Would a railroad use the same number for a locomotive as for a piece of rolling stock or did every piece of equipment on the roster need to have a unique number. I just realized I duplicated a locomotive number in my boxcar fleet that I just decaled and need to know if I should renumber the boxcars into a different series. Also, could freight car and passenger car have the same number.
I know the larger railroads used five and six digit numbers for their freight cars but some of the RTR rolling stock I have of smaller roads use three and four digit numbers like one would find on a loco.
It would be poor practice to use a car number the same as a locomotive number, although I suspect there were plenty of passenger car numbers that technically matched locomotive numbers as there are few opportunities to actually confuse cars with engines in reports.
On a small-prototype road that doesn't explicitly have to conform to UMLER I'd see little problem in having a boxcar, a passenger car, and a locomotive all with the same number. Just be sure your "internal procedures" wouldn't allow confusion of one with another in any operational order or report.
The Monon had boxcars in a series starting with number 1. CSX numbered their first AC locomotives starting with number 1.
Why would a small prototype road want to duplicate numbers on different types of equipment when there are so many numbers available for a small amount of equipment?
Mark Vinski
Yes, railroads would have both cars and engines the same number series, up until the railroads computerized and AEI (automatic equipment identification) was implemented (1980's, 1990's). After that no, every piece of equipment would have a unique initial and number.
The MKT (not computerized) merged with the MP (computerized) and the MKT had a series of 2 bay covered hoppers in the same series as some engines. It drove the AEI batty until they renumbered the cars. Engine MKT 100 would go by a reader in Kansas and the computer would think it was there, then MKT 100 would go by a reader in Texas and the computer would move MKT 100 to Texas, then the engine would go by another reader in Kansas and the computer would move it back there. Etc., etc.
Dave H. Painted side goes up. My website : wnbranch.com
In my opinion, if you're making the effort to number things, why not do it accurately, with no duplicates?
I have several hundred freight cars and over a hundred pieces of passenger equipment, plus lots of locomotives of all types, and while many of them have been sold or given to friends, there is not a duplicate number in the lot.
Many of those freight cars represent those of real railroads, and I've attempted to use numbers specific not only to the cars on which they appear, but also to the numbers that were appropriate in my late '30s modelling era. If I altered my layout, f'rinstance, to the mid-'50s, many of those '30s-era cars would need to be renumbered, as were their prototypes.
Most of my home-road freight cars on the current layout fit into blocks of 99 four-digit numbers, with not all necessarily filled, so gondolas, whose numbers all begin with the numbers 4 and 5, theoretically would be 4501 to 4599. On the current layout, there are 32 in-service, with another 8 gone to other owners. Those 8 car numbers will not be re-used.
However, I have more than a hundred home-road boxcars in-service (some, as with most types, now belonging to others.
I began with numbers in the 7800-series, which generally covered steel boxcars of 40' and 50' lengths. When I backdated my layout, I began buying more cars representing earlier types - single sheathed and double sheathed wooden cars, along with shorter Dominion-Fowler 37' cars, so created some new number series to accommodate and differentiate them from those originally in-use.
While all of the 7800 series cars are gone elsewhere, I had also created, for much newer cars on a '70s/'80s layout, a 7700 series and 7900 series - all of those cars were sold, too, but when I decided to substantially alter some Athearn BlueBox 40' boxcars for my current layout, I placed them, too, in the 7700 series, taking great care to not re-use numbers that are on those modern, and long-gone cars.
When Accurail came out with their version of the Dominion-Fowler cars, I bought a bunch, and created the 1100 series for them, using random numbers to represent an order of 99 cars, even though I have only 10.
I intend to scratchbuild five 40' doubledoor automobile cars, which, I hope, will be board-by-board (styrene boards) construction, with exterior brass framing, scratchbuilt Hutchins ends, and Viking roofs. Because these cars will be rather unique, I've created the 1400 number series for them, modelling only the five of a supposed ten car order.
If you can live with duplicate numbers, you're certainly free to do so.
It, however, would drive me nuts....even though there are many who say that it's within walking distance.
Wayne
doctorwayne In my opinion, if you're making the effort to number things, why not do it accurately, with no duplicates?
In some cases, I have taken equipment lettered for other roads and rebranded them by removing the roadname and decaling them for my fictional road, leaving the original number in place. I did that with a Mikado, leaving the 1053 road number on the cab. I used the 1000 series to number a set of undecorated boxcars I was lettering for my road. I have about 15 such boxcars and when I created the decal sheet, I just picked numbers at random from the 1000-1099 series and happened to pick 1053 which I applied to one of the cars. I realized the duplication after completing the decaling. Now I'm adding another Mikado to the roster and I wanted to put it in the same number series as the existing one.
Based on what a previous poster has written, it seems it would not be inaccurate to duplicate numbers between locos and rolling stock. As long as it is plausible that a prototype railroad would do it, I'm OK doing it on my road.
50+ years ago, the Minneapolis, Northfield and Southern Ry. had a locomotive 15 (Baldwin DRS 6-6-1500) and a bay-window caboose 015, that would sometimes run together on a train.
One of the reason large railroads today sometimes re-use the reporting marks of previous railroads they had bought or merged with decades ago was to avoid having duplicate entries. So Union Pacific might have a UP 123456, CGW 123456, MSTL 123456 etc. on the roster at the same time. (BTW these are not 'old' cars from the prior railroads retaining their original ID's, in the cases I'm aware of they're cars built for UP long after the railroad who's initials are being used had been taken over by UP.)
Interesting topic. The Santa Fe certainly had duplicate numbers for locomotives and passenger equipment. Examples that come to my mind are numbers 3000-3019, used for heavyweight chair cars (which ran into the 60s) and FM H16-44s; 2800 numbers used for lightweight chair cars and GP7s; 3100 numbers used for lightweight chair cars and class 3100 Mikados (which ran until 1954).
DrW Interesting topic. The Santa Fe certainly had duplicate numbers for locomotives and passenger equipment. Examples that come to my mind are numbers 3000-3019, used for heavyweight chair cars (which ran into the 60s) and FM H16-44s; 2800 numbers used for lightweight chair cars and GP7s; 3100 numbers used for lightweight chair cars and class 3100 Mikados (which ran until 1954).
This was what I was hoping was the case. That it would at least be plausible that different types of equipment could share the same number series. It will save me a lot of renumbering. Had I given it more thought from the start, I would have probably assigned unique series to each type of equipment, but not having done that, it's good to know what I am doing is not a stretch.
Keep in mind, it's really only an "issue" on equipment used in interchange service. If a railroad has an engine and a caboose and a snowplow that all have the same number, it's fine if they never go off the railroad's tracks. But cars that they interchange with other railroads have to have their own unique reporting marks (initials and number).
wjstix Keep in mind, it's really only an "issue" on equipment used in interchange service. If a railroad has an engine and a caboose and a snowplow that all have the same number, it's fine if they never go off the railroad's tracks. But cars that they interchange with other railroads have to have their own unique reporting marks (initials and number).
The rolling stock will all have unique numbers but some of those are going to be in the same series as the locos.
wjstixKeep in mind, it's really only an "issue" on equipment used in interchange service.
Its not even that, its only an issue if the railroad is computerized and using AEI. Human beings are smart enough to give out the difference between an engine and parlor car, computers not so much. Even if the ATSF had an engine and a passenger car both numbered 3015, and they were both off line, precomputer they never crossed paths in any record keeping. The engines were tracked by the power group for horsepower hours records and they didn't deal with passenger cars. The passenger car dept tracked the passenger cars for whatever records they had and didn't care about engines. Its only when computers started putting all the equipment in one inventory that it became a problem.
dehusman wjstix Keep in mind, it's really only an "issue" on equipment used in interchange service. Its not even that, its only an issue if the railroad is computerized and using AEI. Human beings are smart enough to give out the difference between an engine and parlor car, computers not so much. Even if the ATSF had an engine and a passenger car both numbered 3015, and they were both off line, precomputer they never crossed paths in any record keeping. The engines were tracked by the power group for horsepower hours records and they didn't deal with passenger cars. The passenger car dept tracked the passenger cars for whatever records they had and didn't care about engines. Its only when computers started putting all the equipment in one inventory that it became a problem.
wjstix Keep in mind, it's really only an "issue" on equipment used in interchange service.
Speaking as someone who programmed mainframes for a living, it shouldn't have been too hard to tell the difference between a car and a loco. You would simply need to have a code for the equipment type and add that to the number to create the identification key. Obviously if you are only using the number, you wouldn't be able to distinguish one from the other.
John-NYBWSpeaking as someone who programmed mainframes for a living, it shouldn't have been too hard to tell the difference between a car and a loco. You would simply need to have a code for the equipment type and add that to the number to create the identification key.
The problem being, of course, that the clerk or whoever puts the number into the computer might assign the wrong code to the entry if he or she doesn't get complete or correct information. For example, in the case of a report that 1111 has derailed and been set out at XXX for attention. There you sit at your terminal with that to go on, and you HAVE to put in a key, on some mainframes a key that can't subsequently be changed or erased. This is almost a poster child for the kind of situation we used to refer to as GIGO.
A similar bit of fun comes in with systems that require a certain number of digits in a field, or that pad out mandatory formats ambiguously. We had an example quoted a few months ago, where a field for locomotive number was something like 8 digits long (IIRC just for the 'locomotive number' part of the field structure). Whether this happily accepts a car-length number without flagging the operator is a whim of the programmers... it wouldn't have been a problem in the olden days of careful design and testing, but in the modern "Internet-style" world where bugs are addressed in production as they 'emerge' there could be some interesting consequences...
Making the numerical codes themselves unambiguous is a good way to get around this. Whether or not the internal system representation of a locomotive number is padded out or in hex or whatever, entering it as four digits with the code for 'motive power' would register it appropriately, and entering more than four digits would put up a flag for the operator to check. Similar for six-digit numbers for cars, perhaps with added checking by numerical sequence.
I believe Amtrak uses five-digit numbers for much of its stock precisely to establish that as a convention to help distinguish passenger from freight equipment...
John-NYBW Speaking as someone who programmed mainframes for a living, it shouldn't have been too hard to tell the difference between a car and a loco. You would simply need to have a code for the equipment type and add that to the number to create the identification key. Obviously if you are only using the number, you wouldn't be able to distinguish one from the other.
The computer systems have all sorts of detailed information on the equipment. The ID tags physically attached to them do not.
Chris van der Heide
My Algoma Central Railway Modeling Blog
cv_acr John-NYBW ... it shouldn't have been too hard to tell the difference between a car and a loco. You would simply need to have a code for the equipment type and add that to the number to create the identification key. Obviously if you are only using the number, you wouldn't be able to distinguish one from the other. The computer systems have all sorts of detailed information on the equipment. The ID tags physically attached to them do not.
John-NYBW ... it shouldn't have been too hard to tell the difference between a car and a loco. You would simply need to have a code for the equipment type and add that to the number to create the identification key. Obviously if you are only using the number, you wouldn't be able to distinguish one from the other.
... it shouldn't have been too hard to tell the difference between a car and a loco. You would simply need to have a code for the equipment type and add that to the number to create the identification key. Obviously if you are only using the number, you wouldn't be able to distinguish one from the other.
I think you're mistaking his point. On pre-UMLER computer systems there has to be a unique 'key' for a set of records, by which all the subsequent information is indexed. That wouldn't be something as ambiguous as the decimal number painted on the equipment; if you used that for the 'key' all the records for that 'number' would be run together indiscriminately in storage and confused in retrieval, with a result perhaps difficult to predict but never, ever good.
As he noted, the 'key' would need to include additional specific information -- but on older systems, where bytes were precious, it would be kept to a minimum, perhaps as he said a simple one- or two-character code for equipment type ... and that would have to be entered both when the original key was created and to link any subsequent record to what the computer is tracking.
Overmod cv_acr John-NYBW ... it shouldn't have been too hard to tell the difference between a car and a loco. You would simply need to have a code for the equipment type and add that to the number to create the identification key. Obviously if you are only using the number, you wouldn't be able to distinguish one from the other. The computer systems have all sorts of detailed information on the equipment. The ID tags physically attached to them do not. I think you're mistaking his point. On pre-UMLER computer systems there has to be a unique 'key' for a set of records, by which all the subsequent information is indexed. That wouldn't be something as ambiguous as the decimal number painted on the equipment; if you used that for the 'key' all the records for that 'number' would be run together indiscriminately in storage and confused in retrieval, with a result perhaps difficult to predict but never, ever good. As he noted, the 'key' would need to include additional specific information -- but on older systems, where bytes were precious, it would be kept to a minimum, perhaps as he said a simple one- or two-character code for equipment type ... and that would have to be entered both when the original key was created and to link any subsequent record to what the computer is tracking.
You could created a unique key as simply as assigning a one letter code for each type of equipment. For example, L for locos, F for freight cars, and P for passenger cars. That could be a prefix so loco L1000, freight car F1000, and passenger car P1000 would all have a unique key.
I have really messed this up.
I keep track of all my equipment by number. So all the equipment in the Fleet Of Nonsense has a unique number, no matter what railroad it belongs to.
I have been very good at this.
With reporting marks... not so much.
I have two railroads with CMR initials and two with CN initials.
Also, I have two different initials for the ATLANTIC CENTRAL, AC and ACR, and two for the MANCHESTER CENTRAL, MC and MCR.
But... I have not duplicated any number!
-Kevin
Living the dream.
John-NYBWYou could create a unique key as simply as assigning a one letter code for each type of equipment. For example, L for locos, F for freight cars, and P for passenger cars. That could be a prefix so loco L1000, freight car F1000, and passenger car P1000 would all have a unique key.
The specific problem here, though, is that unlike mainframes with alpha terminals, there is no way to program ASCII characters into decoder data fields.
Now, I had long and at times very painful experience with medical-research data that had been 'compressed' into a 127-character record structure ... where only numeric data could be entered to signify information in a coded format. Where this got to be fun was that many of these indicated data source fields in which there were four separate ways to distinguish 'information unknown' (two were 'null' and decimal 'zero'; the other two being 8 and 9 for different not-present-in-the-record or data-analysis-not-conducted-so-not-reported) and you had to remember which of the positions represented actual data as opposed to metadata.
In the current case, the additional 'code' parts of the key would have to be placed in specific fields, as above, to be reasonably programmed by humans with 'senior moment' memory characteristics (which the OP and I probably share) and this would require the programming scheme to pad out anything with enough characters to permit the 'longest alternative number' and then place the numeric whatever-it-is code precisely in the next positions. (Note that this is a potential way to address the issue of 'multiple locomotives with the same number' on a club layout or during visiting sessions: you put the 'regular' number in for your base engine, something like <number> and then however many zeros it takes to get to six digits (to keep cars with special-effects decoders out of confusion) and then a one-or two-digit sequence number that reflects the working distinction between locomotives. This is probably about the most 'memorable' convention that would work with existing decoder fields that won't take leading zeroes (you New Haven aficionados will have to decide how to deal) and cover all the bases.
Now, wouldn't it be nice if you had some equivalent of the Alt-xxx keyboard convention that could place an eight-bit ASCII value in a decoder CV spot? Then all you'd need would be some complecticated table cheat-sheet reminding you what could and couldn't be stuck in already complecticated CV structures...
... or some adjunct to decoder programming that uses simple keyboard codes for input and translates them appropriately, as computer systems have done for well over half a century...
Speaking of decades-old, no one seems to have mentioned JMRI in this context. Shouldn't it be possible for him to boot up some version of that, connect his decoder to appropriate "track" or wiring terminals, and run the necessary routines or whatever they call them to human-translate all the CVs sensibly, like the problem in the other thread?
SeeYou190With reporting marks... not so much. I have two railroads with CMR initials and two with CN initials.
When the Conrail split-up (divestiture) was in effect, some of the equipment going to CSX was given reporting marks NYC and likewise equipment going to NS was given PRR reporting marks. So thirty years after the NYC and PRR reporting marks became "obsolete" in the Penn Central merger they were now resurrected again.
So in your case you may claim to have a lease agreement or perhaps "shared assets" as Conrail has with some CSX/NS terminal areas or simply a corporate takeover where the reporting marks of the "new" owner were applied.
Don't sweat it Ed
gmpullmanDon't sweat it
It is too late to sweat it now!
I really doubt any visitor to my layout would ever pick up that out of 150+ custom painted freight cars that a couple of them share the same reporting initials.
I will be the only person ever bothered by this.
A lot of my freight cars are from Accurail kits and and in a few cases that means I have duplicate numbers. In some cases I could fudge part of the number, for example scraping away part of an 8 to make it a 6 but that wasn't always possible. I've developed an operating scheme that doesn't require unique numbers. My railroad is largely a bridge carrier so much of the traffic simply passes from the east end staging yard to the west end. Each staging yard represents 3 seperate locations. I have freights from each of the 3 east end locations. The last 6 cars in each are designated as bridge traffic. When they enter my classification yard they get shuffled into the 3 westbound freights. Again, these bridge cars are placed at the year of these trains. I can arbitrarily decide how many of the bridge cars go into each of the three westbound freights as long as they total 18. One day it might be 7-4-7, the next day 5-8-5. It works the same way going in the opposite direction. Because I am not tracking individual cars, it doesn't matter if there is a dupicate number within the cars designated as bridge traffic. This also allows me to put just about any kind of car I like among the bridge traffic without being concerned about having a modeled industry to accept them. For example I don't have a modeled industry to accept tank cars or covered hoppers but a lot of those appear among the bridge cars. I combine this system with a car-card system to track the cars that do come and go from the modeled industries. Car within the car-card system have to be unique. They can duplicate a number within the pool of cars assigned to the bridge traffic.
To facilitate the difference between loco and car numbers the Southern Railway instituted a check digit with each locomotive number.
http://southern.railfan.net/check.htm
Rick Jesionowski
Rule 1: This is my railroad.
Rule 2: I make the rules.
Rule 3: Illuminating discussion of prototype history, equipment and operating practices is always welcome, but in the event of visitor-perceived anacronisms, detail descrepancies or operating errors, consult RULE 1!
John-NYBWA lot of my freight cars are from Accurail kits and and in a few cases that means I have duplicate numbers. In some cases I could fudge part of the number, for example scraping away part of an 8 to make it a 6 but that wasn't always possible..
Accurail is one manufacturer that offers renumbering decals printed on car-color background available for a small fee.
I have used these every once in a while. I wish more manufacturers would follow suit.
Good Luck, Ed
dti406 To facilitate the difference between loco and car numbers the Southern Railway instituted a check digit with each locomotive number. http://southern.railfan.net/check.htm Rick Jesionowski
Not quite. Check digits verify you entered the number in correctly and didn't mis-type a number.
Overmod I think you're mistaking his point. On pre-UMLER computer systems there has to be a unique 'key' for a set of records, by which all the subsequent information is indexed.
I think you're mistaking his point. On pre-UMLER computer systems there has to be a unique 'key' for a set of records, by which all the subsequent information is indexed.
I'm not sure what point you think I'm missing. The 'key' is the reporting mark (or at least the 4-digit numeric code corresponding to the reporting mark, which is listed in the equipment register) and the car number. That's not "ambiguous". That combination should be unique.
The "tag" on the car just has that key in it. From the late 1960s to late 1970s the KarTrak ACI system used a plate with a coloured bar code to encode the owner and car number which would be read by a trackside optical scanner. The modern system uses electronic RFID tags.
I myself also work with computer software development and am quite familiar with data storage and retrieval.
Overmod The specific problem here, though, is that unlike mainframes with alpha terminals, there is no way to program ASCII characters into decoder data fields.
The specific problem was solved over 50 years ago with the encoding design of the KarTrak ACI labels.
1 - a four digit code corresponding to the owner/reporting mark
2 - the car number, padded with leading zeroes out to six digits (eg. 123 -> 000123)
If you wanted to include a car type in the encoding scheme someone, just translate that to a 0-9 digit instead of a letter.
gmpullman John-NYBW A lot of my freight cars are from Accurail kits and and in a few cases that means I have duplicate numbers. In some cases I could fudge part of the number, for example scraping away part of an 8 to make it a 6 but that wasn't always possible.. Accurail is one manufacturer that offers renumbering decals printed on car-color background available for a small fee. I have used these every once in a while. I wish more manufacturers would follow suit. Good Luck, Ed
John-NYBW A lot of my freight cars are from Accurail kits and and in a few cases that means I have duplicate numbers. In some cases I could fudge part of the number, for example scraping away part of an 8 to make it a 6 but that wasn't always possible..
Didn't know that. Good information. Thanks.
By the way, the AEI standard has a 14 digit car initial number field. There are Mexican and European cars and engines with more than the US standard 10 characters.