• Please review our updated Terms and Rules here

Anybody for NCR?

R

roberttx

Veteran Member
Joined
Aug 6, 2015
Messages
1,133
Location
Texas
This weekend I bought a whole bunch of NCR parts - boards, a backplane, what I think may be magnetic core memory, a disk pack, parts of a disk drive, power supplies and more.

Not all of it has made it home yet - I have to collect some more on Monday. As of this minute, I haven't even Googled any of the part numbers and have no idea which machine or machines I have parts from.

So, I figured that y'all might like to join in the fun. I'm going to post pictures and part numbers here, as I go along. Any help identifying things would be gratefully received.

Here's the backplane - or backpanel as they call it. The slots have 50 pins a side and are just under 6.5" long, not counting the plastic ends.

ncr1.jpg

The polarizing keys are in different places, as are the matching keyways on the boards. This limits which rows a board can be put into.

ncr2.jpg

On the back, we see that the sockets are wire wrappable, the rows are labeled A through P (with no O) and the pins are numbered 1-50, 52-102, 201-250 and 252-302. It was made by Elfab.

ncr3.jpg

Here's the NCR label, with our first part number - "348-0006213". Googling that gets nothing, but I noticed that the vertical text actually reads "6082 Backpanel". Gooling NCR 6082 computer takes us to this page, where we learn that the 6082 was a processor for an 8100 or 8200 series computer. Some progress, perhaps?

ncr4.jpg

Finally, here's the power connector, if anybody is interested.

ncr5.jpg

A quick break, then we'll start on the boards.
 
This, I think, may be the CPU. It has 16 LEDs, possibly suggesting a 16 bit word length.

ncr6.jpg

I base this on the words "One Board Processor", silk screened on the component side.

ncr7.jpg

Here's the back.

ncr8.jpg

And a close up of the label.

ncr9.jpg

Again, googling is unfruitful. Looking at the big ICs, we see four AM2901BPC Bit Slice Processors (4 bit, 14 MHz) and an AM2910DC 8 bit slice microprogram sequencer, all dated 1978. The numbers on the other two 40 pin ICs didn't turn up anything in google.
 
ncr10.jpg

Moving a little faster, clockwise from the top left, I have two of the same board with no description in the silk screen and a copyright date of 1981. It has three WD TR1602P-00 UARTS, labelled CH0, CH1 and CH2, with three banks of 4 LEDs, labelled CR1 through CR12. There are no headers, so if it's I/O it's only part of it. All I can say for now is that it's a three channel something.

The next board is silk screened as a "Memory Control Interface" and has "M118-02-STD" "C/N 1-2" and "P/N 348-0001297 C" on a sticker on the back. The final two boards both have "MOS Array Board" on the silk screen. One has a lot of TMS4116-20NL DRAMs and the other a lot of MK4116P-3 DRAMS, so I guess that they are memory boards.
 
For a brief break from boards, here's an I/O connector. We should expect to see a 20 pin header, somewhere.

ncr11.jpg
 
Clockwise from the top left again, we have another RAM board, this one using MK4027J-3 DRAMS, "Int Disk Controller A", "Int Disk Controller B" and "Multi-Chan Com Adptr". This last has three WD TR1602-B UARTS and three banks of four LEDs, labelled CHN0, CHN1, CHN2, so it would seem to be a different version of the two on the top left in post #3. So now we know what they are.

ncr12.jpg
 
The last of the big boards in this first box: another RAM board, this one using MK4116N-3 DRAMS, "Cassette Transport Interface Logic" and "Cassette Data Interface Logic". The other boards had 1M17 model prefixes and 348 part prefixes, but these two have M64 and 095 instead, so perhaps they are for an earlier machine?

ncr13.jpg
 
I said the last of the big boards, because there were also some smaller boards in there.

ncr14.jpg

This one has a lot of 7400 series logic and some ICs that I don't recognize, along with a WD TR1602B UART. The edge connectors are 80 pin, 4" long.

ncr15.jpg

It's dated Apr 29 1980- on the back.

ncr16.jpg

Sadly the part number, "704-4670097", tells us nothing.
 
The final two boards in this box are connected by a ribbon cable. One has an Intel C8080A CPU, a pushbutton switch and six LEDs, but says "Interface-RS 232", the other is just transistors & passives, so possibly some kind of industrial controller?

ncr17.jpg

ncr18.jpg

ncr19.jpg

ncr20.jpg
 
Chuck(G) said:
My guess would be a Century 8200--very popular mid-70s mini.

e.g.

The technology looks about right for the period.
Click to expand...

Thanks Chuck. Some of the boards have a copyright date of 1981 on them. Would NCR have been designing new boards for the Century 8200 that late?

Also, I suspect that the cassette interface boards, at least, come from an earlier system. I think I may have parts from two computers here, plus the smaller boards.
 
The 8200 was introduced in what, 1975? The 4116 DRAMs would have been current technology then.

But the Criterion 8400 or 8550 dates from 1976, so who knows?
 
Now, I promised you magnetic core memory and I haven't forgotten. Here it is, still in the box.

ncr21.jpg

The back.

ncr22.jpg

It's dated 1974 and is labelled C605-8Kx18. An 18 bit word length? Googling suggests that it came from a 605 processor, which could implicate anything from a Century 150 through 399, 499, 8200, 8250 and 726.

ncr23.jpg

This sticker says 315-0522181. Could it be for a 315? But wikipedia says the 315 had a 12 bit word length.

ncr24.jpg

The M17 is also mentioned in the thread linked to above, so I think that clinches it. It's not for a 315, but for a 605.

ncr25.jpg

However, the keyways in the edge connectors do not match up with any of the slots in the backplane. I think that this confirms that I have parts from at least two different processors - a 605 and a 6082.
 
OK, that's all the pics for today. I have a lot more stuff, some of it at my workshop (including a disk pack, some kind of hours meter, a piece of test equipment, an NOS power supply, punched paper tape, PCB drawings, a NEAT 315 manual and probably other stuff that I've forgotten) and some that I'm picking up tomorrow (disk drive parts, another PSU, most of an acoustic coupler, teletype parts etc.)

It all came from the estate of a retired NCR engineer and was sent to auction. When I went to the preview the auctioneer, who knows what I like, said "Bad news, Robert. You're going to spend a lot of money tomorrow." And he was right.

More pics to follow, as I'm able. I have a busy week, followed by scheduled cataract surgery, so it may be hit or miss for the next few weeks. As always, though, I'd be grateful for any info on what I have.
 
OK, I found a minute to sneak a quick one in. These appear to be some custom ICs. They're a 14 pin package and the printing reads "NCR 72 603".

Does anybody have a clue as to what they might be?

ncr26.jpg
 
roberttx said:
This weekend I bought a whole bunch of NCR parts - boards, a backplane, what I think may be magnetic core memory, a disk pack, parts of a disk drive, power supplies and more.

Not all of it has made it home yet - I have to collect some more on Monday. As of this minute, I haven't even Googled any of the part numbers and have no idea which machine or machines I have parts from.

So, I figured that y'all might like to join in the fun. I'm going to post pictures and part numbers here, as I go along. Any help identifying things would be gratefully received.

Here's the backplane - or backpanel as they call it. The slots have 50 pins a side and are just under 6.5" long, not counting the plastic ends.

View attachment 1058786

The polarizing keys are in different places, as are the matching keyways on the boards. This limits which rows a board can be put into.

View attachment 1058787

On the back, we see that the sockets are wire wrappable, the rows are labeled A through P (with no O) and the pins are numbered 1-50, 52-102, 201-250 and 252-302. It was made by Elfab.

View attachment 1058788

Here's the NCR label, with our first part number - "348-0006213". Googling that gets nothing, but I noticed that the vertical text actually reads "6082 Backpanel". Gooling NCR 6082 computer takes us to this page, where we learn that the 6082 was a processor for an 8100 or 8200 series computer. Some progress, perhaps?

View attachment 1058789

Finally, here's the power connector, if anybody is interested.

View attachment 1058790

A quick break, then we'll start on the boards.
Click to expand...
They look like parts from the venerable 605 processor that powered the 8200 computer system and the supermarket scanning system and the 399/499 accounting systems and other NCR systems. The backpanel was wirewrapped by a 315 model mainframe and could be customized for different configurations and different boards.
 
johnnyd said:
They look like parts from the venerable 605 processor that powered the 8200 computer system and the supermarket scanning system and the 399/499 accounting systems and other NCR systems. The backpanel was wirewrapped by a 315 model mainframe and could be customized for different configurations and different boards.
Click to expand...
FYI
I just scanned a bunch of boards and a drawing set for a 399 http://bitsavers.org/pdf/ncr/ncr-399
There is also a post from someone who has a 725 https://forum.vcfed.org/index.php?threads/ncr-725-ncr-605-processor.53634/
Did you ever figure out what the rest of the documentation was from?
I'm guessing the board with the 2901's is a single board version of the 605 since it has four 2901s

Studying your pictures and mine, you can identify what boards go where by how they are keyed.
The annoying thing is those connectors are >>102<< pins so they are going to be very difficult to find.
The ones keyed with 10 on the left and 20 on the right are I/O controllers.
I'm pretty sure that 2901 board and the semiconductor memory boards go into that backplane and are an upgraded 605 style processor
 
Last edited:
Those boards look similar to what is in the RICM's NCR 8250 system.
www.ricomputermuseum.org

NCR 8250

This NCR 8250 POS system was donated by Ames Department Stores in November of 2002. From the labels it looks like the processor chassis is a model 2520-0090. The Processor-1 board contains four Texas Instruments 74181 bit-slice ALU chips so this is a 16-bit processor.
www.ricomputermuseum.org www.ricomputermuseum.org
 
Did RICM get any documentation or media with it?
It's a drag when the only surviving software is on the system disk.
 
You must log in or register to reply here.
Back
Top