Stealth Locos & Strange Chimneys.

Updated: 1 July 2010
John Stevens; better pic
Back to Home PageBack to The Loco Index

There were several reasons for strange chimney designs. Here are some:

  • SMOKE DISPERSAL To disperse smoke & steam which obscured the driver's view.
  • THERMAL EFFICIENCY To increase efficiency by extracting more heat from the combustion gases.
  • STEALTH To make the locomotive exhaust less conspicuous- the Stealth Steam-engine.
  • SPARK CONTROL To control the emission of sparks and cinders.
  • FOR REASONS UNKNOWN Sometimes it is hard to guess what the designer was aiming at.


Left: Square chimneys like this were unique to Belgian locomotives.

The flat chimney front (which would not have caused that much air resistance at the modest speeds of these locos) was designed to clear smoke from the boiler to improve visibility. The front lip at the top of the chimney, known as a visière or "visor" was supposed to improve the draught.

This is a type 25 locomotive built in 1895 by the maker Franco-Belge. Photo date unknown.

Square chimneys were fitted in the period 1884 - 98.


Left: Scottish louvred chimneys

This design, intended once again to lift the smoke clear of the boiler, was typical of Highland Railway practice in the 1890's.

It did not work.

When No 14393 Loch Laoghal was photographed it was owned by the LMS Northern Division.


Left: Prussian S3 No 237 with two "stove pipes" either side of the steam dome.

This modification followed complaints from the crew about smoke obscuration.

This photo appears to have been retouched in a less than subtle manner, hence the black background. Even in well-regulated Prussia, it would be hard to get the coal in the tender quite that flat...

Above: Prussian S3 No 237 again, showing more clearly that there were two "stove pipes", running each side of the steam dome.

Presumably this arrangement did not help the draught, for there appears to be some sort of valvebox over the normal chimney position that would allow the exhaust to be sent straight out in the usual way when maximum steam generation was needed. A horizontal control rod appears to lead back to the cab.
Since locomotive aesthetics seem to matter, the stove-pipe solution did not find favour. R P Wagner's deflector plates were preferred.


Left: A Hungarian MAV328 501 with special chimney: 1920

The idea being, no doubt, that air would enter the openings at the front of the cowl and be deflected to blow the chimney smoke upwards. The 2C h2 locomotive was built by Henschel of Germany; the MAV number indicates that the locomotive operated in Hungary.

No further facts known, but it looks as though the amount of air that would be so deflected would be completely inadequate for the job.

From Dampflokomotiven und ihre Bauteile by Franz P Flury
ie Steam locomotives and their Components


Left: A German DR57 with special chimney: 1910

For quite I while I suspected that the pipes running into the chimney were for steam jets intended to raise the smoke clear of the boiler; presumably only intended for occasional use, such as when dealing with adverse wind directions, or a lot of steam would have been wasted.

However, Hans-Dieter Jahr tells me (8 Jan 09) that the pipes are actually for a feedwater-preheater built into the unusually wide chimney, and only the cowling is concerned with smoke management.

From Dampflokomotiven und ihre Bauteile by Franz P Flury
ie Steam locomotives and their Components


Left: A German DR01 with a ramp deflector behind the chimney: 1926

Locomotive built by AEG.

From Dampflokomotiven und ihre Bauteile by Franz P Flury
ie Steam locomotives and their Components


Left: A German DR43 with air-deflecting ramp in front of the chimney: 1928

Locomotive built by Schwarzkopf.

From Dampflokomotiven und ihre Bauteile by Franz P Flury
ie Steam locomotives and their Components


Left: A Norwegian C27 with vertical deflectors each side of the chimney: 1912

This reminds me irresistibly of an alarmed Dogbert.

Locomotive built by Thunes. (?)

From Dampflokomotiven und ihre Bauteile by Franz P Flury
ie Steam locomotives and their Components


Left: This engine, with its exceedingly lofty chimney, is the Anglet, built by Anatole Mallet in 1876.

The Bayonne-Anglet-Biarritz line in France used double-decker coaches, and so a very tall chimney was required to lift the smoke clear of the upper deck. The Anglet was the first French compound locomotive.

This photo was taken at Bayonne, home of the Bayonet.

Left: Another view of Anatole Mallet's compound.


Left: The "double-chimney" locomotives of Joseph Beattie. LSWR, 1855

This design was a fast goods engine for the London & South Western Railway, designed and built by Joseph Beattie at Nine Elms in London. They were the first engines to use coal- rather than coke- on the LSWR.
What appears to be a small secondary chimney in front was actually a vertical jet condenser that formed part of Beattie's first type of feedwater heating apparatus.

Beattie produced a number of engines with feedheaters, and it is not known which of them is portrayed above. No 135 "Canute" had a single feedheater column, while No 167 "Atalanta" had two, disposed symmetrically in front of the actual chimney. These engines were reputed to have the highest thermal efficiency of any in the country, though the accuracy of the tests on which this claim was based is dubious.

Joseph Hamilton Beattie, (1808-71) was Locomotive Superintendant of the LSWR from 1850 to 1871. He took out his first patent for feedwater heating in 1854.

Above: Detail from a painting by Cuthbert Hamilton Ellis of Beattie's locomotives at Nine Elms depot.

The "Medusa" at the front has one extra "chimney" that is really a feedheater, but the unnamed locomotive behind it has two columns in front of the real chimney. Note extra pipework at the sides of the smokeboxes. At the moment it is not clear whether the one-column or the two-column version came first.

Left: The "Vulture" 2-4-0 engine of the L&SWR.

This engine design was introduced in 1856; note that it looks very similar to the "Medusa" shown just above. The single vertical column in front of the real chimney was the jet condenser; part of the exhaust steam entered this and was condensed by a jet of water, the result being hot feedwater. Some of this flowed back to the water in the tender, warming it up, and the rest going to the feedpump, which pushed the hot water through a tubular heat exchanger in the smokebox to heat it further and thence to the boiler.

From an article "Locomotive Feed-water Heating and Boiler Feeding" by George Willans in the the journal The Locomotive 15th Feb 1921, p44. Presumably he wsa a relative of Peter William Willans, designer of the famous high-speed steam engine.

Trials were conducted which appeared to show that the feedheating apparatus gave a 20% saving in fuel consumption, and sometimes as high as 33%. Whether anyone believed these figures at the time is unknown, but certainly George Willans wasn't buying it.

The column technology was ultimately replaced in 1858 by more conventional heat exchangers at the side of the boiler; this remained the standard feedwater heater on the L&SWR, but was gradually discarded after W. Adams took over in 1878. Clearly Mr Adams wasn't buying "33%" either.

Left: The "Atalanta" 2-4-0 locomotive of the L&SWR.

This gives a better view of the two-column version.

Note that the middle gentleman on the footplate is wearing a stove-pipe hat.

Left: M Petiet of the Nord strikes again. 186?

A duplex-drive locomotive on the French Nord network. The hot gases passed through the boiler tubes as usual and then turned back to pass through the steam-dryer on top of the boiler. This was before the introduction of the Schmidt superheater in 1898. It is thought that the amount of extra heating of the steam was in fact quite small.

For some reason I always think of elephants when examining Petiet's designs.

For more strange chimneys prompted by a desire for thermal efficiency, see also: Petiet's Experiment and Franco-Crosti Boilers.

Above: Locomotive with preheater for the Egyptian State Railway, by F H Trevithick: 1902

This bizarre construction, which could have been inspired by Petiet's designs, was a modification of an existing 0-6-0 to test the possibilities of feedwater heating rather than steam drying. The cylindrical heater a was 4.25 m long with a diameter of 685mm, and was connected to the smokebox by a 450mm wide pipe. The feedwater was heated by both the combustion gases, which passed through 91 smoke-tubes a1 of 47mm diameter, and the exhaust steam, which was passed through in a pipe c of 230mm diameter to a horizontal blastpipe d. As a result of this construction the combustion gas heating area was 58 m2, but the exhaust steam heating area was only 3.3 m2. The feedwater temperature was raised to 133 °C.

Judging by the height of this assembly, the loading gauge in Egypt must have been a good deal higher than in England; perhaps there simply weren't any overbridges. It is probable that this bizarre form of construction was used simply because it was the cheapest way to try out feedwater heating. Note the fancy pelmet around the cab roof; very tasteful. This sort of decoration seems to have been common on Egyptian locomotives. The locomotive weighed 33 tons without the tender.

The project was designed by F H Trevithick, a grandson of the great Richard Trevithick. He was chief mechanical engineer of the Egyptian State Railways from 1883 to 1912. The source gives his name as F R Trevithick, which caused some problems in tracking him down.

From Zeitschrift Des Vereines Deutscher Ingenieure (The magazine of the Association of German Engineers) for 10 May 1913

Above: Patent drawing of locomotive with preheater for the Egyptian State Railway, by F H Trevithick: 1902

The indefatigable staff of the Museum have answered a few of the questions about this remarkable design by finding the drawing for a German patent. (K13, No 134562) The patent appears to have been taken out in 1902, a long time before the 1913 article which showed the photograph above.

The blast-pipe d is at the rear of the heater, above the front of the cab. A U-shaped pipe can be seen running down from the heater, to the left of the smoke-box b, and terminating in what was presumably a non-return valve feeding the boiler. The cylindrical volume h wrapped around the heater acts as a substitute for the removed steam-dome. Note that in the photograph above the steam-dome was retained; there are several differences been the patent drawing and the photograph, as the patent represented a purpose-designed locomotive rather than an improvisation.

There was a problem; the exhaust steam tended to condense in the feedheater, and although drain pipes were fitted, some of the water was blown out of the chimney. In later trials the feedwater was heated by the combustion gases only, and a feedwater temperature of 115 °C was achieved.

From Zeitschrift Des Vereines Deutscher Ingenieure for 20 Dec 1902

Left: The Pepperpot Special; another preheater locomotive for the Egyptian State Railway, by F H Trevithick: 19??

F H Trevithick's next experiment had six feedwater heaters arranged in a saddle formation over the smokebox. Each heater was 0.91m long, and composed of 31 tubes of 47mm diameter. The total heating surface of tubes and cylinder shells was 31.5 m2. The cross-sectional area for passage of the exhaust gases was 2400 cm2. The feedwater flowed sequentially through the six cylinders and reached an average temperature of 110°C.

Many thanks to Felix Brun for helping with translation from the German.

From Zeitschrift Des Vereines Deutscher Ingenieure for 10 May 1913

Above: The Pepperpot Special again, with its remarkable angled chimney. Heaters partly sectioned.

There seems to be another little tubular heater perched just above the first driving wheel. What looks a steam feedpump can be seen at the front of the cab, (injectors don't work with hot water) just below the safety valves, and I suspect this extra little heater was intended to make use of the exhaust from the feedpump for more feed heating.

From Zeitschrift Des Vereines Deutscher Ingenieure for 10 May 1913

Above: F H Trevithick's swinging chimney. Another of his preheater designs for the Egyptian State Railway

In this design the aiim was to secure better flow of the exhaust gases by mounting the preheater and blast pipe on the smokebox door. The blast pipe was under the base of the chimney in the usual way. The big pipe just under the smoke box carried the cylinder exhaust to the blast pipe, through some sort of coupling. The feedwater being preheated entered and left the swivelling section through conical joints in the centres of its two hinges, sealed by stuffing boxes. You can just see the pipes in the right photograph. There was also a connection for the steam blower but it is not clear where it was.

Trevithick claimed the boiler steamed easily with good superheating and preheating, without needing a high gas flow rate. During starting the steam blower easily maintained the gas flow. He said all the pipe connectors worked well. The smokebox was originally fitted with a spark screen that was later removed; presumably that it is it shown hinged to the preheater in the rightmost picture.

Many thanks to Felix Brun for helping with translation from the German.

From Zeitschrift Des Vereines Deutscher Ingenieure for 17 May 1913

This fascinating excursion into locomotive design in Egypt has underlined the fact that there seems to be very little documentation on the Egyptian railways. Motor Books had nothing...


Left: Chinese stealth chimney

This is an armoured train used by the Chinese army, photographed near Ho-Nan. Photo date unknown. The chimney has been redirected to ground level in an attempt to dissipate the smoke and make the approach of the train less obvious.
The dustbin-like thing at the left is a rotating machine-gun turret.

See also Polish armoured trains, some of which had similiar stealth chimneys. Others had even stranger versions... see below.


Left: The locomotive of the Polish armoured train "Saper"

A kkStB series 97 locomotive armoured in late 1918 by the Zieleniewski factory in Cracov. This locomotive was used in 1919 as part of the Polish armoured train P.P.17 "Saper".

The chimney opening seems to be facing forwards, which not have helped the draught. Unless the loco was designed to normally travel backwards?

Reproduced, by permission, from Michal Derela's superb armoured-train website:
Do not miss!


Left: Another armoured locomotive with a stealth chimney

Armoured locomotives series MAV 377 were the standard locomotives of Austro-Hungarian WW1 armoured trains. This captured MAV 377 powered the Polish armoured train "Pilsudczyk" during the Polish-Soviet War. (February 1919 - March 1921) Jozef Klemens Pilsudski was a Russophobic military hero who was chief of state of the newly reconstructed Poland.

It is not too clear how the original chimney connected with the down-pipe.


Left: Triple stealth chimneys

This was a stealth experiment by Bullied; the idea was to disperse the exhaust quickly to make the train less visible to enemy aircraft. There was no significant improvement and it was quickly removed.

The loco is 4-6-0 No 783 Sir Gillimere of the "King Arthur" class, on the Southern Railway in Great Britain.

Photographed in the mid 1930's.


Above: The Adler locomotive of 1854 had a conventional chimney- unconventionally placed halfway along the boiler.

The Adler was one of the Eagle Class on the Baden State Railway, one of the pre-unification German state railways. It is a Crampton locomotive; ie it has a single pair of drivers at the rear. This was a popular configuration in Europe until heavier trains made its lack of adhesive weight too obvious.
The Adler had what was called a 're-entrant smokebox'. The firetubes ran to the front of the boiler as usual, but the combustion gases then doubled back over the top of the boiler to the base of the chimney. This apparently gave good spark-suppression at the expense of good steaming.
The Crampton layout, with the driving axle at the rear, allowed the cylinders to be halfway along the boiler and therefore conveniently placed for directing the exhaust up the chimney in the usual way.

Above: This monstrosity was another Crampton, and was built for the Camden & Amboy railway in the USA, in 1850.

The John Stevens shown above was designed by Robert Stevens, and dutifully named after his father; it was built by Norris Brothers. It was the first locomotive ever fitted with a six-wheel bogie; the next was in 1939, showing that this was something of a minority taste. The six wheels supported the boiler but left far too little adhesive weight for the rear 8-foot drivers. The locomotive was remodelled in 1856 and I can only hope it looked a bit less ugly afterwards. This is believed to be the oldest photograph of an American locomotive.

However, what concerns us here is the enormous chimney, designed to suppress sparks on the combustible grasslands of America. The two sideways projections at the bottom are for the removal of cinders.


Above: This South American 4-4-0 locomotive had a cowled chimney.

It is pictured here at the head of a tourist train between Tocna and Arica, in the desert Tacna territory, which was disputed between Peru and Chile until 1930. The motivation for the cowl is unknown; possibly it was meant to throw the smoke sideways away from the cab.

"The British Steam Railway Locomotive" Vol I, E L Ahrons. Pub Ian Allan.
"The Complete Book of Locomotives" Colin Garratt

Back to Home PageTop of this pageBack to The Loco Index