully the core itself is OK and the issue is with the write being done by the G235 (fix or replace it). It may be that the -15 is too low (?). The -15v is reading as -17.16 off the backplane. The docs say I need to be within +/-5% (no more than -16.5v). I understand that the h745 is +15 and -15v but I don't know which voltage is effected by the one adjustment pot on the h754 regulator. If it's not for the -15v line then I have to disassemble everything and make adjustments to the h765 itself directly. Ug.

I read the entire 122 page core RAM manual and read about how the -15V line is used. I would think all lines would be bad and not just one if the -15V was too low and not just one line. I am getting good DC LO / AC LO and the CPU is running properly when run the CPU test. I will send an update when I have more to report.

Bill
 
Last edited:
This discussion got me interested in a blinky-light PDP-11.
I looked through my pictures of the RICM collection and can't find any pictures of the 11/05 or the 11/10.
I will take some pictures and post them on the RICM WWW page.

I started work on the DW8E Omnibus expansion chassis that is in the RICM's PDP-12.
The DW8E, PDP-8/M, PDP-8/F, and the 11/05 all share the same power supply, but in a slightly different assembly.
There is a lot more information on the power supply in the 11/05-11/10 manual than there is in the PDP-8 manuals.
Since there is no processor in the DW8E chassis the AC LO and DC LO signals are no used.
I reformed the capacitors, and powered it on, and the power supply worked fine for a few minutes, then there was a flash, and the +5V is at about 1V.
I need to check the pico fuses on the end of the regulator card.
 
With all of the fantastic stuff to play with at the RICM we can forgive you if you didn't prioritize any 11/05 or 11/10 pictures, given you have a lot else going on :)
 
So many toys, and so little time. I need to win the PowerBall so I can retire now and spend my days working on these systems.

Once the PDP-12 is fully functional we will pick a new project. We have two 11/45s, one with core and one with semiconductor memory. Both show signs of life, but are nonfunctional. We have 11/35, 11/05, and 11/10. All are empty except for the processor boards. We have a TC11/TU56 that could go with any of these 11s. We have an empty OEM PDP-11. It must be really early because it doesn't say 11/20 on the tag. It would be nice to get backplanes and processor boards for the 11/20. We have an 11/70 with RM03 disks and 768kB of core. We have a programmer's front panel for it, to make it interesting, but this would likely be a challenging project. I am leaning toward the 11/35 in a short cabinet with the TC11/TU56 as the next project.
 
Hi All;

M-Thompson, "" We have two 11/45s, one with core and one with semiconductor memory. Both show signs of life, but are nonfunctional. We have 11/35, 11/05, and 11/10. I am leaning toward the 11/35 in a short cabinet with the TC11/TU56 as the next project. ""
I would vote for one of the PDP 11/45's, But, I will not complain at all if You go with the PDP 11/35.. And I am prejudiced, though..

THANK YOU Marty
 
Found and fixed the RAM problem. The G114 was bad. I thought I had ruled this out but apparently I was mistaken. The card is fixable (probably a bad cap or transistor that can be targeted using a probe), but for now I will use a G114 from spare parts. NEXT: Set up teletype to load programs, perhaps add a serial card and so on. The goal however, to set up a test system for card testing is done / complete. Even though this backplane is wired for non-parity RAM, you can still test all but the h217C and M7259 (or other parity RAM) cards. The remaining companion cards are the same.
 
Last edited:
Hi All;

Bill, "" Found and fixed the RAM problem. The G114 was bad. I thought I had ruled this out but apparently I was mistaken. The card is fixable (probably a bad cap or transistor that can be targeted using a probe), but for now I will use a G114 from spare parts. NEXT: Set up teletype to load programs, perhaps add a serial card and so on. The goal however, to set up a test system for card testing is done complete. ""
Congratulations on Both Counts.. "" The G114 was bad. ""
Please, not only keep us posted on this, but, I would Like to see and know of the progress You have fixing this Board.. I have some and I would like to know How to fix them.. That is once I get back to my PDP 11/45 and get it running again..
"" The goal however, to set up a test system for card testing is done complete. "" Great, makes me sort of wish, I still had mine and I maybe could have sent You the Boards for You to test, and then you could have let me know the results.. Oh, well, they went to a Good Home..

THANK YOU Marty
 
Hi All;

M-Thompson, "" Got a schematic for the G114 board? ""
Yes, I think I did, otherwise, I would need to Reload one, as well as the rest of the Memory modules..
Before, when it showed up Bad, I just Tested all of each Kind that I had, and Marked them as Good or as Bad.. I didn't and I don't know enough to figure out what was/is wrong with what I have.. If, I remember correctly, I have about 5 each of the different boards, with about three of each Bad..
But, before, I can do anything, I would need to get the PDP 11/45 running again and get the necessary Power Supplies that I would be need for Core Memory.. Namely, --5 and +20 Volts Switchers..

THANK YOU Marty
 
You must log in or register to reply here.
Back
Top