Operating the original Fenditavalat
The original layout was dismantled many years ago. Few photos were ever taken of it and none survive. The photos in this article have been restaged on the current version using the original stock as far as possible.
Fenditavalat as first built was a small terminus based upon a typical NG railway of local interest but it was operated using the principles outlined in the main articles. The model was a small terminus to fiddleyard layout typical of the layouts constructed by many modellers faced with space constraints. To save readers from having to access too many different pages on the website some material is repeated but is presented here for convenience.
The original layout simply had a loop with two sidings leaving it in the trailing direction and another one entering a dairy. The trailing sidings crossed the run round loop via a diamond to make use of the space in front of the fiddle yard. From memory, the capacity of the loop was four wagons and the yard held perhaps eight vehicles. The dairy siding acted as open staging in that the track ended at a gate giving access to the dairy and wagons were placed onto the siding by hand. The railway remained in this form for around fifteen years and was operated by a limited roster of rolling stock consisting of four locos and three railcars, a single carriage and around twenty wagons.
The "railway of local interest" nature of the original was personified by the passenger stock ~ all one of it! The model was based on a photo published in a book about European minor lines by Oakwell Press during the seventies. At £3.50 this seemed hideously expensive at the time but provided much of the inspiration for the early AFK. The coach was based on a Dutch prototype and I later learned that there should be access doors in the ends, although I did not appreciate this at the time. The coal fired outside stove was added at a later date but although I cannot now remember with certainty which railway inspired this modification, I think that it was the Portuguese NG. The bucolic nature of the line was inspired by WJK Davies account of a trip along one of the French NG lines which had to resort to asking passengers not to urinate from the train's balconies as the train wended along! Passengers casually sat on the steps as the AFK counterpart made its stately progress.
The old fellow on the left enjoys a quiet smoke whilst the man in the middle reads the local paper. The one on the right takes another swig from the beer bottle and will soon doubtlessly be urinating from the steps as the train rolls along! The figures are home made from styrene shaped with a soldering iron which accounts for their unique poses and their roughness. The original painted lettering of the right hand photo contrasts with that applied by pen many, many years later. As the AFK evolved 'practices' were tightened up and this remains the only coach where the passengers behave in such a cavalier manner.
Although the timings of the trains varied slightly over the years, depending upon which theoretical route was in favour at the time, the basic pattern remained the same. Fortunately this was recorded in an article in Narrow Lines during the early 1980s and is appended here. The CFS did not exist at this time and Fenditavalat was considered to be a relatively minor settlement at the head of a remote valley.
The timetable is quite typical of the prototype NG local interest lines in that it initially appears substantial. A close inspection reveals that it follows the basic three passenger trains each way with goods trains three days per week. There was an additional train on Wednesdays, which was market day at Lacono, the outward bound stock of which worked into Fenditavalat late on Tuesday evening. The steam engine working the goods trains towed the railcar each way to avoid a light engine working and save railcar fuel. It was also timed so that any shunting took place without interference from other trains. Looking at Thursday, the quietest day, this gives the following service.
The 07.56 railcar waits to depart for Lacono The driver will collect the fares before departing. As has been noted elsewhere with hindsight I suspect that the front railcar should have been a six wheeler but information was hard to come by when the model was built!
In practical terms an operating session on Thursday works out as: run the railcar out to the fiddle yard then run it back before returning to the fiddleyard yet again. Lift it off, and place loco, carriage and fourgon onto fiddleyard with perhaps one van in front of the carriage. Run this into town, drop van in yard, maybe pick up an open, run round train and depart, lift train off and put railcar back on for final run from fiddleyard. This might be prototypical but it is also very dull if this format is followed continuously.

The standard practice on most model railways is to operate one day's timetable but it was quickly realised that it would be better to operate the whole week's timetable to lengthen the session and to provide more interest. This also allowed the occasional trains that appeared in timetable, such as the 05.13 on market days, to run on the appropriate day rather than at every session, "because today is always Wednesday". For much of the layout's existence I was a teacher and I had a large number of sporting commitments which left little time for railway modelling. The timetable offered the option of running a short session, based around one day's events, such as the sequence described above, which was coherent and self contained. Thursday's session could be run through in the lunch break before returning to school for the afternoon, which helped to retain some sanity!
The 05.13 Wednesdays only train to the market at Lacono added a little spice to the basic programme. The cattle were loaded into the vans using the wheeled ramp visible in the photo. The loco body is the original, although smartened up, but the chassis was changed long ago. The loco is now the Jakarutu branch rack engine. The coach saw intensive service ~ it had to as it was the only one!
It did not take long to realise that each week could be made to progress throughout the year which would allow a meaningful representation of seasonal traffic. Forestry goods were more important in winter and dairy goods became more important during the summer due to transhumance. Each month was allocated five days and a specific date was taken from the calendar and given to each day. In the initial stages the traffic generator was a somewhat haphazard affair but eventually it was formalised into a table based upon the probabilities of traffic being offered. These chances were determined by rolling war gaming dice which are generally available in 4,6,8,10,12 and 20 sided formats.
This is a reconstructed traffic generator for Fenditavalat although the current version is derived from this original. The Marronĝaco spirit traffic was added later and the values here are probably set higher than the original.
A fairly typical starting point for the day's work on the original layout was two vans and an open wagon. There were only two short sidings in the original yard which restricted capacity. Another van is in the dairy (i.e. off the layout).
The original generator for Fenditavalat is no longer available but it forms the basis of the current generator. It has been retrograded as far as possible to illustrate the techniques originally used. The generator for incoming traffic has long since disappeared and the one illustrated was thrown together for the purposes of illustrating this article. The loadings on the Fenditavalat generator are set too high for the original timetable. Three goods trains with a four wagon limit and two mixeds restricted to two wagons each gives a maximum capacity of 16 wagons per week, without running an extra goods train. The current capacity is 16 per day (80 per week) excluding a dedicated cement train which runs every other day. The generator obviously needs to be balanced with the train loading capacity so that it does not overwhelm the service but still generates sufficient traffic to maintain interest.
The six dice have been thrown and the values read off the table to generate the loads recorded on the postcard. The original generator was simply a scruffy piece of paper but as the layout has grown there are now 20+ pages. The system is kept in a computer file to allow for easy alterations to the values as new traffic is added or alterations are made to existing flows in the light of experience. The hard copy is kept in a ring binder.
The original layout quite happily got by with scrappy paper notes similar to this one. This represents the traffic generated for Monday 6th October. It took less than five minutes to create. Note that an empty has been added to the incoming train to keep sufficient vehicles on hand to allow for future traffic demands.
The railcar was towed into Fenditavalat to save fuel costs on Mondays, Wednesdays and Fridays. The railcar and fourgon then returned to the junction leaving the loco plenty of time to assemble its train unhindered.
Once the day had been worked through the system could be 'reloaded' by throwing the dice for the following day's traffic. With such simple paperwork and a restricted amount of traffic this could be done in less than five minutes so that the system was virtually seamless. The only concern for the operator was to make sure that sufficient empty vehicles were on hand to meet anticipated demands and this was easily done by adding empty vans to incoming trains on an ad hoc basis.
The loco drops into the yard to collect two vans loaded with blankets and gloves. The Marronĝaco spirit store (the white building) was added to the second version of the layout but was not present on the original.
Having pulled out of the yard the loco retrieves a van from the dairy. The new version is slightly deeper than the old one.
The train is made up as per the instructions and is ready to leave. The first layout had a much shorter loop and could only accommodate four wagons without a fourgon. At the time I assumed that all the wagons were fitted with an automatic brake but later research revealed the fallacy of this assumption. Additional braking could have been applied from the guerite or vigié provided on the last wagon which amounted to little more than an elevated seat adjacent to a hand brake applied by the brakeman. The original layout's stock included a number of vehicles fitted with the brakehouse but the extra constructional time meant that it was omitted from many of the later batches. Yet another area where the model diverges from the prototype!
The system was simple to set up and worked well enough to sustain my interest in the layout during its life time. The basic features were incorporated into the new layout as is explained in Section 5. The only faint misgivings that I had about the original system was that the needs of Fenditavalat were considered in isolation and were probably skewing the operation of the 'total system'. Although Fenditavalat has maintained its importance on the current version of the AFK any problems, such as wagon shortages, now have to be considered in the context of the needs of other stations and, being at the end of a long single line Fenditavalat often has to react to problems caused elsewhere in the system. These problems occur quite naturally on a larger system and the AFK also uses probability tables to simulate events such as snowfall or late running mainline connections. Although these factors affect train running I am not too sure how they could convincingly be simulated upon a small layout without them becoming too contrived. Fenditavalat was a simple layout operated in a simple but satisfying manner.
Return to Section  2 (History of the model)
Return to operational articles index
Return to Section 5 (Load generation)