Steam Locomotive Development on the Ffestiniog Railway
P. Girdlestone, formerly Works Superintendent, Ffestiniog Railway, and now Project Engineer, Hugh Phillips Engineering Limited, describes in detail the conversion of FR 2-4-0 locomotive "Linda" to the Gas Producer Combustion System of coal firing
Ffestiniog Railway 2-4-0 saddle tank "Linda" ex-works at Boston Lodge in March 1985 at the commencements of tuning-up trials. Left to right: Driver Bell, Fireman Brooks, the author, and, leaning on the cabside, Jo Clulow, "coal bunker maker extraordinaire", and driver of the first successful working to Tan-y-Bwlch. © G. Rushton
Details of the rebuilding of Ffestiniog Railway Hunslet 2-4-0 Linda were first given in the RM in August 1985 (page 365), and this article is an endeavour to cover the reasons behind the rebuild, a description of its main features, and the story of the introduction of the engine into service and its first year in traffic.
Because of the forestry fire risk, all FR engines were converted to burn oil in the early 1970s, a move which brought considerable advantages in availability, power output, and reduction in labour requirements. These advantages meant that any reversion to coal firing would be difficult to justify, even if it was cheaper, unless it enabled operation within the intensive service that the FR worked without needing any extra preparation, turnround or disposal time. The severity of the requirements was further compounded by the fact that train loadings and speeds had significantly increased since coal firing was last the norm, and FR single (as opposed to double Fairlie) engines are worked very hard in relation to their size. The writer has always had a deep interest in steam locomotive design and this interest naturally encompassed the work of L. D. Porta in Argentina and subsequently that of D. Wardale as news became available of events in South Africa. The development work carried out by these two was seen to have particular relevance to FR conditions, and despite the potential problems, it had become obvious to the writer that a cheaper alternative to 100% oil burning was possible.
With coal showing prominently in the tender, "Linda" leaves Tan-y-Grisiau with the 14.25 train from Blaenau Ffestiniog to Porthmadog Harbour on Bank Holiday Monday, May 16, 1986. © L. Goddard
In 1983 the writer presented a report entitled 'Fuel and Economy Options for FR Locomotives'. The main points of this were:
Chief Executive Dick Wollan agreed immediately with the course of action suggested, and took the report to the FR Board which endorsed his decision. Allan Garraway, General Manager, also gave it his wholehearted support. Unfortunately, these two effectively departed the scene very soon afterwards, but the Board decision meant that the project went ahead and the new broom did not sweep it away with the austere financial circumstances that it swept in.
While all the deliberations had been going on, design work had been proceeding for some time, a balance having been struck between essential and desirable modifications. These would enable the incorporation of the GPCS and also make some improvement to a thermodynamically primitive engine. The steam locomotive is a highly integrated machine and each modification has a ripple effect throughout its structure and needs careful study and incorporation - the change one thing at a time mentality has little place except for those with poor understanding of the subject. Linda was chosen for the conversion because it was due for a major overhaul, and also because of all FR locomotives it presented the fewest problems - although these were still numerous.
The principal modifications made and the reasons for them are:
Gas Producer Combustion System (GPCS)
The most serious waste of fuel in the coal fired steam locomotive is coal particles carried over unburnt or partially burnt at high power outputs because of the destructive effect of the airflow through the firebed. The GPCS reduces this loss by gasifying the coal and burning the gas produced in secondary air admitted over the firebed. Airflow through the fire is therefore reduced to about 30-50% of the total needed in normal combustion at the same rate. The modifications necessary were a lowered grate to allow the correct depth of firebed for optimum gasification, pinhole rocking rate segments of 10% airspace, blanking plates at the firebox wall, and secondary air tubes passing through the firebox walls.
The restricted airspace grate serves to meter the primary air flow and helps avoid disturbance of the firebed, while the sealing plates at the firebox walls avoid channelling. A portion of the exhaust steam is piped to the ashpan, where it mixes with the primary air. This cools the firebed below the ash fusion temperature and prevents the formation of a clinker. Additionally, it reacts chemically with the coal, producing gas which is burnt in the secondary air. The fraction of heat liberated in the combustion zone is therefore increased and so the amount of secondary air which can be admitted. The firehole door is also used to deflect air into the main combustion gas stream which is directed by an extended arch.
Spark arrestor
Although coal particle carryover is much reduced, it still exists of course, and so a spark arrestor is an essential fitting, especially when running through forestry. A modified Master Mechanics self-cleaning type was fitted.
Exhaust system
The use of the GPCS and a spark arrestor entail a heavy draught loss, and so a very efficient exhaust system (the function of which is to pump combustion gas through the boiler) was essential. Of all systems for which there is test data available those developed by Porta, the Kylpor and Lempor, are the most efficient. The Lempor was chosen because of its simplicity and it made the locomotive immediately recognisable with its very distinctive shape. A specially shaped exhaust manifold was also fitted.
"Linda" at Porthmadog Harbour Station. Note the pipes carrying exhaust steam to the ashpan along the saddle tank. © N. Gurley |
 |
Other modifications
There were many minor changes. Of these the most important was an increase in the superheater element bore diameter to reduce pressure drop from regulator to steamchest. The snifting valves were also removed and crews instructed to drift the locomotive in mid gear. Drifting steam in the form of lubrication atomiser steam is supplied and the positive pressure generated in the steamchest helps to avoid the formation of carbon caused by hot smokebox gases and oxidisation by cold air. The atomisation steam was controlled by the draincock lever.
The locomotive entered the works in November 1983, but because of various problems did not emerge until March 1985. However, this was a convenient time of year as any service trains were at weekends, and if running through possessions could be negotiated with the Permanent Way Engineer, then testing could take place on weekdays without the trauma of the timetable to worry about. So on March 5, 1985, the first FR coal burner for thirteen years was steamed at Boston Lodge Works and took its first tentative trip on the main line. Tests quickly - went from light engine to freight train working, the first really successful trip being an engineering working to Tan-y-Bwlch. Weekend passenger train work was then undertaken, initially with diesel assistance to ensure the timetable was kept to in case of problems. These first trips enabled various defects to he quickly pinpointed and the working of the test instrumentation to he checked. Modifications found necessary were:
By the time these, and other minor changes, had been made the locomotive was capable of hauling the normal maximum timing load for this class of engine of eight coaches, and after Easter entered regular service. Although the pressure of service eventually precluded modification, testing and minor change carried on until the middle of May. Latterly this was more concerned with component performance as regular service took its toll. However, most items stood up well to their arduous conditions of service. The spark arrestor performance was naturally carefully monitored and two instances of fire, one of which was definitely attributable to the locomotive, necessitated a change in design. This principally involved reducing the mesh opening area.
After this, during a period of very dry weather when a succession of very heavy trains was worked, no more trouble was experienced. Around this period, the free area through the grate was also reduced both generally and locally to help avoid the very destructive effect of primary airflow. It should be noted though, that conditions were very abnormal, in that at the heaviest rates of working in normal service firing rates of up to 140 pounds per square foot of grate per hour were necessary because of the very small grate area, coupled with smokebox vacuums of 8-9 inches w.g. Even at these rates though, coal particle carryover was minimal compared with what would be expected with a conventionally fired locomotive.
Because it had been so long since the FR had regularly used coal-burning locomotives it was necessary to start from scratch with crew training, especially because the GPCS demands rather different firing techniques. The enthusiasm brought to the job after the writer reluctantly relinquished his regular place on the engine after about a month to get back to more mundane work was remarkable, and very gratifying. Seamus Rogers regularly drove the engine throughout the season and kept it in beautiful condition - despite our occasional arguments about overtime and the colour of nameplates! The regular firemen, who trained many others, were Graham Byrom, Terry Turner, and from Switzerland Louis Bergner. To these few is owed a lot in ensuring the reliable performance of the locomotive throughout the period of test and change and before a sufficient pool of crews became available. Of the latter, even those who initially doubted their ability often became very good - even desk-bound accountants who found the going hard after oil firing!
Because of operating requirements and the fact that it was used indiscriminately with oil-fired units, no concession could be made to the fact that the locomotive was coal fired. In fact, this was not necessary, and it is illustrative of the benefits of a modern combustion system to show how well the engine blended in with its oil-fired counterparts. Lighting up was achieved by using compressed air to operate the blower in the same amount of preparation time, and turn-round times were not affected as there was no clinker whilst ash could be carried in the firebed, usually needing removal only if the engine had to work a third trip. Depending on the quality of the coal and rate of working, the spark arrestor mesh sometimes needed brushing down to remove char plugging.
Further, after muscles had developed, coaling could be achieved in the same or less time than oil fuelling, despite no special facilities initially being provided. Combustion was good, and smokeless operation the norm in common with all gas producer locomotives, although smoke can be provided to order. Tube cleaning was only carried out at washout when the arrestor screens were removed but was found to be only really necessary to check for the odd tube blockage. This was in stark contrast to the oil-fired engines whose tubes need cleaning virtually every day by brush and by sand. Towards the end of the 1985 high summer service the locomotive was worked to the most intensive roster, running 500 miles per week, hauling heavy trains sometimes made up to the equivalent of nine bogie coaches. These loads needed skilful operation, but their haulage was achieved burning low grade coal.
 |
"Linda" arriving smokelessly at Blaenau Ffestiniog station in October 1985. © N. Gurley |
As to the reason behind the rebuild, that of economy, it is pleasing to record that economies were made. By August, at the prices then current, running costs were 73% of those that would have been incurred burning oil, compared to the 70% predicted two years before. However the other advantage for the FR in possessing coal burning power was the fact that it released amounts of very cheap waste oil that could be used in higher percentages on other engines, thus reducing the oil-fired costs as well. The FR has never been able to collect enough waste oil to satisfy more than about 30% of its needs and so for the cheapest running costs would probably operate a mix of coal and oil fired units.
It remains at this point to see what problems were experienced in service, and what direction further development should have taken. The principal problem experienced was connected with the characteristics of the coal used, and this came to light during the summer when some steaming problems began to be experienced. It was decided to use a washed singles coal because the sizing was suitable and also because it would have a low ash content, the latter being important because of a very small ash carrying capacity (a problem incidentally with much more modern engines as well). It also had one of the lowest ash fusion temperatures of any British coal, which is a measure of its clinkering tendency.
After some investigation into the problem it was found that the swelling tendency of the seam being mined had risen and this caused the fuel to agglomerate into larger lumps which had a detrimental effect on the gasification process which requires a deep firebed of small coal. The attitude of the National Coal Board was most helpful and another batch was supplied to replace that causing problems. Another colliery was also identified where the characteristics were very consistent and it was planned to try this next. In larger locomotives these minor variations in quality would not be important and help to show the problems experienced with both liquid and solid fuel in locomotives with fireboxes too small for the power outputs demanded of them. Indeed, it became obvious to anyone who saw the firebed completely fluidised at high power outputs before the grate free area was reduced from the original figure, that coal fired conventionally at these outputs would have been a disaster.
Further developments would have included a feedwater heater. Despite its obvious benefits this was initially rejected from the original proposals for the reason of timescale. These benefits would include reduction of the high grate loading, increase in boiler efficiency, and would leave the engine less susceptible to minor changes it coal quality and firing technique. It would also produce further economy of 5% overall. Superheater performance was improved with the GPCS but further improvement was necessary to bring the temperatures at least up to modern conventional maxima.
It was clear to the writer that the success obtained with Linda was capable of being exploited further - indeed Linda could be regarded as a pilot project producing the groundwork for more extensive schemes First thoughts were focused on the possibility of utilising the power bogie from South African Railways two foot gauge Garratts to produce a modern conventional form locomotive of exceptional power and economy, and some investigation was carried out into acquiring such units at minimal cost. It was clear that fuel costs could be halved and maintenance costs drastically reduced by adopting modern techniques and design so painstakingly researched and tested over the last thirty years. The costs of so doing should be related to the reduction of repetitive and time consuming maintenance and fuel costs - Linda's modifications virtually paid for themselves in the first year of operation. However, it became clear that the FR saw the answer to its problems in part dieselisation, and other influences were allowed to prejudice the chance to tackle the problems of the performance of its steam locomotives positively, and the writer left FR service towards the end of 1985.
For a short period at the end of 1985 Linda was reconverted to burn oil because the fuel budget for that year was almost spent and a quantity of waste oil remained in stock. Therefore, for that reason (and also the desire to "see how it goes"), the oil firing equipment was re-installed (for the ease of which the conversion was specifically designed). For the start of the 1986 daily train service the locomotive was back in service burning coal, and after various faults separate from the combustion system were attended to, gave some good accounts of itself burning coal from Lea Hall Colliery which was the fuel selected to replace that originally chosen. Apart from a sheet metal tube to cover the distinctive shape of the Lempor chimney, and increased coal space to allow operation for three trips without refuelling, the engine remained in much the same condition as originally rebuilt.
Unfortunately, because of the drastic (and undoubtedly temporary) reduction in oil prices, and for other reasons that are, perhaps, less clear, Linda was due to have reverted to oil firing in July of this year.
The writer would like to record his pleasure in his involvement in this project, that, through technical innovation, gave fresh business to the NCB through the replacement of oil by coal, and also helped to show that the ultimate potential of the Stephensonian steam locomotive is far from fully realised whatever its physical size. It is to be hoped that other administrations will prove more receptive where circumstances dictate conditions to be favourable towards its development.
Ffestiniog Railway 2-4-0 "Linda" at Porthmadog Harbour Station on April 20, 1986, awaiting its next turn of duty. © P. Girdlestone
It remains only for the writer to make note of the help and advice given by his good friends Dante Porta and David Wardale, his appreciation of those who gave their support, and his thanks to Paul Dukes, Mechanical Engineer, Ffestiniog Railway, for permission to quote various statistical information in this article.