• Please review our updated Terms and Rules here

Saved an 11/73, help me revive it!

M

mechaniputer

Experienced Member
Joined
May 23, 2015
Messages
59
I'm a CS grad student at an engineering university and there's a recycling area for old electronics on campus that I check frequently. I'm into all sorts of old computers and I have a meagre collection of mostly Apple 2 stuff, but I've always been most interested in the minicomputer era, and DEC in particular. Admittedly I grew up in the Win95-XP era, but computing history has been a hobby of mine for quite awhile. The number one machine that I most wanted to have a working example of has been the PDP-11, but I'd never had the chance to get one until today.

Anyway, I was checking the electronics graveyard and saw a large enclosure labeled "Kevex", and I quickly determined that it was a controller for an electron microscope from the late 80's. Knowing that the PDP-11 was a common embedded controller during that time period, I took apart the enclosure and sure enough, I found an 11/73.

I see that another person had a similar find back in December: http://www.vintage-computer.com/vcf...abilitating-a-little-PDP-11-73-what-do-I-need I have mostly the same equipment but I have a total of 7 boards.

Not sure if I am identifying these boards correctly... I think that I have:
CPU card
2 megabyte ram card
4-port serial card
DAQ card
Kevex microscope controller card
SCSI card
MDB MLSI-DRV11C (What is this thing anyway?)

Here are some pictures from my terrible phone camera: http://imgur.com/a/jnH1p

I'd love to get this thing running in some capacity so I'll be researching things from that other thread to get this thing powered and see what works!
Currently, I need to get the cards in the right order, get it powered somehow, and set up a terminal to play with.
If possible I would ultimately like to get it running some very old Unix. But first things first!

Edit: I've been replying, but due to spam protection my posts seem to be taking in excess of 24 hours to become visible. Hopefully my posts will be visible soon.
Re-Edit: My posts have shown up!
 
Last edited:
Most excellent find! My 11/73 was also found when I was in grad school, and it controlled some kind of spectrometer. However I graduated 15 years ago!

Which NY engineering school? There are a few, but none near me in Westchester.

You can pretty much run any OS you want on this machine. It probably ran RT-11. Do you have the SCSI disk that would have been on it?

DRV-11 is general purpose I/O. Perhaps the microscope used that too.

You have a lot of reading to do in the near term.

Lou
 
What brand / model SCSI controller is that?

For a minimal system you would need the M8192 KDJ11-A CPU card, the 2MB memory card, the 4-port serial card, and the SCSI controller, plus a SCSI device.

The M8192 CPU doesn't have any boot ROMs but the SCSI controller should have the capability of booting from an attached SCSI device.

For power I think you should only need a +5V and +12V supply, but there are also some signals that usually come from the power supply such as a power OK signal and a Line Time Clock signal.
 
I'm at Clarkson University in Potsdam. I don't have a hard drive for it, but it has a pair of 8-inch Flexible Disk Cartridge Floppy drives... I had not seen those before. I'm guessing that the hard drive was removed before I got to it. This week I'll try to scrounge a PC PSU for it, then construct a 10-pin control board and then a serial adapter to see if I can make it do something over serial. Not sure if I have parts lying around for those things, which might become the limiting factor as I don't currently have much room in the budget.
 
Congratulations! Yeah, embedded systems of a certain age can be a great resource - I upgraded my 11/23 with 128KB to an 11/73 with 4MB with bits from a mass spectrometer system that wound up in my local recycle center. (Still need to get the SCSI controller it had set up in mine, though.) I believe for Unix you can run 2.11BSD and a couple different System 7 versions.
 
gslick: Not sure and I'm not near the machine right now. I'll look tomorrow. My plan is to start by testing it with ODT mode and go from there. I can emulate disks if needed by following instructions on this page:
http://www.diane-neisius.de/pdp11/index_E.html
I'm pretty much going to be trying to follow what that page says to do, starting with ODT mode, so if there are any pitfalls for doing that with an 11/73, let me know.
 
mechaniputer,
You found a great system to work with! As gslick said the SCSI controller should have the ability to boot a SCSI device. If the SCSI disk for the system is non-functional, there are two ways to think about getting started. One is that the 4 port serial card likely acts like a DLV-11J. It can have one port act as the console and allow you to interact with ODT (Octal Debugging Tool). Another port can be configured to interact with a laptop (or a Raspberry Pi) running software to act as a TU58 tape drive. From there you can boot diagnostics like XXDP.

The other approach is to buy a SCSI2SD card (see http://www.codesrc.com/mediawiki/index.php?title=SCSI2SD ). This card has a 50 pin SCSI interface that with a flat ribbon cable will act as a disk drive to your SCSI card. The disk it uses is a microSD card. The microSD card can be loaded with DEC operating systems like RT-11, RSTS, or RSX11M+. I have a PDP-11/83 with an Emulex UC07 SCSI card and it uses a couple SCSI2SD devices so I can have a system disk and a data disk. The RSX11M+ operating system in its last release could handle disks up to 8 Gigabytes, but it is convenient to use 150 MB RD54 sized disks and if you really need space a 1.5 GB RA92 sized disk.

The great thing about the SCSI2SD approach is that you can use Simh to emulate a PDP-11 on an ordinary Windows, Linux or Mac OSX PC download the disk images from places like ftp://ftp.trailing-edge.com/pub/ testing that everything works fine under Simh, even doing a SYSGEN to build an OS like RSX11M+ customized for your hardware. Then you can use the Unix/Linux dd disk copy utility to load the SD card and then boot it on your 11/73. Much of the documentation that you will need for the DEC operating systems can be found at http://www.textfiles.com/bitsavers/pdf/dec/pdp11/

Also look through the earlier threads in this forum for info on SCSI2SD and the TU58 emulators. I highly recommend utilizing RSX11M+ as it will take full advantage of the capabilities of the PDP-11/73 and the RAM that you have allowing multiuser access through the serial card and through ethernet if you add an ethernet card down the road. If you prefer UNIX 2.11BSD, it will also run on it just fine and also provide a multiuser experience as commodorejohn points out.

Good Luck and Enjoy!
Mark Matlock
 
Wow! Old computers, Snackwell cookies and Skippy peanut butter! You'll fit right in here!

Two sites I found particularly helpful are this one (of course), and a good site for getting a minimal configuration up with some jury rigged additions is here:

http://www.diane-neisius.de/pdp11/index_E.html

There are others of course, but part of the fun is in the searching. ;-)
 
You can certainly run 2.11 BSD on that system. If the SCSI card is a CMD CQD type interface, you will need the patched boot sector to boot the system successfully.
 
mechaniputer said:
gslick, that was a very good question. I guess it's an interesting one, a TDL-12 with v 1.9 ROM.
I see that you've had some fun with one of those here: http://www.vintage-computer.com/vcf...stems-TDL-12-QBus-SCSI-controller-information
I'd appreciate any tips with regard to getting that to work!
Click to expand...

The TDL-12 is an interesting SCSI controller, mainly because it emulates an RL02 controller instead of the more typical MSCP controller. That is useful if you want to run an older version of an OS which does not have MSCP support, but the drawback is that it limits you to 10MB logical RL02 disk sizes. You can map up to 4 logical RL02 disks to a single SCSI device so you can have a total of 40MB online at a time. Anything beyond the first 40MB of a SCSI hard drive will be ignored by the TDL-12.

There is a TDL-12.pdf manual on this page:
https://sites.google.com/site/glensvintagecomputerinfo/td-systems/tdl-12

The firmware version I have on my controller is V1.80 so it would be interesting to get a dump of the V1.90 firmware EPROM on you controller sometime for comparison if you have the means to dump that.

If you have a spare SCSI hard drive and a system with a SCSI host controller you can use to 'dd' disk images to the SCSI hard drive that would be the easiest way to get things going.

-Glen
 
gslick said:
The firmware version I have on my controller is V1.80 so it would be interesting to get a dump of the V1.90 firmware EPROM on you controller sometime for comparison if you have the means to dump that.

If you have a spare SCSI hard drive and a system with a SCSI host controller you can use to 'dd' disk images to the SCSI hard drive that would be the easiest way to get things going.
Click to expand...

I don't currently have the ability to get a dump of that, but if/when I can, I'd be happy to share it.
I hadn't realized that just any SCSI drive would work, but that looks to generally be the case! That's nice. I'll probably do that.

Question: Just out of curiosity (don't worry, I haven't broken anything yet) what happens if I put one of the cards in the wrong slot? Will that usually cause damage?
Also, to make extra sure, the CPU always goes in the first slot (such that the chips face "outwards" on the chassis), the memory always goes in the 2nd slot, the serial card always goes in the 3rd slot and the SCSI controller should (always?) go in the 4th slot? Is all of that true regardless of which backplane I have?
 
If you have a dual wide QBus backplane there shouldn't be anything that can get damaged by placing cards in the "wrong" slot. That is mostly only an issue with quad wide QBus backplanes where you may have a mix of Q/Q and Q/CD cards and backplanes.
 
Given all the "tricks" that DEC performed on the Qbus over it's life span, not to mention the ones done by OEM vendors then and up to the present, you might want to take voltmeter in hand and check out how your particular backplane is wired. They may look similar, after all a Qbus connector is a Qbus connector, but they left the factorys in quite a few different configurations. Then the field modifications began. Saying how yours might be wired today would be an educated guess at best. Here's one thread you might want to look at. There are many others that discuss card order, dangerous combinations, and 16/18/22 bit addressing. It can be an interesting hobby. ;-)

https://groups.google.com/forum/#!msg/alt.sys.pdp11/nGIErC8Bu2M/1IxfEy6wuakJ
 
Thanks, I'll be doing some reading there.

Another question: I've got the serial card mostly set up as per Diane's PDP-11 page, but I don't know what the interrupt level should be set to. Right now mine is set to 7 but default for the card seems to be 4 according to the manual.
Mine is a different card, but I found my manual here starting at page 34, since it's the newer revision: http://www.scotnet.co.uk/iain/DEC/11-73/IM983_GTSC304.pdf

I'm hoping to have enough parts together to have this running this weekend, maybe sooner. I'll keep you guys updated with progress also.
 
mechaniputer said:
I've got the serial card mostly set up as per Diane's PDP-11 page, but I don't know what the interrupt level should be set to. Right now mine is set to 7 but default for the card seems to be 4 according to the manual.
Click to expand...
Set it to 4 if you want to be able to use a standard operating system driver without issues.
 
One unusual thing that I noticed is that the 10-pin connector on the backplane seems to not have been connected when I found it. Indeed, when I looked at the pile of cables from dismantling everything, there were no cables that would clearly fit it, and the header itself was covered in dust that would not have been there if it had been connected.

Is there some way that it could have run with nothing being connected there? Seems weird.
 
Last edited:
mechaniputer said:
One unusual thing that I noticed is that the 10-pin connector on the backplane seems to not have been connected when I found it. Indeed, when I looked at the pile of cables from dismantling everything, there were no cables that would clearly fit it, and the header itself was covered in dust that would not have been there if it had been connected.

Is there some way that it could have run with nothing being connected there? Seems weird.
Click to expand...

A 10-pin header connector for the backplane is likely for some of the control signals:
BPOK, BEVENT, SRUN, BHALT, BDCOK

For example see the H9275-A backplane Figure 5 on page 381 of the PDP-11 Microcomputer Interface Handbook 1983-84:
http://www.bitsavers.org/pdf/dec/qbus/EB-23144-18_QbusIntrfs_1983.pdf

The BEVENT Line Time Clock 60Hz signal is usually generated by the power supply but there could be something else in the system that generates it. Some software might not operate correctly without the BEVENT LTC signal.

The BHALT signal would usually be generated by a push button on a front panel interface.

The BPOK and BDCOK signals would usually be generated by the power supply in DEC systems, and sometimes by power supply monitoring boards in 3rd party systems. I think one of those two signals is also usually connected to a restart push button on a front panel interface.
 
gslick said:
A 10-pin header connector for the backplane is likely for some of the control signals:
BPOK, BEVENT, SRUN, BHALT, BDCOK
Click to expand...

I know what all the pins are for, but what I'm saying is that when I found it, it did not have any of those connected (unless I missed something like a single wire going to one or two of them, but I don't think there was based on the amount of dust on all of the pins). I don't think it could have run like that.
 
You must log in or register to reply here.
Back
Top