Of operators and orders, back in the day

1512 views
53 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    May 2003
  • From: US
  • 18,130 posts
Posted by BaltACD on Thursday, January 23, 2020 4:11 PM

The Computer Automation boxes I mentioned had 32K Words of Core Memory to run the show.  Those 32K Words were programmed by Kleinschmidt using Assembly and the compiled program was distributed on punched paper tape - several thousand feet of punched paper tape.  At times, floor dust could 'contaminate' the tape and give a false character when read in.  Why magnetic tape wasn''t used I have no idea.

Once the Tandem equipment was installed, program updates could be distributed over our private network by electronic means.  By that time, all the computer sites were linked together on our own private network.

  • Member since
    December 2001
  • From: Northern New York
  • 20,461 posts
Posted by tree68 on Thursday, January 23, 2020 3:39 PM

And to think, you carry more computing power (and memory) in your pocket today...

The disk units for the 4331 and 4361 I worked on were all of 592 Mb...  They were the size of dishwashers, and I think there were at least a dozen of them.  Nice warm place for a nap, though...

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

  • Member since
    July 2010
  • From: Louisiana
  • 1,963 posts
Posted by Paul of Covington on Thursday, January 23, 2020 2:11 PM

Erik_Mag

Paul,

The SS-80 and SS-90's were late 1950's machines, the difference between the 80s and 90s was that the 80s used cards with standard 80 column Hollerith format and the 90s had 90 columns of round holes arranged in two rows of 45 columns each. Both of the SS machines used ferrite core logic elements with a "radar" tube (4CX250B?) providing the high power clock. Memory was a drum with 5K words of 40 bits, with a packed decimal format (i.e. 40 bits = 10 decimal digits).

I think Rohr was still using an "SS" series computer in spring 1969, though the main computer was an IBM 360.

 

   Ah, yes!  We just called it the Solid State, and there was the 80 and 90.  I didn't make the connection with "SS."   That was a monster.  I never worked on one, but when I hired on, the La. Highway Department was still using one.  And it always struck me as unusual that the term "Solid State" referred not to transistors, but magnetic amplifiers.  There were still tubes in the printers and other peripherals.

_____________

   "A stranger is just a friend you ain't met yet."  ___ Dave Gardner

  • Member since
    January 2019
  • 389 posts
Posted by Erik_Mag on Thursday, January 23, 2020 1:51 PM

Paul,

The SS-80 and SS-90's were late 1950's machines, the difference between the 80s and 90s was that the 80s used cards with standard 80 column Hollerith format and the 90s had 90 columns of round holes arranged in two rows of 45 columns each. Both of the SS machines used ferrite core logic elements with a "radar" tube (4CX250B?) providing the high power clock. Memory was a drum with 5K words of 40 bits, with a packed decimal format (i.e. 40 bits = 10 decimal digits).

I think Rohr was still using an "SS" series computer in spring 1969, though the main computer was an IBM 360.

  • Member since
    July 2010
  • From: Louisiana
  • 1,963 posts
Posted by Paul of Covington on Thursday, January 23, 2020 1:19 PM

   Thanks, Overmod.  I remember being told about the dual systems.

_____________

   "A stranger is just a friend you ain't met yet."  ___ Dave Gardner

  • Member since
    September 2003
  • 10,536 posts
Posted by Overmod on Wednesday, January 22, 2020 8:14 PM

Paul of Covington
Did they call their model "Non-Stop", or something like that? 

That's the Tandem system (NonStop) that Balt talks about later.  Two complete systems running in parallel, so failover was immediate with no loss of essential data; a system could be scaled to thousands of processors (in pairs) for tremendous reliable throughput.  This was one of the great innovations of the mid-Seventies.

  • Member since
    May 2003
  • From: US
  • 18,130 posts
Posted by BaltACD on Wednesday, January 22, 2020 5:41 PM

Paul of Covington
   Ah, the good ole days!

Actually they were - I had a boss that respected and compensated me for my abilities.  Would travel to Chicago quarterly to plan our goals in concert with Kleinshcmidt for improvements and enhancements to our systems that were installed in Baltimore, Richmond, Newport News, Clifton Forge, Cumberland, Willard, Walbridge, Russell, Columbus Philadelphia, Pittsburgh.

  • Member since
    July 2010
  • From: Louisiana
  • 1,963 posts
Posted by Paul of Covington on Wednesday, January 22, 2020 4:59 PM

   Ah, the good ole days!

_____________

   "A stranger is just a friend you ain't met yet."  ___ Dave Gardner

  • Member since
    May 2003
  • From: US
  • 18,130 posts
Posted by BaltACD on Wednesday, January 22, 2020 4:54 PM

Paul of Covington
 
BaltACD
Kleinschmidt was the computer OEM for the computers we used, using Computer Automation boxes as the starting point with 'mini' computers (mini in the sense that they would only fill a closet, not a entire room). 

   Did they call their model "Non-Stop", or something like that?  I remember seeing one and was told that it was used by the railroads.

No.  In their original incarnation, they used the Computer Automation boxes and 10MB (yep megabyte) disk drives.  Kleinschmidt programmed the box in Assembler.  The 12 inch platters on the 10MB disks would crash every month to six weeks, after a period of time the disk drives were upgraded to 100 MB units consisting of multiple platters which didn't crash as frequently.  The Assembler programing was progressed to allow end user neumonic style end user application programing to better support the railroad functions we were using.  The end user programming allowed end user iterative file manipulation to get to the end results, initially the file manipulation took place on the 'zero' disk platter or 'scratch pad' - some applications could grind on for minutes.  Later a 'disk emulator' was developed that allowed 'scratch pad' manipulations to take place on a all memory chip 'disk', with a mega increase in processing speeds.  Ultimately the system was reconfigured to run on Tandem computer systems, with real time back up.  After several years of operation on the Tandem systems the applications were integrated into the CSX Main Frame computer system and the department I worked for was eliminated.

  • Member since
    July 2010
  • From: Louisiana
  • 1,963 posts
Posted by Paul of Covington on Wednesday, January 22, 2020 3:18 PM

BaltACD
Kleinschmidt was the computer OEM for the computers we used, using Computer Automation boxes as the starting point with 'mini' computers (mini in the sense that they would only fill a closet, not a entire room).

   Did they call their model "Non-Stop", or something like that?  I remember seeing one and was told that it was used by the railroads.

_____________

   "A stranger is just a friend you ain't met yet."  ___ Dave Gardner

  • Member since
    July 2010
  • From: Louisiana
  • 1,963 posts
Posted by Paul of Covington on Wednesday, January 22, 2020 3:14 PM

Erik_Mag
My sister worked on actual Kleinshmidt machies when she was in the Army ('75 - '78) and I'm pretty sure I saw one in the basement of Corey Hall at UCB in '72 - it was part of a Univac SS-90 the computer club was working on.

   I worked for Univac/Unisys from 1965 to 1995, and SS-90 doesn't ring a bell.  Was it a military model?  They were sort of separate animals, sometimes variations of commercial ones.

_____________

   "A stranger is just a friend you ain't met yet."  ___ Dave Gardner

  • Member since
    May 2003
  • From: US
  • 18,130 posts
Posted by BaltACD on Wednesday, January 22, 2020 1:50 PM

jeffhergert
On the subject of handing them up at night, an operator I knew well once told me that after doing it for so many years he could tell when the rear end was 10 or 12 cars from him by the way the sound of the train changed.  Balt or Lith (or any one else), did you ever notice that?

Jeff  

Depended on the location - some places had a small dip or rise and you could hear the slack action change.  For myself, the worst place to hand up orders was for Eastbounds at North Vernon, IN.  Trains had to make a Statutatory Stop and observe the target on the railroad crossing at grade with the New York Central N-S branch line.  With the train starting from a stop, just West of the crossing and the operators hand up location - handing up to the head end was as easy as it got.  If one was handing up to one of the premier, high powered manifest trains - findind the rear end got difficult.  The general grade was descending Eastbound, and the high powered (long) manifests would be chanting in Run 8 soon after passing the station and accelerating the train for all it was worth down the slight grade.  Did I mention, railroad crossing at grade, trains crossing the diamond pound the track structure into the ground - one axle at a time like a 25 ton power hammer on each point of the diamond - pounding track structure into the ground creates dust, the ever increasing train speed creates wind, wind blowing the dust from the crossing into the operators face as he is searching to the West to catch a glimpse of the markers so you can tell where the rear end is.  I only worked 3rd trick at North Vernon for a week or two - had I worked it longer I would have invested in a set of goggles to let me peer into the wind blown dust with more eye safety.

I did miss getting the orders to the rear of a Eastbound one time - they stopped where the Conductor could access the wayside telephone and contacted me on the Block Line - I was able to read the Conductor the contents of the orders that had not been properly delivered.  The train the proceeded.  The Train Dispatcher was informed of my mistake so he could enter the delay on the Train Sheet and use it in figuring other meets down the line.

  • Member since
    March 2003
  • From: Central Iowa
  • 5,230 posts
Posted by jeffhergert on Wednesday, January 22, 2020 10:47 AM

I think most roads used two sided carbon paper.  The Uniform Code roads, 1968 - RI, MP, SSW, MKT and some terminal companies made 5 copies for a train.  One office copy, two for the engineer - one set for the fireman and/or brakeman to read and a set for the conductor and rear trainman.  (They, along with some other roads, delivered orders and clearance on the fly with the T-O signal in the Stop position.)

I read that the D&RGW copied 4 for a train.  As I recall; office copy, engineer's copy, and one each for the conductor and rear trainman.

On the subject of handing them up at night, an operator I knew well once told me that after doing it for so many years he could tell when the rear end was 10 or 12 cars from him by the way the sound of the train changed.  Balt or Lith (or any one else), did you ever notice that?

Jeff   

  • Member since
    May 2003
  • From: US
  • 18,130 posts
Posted by BaltACD on Tuesday, January 21, 2020 10:11 PM

Deggesty
The Southern RR train orders that I saw were made with two-sided carbon paper, and whether typed or handwritten were easy to read

You had to use as much pressure on the stylus when hand writing the orders as you could muster when doing the 'copy 13' or 'copy a bunch' without tearing through the top copy of the order.  Since the ALL CAPS billing machine typewriters were all manual - you needed to strike the keys with force when copying a high number of copies.  As a neophyte typist at the time - it took me some time before I was able to type the orders and keep up with the Train Dispatcher in doing it.

When copying 3, 5 or 7 copies, a reasonable amount of pressure would produce readable copies either hand written or typed.

  • Member since
    January 2019
  • 389 posts
Posted by Erik_Mag on Tuesday, January 21, 2020 10:01 PM

BaltACD

Kleinschmidt, I believe, was the original holder of the Patents on most forms of teletype.

You are correct, although the correct term would be teleprinter as "Teletype" was a Western Electric trademark. OTOH, it's pretty much a generic term, especially with "TTY" as an abreviation.

My sister worked on actual Kleinshmidt machies when she was in the Army ('75 - '78) and I'm pretty sure I saw one in the basement of Corey Hall at UCB in '72 - it was part of a Univac SS-90 the computer club was working on.

  • Member since
    December 2001
  • From: Northern New York
  • 20,461 posts
Posted by tree68 on Tuesday, January 21, 2020 9:04 PM

Lithonia Operator
I rarely had to make more than four copies. But when I had to do six, the lower ones were quite faint.

Used to have that problem even with six part computer paper - and that was printed on a machine meant to do that.  No worries about not striking a key hard enough or pressing down hard enough.

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

  • Member since
    May 2019
  • 859 posts
Posted by Lithonia Operator on Tuesday, January 21, 2020 7:40 PM

I rarely had to make more than four copies. But when I had to do six, the lower ones were quite faint.

We made two for the train, one for the station, and one to go to HQ. So if two trains were involved, the number jumped to six. I never took an order for more than two trains.

  • Member since
    August 2005
  • From: At the Crossroads of the West
  • 10,720 posts
Posted by Deggesty on Tuesday, January 21, 2020 7:28 PM

The Southern RR train orders that I saw were made with two-sided carbon paper, and whether typed or handwritten were easy to read

Johnny

  • Member since
    June 2011
  • 800 posts
Posted by NP Eddie on Tuesday, January 21, 2020 5:23 PM

This in an addition to train order hoops.

Coon Creek is in Coon Rapids, MN and is where the present day BNSF Hinckley Sub ends and enters the BNSF Staples Sub. There was an interlocking at that point because the GN entered the joint NP-GN line from Northtown to St. Paul, MN.

After the 1970 merger, Marcella was the night operator at Coon Creek and placed train orders in the delivering machine for the crew to pickup. The delivering hoops snapped back and Marcella had a broken arm. I never did know severe her injury was.

Ed Burns

  • Member since
    May 2003
  • From: US
  • 18,130 posts
Posted by BaltACD on Tuesday, January 21, 2020 3:58 PM

Paul of Covington
   While I was making my last entry, Balt made his, saying that they did not use Teletype equipment.   Oh well.  

When Chessie installed their Terminal Service Center concept in about 20 terminals throughout the system, during the late 70's & early 80's - Kleinschmidt was the computer OEM for the computers we used, using Computer Automation boxes as the starting point with 'mini' computers (mini in the sense that they would only fill a closet, not a entire room).

Kleinschmidt, I believe, was the original holder of the Patents on most forms of teletype.  My boss and I would visit their plant, which was two buildings of about 200 feet by 600 feet each and seemed to have every form of metal working machienry ever made to construct teletypes - all idle - as the industry had moved away from teletypes and the company was in litigation with the US Government about responsibilities for prior contracts.  Instead of having about 2K employees using all the machinery, they employeed about 50, doing the computer work for Chessie System and also doing 3rd party rail logistics tracking for a variety of customers using AAR supplied car movement data.

https://www.kleinschmidt.com/ks/company

  • Member since
    March 2003
  • From: Central Iowa
  • 5,230 posts
Posted by jeffhergert on Tuesday, January 21, 2020 3:55 PM

Once typewriters became available, at first they were not to be used for train orders. TOs had to be hand written.  Eventually, speed restriction orders were allowed to be typed, all others still needed to be hand written.  Then the restriction was lifted and all train orders could be either typed or hand written.  You have to check the various rule books to see when the changes happened.  And I think most, if not all, had a requirement that typewritten orders had to all-cap 

One operator I knew always made an extra copy because the typewriter tended to tear holes in the train order paper.  They were nicknamed "flimsies" for a reason.  Why I didn't "rescue" from the waste basket more of those extra copies I don't know.

I suppose it depends where you are, but "board" for certain fixed signals is still used.  Most of us call the temporary flags for restrictions "boards" - as in "yellow board 40" for one for a 40 mph slow.  Or even "40 board."  In the distant past, I've heard board, used for block signal indications, as in "green (or clear) board" when viewing a green/clear block signal.

Jeff  

 

 

 

  • Member since
    July 2010
  • From: Louisiana
  • 1,963 posts
Posted by Paul of Covington on Tuesday, January 21, 2020 2:34 PM

   While I was making my last entry, Balt made his, saying that they did not use Teletype equipment.   Oh well.   

_____________

   "A stranger is just a friend you ain't met yet."  ___ Dave Gardner

  • Member since
    July 2010
  • From: Louisiana
  • 1,963 posts
Posted by Paul of Covington on Tuesday, January 21, 2020 2:27 PM

   Besides the legibility issue, there was also the slow transmission rates to consider.  The early equipment at each end was mechanical and limited in speed, so to maximize throughput, the character size was kept short.  Five-bit and six-bit characters were common.   Six bits gave you 64 characters, so you had room for the alphabet, numerals and special characters but little else.  Five bits gave you just 32 characters, so you had room for the alphabet and little else, but one of those characters was a shift character that made the same codes into numerics and most special characters.  If you had a lot of text, the five-bit mode was faster.  Expanding the character size to include lower case would have been a luxury that would have slowed them down.

  Thinking about those Teletype model 35 printers, I remember watching them in amazement as all the levers and dogs and sissor mechanisms flew back and forth.  And they were a nightmare to work on.

_____________

   "A stranger is just a friend you ain't met yet."  ___ Dave Gardner

  • Member since
    May 2003
  • From: US
  • 18,130 posts
Posted by BaltACD on Tuesday, January 21, 2020 2:15 PM

Overmod
 
Euclid
What I want to know is whether typing train orders in all caps meant that they were yelling at the crew. 

The convention of all caps being perceived as 'shouting' in the current Internet sense wouldn't apply to train orders.  I suspect that orders received by any early Teletype system (which only rendered capital letters) would be perceived only as 'mechanically clear' as opposed to our modern view, and it seems to me that reading orders in 'all caps' is clearer, especially in low cab light at night, than resolving mixed upper and lower case; it might be easier too to read the 'lowest' of multiple copies where the print might be considerably muddied in fine detail.

I would defer, of course, to actual operators given different instructions on how to type up orders.

It is possible that the European convention of expressing emphasis -- which is increased space between letters and words -- might be applied to text in all caps to signify particular importance.  It would certainly be less ambiguous.

I believe the B&O Rule Book specified that typed orders had to be in ALL CAPS.  

Most Agent/Operator positions had 'billing machines'  typewriters the only typed in CAPS.  Secondly, as I recall, one could only get legible copies of orders up to 13 - no matter if hand written (and you had to be heavy handed with the stylus) or typed.  The carbons used in copying train orders (at least on the B&O) were double faced.  The normal way of setting up train order pad - for copying 3 copies was 'order-carbon-order-carbon-order'; to copy 4 or more the pad would be set up 'order-carbon-order-order-carbon-order'.  The offices I worked would have two 19 pads set up; one for 3 and one for 4 or more.  When hand writing orders a sheet of metal would be inserted in the pad after the number of copies needed.  When typed, the platten became the backing for the writing surface.

The primary reason for having all caps is that lower case letters when typed through enough orders and carbons to create 13 copies - the lower case letter begin to look like black dots after about the 5th copy.

When I was working as a operator - operators positions did not have teletype or copying equipment.  In later years, when I was in Management - operators did use copying equipment to replicate the 'daily bulletin orders' that were to be given to every train in a particular position.  The Dispatcher, when transmitting these orders would tell the operator '19 Copy a Bunch' or words to a similar effect.  Without copying equipment, when a Operator had to recopy a bulletin type order, he had to read the recopied order back to the Train Dispatcher, who would underline the repeat in his train order book on the original order.

Train Dispatchers DID NOT have teletype equipment to transmit Train Orders.

  • Member since
    August 2005
  • From: At the Crossroads of the West
  • 10,720 posts
Posted by Deggesty on Tuesday, January 21, 2020 2:11 PM

Overmod

 

 
tree68
Back in the day operators usually had pretty good cursive handwriting (truly beautiful, in some cases), but capitalization was arbitrary no matter if they were copying off the telegraph or off the phone.

 

Some of you may find this discussion of interest regarding clear 'hands' for civil service, telegraphy, and then preparation of library reference cards, all of which involve an intersection between speed and legibility.

 

I have seen examples of 18th century handwriting in the court records of Lancaster County, Virginia. I cannot read the record.

Johnny

  • Member since
    August 2005
  • From: At the Crossroads of the West
  • 10,720 posts
Posted by Deggesty on Tuesday, January 21, 2020 1:58 PM

Lithonia Operator

I guess I have been operating under a misunderstanding for decades. I alwas thought that if an engineer says, "We caught a red board in Podunkville," that simply means he encountered a red signal, and had to stop. Any red signal; Block signal light, block signal semaphore, CTC, interlocking, whatever.

Apparently I have been wrong for a very long time.

 

I have long understood that lineside signals were referred to as "block" signals; thus, a red lineside signal was called a "red block" while only train order signals were called "boards." I was corrected by a conductor after speaking of a "red board" when I referred to a lineside dignal.

Johnny

  • Member since
    September 2003
  • 10,536 posts
Posted by Overmod on Tuesday, January 21, 2020 12:56 PM

tree68
Back in the day operators usually had pretty good cursive handwriting (truly beautiful, in some cases), but capitalization was arbitrary no matter if they were copying off the telegraph or off the phone.

Some of you may find this discussion of interest regarding clear 'hands' for civil service, telegraphy, and then preparation of library reference cards, all of which involve an intersection between speed and legibility.

  • Member since
    May 2019
  • 859 posts
Posted by Lithonia Operator on Tuesday, January 21, 2020 12:46 PM

I believe Overmod is correct. Caps are simply easier to read, and made a clearer image on the carbon copies near the bottom of the stack.

My admittedly very fuzzy memory seems to recall that everything was in caps. I don't think we used lower case for anything. I don't remember ever using the shift key. I think it's possible that the typewriter didn't even have lower case, but maybe that's unlikely. But maybe somehow lower case was locked out mechanically. The typewriter was a big heavy beast, and took serious finger strokes to get good copies.

I regret that I never once made an extra copy for a souvenir. 

  • Member since
    December 2001
  • From: Northern New York
  • 20,461 posts
Posted by tree68 on Tuesday, January 21, 2020 12:41 PM

Remember, too, that Morse code has no "case."  

When handling radio traffic, be it on nets or in radio messages, I slash the diagonal line on Z and 7 so as not to confuse them with 2 and 1.  And I print, usually in all caps.

Back in the day operators usually had pretty good cursive handwriting (truly beautiful, in some cases), but capitalization was arbitrary no matter if they were copying off the telegraph or off the phone.

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Search the Community

Newsletter Sign-Up

By signing up you may also receive occasional reader surveys and special offers from Trains magazine.Please view our privacy policy