A new fossil record

[LadderLogic]

I'm waiting for: "I had to repair a Jacquard Loom the other day ..."

I at least have seen one.

 

[Bitmore]

O.K. guys when I get back to the home office I will show you (pictures) of the ultimate PLC, the PLC 1. Yes Ken, we tried to sell it back for the museum but they didn't want it (we have 2 of them). It is used to control the security lighting.
It, the PLC, itself is about 2' by 2' by 3', and wait till you see it!

bitmore

 

[OkiePC]

My first job as a PLC guy was working with these:

http://www.applied-automation.com/products/honeywell/ipc300.shtml

We had three racks of I/O per machine, and they were really cantankerous. You could corrupt the program with a faulty ribbon cable in the I/O!. I had 19 tire assembly machines with those antiques. This was in 1996...The Fonzie method of repair became a standard first step. (you know, smack it firmly with a closed fist). I also got real good at folding paper and stuffing it up against ribbon cables. We had four of the loader monitors that we had to keep going, and 16 different programs that were loaded by cassette tape to change the tire construction. My first programming task was to make one program that would build all the different tires. It took me over a month to get that done and tested.

All 36 of my machines had mechanical drum steppers for the belt/tread assembly. They were called "Tenor Steppers" and we had a lot of fun with those too. I spent many months swapping them for SLC 5/04s and got pretty good at panel wiring, since I only had four hours of downtime per machine to complete that job.

I can't find any pictures of the old tenor steppers though...

 

[robertmee]

Anyone ever use or remember Sprint Logic....I believe it was early 70's stuff. I tried googling it but of course all that comes up is Sprint (the telephone company) and how they behave illogically

The Sprint Logic I remember consisted of these little logic cards. Maybe the size of a credit card. Each card was an OR gate or an AND gate or some other boolean logical function. You had a huge rack/backplane (like 8' tall by 8' wide) that you plugged these cards into to make up your logic. Think of every XIC or XIO being a Card. One of those babies goes out and imagine the troubleshooting of that!

 

[gas]

robertme
We had a milling machine with a huge cabinet full of those cards. Any dust, change of temp or phase of the moon would give us fits. There were also racks in the door and whenever you opened the door some of those cards would loosen creating another problem.

 

[robertmee]

robertme
We had a milling machine with a huge cabinet full of those cards. Any dust, change of temp or phase of the moon would give us fits. There were also racks in the door and whenever you opened the door some of those cards would loosen creating another problem.

Sounds like them! I'd love to see a picture of them again. One of the earliest logic processes that I cut my teeth on.

 

[PeterW]

My first job as a PLC guy was working with these:

http://www.applied-automation.com/products/honeywell/ipc300.shtml

We had three racks of I/O per machine, and they were really cantankerous. You could corrupt the program with a faulty ribbon cable in the I/O!. I had 19 tire assembly machines with those antiques. This was in 1996...The Fonzie method of repair became a standard first step. (you know, smack it firmly with a closed fist). I also got real good at folding paper and stuffing it up against ribbon cables. We had four of the loader monitors that we had to keep going, and 16 different programs that were loaded by cassette tape to change the tire construction. My first programming task was to make one program that would build all the different tires. It took me over a month to get that done and tested.

All 36 of my machines had mechanical drum steppers for the belt/tread assembly. They were called "Tenor Steppers" and we had a lot of fun with those too. I spent many months swapping them for SLC 5/04s and got pretty good at panel wiring, since I only had four hours of downtime per machine to complete that job.

I can't find any pictures of the old tenor steppers though...

We had these as the first PLC's in the Ford plant I was working in, prior to that they were experimenting with logic gate technology.

We had IPC300's and another model (190 I thionk), plus some Allen Bradley's, a few 1774's and what they called Mini PLC's, which eventually became the PLC2/20's and 2/30's.

 

[TConnolly]

The first controller I worked with was a DEC PDP8A. 4K of 12 bit words in mag core memory. No microprocessor, CPU was a double stack of two sided boards populated with gates. The instruction set had just 8 instructions. Boot sequence had to be entered by hand with a switch panel on the front. Then program could be loaded with paper punch tape. The DEC PDP was primarily a mini-computer but there were lots of interface boards available that made it a popular controller.

Primary troubleshooting tool was a good hard rubber eraser and a solid flat surface to bang the mag core memory cards against. That was usually sufficient to solve most problems. Jostling all those ferrite donuts could miraculous cure untold numbers of mysterious memory errors.

 

[robertmee]

The first controller I worked with was a DEC PDP8A. 4K of 12 bit words in mag core memory. No microprocessor, CPU was a double stack of two sided boards populated with gates. The instruction set had just 8 instructions. Boot sequence had to be entered by hand with a switch panel on the front. Then program could be loaded with paper punch tape. The DEC PDP was primarily a mini-computer but there were lots of interface boards available that made it a popular controller.

Primary troubleshooting tool was a good hard rubber eraser and a solid flat surface to bang the mag core memory cards against. That was usually sufficient to solve most problems. Jostling all those ferrite donuts could miraculous cure untold numbers of mysterious memory errors.

We ran an entire plant on PDP11's and uVaxes. Fun days

 

[rguimond]

We just took out our last Tenor Stepper last year. Used to open/close valves during CIP.

 

[Mike Lamond]

The chief engineer at my first and second (long story there) jobs told me about working on DEC systems when he was first out of college, in the early 1970's. It seems that the first AC input modules pulled about 3 watts per point, which became very obvious when they did a job with nearly 4000 I/O. To deal with the heat output, the plant left the I/O cabinet doors open and built a break room around them.

Other extinct species:
Does anyone remember a Sylvania PLC? About the time I started my first job in 1986, the company used these in several natural gas odorizer feed control panels. They were programmed in SYBIL (SYlvania Basic Instruction Language) on a Kaypro II CP/M portable repainted in Sylvania blue.

We put a bunch of IDEC FA2J PLCs in small water and wastewater treatment plants across New York State during the early to mid-1990s, before switching to Modicon and sometimes A-B. I wonder who's supporting them now.

I last worked on a PLC-2 in 1995 and upgraded two others to PLC-5 processors in 1997-98.

 

[OkiePC]

. . .To deal with the heat output, the plant left the I/O cabinet doors open and built a break room around them.

That's funny...

 

[RMA]

The first controller I worked with was a DEC PDP8A. 4K of 12 bit words in mag core memory. No microprocessor, CPU was a double stack of two sided boards populated with gates. The instruction set had just 8 instructions. Boot sequence had to be entered by hand with a switch panel on the front. Then program could be loaded with paper punch tape. The DEC PDP was primarily a mini-computer but there were lots of interface boards available that made it a popular controller.

The 8A, if I remember corrrectly was actually the last of the 19" rack PDP 8s, must have been mid to late '70s.

My first process control project was in 1969 and it used a PDP 8E (shown here, at the bottom of the page) with a whole 12 kB of memory to control the hydraulics and hangar temperature of the system that "flew" Concorde 001 on a simulated flight to New York and back every day. The system ran every day until Concorde was withdrawn from service.

After 1972 our systems were based on the 16 Bit PDP 11. The last four of 14 Systems which I commissioned for Bayer in Germany in 1980 (and re-visited on average every four years) were finally shut down in 2003 when Bayer sold the factory to Rüttgers Chemicals. I maintained those systems right up to the end.

One of the nicest features of the early versions of both the PDP 8 and the 11 were the "piano" keys for data input and the rows of LED for Register data output. You got a good feel of what was going on simply by watching those LED's flashing. Data input with both hands (for test programs in machine code) was at least five times faster than with the subsequent calculator style keyboard (as used on the 8A).

Primary troubleshooting tool was a good hard rubber eraser and a solid flat surface to bang the mag core memory cards against. That was usually sufficient to solve most problems. Jostling all those ferrite donuts could miraculous cure untold numbers of mysterious memory errors.

The bit about the hard rubber eraser was just as true for the PDP 11 and I still have one in my tool kit till this day!

 

[leitmotif]

Did any of you guys put in the propulsion control systems on Roman galleys??

Dan Bentler

 

[Ken Roach]

Excellent stories ! Any time I start feeling like I've seen it all, I can always come back here.

It turns out there was nothing wrong with the Mini PLC-2. These models were built before EEPROMs, so you either had to keep the batteries refreshed (twin D-size alkalines!), keep the power up 24x7, or have a re-loader ready to go. Nobody could remember the last time they'd used the Apple II downloader, which would probably work if they had a good serial card and knew how to configure it.19

I manually re-entered the program in AI-2 (vintage 1986-1996) and had to get an RS-232/PLC2 cable because I couldn't get the 1784-PCMK to work in DOS reliably on a x486 NEC VersaPro laptop.

The challenges I had just going back to 1992 technology (1784-PCMK and DOS, with Cardware and EMS managers, configuring config.sys and autoexec.bat files) were instructive.

The customer has no plans to replace this controller or upgrade it; they plan to run every machine in the place to failure.