Return to index page

Gas Works, Coke and Smokeless Fuels

Prior to the development of natural gas supplies in the late 1960's and early 1970's gas was made from coal in a local gas works. The town gas works is a popular adjunct to model railways set in any period up to the 1970's, they were built in sizes to suit almost any size of layout and generated an interesting and regular traffic for the line. In the 1930s there were over a thousand gas works in the country, this figure had fallen slightly by the time they were nationalised in 1949 but then dropped dramatically with only about 500 still in operation by the later 1950s. Originally they used coal to make the gas, but in the later 1950s and through the 1960s the cheap cost of oil meant that many switched to using oil to make 'synthetic town gas', becoming what were known as 'oil reforming plants'. This is worth considering as it gets round the problem of loaded and empty coal and coke wagons (if that is a problem for you). Most coal gas plants closed in the later 1960s, some coal based works and the oil plants continued into the 1970s when North Sea Gas took over.

Coal consists of carbon mixed with a range of impurities, if it is heated in a sealed container with no air (pyrolysis) the various impurities in the coal are driven off as a thick dirty smoke to leave almost pure carbon in the form of coke. The smoke was for many years merely an undesirable by-product but over time people realised it had potential uses. If the smoke from the retort is allowed to cool tars and oils are condensed out, leaving a fairly clear gas which can be used as a fuel, although a very smelly one.

This process was developed by a William Murdock in the 1790's, with additional filtering to reduce the smell and his 'coal gas' was soon developed for domestic and industrial use. He set up plants to light the entrance to the Manchester Police Commissioners premises in 1797, the exterior of the factory of Boulton and Watt in Birmingham and a large cotton mill in Salford in 1805. He obtained the gas from coke works (associated with steel works) and supplied it in cloth bags. The first large scale commercial operation, supplying gas from a purpose built 'gas works' to light Westminster bridge in London via wooden pipes, was set up in 1813 by the London and Westminster Gas Light and Coke Company. The law had to be changed to allow the roads to be dug up to lay underground pipes and with various other problems to solve it was the mid 19th century before the industry really took off. Development in the later 19th century was rapid, by the 1870s a 'night shift' was possible using factories illuminated by coal gas. Up to the time of the First World War the gas produced was often referred to as 'illuminating gas' as its main use was in lighting (as people already had a coal fire for heating they tended to have a coal fired 'range' for cooking). After the First World War the term 'town gas' became the norm. It was only in the mid 1920s that anyone set up a proper establishment for testing new designs of gas equipment (set up by the Gas Light and Coke Company next door to their Nine Elms Gas Works), allowing the establishment of proper standards for safety.

The tar and oils that were condensed out of the gas were initially a waste problem for the coke works but by the time the gas industry became significant in size people were already looking at these tars and oils to see if they contained anything of value. With the water content reduced (it was simply left in a pit to settle out) the tar was found to be useable as a protective 'paint' on buildings (rammed earth cottages had a black strip of tar around the base of their whitewashed walls where the rain might cause most damage). By the early nineteenth century someone had developed a method of producing 'roofing felt', essentially a woolen fabric coated in tar and then covered by sand. Roofing felt was popular in Germany and the USA throughout the nineteenth century but was less common in Britain other than on flat roofed buildings. The tar in the pit or well at the gas works could be pumped out, either into a header tank for loading railway wagons or just to a 'farm pump' type pump to be sold by the bucket full (you could still take a bucket along to buy the tar in the 1950s, possibly the 1960s). For a description showing how to model the hand pumps used see also 'Appendix X -Wargaming - Agricultural, Industrial and Domestic Clutter'. By the later 19th century specialist firms were operating to process the coal tar and recover the valuable by-products it contained.

The unrefined naphtha, a deep amber to dark red liquid, was often shipped out for further processing in drums from smaller gas works, or in tank wagons for larger establishments. Coal tar naphtha also contains some of the more valuable light distillates usually lumped together under the general term 'Benzol'. Coal tar distillers were already taking the tar from the coke works and reprocessing it to recover the valuable fractions, the residual pitch being used for waterproofing or road surfacing. About the start of the 20th century the gas works changed their process, using greater heat in the retorts, and this reduced the number of useable by-products in gas works tar (the stuff from the coke plants was rather better in this respect).



Some works were very small indeed, virtually every town had a gas works and not all towns were large. About half the gas works in the country employed 10 people or less, although the really large works could have up to 300 people working on site. At Seascale in Cumberland in 1939 the entire staff of the local gas works was one man, Mr. Lee, who also had to maintain the local gas meters and appliances. He crushed, bagged and carted the coke from the ovens for sale in the area and also sold barrels of unrefined coal tar. These small works continued in use right up to the change over to North Sea or Natural Gas, in fact the smaller establishments in remote towns were the last to go, outliving their larger more modern brethren. The last functioning coal gas works in Britain was at Muirkirk in Ayrshire which closed in 1977 when the town was connected to the new gas grid.

The small town gas works at Fakenham, about thirty kilometers North West of Norwich in northern Norfolk, is now a listed ancient monument. This works has now been used for the range of Hornby 'gas works' models in their Scaledale OO scale range. When it closed in 1965 the works was partially preserved as a museum, operated on a volunteer basis and normally only has open days on Thursdays during the summer months (May to early September). For further information on open days for the Fakenham Museum of Gas call: 01328 863 150 (Summer months and daytime only). There is a very nice little gas holder dating from the 1880's and a ground floor retort house dating from the 1840's (the building is almost as high as a two-storey house). There were at one time three gas holders on the site, the preserved example is the first and smallest of these.

In the Scottish Borders the 1839 Biggar Gas Works has been preserved as it was when it closed in 1973. The museum has a web site at http://www.biggarmuseumtrust.co.uk and the OS ref is NT 038377. There are regular bus services from Peebles and Lanark, the museum is open from about Easter to mid October, t the time of writing the opening hours are Monday to Saturday 11am to 4.30pm and Sunday 2pm to 4.30pm.

There is a large section of the Museum of Science and Industry in Manchester dealing with gas supplies.

The many small to medium sized establishments represent a viable option for inclusion on a model railway and the books by Peter Denny on his Buckingham Great Central layout contain valuable drawings and photographs of his models. Mr. Denny also wrote a very detailed article in (I think) July 1989 Railway Modeller Magazine on the design of small and medium sized Gas Works which you may find being sold at a bookshop or model railway exhibition. If you belong to a modelling club this may well be in the club library.



During the 1930s experiments were made with introducing hydrogen into the tar to produce something akin to fuel oil (the process was called 'gasification'), ICI built a 'petrol' plant to make the stuff in the mid 1930s and the Germans depended heavily on this 'ersatz' fuel during the Second World War. However this method of making petrol was expensive and was not otherwise exploited in the UK (see also Petroleum and LPG for details of the ICI works). More recently there has been a resurgence of interest in this process, but this is mainly based in the USA.

---

Overview of the gas production process

Taking a simple small rural gas works as an example we can follow the basic gas production process. The Hornby range now includes all the buildings required for a small gas works of this type (based on Fakenham), although these small works were often not rail connected and those that were generated little railway traffic.

Fig ___ Flow diagram for a small gas works

Incoming: Coal Iron sponge Outgoing: Coke, probably sold locally, delivered is small horse cart or small flat bed motor vehicle in sacks Tar (probably in barrels, possibly in tank wagons for tar distillation plants) Ammoniacal Liquor (only a possible, probably in barrels) Spent oxide (for sulphur recovery)

Briefly the coal is placed into the sealed retorts in the retort house, which (in a really small works) are heated by more coal or by burning some of the coke produced by the works. This drives off all the volatile components as thick oily smoke, leaving behind coke in the retorts. One retort at a time is then opened and the coke raked out into a two wheeled iron barrow, it is still hot and catches fire but it is immediately wheeled to a raised tank of water which is sprayed onto the coke to stop it burning, it is then wheeled to the coke pile for sale. The illustration upper left below is the preserved barrow at Fakenham, this has solid sides but some were made of riveted bars (on the right below) the figure gives the scale of the sketched barrow and rake.

Fig ___ Iron coke barrow and 'coke rake' for a small gas works


Meanwhile, inside the retort house, the oily smoke is passed into a 'hydraulic main' which is tube partly filled with water (which seals the ends of the feed pipes preventing air for reaching the retort). The smoke is then passed outside to a condenser to cool down, at this stage a lot of the volatile components condense into a dark amber liquid called 'liquor', which is drained off from the condensers into a tar pit or 'well'.

The gas leaving the condenser is pumped out by the 'exhauster' to maintain a flow through the system and then passes to the 'ammonia washer' where it is bubbled up through water which absorbs the ammonia in the gas. This water ends up as a fairly concentrated solution known as 'ammoniacal liquor' and is drained off, often into the same pit as the tar (where it floats on top of the tar).

The gas still has a lot of impurities in it however and the next stage is the 'purifier'. The purifier is required to remove the sulphur dioxide from the gas, in the early works they passed it over 'slaked lime', the resulting 'foul lime' could not be reprocessed and was dumped, where a lot of the sulphur dioxide escaped and produced a nasty rotten eggs smell. After about 1850 they gradually changed over to using a metal box containing planks covered in stuff called 'iron sponge'. The original iron sponge was also called 'bog ore', it was peat with a lot of iron oxide (rust) in it, this reacts with the hydrogen sulphide in the gas, turning it into iron sulphate and water (the water has to be drained off from these tanks at intervals). Later they changed to using wood chippings or wood shavings impregnated with the iron oxide. If air got into the purifier when it was in use it was likely to catch fire. There will be at least two purifier tanks even at the smallest works. At intervals one will be isolated using valves on the pipes and the heavy lids lifted off, the 'spent oxide' is then dug out and spread out inside a low walled, roofed enclosure where it reacts with the air, the iron oxide reforms and the sulphur crystalises out. This produced a rather unpleasant 'rotten eggs' smell as there was some hydrogen sulphide still in the mix. The 'regenerated' iron sponge was put back into the tanks until the proportion of sulphur reached about 40 percent, at which point it was processed to recover the sulphur (usually as suphuric acid). The sulphur rich spent oxide was shipped out (I believe in barrels) to have the sulphur recovered, meanwhile the empty purifier was re-filled with fresh 'iron sponge', the top is bolted back down and the valves are opened.

The gas emerging from the purifier is ready for sale, by this stage the gas is mostly just hydrogen but with some carbon dioxide and some carbon monoxide in it. The carbon monoxide is a poison, which is how people used to kill themselves by putting their head in the oven, the gas is still not as pure as it might be however and larger works subjected it to further processing.

The gas is then dried and metered before passing to the gas holder (the big tanks associated with gas works), from where it passes to the mains, sometimes with a booster pump to increase the supply pressure a bit.

The same process happens at a larger works although these would usually add a second stage of washing before the ammonia washer using a powered 'livsey washer', in which rapidly rotating brushes generate a mist which recovers more of the ammonia from the gas, again producing ammoniacal liquor. Also from the mid 1930s most larger gas works would pass the gas through a 'benzol' plant before passing it to the gas holders. In this plant it is bubbled up through petroleum oil to extract some of the very light 'fractions' in the gas which would be difficult to condense out (they do however dissolve in the oil and can then be recovered from it by distillation).

Fig ___ Typical small gas works

---

Gas Works Structures and Plant

At the gas works the sealed containers used to cook the coal are called retorts, originally made of metal they changed to moulded fire clay in the 1830's. In the smaller works the retorts and associated firing holes were all on one floor, in larger works the retorts were on an upper floor with the men stoking the fires on the ground floor below. This resulted in a taller building, although they could be surprisingly narrow. This is probably the best type for a model railway, large enough to be interesting but small enough to fit into a layout. The evolution of coal gas retorts is illustrated in Fig ___, note that the cross sections shown are generalised to show the basic changes, they are not accurately based on specific prototypes.

Fig ___ Gas retort buildings


Fig ___ (1) shows the early form of retort house, there were between five and seven retorts (A) arranged rather like a bank of torpedo tubes in a submarine, each about ten to twelve feet long. The coke used to heat the retorts was shovelled into the furnace (B), which gas works people called the 'producer'. Ash from the producer fell through the fire bars and collected underneath (C), the floor of the retort house was made up of removable plates (D) to allow the ash to be periodically shovelled out. The gas was drawn off from the retorts through pipes (E) which fed a single main pipe called the 'foul main' which lead out to the purification plant.

Once all the gas had been removed the hot coke from the retorts was dug out using long handled rakes and tipped into metal wheel barrows (called coke barrows). These were then quickly wheeled to a small brick or stone shed where they were sprayed with water to cool the coke down. An example is shown in Fig ___ (2), in this case there is a small water tank on top of the structure, fed through a small pipe to the right. The coke would then be crushed to break up the larger lumps, graded (typically probably using angled bars tapering outwards to give a rough small medium and large selection), stockpiled and sold. This type of retort remained in use in more rural areas right up to the end of coal gas production in the UK.
In the mid nineteenth century, as the size of retorts increased, mechanical stoking was introduced but this was not a great success. In the 1880's someone thought of building angled retorts, filled from the top and emptied from the bottom with the assistance of gravity. These were not very common as many works were already up and running but where capacity was increased they were adopted, Fig ___ (3) shows the general arrangement. The angled retorts (A) are supplied with coal from bunkers (F) in the upper part of the building. The coal is wheeled in from the stock pile in small trucks (B) which are then hoisted up the outside of the building. A separate 'charging stage' (C), also fed from the hoist, allowed the crew to stoke the 'producer' to heat the retorts. The ash from the producer fell through the fire bars to an ash pit (D) in the base of the building. The hot coke was emptied into small trucks (E) or onto a conveyor belt, quenched with cold water, graded for size and stockpiled.

The next important development in the later 19th century was the use of regenerative heating, in which some of the gas was used to heat the retorts. This was not terribly efficient and a more advanced form was developed in which some of the coke from the process was re-used to make 'producer gas' and then 'water gas' to fire the system. Later additions were an oil-spray to make 'carburrated water gas', the oil being supplemented by about 1900 with benzol vapour (recovered from the tar at larger works). For more on these gasses see also Appendix One - Fuels.

The demand for gas increased over time and the solution was to build bigger retorts, Fig ___ (4) shows a cross section of a retort house dating from about 1910. The retort in this case was made of metal and on the plans it is called a 'chamber oven' (A), the building was still called a retort house however. The coal is supplied by a conveyor system (B) from the stockpile and fed into a crusher (C). A gravity bucket hoist (D) lifts the coal into the upper part of the building where it is distributed to the storage bunkers (F) using small tipping trucks (E) or a conveyor system. Small tipper trucks feed the vertical producers (G) through the top and the ash is extracted at the bottom (H). When all the gas has been extracted the coke is emptied with the assistance of a powered ram (I) via a power operated door into a quenching trough (J) and from there it is fed into tipper trucks (K) to be hauled away, graded and stockpiled.

Fig ___ (5) shows a highly mechanised retort house developed after the First World War, the example shown dates from the 1930's. The retorts (A) are vertical and they are heated with carburrated water gas from an on-site plant. Unlike the older designs the retorts operate continuously with the coal passing down through them changing to coke by the time it reaches the bottom. Coal is supplied via lorry or railway wagons to the loading point (B). Here it is crushed by the machinery (C) then hoisted in a gravity bucket conveyor (D) to the bunkers (E) at the top of the building. The hot coke is mechanically extracted at the bottom of the retorts and is carried away on conveyor belts (F) to be quenched, sorted for size and stockpiled. Conveyor belts under the floor of the building (G) move the coal from one side to the other and feed the hoppers at the top.

Fig ___ (6) shows a coke grading plant which might be built onto the end of the retort houses shown in Fig ___ (3), (4) or (5). The coke is first quenched with cold water. A gravity bucket conveyor lifts the coke into a hopper (A) at the top of the building. The coke is then fed through a series of rotating screens (B) which grade it for size, emptying into the hoppers below. From the hoppers it can be bagged in a bagging plant (C) or dropped straight into road or railway vehicles (D). As a rough guide a hundred tons of coal would typically yield seventy tons of coke.

One element not shown in the above was the 'economiser' or 'exhaust gas boiler', used to generate steam using the exhaust gasses from the retort heating system. This adds a smaller chimney to the roof of the retort building itself.

I should also mention that at larger works they often added 'carburrated water gas' to the gas being fed into the mains to raise its calorific value. The process of making this gas produced a tar of its own, from the break down of the oil used, but this was produced in small quantities and a lot less toxic than the coal tar.

Oil-Gas Plant

In the later 1950s and on through the 1960s, the low price of oil lead many gas works to change over to an oil based production. This could use several lighter petroleum oils as well as LPG, reacting these with steam in the presence of a catalyst. The resulting gas did not smell so a smell was added to match (near enough) the smell of the coal produced town gas. The gas was passed from the retort into a small has holder, it then had the carbon dioxide level reduced using stuff called triethanolamine and (as with carburrated water gas) a quantity of tar was produced, which they washed out of the gas before feeding it into the main gas holders.
I believe that at some gas works they added these new system by erecting a new building to house them (enabling coal gas production to continue). I have not yet raced details of how the oil or LPG was supplied to these works, however for modelling purposes you can add a more modern building to your gas works and justify the arrival of train loads of tank wagons.
The remaining plant before the gas holders discussed below is associated with coal gas production.

Condensers

The simple early condenser was just a set of vertical iron pipes in the open air, the gas passed upwards through these and the tars condensed into a liquids, tapped off at the base and fed to the tar pit. The pipes had to be fairly small diameter to maximise the cooling surface and there would be access points to allow the pipes to be cleaned (the gas at this stage was still leaving oily deposits). There were many variations on this basic idea, when I built mine many years ago I only had a photo of a type with angled pipes, it was terribly difficult to model and proved much too fragile. The simple pairs of vertical pipes type (as used at the preserved Biggar works) is much easier to model. By the mid 1930s there was another small type in use, as seen at the Fakenham works, consisting of a tall rectangular metal box with bolted on ports on the side for cleaning. This is dead easy to model but doesn't look terribly impressive. At larger works they sometimes used a rather similar box type condenser but set up in banks rather than just individual units as at Fakenham. At some of the larger works the condensers were contained in a building with slatted walls, which makes modelling easy. The condensers at the large Stockport works looked a lot like ammonia washers, however they are actually 'water cooled tubular gas condensers'.
Some condensers were 'condenser-washers', these had cold water sprays to help condense the tars and also removed some of the ammonia. The resulting liquid was drained off into the tar pit (or tar tank) to join the ammoniacal liquor from the ammonia washers proper.

Exhauster

This pumped the gas out of the condensers and into the ammonia washers, I believe they were usually housed in a small building.

Livsey washer

This was a horizontal cylinder containing rotating brushes and half filled with water. A motor at one end rotated the brushes, which formed a mist of water droplets that absorbed any residual ammonia in the gas.

Ammonia Washers

From the 'purifiers' the gas was passed to the ammonia washer in which it was bubbled through water into which lime had been mixed to remove the ammonia and some more of the impurities (most of which were of no or limited value). The size of the ammonia washer was entirely dependent upon the size of the job it had to do. At a really small works it might be an iron tank perhaps four feet square by three feet high, the gas arriving and departing through eight inch iron pipes. At a medium size works it was more likely to be a couple of vertical cylinders perhaps XXX feet in diameter and XXX feet high with gas pipes perhaps a foot in diameter and at a large works it would be a very substantial structure with the gas passing through pipes perhaps eighteen inches in diameter.



Purifiers

As noted above at a smaller works these were simply large iron tanks containing planks of wood covered in 'iron sponge', the gas is passed into the top of container, and passes down through the sponge, the hydrogen sulphide reacting with the iron oxide to produce iron sulphide and water (the water has to be drained off occasionally). The oxide could be regenerated by exposing it to the air until the sulphur content reached a point where it was worth recovering.
Smaller works shipped out the 'spent oxide' to have the sulphur recovered however in a larger works they would do this on site. As I understand it to recover the sulphur (and 'regenerate' the iron oxide) at the larger works damp air was allowed into the container under controlled conditions, re-oxidising the iron and leaving the sulphur as crystals. The 'spent oxide was then roasted to produce sulphur dioxide which was converted to suphuric acid using the 'lead chamber' process (see also 'Lineside Industries - Chemicals and Plastics' for more on this).

At smaller works the tanks sat on the ground, they were typically about two feet high and made of iron. They all seem top have had a simple overhead gantry to lift the heavy lids from the tanks, or in some cases to shift the containers of 'spent oxide' to the regeneration area. In a larger installation the iron tanks containing the oxide were often mounted on a low raised platform. There would be several banks of filters and each in turn would be un-coupled from the gas lines and lowered to ground level.
I seem to remember being told at school that the acid could be recovered directly from the spent oxide by using damp air to regenerate the oxide. As I remember it, possibly wrongly, the damp air was fed in and the sulphur converted to sulphur dioxide which then reacted with the water to form a mild sulphuric acid. However as all this would be done inside a building the details are not that important in the present context.

In the 1930s several larger works switched to alternative methods of hydrogen sulphide removal, using liquids that would absorb the stuff. Various liquids were used, some could be regenerated using air, others required steam, however I am not sure how the sulphur was recovered from these liquids.

Benzol Plant The illustration below gives a general idea of the size and appearance of a Benzol plant at a gas works. As benzol is made up of lighter fractions it boils off with the other gasses but doesn't get condensed or washed out in the subsequent processing of the gas. To recover it they bubbled it through towers containing petroleum oil, in which it dissolves. The towers are on the left in the illustration below. The oil is then run through a steam distillation plant, with its small fractionating column, to separate out the benzole from the petroleum oil, the oil is then returned to tanks and recycled into the towers.

The photograph below shows the oil-filled columns used to recover the Benzol at a large gas works.

Fig ___ Benzol columns




Benzol is a mixture containing mostly benzene but also toluene, anthracene, xylene and the aniline so important to the dyestuffs industry. Benzene is highly inflammable and was used as an additive to (and in some cases as a substitute for) petrol in motor engines, National Benzole was a company trading petrol with this additive. Aniline is colourless but turns brown on exposure to air and light, it is one of the most important of the organic bases and as well as dyestuffs it is used herbicides, fungicides, photographic chemicals, in the making of urethane foam and pharmaceuticals. Over half the dyes used today are the nitrogen based 'azo dyes', the basis of which is naphtha. Since the 1930's however these chemicals have increasingly been obtained from petroleum oil at oil refineries.

As the benzole was being recovered by fractional distillation it was possible to separate out several 'cuts' and ship out either a mixture (benzole) or tanker loads of benzene with drums of the remaining fractions. The larger gas works all had this kind of plant and after World War One there was a national organisation (National Benzole) set up to sell the stuff on. Both benzole and benzene are Class A liquids so for a pre-war layout a National Benzole Class A tank wagon could collect the product from the plant. In the post World War Two war era, up to the 1970s, there was a fleet of rail tanks operated in the rather plain livery of 'Benzole Producers Ltd' from many of the larger gas works sites. In the early 1950s their plain silver Class A tanks had either BENZOLE or BENZENE written in large lettering on the sides with the company name and 'home' depot details in the lower left of the tank in small lettering. Full details on these tanks can be found in Mr R. Tourret's book Petroleum rail tank wagons of Great Britain' - See bibliography for details.

Fig ___ Benzol Producers rail tanks







Drying and metering



The meter was a simple device, I believe the most common design was a 'paddle wheel' half immersed in water, the gas was fed in at one side and passed over the water to the other side, taking the paddles with it, a shaft then feeding a mechanical counter. An alternative design has a screw thread half immersed in water, the gas passed through the upper half, forcing the screw to turn, the axle of the screw then operating a simple mechanical counter (see also the extra illustrations provided by Mr Franz via the clickable link below for a cut-away illustration of this apparatus)

Gas Holders

Gas holders are often called 'gasometers' although there is in fact no such word. These large gas holders are perhaps the most distinctive feature of a gas works. Inside the base of the gas holder there is a water bath and the gas holder sits in this rather like an up-turned paper cup sitting in a bowl of water. The gas is trapped above the water and as the holder fills with gas the floating 'cup' is lifted up. The weight of the metal gas holder provided enough pressure to feed the mains (a booster pump was often required to supply mains pressure as demand and the area being served increased).

Up to the mid 1830's the gas holders had to be encased in a brick building by law. These buildings were usually hexagonal in form, often with a prominent roof ventilator, and a surviving example may be seen at Warwick although the tank holding buildings are now offices. This crude secondary containment was actually highly dangerous, in the tank there is no air so the gas cannot explode, but trapped between the tank and the building an explosive gas-air mix could form if the tank leaked. If the mix was ignited the brick building would in effect become a very large grenade. I have found no evidence of any of these encased gas holders lasting later than the 1850's, for one thing they were all rather small and the demand for gas was growing rapidly.

Early gas holding tanks were a single open-bottomed metal container floating on water as a seal, examples of this type remained in use at rural gas works right to the end of production. By the 1840's however telescopic tanks were in use, with additional sections (called 'lifts') to increase the height of the tank without increasing the depth of the water tank. These early single and multiple lift tanks had supporting masts or 'guide columns' spaced around the sides with rails for small guide wheels on the top of the tank, hence they were known as 'guide framed tanks'. Early gas holders were fairly small (perhaps 40 feet or 12m across) and had only three guide columns arranged around them, gas holder of this type remained in use right to the end of coal gas production. The example below left is from an illustration of the original London gas works in Westminster, that on the right is based on the gas holder at the preserved Biggar gas works in Scotland. This is probably the easiest type to model, the vertical posts could be replaced with pairs of H section girder (as on the Westminster example) to really simplify the thing.

Fig ___ Early type of gas holder


As demand and hence the size of gas holders increased additional columns were added. The supporting frame designs were quite varied, on larger tanks there might be twelve columns, connected together at the top and one or two points on the ides with horizontal trussed frames and with diagonal metal rods forming X shapes between them. Having said which some were very minimalist indeed, the preserved tanks at at least one gas works have only the original three masts around a single lift, with the bracing cutting across the top of the tank. Although this is real it just doesn't look real, the most simple of the large tanks I have traced was the holder in Dalbeattie, Dumfries, Scotland.

Fig ___ Dalbeattie gas holder


This tank has only five guide columns, inset from these is a rail for the guide wheels on the tank sections. There is no cross bracing on the masts other than the trussing at the top. The tank sits in a metal caisson partially sunk into the ground which contains the water. Some tanks were rather more complex and represent something of a modelling challenge. The example shown below is quite a large town type in Salford near Manchester, in British N the diameter should be about eight inches but you can reduce this quite a lot and it still looks acceptable.

Fig ___ Gas holder


The capacity of this tank is 1.9 million cubic feet, or 53,704 cu m. As each section is lifted by the gas the weight and hence the pressure increases, with one section lifted the pressure is 10mb, with all four lifted it is 23mb. The base of the tank is set into a concrete base, when empty you can step directly on to the top of the tank, other tanks had a fixed base extending above ground (as shown on the spiral lift tank below).

As demand increased smaller multi-lift tanks were sometimes modified by adding more lifts, this meant they extended above the original frame and they were usually fitted with rope guides to keep them steady.

The external supporting frame was sometimes painted in a light colour, typically something similar the 'light earth' or even 'sand' in the standard modelling paint ranges but dark green seems to have been the most common colour for the entire apparatus, with some changing to a mid grey in the later 1930s. When tanks were in daily use for several years the lower sections could not be painted and went rusty, hence you may see pictures of 'rust coloured' tanks, but if you look at the top part you get a sense of the actual colour it should be.

In 1888 a new design appeared which used spiral tracks on the tank sections, these tanks did not have the side mounted supports. The example shown is in Sale south of Manchester, again the diameter should be roughly eight inches in British N but six inches looks acceptable. Note how the side staircases are fixed, there is no inner handrail so as the tank sections move you can always step onto the walkway. The stairs are narrow, only perhaps fifteen inches wide. This tank holds 2 million cubic feet of gas (54,000 cu m) and has four sections or 'lifts' which rise up from the base. The base of the tank is set in a pit, presumably to contain the water if it leaked, this can be seen in the lower left of the photos below.

Fig ___ Spiral track gas holder


Note that on the 'spiral track' type tank the rails usually (but not always) run in alternate directions on each section. The access ladders, which are very narrow, perhaps 15 inches or 40cm, are supported on triangular frames mounted on top of the sliding sections and off-set round the tank so as the tank sections spiral up and down access to one ladder is not blocked by the ladder on the lower sections.

In the late 1920's an alternative kind of gas holder appeared which did not use the water seal, these consisted of a tall vertical tube with a close fitting piston inside. These 'dry' gas holders were always large, typically a third again as high as any adjacent telescopic types, and I was not able to get a decent photo of any before they were demolished. They were encased in very tall buildings which appear circular from a distance but which were actually multi-sided. The sides appear corrugated but in fact the ribs were widely spaced, however for an N Gauge model you can get away with using Slaters OO scale corrugated plastic card sheet. There were access ladders on the sides as shown and often one or two pipes about a foot in diameter running down the side. There were usually windows set into the top section for maintenance workers.

Fig ___ 'Dry' storage gas holder


In British N scale (1:148) the dry type tank which used to stand at Stockport would be eleven inches in diameter by eighteen inches high. Even with modellers licence to keep the look of the thing it would need to be a minimum seven inches in diameter by a foot high. This type of tank was also a regular feature at larger steel works, they often had the name of the works painted on them in large white lettering.

The 'dry' type tanks (as I remember them) were usually painted green although one which remained in use in Salford near Manchester until 2002 was a light grey-green colour for the last few years. Both the earlier 'wet' types were usually painted green, some were painted light grey and a few were very dark, almost black (that could have been due to rust and accumulated atmospheric pollution).

The gas was often pumped into smaller gas holders in the chain of processes so the flow could be controlled, these small tanks were mainly of the single-lift (masted or spiral) type.

At a more rural gas works the gas holders would be proportionately smaller, you can get away with a couple of holders about three and a half inches in diameter for a small country town gas works.

The change to North Sea Gas in the 1970s did not make these gas holders totally redundant, they were used for storage and pressure control for the new service. They cost a lot to maintain however and by the late 1990s there were only about five hundred still in existence. The privatised gas distribution company Transco has invested in new pipeline technology which (they feel) eliminates the need for these gas holders and they plan to demolish all of them by 2010. Some have been 'listed' for preservation as monuments but most will have be demolished in the first few years of the twenty first century.

From the tanks the gas was passed through cast iron pipes, coated with a black enamel material made by dissolving coal dust into coal tar, to the homes and factories. Difficulties in sealing lengths of pipe lead to quite frequent explosions in the streets due to gas leaks in the 1820's and early 1830's. The gas from the early works still contained a lot of impurities, it produced an unpleasant sulphurous smell and soot when it burned. By the 1850's the process was much better understood and the filtering had improved but no one has yet found a way to get all the smelly sulphur compounds out of coal gas.

By this time people were making considerable use of the by-products extracted from the coal tar and from the gas itself. The hot smoke was first passed through a water or air cooled condenser which removes the oily coal tar. A pump called an exhauster then passes the gas to a filtering bed to remove the smelly hydrogen sulphide (which gave early gas supplies the smell of rotten eggs). The original water and lime mix was replaced by a dry lime filter in the 1830's and this in turn was replaced by iron oxide filtering in the 1850's. The iron oxide was actually a mixture of iron oxide (rust) and wood shavings, called Iron Sponge or Iron Mass. This stuff is still used today to remove hydrogen sulphide from sewer and land-fill gas so the gas can be burned (if the H2S was not removed it would form sulphur dioxide, which turns to sulphuric acid when mixed with water in the air).

The gas is then bubbled through water to remove trace acids and nitrates, of which ammonia was the most valuable. Water will absorb over seven hundred times its own volume of ammonia and the resulting liquid was called Ammoniacal Liquor. There were still impurities in the gas, some of which were worth recovering. The larger works would often have a Benzol plant to extract a range of oily hydrocarbons. The benzol mixture from town gas works was of poor quality and was mainly used as an additive for petrol, some being used at the works itself to boost heating value of the 'water gas' used in the retorts. The gas would then be dried and pumped via a meter into the gas holder.

Gas Works Coal Supplies

Gas works used coal in small sizes, typically 8 inches (400 mm) cube down to dust. This was often carted from the local railway yard in horse drawn carts but even some quite small gas works boasted a siding or two. At my local town gas works a street tramway was used with a gas works locomotive hauling coal wagons up the middle of the road from the station yard to the works. Most gas works coal was supplied as a rough mixture of small coals mixed with coal dust, big lumps had to be broken up and later gas works often had a 'coal crusher' to pre-process the material.

In smaller gas works the coal was wheel barrowed or shifted in small hand-trammed tubs on a narrow gauge track way to the retort house. A man with a shovel would then throw the coal into the retort pipes. At smaller works the coke was handled using long handled shovels and rakes. It came out of the retorts red hot and in single storey works it was moved in slatted metal 'coke barrows' to the nearby quenching hut. The cooled coke was then wheeled in the coke barrow to a stockpile.

At larger works in towns the retorts were similar but they used a 'coal charging apparatus' to load the coal into the retorts (see also the extra illustrations provided by Mr Franz via the clickable link below for an illustration of this apparatus) . Where the retorts were on an upper floor, above the gas generator equipment, the coal was hauled up the side of the building in 'skips' and the coke typically emerged on a raised narrow gauge railway running across to the coke storage area.

Only some coal is suitable for making 'town gas', in the UK the most useful bituminous coals came from the Newcastle and Durham field, South Yorkshire, Derbyshire and Barnsley districts. Other coal was used but wagons arriving at a gas works would be most likely to be from one of these areas.

Some of the resulting coke was used as fuel by the gas works to heat the retorts, often this was loaded directly from the retort using a simple wheeled chute. Coke firing remained the norm at existing small gas works but in the 1870's there was a shift to using producer gas, made on site from the coke, to heat the retorts. This would require either another small building, with a chimney, to house the gas plant (see also Lineside Industries - Prototype industrial ancillary structures) or it might be housed inside the retort house, with the retorts themselves on a raised floor above. This kind of works remained in use right to the end of coal gas as the small communities supplied from such works were the last to be changed over to North Sea Gas in the 1970's.

The illustration below shows a 'carburrated water gas' (CWG) plant, a developed form of producer gas which uses superheated steam and an oil spray to add calorific value to the gas produced. This is the sort of plant that was added to some gas works and was used in other industries to make their own gas from coke.


Fig ___ Producer gas plant




By-products



The residual coke from the gas works was rather soft and could only be used as fuel, although it has the benefit of being smokeless. It came out of the retorts in large blocks and most would be crushed and sold locally, although some might be shipped out by rail for delivery to a coal merchant (the coke from the gas works was sold as Firemax after Nationalisation of the gas industry). The coke from the retorts was usually small (two inch lumps down to dust), some emerged in larger fused lumps which had to be broken down for sale. Most of the coke from smaller works was just graded and sold locally. In the early days children with small hand carts were regularly seen hauling it away but by the early twentieth century it was being sold in sacks carted away on lorries resembling loads of coal. Some was also sold in bulk and shipped by the wagon load for use as a fuel in factories. Most gas works feature a large stockpile of coke, the bigger the works the larger this would be, but for modelling purposes a scaled down representation is all that is required.

Coke weighs a lot less than coal and it was common practice to add 'coke rails' to the top of a standard railway mineral wagon to enable it to carry a full load (Peco offer a set of rails to suit their wagons). On balance you would see roughly the same number of coke wagons in the gas works traffic as coal wagons (see Volume 1 Fig___).

The coke from a gas works consists of over 90% pure carbon and its porous structure allows it to burn rapidly. As most of the impurities have been removed it is 'smokeless' and much less smelly than coal, so the larger lumps were popular for domestic stoves. Gas works coke was not suitable for use in a blast furnace for iron or steel making as it is too soft, but some was used in refining other metallic ores. Up to the 1940's most coke from the larger gas works was sold to factories as fuel, where some was burnt but more was used to make producer gas and water gas which could be piped through the factory wherever heat was required. Gas works coke was also used (with limestone) to make calcium carbide, used in the production of acetylene gas.

Tar by-products

In smaller works the tar decanted into wooden barrels and sold locally for use as a wood preservative, or you could turn up yourself at the works with a bucket or barrel to be filled. At larger works the tar was decanted into tank wagons to be shipped to the tar distillers for further processing (see also 'Lineside Industries - Chemical Industries - Coal Tar Distillers').
Coal gas tar consists on a range of materials, basically divided into 'Naphtha' (valuable) and 'Pitch' (less valuable). Naphtha is a generic term used for the assorted light distillates of coal tar, wood and petroleum. A ton of coal produced about seven and a half gallons of tar, of which about 60% ends up as 'pitch' and only about 5% constitutes the more valuable elements in the naphtha. One of the first valuable by-products from coal distillation was creosote and by the 1860's the remaining lighter fractions were distilled to obtain benzene, toluene naphthalene & etc. Some of these fractions ended up in the gas being produced rather than the tar and from the 1870's Benzol was being recovered from coal gas at many larger works (discussed below). The process of recovering the useful constituents of the tar is more fully discussed in Lineside Industries - Chemical Industries - Coal Tar Distillers, the descriptions below relate only to processes commonly applied at gas works.
At a gas works the tar was fed into the Naphthalene plant where it is heated (to about 200-250 degrees centigrade). From this you can take the naphthalene direct as a white flaky powder or get heavier oils including carbolic oil. At still higher temperatures (up to 400 degrees centigrade) you can get creosote oil, a yellowish to dark green-brown liquid (also known as tar oil and occasionally as liquid pitch oil). The remaining tar, a thick black oily liquid is called 'pitch', was stored on site in a pit and shipped out in barrels or Graham Farish type tank wagons to be used for tarring roads and as a wood preservative.
By the end of the 19th century some larger gas works had their own tar distillery on site, invariably known locally as 'the chemical works'. The degree of processing varied from site to site and from a modelling perspective it is better to rely on a separate firm as this generates more interesting traffic.

Products from ammoniacal liquor

Up to the later 20th century the ammoniacal liquor from gas works was a major source of ammonia, however its main use at gas works was the production of ammonium sulphate fertiliser. The process used a continuous still with steam fed in at the bottom and the liquor fed in at the top. Some condensed at the bottom as a liquid, some came out the top as an ammonia saturated steam. The stuff at the bottom was treated with a strong alkali (usually lime, delivered in sheeted or 'cottage topped' open wagons) and the stuff at the top was treated with mild sulphuric acid, the gaseous residue was then passed to a condenser where they formed what was called Devils Liquor (as it contained a lot of rather unpleasant substances) and returned to the still for further processing. The end result of all this was ammonium sulphate, a valuable fertiliser.
The sulphuric acid could be made at the gas works as described above, additional supplies would be delivered in glass carboys or, from the 1930's, in de-mountable iron tanks. Ammonium sulphate is a white powdery substance, shipped out in sacks, the area where this stuff was handled would have a lot of white staining on the ground and on any loading bank or platform.



Benzol

In the larger gas works the gas was passed to the Benzol plant where it was re-heated to about 170 degrees centigrade to recover the lighter distillates. A ton of coal only produced a few pounds of Benzol. Benzol is a mixture containing mostly benzene but also toluene, anthracene, xylene and the aniline so important to the dyestuffs industry. Aniline is colourless but turns brown on exposure to air and light, it is one of the most important of the organic bases and as well as dyestuffs it is used herbicides, fungicides, photographic chemicals, in the making of urethane foam and pharmaceuticals. Over half the dyes used today are the nitrogen based 'azo dyes', the basis of which is naphtha. Since the 1930's these chemicals have also been obtained from petroleum oil at oil refineries and by the later 20th century this source had effectively replaced coal based production.







Most of the tank wagons used for both ammoniacal liquor and tar were owned by the customer, but gas works did operate some tank wagons themselves. My local gas works at Altrincham owned some tanks which resemble the Graham Farish rectangular tank wagon although they were of an older curved top design (see also 'Goods Rolling Stock Design - Rail Tanks' for more on these early tank wagons). The Altrincham tanks were painted red with ALTRINCHAM GAS CO. in foot high lettering on the sides, the remainder of the information, 'Return to Altrincham C.L.C. and etc) was on a red painted cast plate mounted on the solebar. There is a photograph of an early wagon owned by this company in Bill Hudson's book Private Owner Wagons Volume One (see Bibliography). Do note however that this early tank wagon dates from the private ownership era, the Altrincham Gas Company was purchased outright by the corporation in 1872.


Flow diagrams for medium and large gas works

Fig ___ Small Gas Works

Incoming: Coal Iron sponge Outgoing: Coke, probably sold locally, delivered is small horse cart or small flat bed motor vehicle in sacks Tar (probably in barrels, possibly in tank wagons for tar distillation plants) Ammoniacal Liquor (only a possible, probably in barrels) Spent oxide (for sulphur recovery)

Fig ___ Medium Size Gas Works

Incoming: Coal Outgoing: Coke Tar (probably in tank wagons) Sulphuric acid (probably in carboys, possibly in tank wagons) Ammoniacal Liquor (probably in barrels, possibly in tank wagons)

Fig ___ Large town gas works

Incoming: Coal Lime for the fertiliser plant Tank wagons of oil for the water gas plant Possibly nitric acid for the fertiliser plant Outgoing: Coke Tar (probably in tank wagons) Possibly some ammoniacal liquor, in drums or tank wagons Sulphuric acid (probably in carboys, possibly in tank wagons) Fertiliser is sacks Benzol in tank wagons

In the larger works narrow gauge tramways were used to move the coal from the stockpile in small wagons, several of these tramways used steam locomotives to haul the coal and coke wagons and also to move other wagons transporting materials about the site.

By the end of the 19th century new gasworks had become generally quite extensive installations featuring the widespread use of mechanical handling aids. The associated processing plant for the by-products would often be on the same site and the whole establishment might cover anything up to a couple of hundred acres. They were often connected to the main railway system by a series of exchange sidings serviced by the gas works internal locomotives. To represent a really large gas works the exchange sidings, two or three tracks each capable of holding perhaps twenty wagons, could be arranged along the rear side of the main line, feeding the 'works' represented in low relief or even painted directly on the back scene.

---

Modelling a Gasworks

Ian Franz has sent in a scan from a school book published in 1949 which shows the various elements of a medium sized gas works. My drawings are based on photographs of the large Stockport works and the small preserved works at Fakenham and Biggar, the plant in Mr Franz's contribution is different.
Click here to open this illustration in a new browser window, it is rotated to allow easier printing

The retort house at a small gas works was often surprisingly small, slightly larger than a domestic two-car garage and about one and a half stories tall with some roof ventilation. There would always be at least two banks of retorts but these were usually in the same building. The Peco N gauge engine shed is a bit long, but the ends, with a new roof and sides from embossed brick plastic card roof would serve, in the kit you get two shorter roof ventilators which would be fine for this job. The side walls, roof and roof ventilator from the kit can be used elsewhere for other industrial structures.

The retort house at medium sized works would be a tall building, typically three or four stories high, with two or more chimneys. There would be few windows but plenty of ventilation in the form of louvered or unprotected openings in the upper walls and probably a louvered roof ventilator similar to that on the Peco engine shed kit.

The roof might well be asymmetric (one side larger than the other) due to the inclined retorts as shown in Fig ___ (4), and the hoist for the coal tubs would be prominent on the side.

A large town gas works vertical retort building is shown below, the steam lorry making a delivery of coal gives a sense of the scale of the building however something similar (scaled down somewhat) might be arranged in low relief at the rear of a layout to suggest a large works. The smaller brick building to the right with its associated chimney is the carburrated water gas plant, the building with the iron tank above in the far right is (I think) the ammoniacal liquor concentration plant.

Fig ___ Large town gas works vertical retort house




In a very small nineteenth century gasworks serving an isolated community the condenser might simply be a six or eight inch diameter copper pipe, mounted zig-zag fashion on the outside of the retort house. At such a small works the condensed liquid tar with all the other impurities still in it might simply be drained into pits to be sold as a wood preservative. Condensers of this type remained in use up to about the 1920s at some older works.

By the mid nineteenth century a fairly standard arrangement of air-cooled vertical pipes was used in smaller works. These pipes were about eight inches in diameter and stood roughly ten feet high. They had couplings on the tops to allow them to be washed through, and one method of making such a condenser is shown in Fig ___.

By the later 1930s a much easier to model water cooled type was in use, consisting of a two or more tall (12 feet or 4m) high rectangular boxes. The example at Fakenham was black with angle bracing on the sides. This is by far the easiest to make (I whish I had known about them when I built mine) and it is offered in the Hornby range of gas works buildings.

The condensers at a medium to large scale works would typically be large water cooled affairs, the sketch below is based on the system at a large gas works (Stockport).

Fig ___ Tar Condensers


The sketch shows both air-cooled and water cooled condensers, the former would be seen at smaller works, the water cooled type is sketched from an example at a large town works. For the air-cooled type take a supply of one inch panel pins, the original pipes were about six to eight inches in diameter. Add a strip of paper about 1mm wide about 2mm from the top as shown (A). Now cut some lengths of any suitable 1mm diameter rod, plastic or wire depending on what you have available (B), all these must be close to the same length. Glue the short rods between pairs of the panel pins (C). The pipes might be arranged in a line (D) or in a double row (E). There will be a pipe leading to the tar tank, often through a raised metal man-hole size cover (F) and an inlet pipe (G), both of which should be no less than 1mm diameter rod or wire.

Glue the pairs to an L section support or to either side of a central support and fill between them with Milliput putty. The base of the pipes was often encased in an iron box. In some installations there was a low brick wall round this unit, in others it was open onto the yard area.

The water-cooled condenser could be simply a set of three or four 8mm diameter tubes with a 1mm and a 2mm diameter pipes feeding each top and bottom (the top 2mm pipe might come out through the top or from the side near the top). Gate valves to control the flow would normally be fitted as described below. The large water cooled condenser is based on the unit at Stockport. As drawn the system would work but I have simplified things somewhat to make modelling a more practical proposition.

The handrails (H) are from Plastruct glued to a 6mm wide strip of 1mm scribed plastic card to represent the walk-way. The raised spindles (L) connected to the gate valves (N) are track pins, the heads representing the hand wheels. The main body of the unit (I) is a tube of between 8 and 10mm diameter, it is 35mm long with the end covered with a disk of 10 thou card or simply filled in with putty. The larger diameter pipework would be 2 or 3mm diameter and one option is to use old ball-point pen refills. These are made of quite a soft plastic and can have a length of brass wire inserted to help when making curved sections. You will need five lengths of the same pipe roughly 12mm long. The main horizontal pipe (R) is 30 to 35mm long and the vertical pipes beneath it are all 12 mm long. The three centre vertical pipes (P) should have a 2mm wide strip of paper wound round the base. Once the tube for the main pipework is selected and the diameter is known you can use a leather punch to make suitable holes in 20 thou card. Once the holes are made use a sharp knife to cut out the shape for the valve housing as shown (K). Three of these are threaded onto the horizontal top pipe (N) and three more are used for the three centre vertical pipes. Measure the distance between the valves and drill three holes in the walk-way to suit. The track pins pass down through these to the valves. If you feel up to it the gate valves on the vertical pipes can be controlled from the upper platform by using track pins passing down through the floor, this was the arrangement on the prototype but on the drawing I have shown hand wheels at ground level.

The platform is supported on struts from the pipework as shown, if you feel it would be easier you can support this on 1mm diameter rod reaching down to ground level. One end of the platform must have a length of brass signal ladder for access. The whole assembly was a dull silver colour with some staining and the bottom 2 or 3 mm of the vertical feed pipes at either end (O) should be painted black. You could use green as the base colour if you wished and a few streaks of 'rust' using an almost dry brush of 'track colour' would be appropriate.

In some locations the water cooled condensers were inside the main buildings or in separate, small but well ventilated, structures. In photographs I found of one medium sized gas works the condensers were housed in small square buildings perhaps fifteen feet high with slatted sides and pipes for the gas and cooling water. These can be made using a length of square section wood perhaps 25 mm deep by 30mm high and 30 mm long, with razor saw cut slots at about 2mm intervals in the long sides. Cut lengths of printed brick paper about 8 mm wide and 30 mm high, fill the edges of the slots in the wood with a little Milliput and glue the brick paper onto the corners to form the supporting columns. Finally add a door to one end and gas pipes from coat hanger wire or something similar.

The ammonia washer at Fakenham is a rectangular metal box, now available ready-made in the Hornby range, it is sketched in the upper left below. A slightly larger works often used tall tubes perhaps three feet in diameter with gas and water pipes top and bottom. They were usually mounted outside and connected with gas pipes perhaps a foot to two foot in diameter and water pipes about three inches in diameter. Biro tube will serve for the gas pipe, soft iron 'florists wire' is suitable for the water pipes. The large set up shown below bottom is based on the equipment at a large town gas works (Stockport)

Fig ___ Ammonia washers


The main body of the unit is a simple box, add strapping to the outside as shown using 10x20 thou strip or thin strips cut from masking tape. The tank and pipework was all green at the Stockport plant, you might paint the tank black to show up the pipework.

The liquor from the ammonia washers was typically run off into the 'tar pit' but at larger works it was also stored in tanks, above or below ground. It was usual to concentrate the ammonia to as much as 25% by volume before shipping it out or using it to make fertiliser on site. The production of sulphate of ammonia fertiliser would involve quite a large building with good ventilation and a chimney or two. The process involved heating a mixture of concentrated ammoniacal liquor and lime with steam, the ammonia boils off and is them bubbled through sulphuric acid (from the spent oxide) to produce saturated ammonium sulphate, which is then crystalised to produce the fertiliser (shipped out in jute sacks, although if stored for a long time it tended to eat away the sacking).

Fig ___ Ammonia concentration plant & fertiliser works



The iron sponge was contained in air-tight rectangular iron tanks, the prototype in smaller works was about two or at most three feet high and was often housed in an open sided enclosure. Above the tanks was a simple gantry to allow the heavy lids to be lifted off the tanks so the spent oxide could be removed and the sulphur recovered. They were usually 'under cover' but the 'building' often had one or more walls missing, some were just a simple roof on metal supports. The Ratio corrugated iron roof unit would do for such a cover, but a flat (angled) corrugated iron roof can be made from card wound with thread. The tanks were fed by a thick pipe, perhaps eight or ten inches in diameter, each tank in the row having its own valve to allow them to be isolated for emptying. The tanks at larger works were inside a brick building, quite tall (about the same as a two storey house), in some the tanks were on a raised floor and were lowered to regenerate the oxide, but all of this took place inside the building.

There was a lot of pipework in a gas works, in the larger works much of this was elevated and carried in gantries to allow road and rail vehicles to pass underneath. Adding such pipework helps identify the place and you can use the smallest Plastruct Fineline truss frames or even signal gantries to carry the pipes.

There would usually be a fairly large stores and workshop complex somewhere about the works. One area would be reserved for storing pipes, typically twenty feet long by four inches in diameter and black in colour. There was often a small crane in the pipe storage area for loading and unloading road vehicles. The modern thick walled yellow plastic HDPE (high-density polyethylene) high pressure pipes for gas were introduced into UK by Wavin/British Gas in 1970 and were used for a lot of the changes made for North Sea Gas supplies. This was transported in coils about six feet in diameter, with a hole in the middle about four and a half feet across.The blue type for water appeared in 1980 and by the end of the 20th century pipes of this type were being used in a wide range of industrial and agricultural applications.

Fig ___ 1970s plastic gas pipes




The Benzol plant resembled a miniature oil refinery, this is not actually difficult to model, although you may have to explain to visitors what it is supposed to be.

Fig ___ Benzol plant







The gas holder itself represents the most difficult element to model, for one thing these tanks were often very big. In British N a large gas holder would be perhaps a foot in diameter and getting on for a foot or so high but as few people ever got very close to a gas holder you can reduce the size a lot a still keep the right general 'look' of the thing for modelling purposes. Fortunately not all gas holders where huge and the smaller kind were the last to go. For a smaller works serving a small town the minimum size for a tank would be about six inches in diameter by about six inches high.

You could add a tank or two painted on to the back-scene, but it looks a lot better if there is something actually there. One option is to build a half tank against the back scene or better yet build it into a corner where you cannot have any railway in any case. Piko offer a gasometer in their range (part number 60013), although it is rather small and of a somewhat continental appearance it would do for a small gas works. Very small gas holders are quite easy things to model, they might be used at a small works as the main holding tank or at a larger works for flow control. Fig ___ shows the small gas holder at Fakenham and a suggestion for modeling it using the cap from a tin of Jiff spray cleaning mouse. This is two and a half inches in diameter by two inches high so the proportions are about right. Do note that these aerosol can plastic tops are not cylindrical there is a very slight taper from the open end toward the top, but in this case we can make use of that feature.

Fig ___ Fakenham Gas Holder

To convert the tank into a gasometer you need to make the top domed and add something round the lower half to represent the water tank base. To dome the top pour in some boiling water and leave it on the draining board for about twenty seconds. This softens the plastic slightly. Tip out the water and holding the top in both hands use your thumbs inside to push the middle outwards. The required curve can be very slight but you will probably need to repeat this two or three times to get it right.

Fig ___ Modelling a Gas Holder


Now take some one inch brown paper masking tape and wind this round the lower part of the Jiff top (A), keep it tight at the bottom edge, do not smooth it down as you go. This is because of the taper of the top, smoothing the tape down results in lumps and folds of tape. Keep winding until you have built up about a millimetre of thickness. Cut three cocktail sticks in half and chamfer the cut ends as shown (B).

Use the tip of your modelling knife to ease the masking tape away from the plastic at the top and insert the cut down cocktail sticks. Adjust the position until all three pairs are arranged symmetrically round the sides (C). Now lay a strip of masking tape down on a smooth surface (I happened to have a cutting board but a china plate would do). Using a metal straight edge and a sharp knife cut strips 1mm or so wide from the tape (D). Lay these strips round the base of the tank on the existing masking tape base, this represents the horizontal frames. I suggest you lay one strip round the bottom edge, another round the top and two more in between. Cut some more strips and lay vertically these between the horizontal bands, you should cut them with your knife where they join and arrange them in a brickwork pattern as shown (E).

Now use a set of compasses to draw two circles on card, if you use plastic card for this you can use dividers and scribe round until you cut through the card. If using cardboard mark the rings and cut with a small, sharp, pair of scissors (F).

Offer the resulting ring up to the cocktail sticks and use a pencil to mark the ring where they fall. Use your modelling knife to make small nicks in the card so it will slip over the sticks and slide down onto the masking tape ring (G).

The access ladder is an N gauge signal ladder glued to the edge of the platform as shown (H).

To add the handrail I suggest you use OO scale etched brass signal ladders, curved to fit the outside of the ring and glued on edge. (I) Alternatively you can do what I did and add an inner handrail of fine wire glued to the cocktail sticks, Health and Safety would not appreciate such an arrangement but I suspect it was not as uncommon as you might think.

Finally you need to add the bracing to the top of the cocktail sticks, you could use etched brass N gauge signal ladders for this (it should be trussed as shown but laddering looks okay). Personally I just added another strip of the masking tape, 2mm wide, round the posts at the point where they start to taper. Paint the upper part of the tank, the cocktail sticks and the bracing between the posts green and the masking tape very dark green. The ladder leading to the platform can be light green or black, the handrails should be light green or light grey in colour. When the paint is dry give the lower section a dry-brush of dark grey and green to bring out the detail of the strips.

This represents the easiest way to obtain an acceptable result, you can use more care and more expensive materials to improve the model but as described it looks the part well enough and would stand comparison with most commercial models.

To model larger tanks one option is to use another tank kit basis, you need to select a model of about the right size for this. Vero, who operated from the former East Germany, used to offer an inexpensive gasometer in HO, the tank from which would give you a reasonable scale tank for an N/2mm scale model. You would need to add the supports and other details yourself. On the prototypes the smallest number of supports I have seen on a larger tank was five (at a gas works in Scotland), the trussing across the tops cut across the top of the tank and there were no trusses further down the tank.

Kibri offer a pair of oil storage tanks (part number B-7466) which can be used for the job. These are three inches or so in diameter by about three inches high, which is somewhat under-size but the end result is acceptable. The tanks are supplied with a spiral staircase up the outside, for a tank with side posts that goes into the 'bits box' as all you need is the tank body and top dome, for a spiral type tank you add the triangular supports. I would suggest the easiest type to model from these tanks would be a three-section spiral-tracked type, suitable for any period after 1888.

Again the lower sections can be made by winding masking tape round the tank, in this case the tank is in three sections so you need two strips of tape one twice the thickness of the other as shown in Fig ___. The spiral tracks can be represented by strips of microstrip (30x30 thou is about right) but this is difficult to glue to the tap, thin strips of post card would be easier to attach. Failing which you can use thread, having built up the telescoping sections measure the circumference of the tank and mark the top edge with five equally spaced points. Drill small holes through each of these points and drill another set of five equally spaced holes at the first joint, roughly in line with original set. Fine thread or preferably monofilament fishing line can then be passed through these holes as shown to suggest the inclined spiral track on the side of the tank. Repeat this with the middle section, keeping the angle of the lines about the same as the top section but inclined in the reverse direction. If using a snap-on aerosol cap for your tank gluing the thread to the tank is a problem (Uhu might work) but by passing them through holes you can secure them to the inside with Araldite and that will hold them tightly in place when it dries.

Fig ___ Modelling a telescopic gas holder

Modelling a big gas works is possible, the preferable location would be at one end of a baseboard where the gas holder can be represented by a simple quadrant section built into a corner.

At one of my local gas works (Stockport) the retorts and coke plant were all contained in a single large building which can be modelled in low relief. Sadly Stockport gas works was not rail connected, the coal was delivered by horse drawn cart, steam lorry and finally diesel tipper lorries and the coke was hauled away in similar vehicles. However for modelling purposes the arrangement at Stockport has several advantages.

The main building at Stockport, housing the retorts and coke plant, was a steel framed building with brick in-fill. The basic shell of the building was retained when the works changed from horizontal to inclined and finally to vertical retorts (in the later 1920s or early 1930s). The structure is interesting and serves well as the basis for a reduced size model. The coal was lifted by bucket chain conveyors to feed hoppers inside the building, these hoppers contained about 48 hours supply of coal. Similarly the coke was handled using conveyors and bucket-chain lifts feeding inclined tubular graders and thence storage hoppers inside the building. The coke was stockpiled outside in large heaps, but for modelling purposes this can be ignored. A coke bagging plant was included but for our purposes bulk loading of railway wagons is of more interest. A water-gas plant was also built-in to this structure, the gas was used for heating the retorts and topping up the coal gas as required.

Fig ___ Suggested large gas works main building

Large water-cooled condensers were used at all the local gas works and the tar was stored either in underground pits or in iron tanks. The ammoniacal liquor was concentrated in a small plant on the site, we can assume this is on the far side of the main building for modelling purposes, and stored in a fairly standard oil tank for bulk loading of tank wagons. The residual pitch from the tar pits would be pumped into a heated container for decanting into tank wagons.

At least one small oil tank is required as oil-mist was sprayed into the gas flowing into the mains to form an anti-corrosion coating on the inside of the pipes (it also removed some of the more troublesome impurities). This allows conventional class B oil tanks to be shunted into the works at intervals. I have included a small gas holding tank of the type described above, the local works had such a tank for flow control purposes and it is so easy to model it is worth including.

The main gas holder would be more difficult to model, probably the easiest would be one of the warterless types built in the 1930's (see Fig ___ above) but a big telescopic tank with external frame helps set the scene.

Optional elements include the benzol plant and the fertiliser plant both of which generate traffic and ancillary elements such as the iron-sponge filter units (these were processed on-site at my local works so involve no additional traffic).

Fig ___ Suggested layout for a gas works


Any gas works would receive regular, typically daily, shipments of coal and occasional consignments of 'iron sponge' in drums, loads of pipes in open wagons and other equipment in vans. Many gas works used carburrated water gas to heat the retorts (see also Prototype industrial ancillary structures) and so would receive regular tank wagon loads of (petroleum) oil. Outgoing cargo would be wagon loads of coke, barrels or tank wagon loads of tar and possibly ammoniacal liquor, occasional wagon loads of spent oxide in drums and quantities of various 'distillates' drawn from the gas in manufacture. If the works is large enough to boast a small fertiliser plant you can justify colourful PO vans.

More recently there have been a number of attempts to obtain gas from coal in other ways. One which found favour in the UK was the German Lurgi process, in which powdered coal is heated and steam is blown through it to produce the gas. The process is similar to that for water gas but the chamber is pressurised and the yield contains a lot of methane and is usually referred to as Synthetic Natural Gas (SNG) to distinguish it from normal coal gas or 'town gas'. The last Lurgi SNG plant, in Scotland, closed in 1985.

Fig ___Sketch of a Lurgi Process plant in Scotland (1970's)


In the 1960's various ways were tried to produce something similar to coal gas using oil. Oil was cheap at the time and the resulting gas could be made without the poisonous carbon monoxide found in coal gas. Over two hundred plants were built round the country but they resembled oil refineries rather than gas works, the only real clue being the associated gas holders. Such a plant would require regular shipments of oil in tankers, they used a blend of fuels so you need a mix of class A and class B tank wagons. The 'works' consisted of modern concrete and pressed metal buildings, a lot of pipe work, several chimneys of the tall spiral finned type and several oil tanks. This is perhaps the easiest type of works to model. As with the coal burning plants these were all closed down by the end of the 1970's.

The change to North Sea Gas saw the introduction of a new type of gas holder designed to operate at high pressure. These pressure tanks were large, perhaps thirty feet (10m) in diameter with domed ends. They were mounted in banks as part of a high pressure gas distribution grid.

The nationalised gas industry was sold into private ownership in 1986. The distribution side of the business, being a national monopoly by its very nature, became a new company called Transco and the purchasers of the former Gas Boards act as customers who sell the gas to the consumers. In the late 1990s the business was further de-regulated allowing other companies to act as middle men between Transco and the consumer.

One odd point that came up when talking with a banker was that current banking practice will only tolerate a lag of three years before gaining a return on their investment. The early gas industry, competing with existing systems and requiring investment by consumers, seldom saw a return in less than five years (the first customer was often the local corporation street lighting and the gas was often supplied at cost as the advertising value and development of the mains supply was worth having). Hence, if the gas supply were being proposed today it would not receive the required funding and we would have to do without it. This three years limit has undoubtedly impacted British industry, whilst German and French engineering firms are booming supplying high tech equipment to the world British manufacturing is still contracting at a steady three percent per year. Perhaps more importantly it begs the question what options today, perhaps relating to matters such as clean drinking water, are not being pursued because of this arbitrary limitation.

---

Coking and Smokeless fuel plants

Coke from gas works was unsuitable for making iron and steel and to meet the demand most steel works had their own internal plant which although working on the same principal as the gas works was built very differently.

Coke was originally made in simple beehive kilns (see Fig ___), and some was made in this way into the 1950's when the practice was ended by the clean air acts. By the 1850's an alternative method used long horizontal retorts forming a series of narrow slots, a system originally developed in Belgium, these did not become the norm in Britain until about the time of the First Word War. The walls of the retorts are made of fire clay and contain small tubes in which gas is burnt to heat the coke. A single retort is typically a couple of feet wide by perhaps fifty feet long and fifteen to twenty feet high. They are top-loaded and typically hold about fifty tons of coal at a single charging. These retorts are formed into banks of twenty to fifty and each in turn is filled, heated and discharged. The discharge is accomplished using a hydraulic ram, mounted in a building along the rear wall of the bank, and pushes the coke out through a hole at the opposite end.

The heat for the retort was originally provided by burning some of the coke and directing the hot gasses through the retort walls but by the time of the First World War the most common method was to use the coke to make water gas or producer gas and burn that inside the retort walls themselves (see Lineside Industries - Prototype industrial ancillary structures' for details of the water gas plant). All the exhaust gasses from the heating of the retorts are taken to a single tall chimney and ranged along the top of the retort are the pipes for tapping off the oily smoke and tar fumes.

Fig ___ Modelling a Coke Oven


Most coking plants had a sloping bank at the discharge side of the ovens, the hot coke was pushed from the oven onto this slope where water was sprayed over the coke to cool it. The coke then had to be picked up and shifted to the grader and thence to the stockpile and wagon loading area.

There is a useful variation at the Moncton Gas and Chemical plant in which a special hopper wagon running on a track beside the ovens catches the coke as it is pushed out, the wagon is then moved under a water spray to cool the coke down and finally empties into an underground hopper feeding a conveyor belt. This requires much less room and as such is an interesting option for a model railway layout.

One oddity worth mentioning was the sheeted coke wagon, these carried specially selected (actually hand picked) coke for foundries and the like. The tarpaulin sheeting prevented the coke from getting wet, which wasted heat at the foundry, and I gather the wagons (and sheets) were usually owned by the customer. The sheets would be branded with the owners name as these were valuable items.

There were once a lot of coke plants, feeding the many steel works, but by the 1980s there were only sixteen in operation and by the mid 1990s this was down to four. The tar distillers who processed the tar from gasworks and coke plants have now all either closed to changed to processing other materials (see also 'Lineside Industries - Chemical Industries - Coal Tar Distillers')

---

Smokeless Fuel Plant

In the later 19th century research began on extracting more of the valuable by-products from coal by varying the temperature and pressure inside the retorts. By the early 20th century there were over a hundred methods used in various locations but two in particular proved commercially viable. The older is the Coalite process, which uses externally heated retorts similar to those used in gas making. The Rexco process is slightly different as it uses internally heated retorts, these two processes recover different quantities of the various distillates from the coal gas and produce large quantities of coke.

As coke has had the impurities removed it burns with a clean smokeless flame, making it popular for domestic fires. Unfortunately most coke is rather friable and so it was usual to grind it to a powder mix it with some of the pitch recovered from the coal and press the resulting mix into briquettes. These briquets are sold as 'Coalite' or 'Rexco' smokeless fuel.

Britain has put more effort than any other country into developing smokeless fuels from coal, they have been on the market since at least the 1920's. The main purpose of the above named plants is however the recovery of by-products, so you should look at having at least two benzol type plants as illustrated above under modelling gas works.

__________________





^
Go to top of page