• Please review our updated Terms and Rules here

70's NCR Retail Terminal Help

NeXT

Veteran Member
Joined
Oct 22, 2008
Messages
8,096
Location
Kamloops, BC, Canada
This has been my project ever since 'rona mania flared up. It's an early 1970's Class 250 electronic cash register manufactured by NCR and using core memory. Could also be a networked terminal. All I know is that at one point it was operating standalone. Likewise it showed up a complete disaster and had to be almost entirely stripped, cleaned, reoiled and repaired.

59DA4734.jpg
CGS_9667.jpg


There's a bit more than jsut spit and polish that went into making this shine again. ;)
CGS_9797.jpg
CGS_9798.jpg


Okay, so here's where the fun starts. It runs a 400w linear power supply that came up seemingly fine after all the big caps were reformed. It was dirty and the power resistors all showed their age but it hasn't so far given me any troubles.
So now it turns on, the cooling fan works and I can trigger the print motor if I push the receipt advance button, but the advance mechanism relies on the digital side and that doesn't seem to want to come up. In fact, nothing wants to come up. The display remains blank, the keypad does nothing and it doesn't make any noises.
I've made a tread previous about this here about the memory. The electronics are really, really early IC tech. Some chips have date codes from 1971! NCR has gone through their parts and relabeled everything to their own internal part numbers and the only databook I found on bitsavers seems to be too new to list anything. I also sanity checked and cleaned and reseated every plug and edge connector in the machine while checking for any obvious shorts or smaller cap issues. thermal camera complains about a 22mfd50v axial in the display section but otherwise I'm not seeing any grossly overheating components. In fact I'm not seeing most of the IC's give off heat, but the scope at least says there's signals moving around at the test points so....yeah no idea if I'm missing volts or these just are not as power hungry as I'm expecting. I have no datasheets on what to expect from the power supply.

CGS_9829.jpg


The only way I've managed so far to make it do much of anything was to either press SP2 on the control unit, then it beeps, displays CLEAR ERROR on the display and after six seconds blanks the display, beeps and lights the indicator again...OR by accident I found that if I grounded my scope lead on the black test point on the control unit and poked my probe at the orange test point on the terminal clock board, it might beep, it might run the motor, it might fire a few relays or it might even light up the entire display. Either way, it doesn't like it when I'm scoping using the wrong ground but it still twitches its leg. ;)

CGS_9881.JPG

CGS_9883.JPG


I don't know how to proceed from here. I have no service docs, no cross references and the boards are all buried under the keypad assembly which makes them hard to probe pins while running so randomly poking around is a massive pain. Does anyone else have good pointers and what to do next? High resolution images of the boards, front and back, are available here --> https://imgur.com/a/nnDAhRy
 
Hi. I was an NCR Field Engineer (repair tech) for 37 years and my greatest accomplishment was to completely avoid getting involved with the Class 250. I was heavily involved with the Class 255 which was really just a dumb terminal attached via 4 wire twisted pair network wiring to the 605 based minicomputer that ran the whole scanning system. The 255 was used in supermarkets. It was attached to the laser scanner but did no processing. Everything was passed through to the 605. The printer was a helical printer and each single character printed by it was timed and printed by the 605.
The 250 was a stand alone cash register that was a nightmare of rom chips. Servicing them was a matter of carrying in a rom burner to the place of business and flashing the errant prom. It used the same mechanical frame as the 255 but the boards under the keyboard were completely different. I do know that on startup the 250 expects to run the printer and get back timing pulses from the stationary wheel that has a step/lug for each character position on the helical print wheels. I meet up with a few guys from back in the day. A couple of them worked on the 250. I could see if any of them would have books sitting around from those days, but it's unlikely they do.
 
Hey johnnyd. Sorry I didn't see your post earlier. I guess replies to threads before the big forum upgrade didn't carry over.

Okay so the existence of the 255 absolutely answers some questions. While I was unseizing the printer (that was a nightmare) I popped out the imprint stamps and the machine seemed to be last working at a religious book shop which would absolutely have no need for a 605 system.
I'm curious about the rom chips. There's nothing socketed in here. The closest thing this has to storage is shift register memory and core memory, which are the two big reasons I've adoped it into my collection now.
I know the lug wheel and pickups you are referring to. Hides in the middle compartment of the printer and has one timing wheel for each letter position on the drums and another to indicate one full rotation. I didn't check if either coil had gone open at the time.
In the time since I made the thread the power supply acted up again. It overheated one fuse and all the rails died, so either the capacitors "deformed" and it didn't like an inrush load or something else is not happy. The main power connector is also being really annoying as the pins are corroded and are not delivering an excellent condition. I also got a few documents for the later 280 which don't answer every questions but parts like the clock board and peripheral controller are the same, so I have *some* documentation now. I still lack anything for the main CPU and memory, display and the power supply.
 
I also was with NCR for 44 years. The NCR 280 was before the NCR 250. They shared much of the design and were build in Cambridge, OH. The lower board on the left side of card bay is the TCU (terminal control unit), 4004 cpu chip on lower card. There was a ROM card attached in early years (1972-1976). Then redesign got the ROM “cans” on one card. The 2 push button switches get you into diagnostics and programming mode. The top card on left is the clock card (M19). I dont have any docs anymore, only what I can find in “the cobwebs of my mind”. FYI the person who worked on 255 and posted about rom burners was confused with the NCR 2140. I did use many EPROM’s and yes we had special machine to “burn” replacements.
 
That's interesting. A 4004? That seems a little late for a 1973/1974 system but I can't argue with that since you know more about this than I do. I really wish I could decode the chip part numbers from what NCR marked them with. Without any idea what any of the IC's are debugging is a complete no-start.
 
Back
Top