• Please review our updated Terms and Rules here

PDP 11/45, Part 3

What’s your Cipher Tape Drive look like? 32k of memory won’t be that bad, what are you planning on running as an operating system? I have three or four Unibus memory cards lying around with no documentation and it’s a real drag trying to set addresses but thru trial and error was able to get a couple going. Think my 11/34 is just 28 or 32K and I run RT-11, DEC Basic and all sorts of other stuff with no issues.
People tend to get all worked up about how much memory but back in the days of those systems programs were smaller nothing like this junk today. My stupid Win7/64 workstations need 8 Gb of ram just to get started.
 
Hi All;
QBus, Thank You for Your inquiry..
The Reason I was hoping for more Memory, was I would have like to get Unix running on my 11/45.. Otherwise, Yes 28K is enough..
Here is an Example of an AutoLoader, it probably is NOT Cipher, most likely HP..
https://www.youtube.com/watch?v=gsiDubVPiEw&list=UUCSr-zB35tShRtwiJBCD1Bw
My Tape Drive is an AutoLoad Cipher M990..

001.jpg

My Spectra Tape Emulator isn't working, its missing a Pal.. (no friends).

001.jpg

THANK YOU Marty
 
Last edited:
Hi All;
With the Assistance of Bruce Ray, I now have the missing pages from a manual that I thought I had Lost.. THANK YOU Bruce !!
So, I have in PALASM file format what the contents of the missing Pal would need to be Programmed..
I have a NeedHam EMP-11 Programmer, that will program 16L8's, So How do I convert from the OLD Palasm type file to what the EMP will take ??
OR can the NeedHam accept the old PAlasm type of file ??

003.jpg 002.jpg 001.jpg

THANK YOU Marty
 
Last edited:
Hi All;
I put in another Power Supply for the --5 volts and for the +20 volts, and I got a Core Board to at least show up..
I can initially, Deposit and Examine it, But, it will not let me do a Tape Load, where as I can with the Semiconductor Memory..
It most likely is one or more of the Voltages is too Low for it to work properly all of the time.. I will look into it later..

THANK YOU Marty
 
Hi All;
An update, I found that the old M7800 Board works fine, but, sometime since I was last able to Load and Run the TU58en, that the M7856 Does Not work..
I have been able to do a Console (I forgot, if its called emulation or simulation) and was able to Run from the proms in the M9312, such as Don's ZZ program..
But when I set up the M7856 for TU58 operation, it wasn't acting correctly.. So, I took out the M7800, and left in the M7856, after resetting the Switches for console operation.. And I get an Address Error Led to come on..
So, that needs to be fixed before going any farther, with TU58 operation..

THANK YOU Marty
 
Hi All;
I got it fixed, a broken wire and a wrong switch position set when trying to use it in Console mode.. And I found it wasn't seeing my USB to Serial unit.. So, I switched to a Straight cable..
I put it all back as it was and I at least get some Output..
There is still some problem, but I think I need to find a Correct file, and set things up and make sure I need or I do not need to keep the cable straight thru or crossed.. I will put it on my Breakout-box and see.. IT's CORRECT ..
But, things are progressing..

THANK YOU Marty
 
Hi All;
I got it working.. I NOW have TU58 !!!!
Another wire broke, fixed it, and also found that things take a little while longer than I remembered to load..

THANK YOU Marty
 
Hi All;
I got it working.. I NOW have TU58 !!!!
Another wire broke, fixed it, and also found that things take a little while longer than I remembered to load..

THANK YOU Marty

Good work! Glad to see you got it up and running. IIRC at 9600b it takes about 2 to 3 min for XXDP to boot to the command line.

My 11/44 at 38.4Kb takes about 30-40sec, and my 11/34 with 115.2Kb modified DL11-W takes only about 15 to 20 seconds.

Don
 
Hi All;
Don, Thank You for Your Encouraging note..
Once I waited, At first I may have not waited enough, everything came out OK..
On a side note, I think I may have a Bad Load from Your site, the 11xxdp.dsk didn't work, but the 11xx_9.dsk did work, only I wasn't sure what to choose from, I will look at Your site for more information.. But, it did come up..
What was the Modification did You do to Your DL11-W to increase its Baud Rate, I don't want to go tooo high, but either 19.2Kb or 38.4Kb would be fine for me.. I assume You replaced the Resonant Oscillator, am I correct, or is there more to it than that..

THANK YOU Marty
 
What was the Modification did You do to Your DL11-W to increase its Baud Rate, I don't want to go tooo high, but either 19.2Kb or 38.4Kb would be fine for me.. I assume You replaced the Resonant Oscillator, am I correct, or is there more to it than that..

I have found that the DL11-W maximum baud rate can be pushed to 19.2Kb (from 9.6Kb) with just an oscillator change. The standard UART can still run at this higher rate. When I tried to push the standard UART to 38.4Kb I found it would not work reliably.

I replaced the DL11-W standard 5.0688MHz oscillator with 20.2752MHz (4X) to multiply all baud rate settings by four. I used a custom one-time programmable oscillator available from DIGI-KEY (they will program the parts to custom frequencies for you).

The modification to push the DL11-W to 115.2Kb is a bit more complex. It requires an oscillator change, a UART part change (faster part), and a wiring change in the baud-rate generation divider logic to bypass some divider stages.

Don
 
Hi All;
THANK YOU, DON, I think I will just go with 19.2Kb, and use if I can find one or have Digikey program some for me for 10.1376..

THANK YOU Marty
 
The modification to push the DL11-W to 115.2Kb is a bit more complex. It requires an oscillator change, a UART part change (faster part), and a wiring change in the baud-rate generation divider logic to bypass some divider stages.Don

Don: Can you share a pointer to where the modification details are published? THX!
 
Hi All;
I now I need to look into my various Power Supplies, and see How to fix the +5 and -5 side of it/them, So that I can Run using Core, that way I can, put something into memory and turn off the Machine and it still would be there.. That way I can load some programs a little at a time into the machine..
One of the Supplies has a good +5 and a bad -5 and the +20 is good as well.. The other one has a bad +5 and a good -5 and a good +20.. The in the machine at present has the bad +5 and the good -5 and the good +20 volt sections..

THANK YOU Marty
 
ak6dn@mindspring.com said:
The modification to push the DL11-W to 115.2Kb is a bit more complex. It requires an oscillator change, a UART part change (faster part), and a wiring change in the baud-rate generation divider logic to bypass some divider stages.Don
Don: Can you share a pointer to where the modification details are published? THX!

Here are the changes I made to the M7856 DL11-W async serial board. The changes are easily reversible if required.

  1. Remove E29 (7486), insert a 14P DIP socket, add a wire jumper from pin 3 to pin 12 in the socket.
    This then connects the oscillator output directly to E27 (7493) pin CLK0, the first stage in the baud rate divider.
    To undo the change remove the jumper wire and put a standard 7486 part in the socket.
    .
  2. Remove OSC1 (5.0688MHz), insert a 4P in 14P DIP XTAL OSC socket.
    Put a 3.6864MHz 5V oscillator in the socket (or 7.3728MHz can work also; this is what I used as I had some in my parts bin).
    Oscillator frequency is 2 x 16 x 115200 = 3.6864MHz to have 115.2Kb be on the 9600b DIP switch selection.
    I used 7.3728MHz, which means 9600b setting runs at 230.4Kb (which does not work), and 4800b setting runs at 115.2Kb (which does).
    To undo the change put a 5.0688MHz oscillator in the socket.
    .
  3. Remove E7 (UART), insert a 40P DIP socket. Put an HD3-6402R or CDP6402C 200Kb capable UART part in the socket.
    You can get them brand new here: http://www.mouser.com/ProductDetail...=sGAEpiMZZMtOXy69nW9rM/iHuz5dAptukISN8zpcOQE= and there are other less expensive sources as well.
    These newer UARTs do not have internal pullups on all digital pins, which the DL11-W design requires.
    Add 10K pullups on pins 34 thru 39 so they will pull high and not float. I used an 8p 10K SIP resistor pack.
    Connect the common lead of the RPAK to a nearby 5V power node.

Don
 
Last edited:
Hi All;
Don, based on what You have and are doing here, I am going to try and keep the existing Resonant Oscillator, and change it so that I can get 19.2Kb.. I will let everyone know what I find and whether I succeed.. I am going to try and play with the outputs of E12, hopefully I can tap at pin 9 instead of pin 8, but, I need to find whether I also need to move from E29 pin 13 from 8 to 9 or just leave it alone..

THANK YOU Marty
 
Hi All;
I have Core memory working.. Now whether all of it works that is another thing, I have not Run any Diagnostics, I just wrote to the first 15 locations and made sure, I got back what I put in..

THANK YOU Marty
 
Don,

Not to butt too much into Marty's thread, but do you think something like this programmable oscillator might make an easier overall solution for folks like me that want higher baud rates on Qbus systems?

https://www.sparkfun.com/products/9116

A VFO is a nice idea (once you get it programmed!), but not sure that 7.407388 MHz (divide 133.333 by 18) is close enough to the correct 7.3728 MHz?
 
Back
Top