Three Cylinder Steam Locomotives
Andrew Waugh
>I have heard reference to some steam locomitve having three cylinders.
>What exactly is a three cylinder steam locomotive? Where was
>the third cylinder located?
Normal steam locomotives had two cylinders, one located on each side
of the locomotive. Three cylinder locomotives had a third cylinder
located between the frames. There were also four cylinder locomotives
with two cylinders located between the frames.
>What was the advatage to having the third cylinder?
Basically power. The maximum width of a steam locomotive is limited
by the loading gauge. This, in turn, imposes a limit on the maximum
cylinder diameter which can be used, which imposes an upper limit on
the amount of force a conventional steam locomotive can exert
(for a given boiler pressure). Adding a third cylinder increases this
force by 50% without infringing the loading gauge, increasing the
boiler pressure (i.e. making the boiler heavier) or increasing the
locomotive length (i.e. by using a Mallet or Garratt design). Three and
four cylinder locomotives were popular in countries with extremely
limited loading gauges (e.g. the UK).
A second advantage was that three and four cylinder locomotives were
much better balanced than two cylinder locomotives and hence gave a
much smaller dynamic blow to the track at speed. This allowed a
heavier locomotive to be built without needing to rebuild the track
or bridges.
There were several disadvantages to three cylinder locomotives. First,
the inside cylinder and motion was expensive to build and maintain.
Second, it was difficult to provide a sufficiently large bearing surface
in the big end of the inside connecting rod for the power being
transferred. This meant that this bearing often ran hot, and the
problem got worse as the locomotives got more powerful. Finally, it was
difficult to provide a suitable valve mechanism for the third cylinder.
A commonly used system was Gresley's conjugated valve gear which
derived the drive for the central valve from the two outside valves.
But the commonly used arrangement of this valve gear gave such
atrocious steam distribution that locomotives which used it were said
to have 'two and a half cylinders'.
andrew waugh
Darryl Van Nort
>In article <3jngu5$j...@newssvr.cacd.rockwell.com>,
>rlpe...@cacd.rockwell.com (ROBERT L PERKINS) wrote:
>
>* I have heard reference to some steam locomitve having three
>* cylinders. What exactly is a three cylinder steam locomotive? Where was
>* the third cylinder located? What was the advatage to having the third
>* cylinder?
>
>
>The third cylinder was located inbetween the two(you see) under the
>boiler in the middle of the frame. Advantages...more horsepower...but
>maybe not necessarily more tractive effort...ther is one t the Franklin
>Museum in Phila. and I think one out in California at the museum( a SP or
>UP engine) The one in Phila was the demonstrator ....form Baldwin I
>think...oh well
>
>--
>jpolaski @mcs.com or jpol...@aol.com
>A photographer, musician,woodworker, graphic designer and computer consultant...Just being a traveler on the information highway!
Just for added clarity...the valve mechanism and piston rod (via a crosshead) were
connected via a cranked axle. Made for interesting maintenance. The crank was
offset from the crank pins ...in fact, they may have been "thirded" instead of
quartered. I believe some of the 3-cylinder locos were compound, with the
3rd (center) cylinder being the low pressure cyl. If I recall, the UP had some, and
the Indiana Harbor Belt had at least two...on the two largest 0-8-0 drag locomotives
ever built. Not only were they three-cylinder locos, but they were equipped with
tender boosters as well. Quite a beefy chooch!
..Darryl Van Nort
dev...@mcs.com
Highballing the Internet Mainline
http://www.mcs.net/~devanno/
Jim Polaski
a...@conger.mel.dit.CSIRO.AU (Andrew Waugh) wrote:
* In article <3jngu5$j...@newssvr.cacd.rockwell.com>
rlpe...@cacd.rockwell.com (ROBERT L PERKINS) writes:
* >I have heard reference to some steam locomitve having three cylinders.
* >What exactly is a three cylinder steam locomotive? Where was
* >the third cylinder located?
*
* Normal steam locomotives had two cylinders, one located on each side
* of the locomotive. Three cylinder locomotives had a third cylinder
* located between the frames. There were also four cylinder locomotives
* with two cylinders located between the frames.
*
* >What was the advatage to having the third cylinder?
*
* Basically power. The maximum width of a steam locomotive is limited
* by the loading gauge. This, in turn, imposes a limit on the maximum
* cylinder diameter which can be used, which imposes an upper limit on
* the amount of force a conventional steam locomotive can exert
* (for a given boiler pressure). Adding a third cylinder increases this
* force by 50% without infringing the loading gauge, increasing the
* boiler pressure (i.e. making the boiler heavier) or increasing the
* locomotive length (i.e. by using a Mallet or Garratt design). Three and
* four cylinder locomotives were popular in countries with extremely
* limited loading gauges (e.g. the UK).
*
* A second advantage was that three and four cylinder locomotives were
* much better balanced than two cylinder locomotives and hence gave a
* much smaller dynamic blow to the track at speed. This allowed a
* heavier locomotive to be built without needing to rebuild the track
* or bridges.
*
* There were several disadvantages to three cylinder locomotives. First,
* the inside cylinder and motion was expensive to build and maintain.
* Second, it was difficult to provide a sufficiently large bearing surface
* in the big end of the inside connecting rod for the power being
* transferred. This meant that this bearing often ran hot, and the
* problem got worse as the locomotives got more powerful. Finally, it was
* difficult to provide a suitable valve mechanism for the third cylinder.
* A commonly used system was Gresley's conjugated valve gear which
* derived the drive for the central valve from the two outside valves.
* But the commonly used arrangement of this valve gear gave such
* atrocious steam distribution that locomotives which used it were said
* to have 'two and a half cylinders'.
*
* andrew waugh
=====
But even though you had more power, you didn;t get a linear proportion
more weight, so I don;t think you'd have a linear increase in "tractive
effort"...but the engine would also use more coal and water( because of
the extra cylinder) ..so was the gain worth it...form maintenance, and
fuel costs reative to availability, repairs and fuel cost?
hmmmm...
Jim Polaski
rlpe...@cacd.rockwell.com (ROBERT L PERKINS) wrote:
* I have heard reference to some steam locomitve having three
* cylinders. What exactly is a three cylinder steam locomotive? Where was
* the third cylinder located? What was the advatage to having the third
* cylinder?
The third cylinder was located inbetween the two(you see) under the
boiler in the middle of the frame. Advantages...more horsepower...but
maybe not necessarily more tractive effort...ther is one t the Franklin
Museum in Phila. and I think one out in California at the museum( a SP or
UP engine) The one in Phila was the demonstrator ....form Baldwin I
think...oh well
--
Darryl Van Nort
>
>But even though you had more power, you didn;t get a linear proportion
>more weight, so I don;t think you'd have a linear increase in "tractive
>effort"...but the engine would also use more coal and water( because of
>the extra cylinder) ..so was the gain worth it...form maintenance, and
>fuel costs reative to availability, repairs and fuel cost?
>
>
>hmmmm...
As others have indicated, the benefits are in starting power and low speed
operation, where having six power cycles per revolution reduces the chances
of slipping. You're right...you don't get a linear increase in TE. In
the case of compound locos, the third cylinder was running off exhaust steam
from the first two, so it wasn't all that costly from a fuel/water standpoint.
If your road was running heavy, slow-speed drag freights, the savings in
maintenance from not slipping (not to mention the savings in fuel and the
temper of the fireman (!)) may have made up for the cost of using them.
..Darryl
-F.KERFOOT
>jpol...@mcs.com (Jim Polaski) wrote:
>>In article <3jngu5$j...@newssvr.cacd.rockwell.com>,
>>rlpe...@cacd.rockwell.com (ROBERT L PERKINS) wrote:
>>
>>* I have heard reference to some steam locomitve having three
>>* cylinders. What exactly is a three cylinder steam locomotive? Where was
>>* the third cylinder located? What was the advatage to having the third
>>* cylinder?
>>
>>The third cylinder was located inbetween the two(you see) under the
>>boiler in the middle of the frame. Advantages...more horsepower...but
>>maybe not necessarily more tractive effort...ther is one t the Franklin
>>Museum in Phila. and I think one out in California at the museum( a SP or
>>UP engine) The one in Phila was the demonstrator ....form Baldwin I
>>think...oh well
>
>connected via a cranked axle. Made for interesting maintenance. The crank was
>offset from the crank pins ...in fact, they may have been "thirded" instead of
>quartered. I believe some of the 3-cylinder locos were compound, with the
>3rd (center) cylinder being the low pressure cyl. If I recall, the UP had some, and
>the Indiana Harbor Belt had at least two...on the two largest 0-8-0 drag locomotives
>ever built. Not only were they three-cylinder locos, but they were equipped with
>tender boosters as well. Quite a beefy chooch!
The 3 cylinder Baldwin at the Franklin Institute in Philadelphia is in fact
a compound. It was an experimental locomotive, and had other interesting
features as well, like a water tube boiler. Most other US 3 cylinder
locomotives post-dated the time when compounding was in vogue, and were
simple expansion engines. As you suggest, the crankpins were at 120 degree
intervals on most if not all of the later designs. You commented that the
valve gear for the center cylinder was driven by the cranked axle. This
was true for a few designs (notably the Reading 3 cylinder Atlantics and
a 10-wheeler, of 1909-1912, which used Joy valve gear operating from
the center cylinder connecting rod to operate the center cylinder valve
gear). However, most other 3 cylinder locomotives used a variation of the
Gresley approach, where the valve motions for the center cylinder were
derived from the motions of the 2 outside cylinder valve gears, which in
turn were developed by a traditional valve gear such as Walchaerts or
Baker. Thus the valve motion for the center cylinder had no direct
connection to the cranked axle. Most later designs had the center cylinder
somewhat higher and inclined relative to the outside cylinders, so the
connecting rod would clear the front driven axle and could be connected
to a cranked axle on the same axle where the outside crankpins were
connected (typically the second driven axle). The Reading 3 cylinder
engines did not follow this approach, but had all three cylinders in
the same plane, and coupled the center cylinder to the first driven axle
(which was cranked), and the outside cylinders to the 2nd driven axle.
Thus the center cylinder had more extreme connecting rod angularity than
the outside rods. BTW, another significant benefit of 3 cylinder engines
was they they had lower peak tractive effort relative to the average,
which helped with slipping (other things being equal).
Frank Kerfoot
f...@hogpa.att.com
Jobst Brandt
> I have heard reference to some steam locomotive having three
> cylinders. What exactly is a three cylinder steam locomotive? Where
> was the third cylinder located? What was the advantage to having the
> third cylinder?
Locally, the SP had an SP-Type (5200 series) 4-10-2 with the third
cylinder in the center between the other conventionally located
cylinders. They were timed to give six stroke per revolution but had
a strongly syncopated exhaust. This arose from the center cylinder
that was higher than the others and apparently sloped at about 15
degrees above the horizontal to clear the the first axles with the
cross head and caused the unusual characteristic exhaust rhythm.
Obviously this does not increase the power of the locomotive because
that is limited by the fire box size, and it doesn't appear to affect
tractive effort but it does. More strokes per revolution means that
the power peaks of each stroke, on average, can be lower and thereby
prevent wheelspin at low speeds where the Johnson bar is nearly at
full forward. Shay locomotives with their 12+ strokes per revolution
of the drivers demonstrates this admirably.
Jobst Brandt <jbr...@hpl.hp.com>
Craig Dewick
> I have heard reference to some steam locomitve having three
>cylinders. What exactly is a three cylinder steam locomotive? Where was
>the third cylinder located? What was the advatage to having the third
>cylinder?
A 3 cylinder is exactly that - it has three steam cylinders instead of
the usual two. Normally the third cylinder was located underneath the
smokebox between where the two main cylinders lived.
The advtanage of a 3rd cylinder is that you then can have at least one
additional power stroke per revolution of the driving wheels. I'm not sure
how the timing worked for the 3rd cylinder, but suffice to say that if a 3
cylinder loco had the 3rd cylinder removed it's maximum tractive effort
would fall by around 30 percent.
Here is NSW, Australia, we had 3 cylinder 'mountain' class steam
locomotives from the mid 19020's. The (D)57 class, as they were known,
were the most powerful steam locomotives in this state right up until the
early 1950's when the (AD)60 class Beyer-Garratts appeared on the scene.
The fact that the (D)57's had a third cylinder gave the locos enormous
grunt, but they consumed fuel and water at a much faster rate that any
comparable 2 cylinder engine. Fortunately for the firemen they were fitted
with automatic stokers which worked most of the time.
The (D)57's used 'conjugating' valve gear for the 3rd cylinder, the main
component of which was a long transverse lever connected between the
rodding of the two main cylinders in a 2:1 ratio.
There was also the (D)58 class of 3 cylinder steam locos, built as a
modification to the (D)57 design during the years following WW2. They were
a dismal failure in service because of continuous trouble with the tuning
of the mechanics controlling the valve gear for the 3rd cylinder. These
locos used a partially geared, partially levered setup to operate the
valve gear for the 3rd cylinder instead of the 'conjugating' gear as
fitted to the (D)57's.
I don't have info on whether any other Australian systems has 3 cylinder
steam locos.
Regards,
Craig.
--
Craig Dewick (cr...@orb.apana.org.au) - Passenger train driver, Sydney, Aust.
Support *independent* media <--> Support your local community radio station
Swimming in the MUSIQUARIUM of life <--> Monday 2 - 4pm, 2SER FM (107.3 MHz).
Always striving for a secure long-term future in an insecure short-term world
Andrew Waugh
>In article <3jo4tr$s...@weever.mel.dit.csiro.au>,
>a...@conger.mel.dit.CSIRO.AU (Andrew Waugh) wrote:
>* In article <3jngu5$j...@newssvr.cacd.rockwell.com>
>rlpe...@cacd.rockwell.com (ROBERT L PERKINS) writes:
>* >What was the advatage to having the third cylinder?
>*
>* Basically power. The maximum width of a steam locomotive is limited
>* by the loading gauge. This, in turn, imposes a limit on the maximum
>* cylinder diameter which can be used, which imposes an upper limit on
>* the amount of force a conventional steam locomotive can exert
>* (for a given boiler pressure). Adding a third cylinder increases this
>* force by 50% without infringing the loading gauge, increasing the
>* boiler pressure (i.e. making the boiler heavier) or increasing the
>* locomotive length (i.e. by using a Mallet or Garratt design). Three and
>* four cylinder locomotives were popular in countries with extremely
>* limited loading gauges (e.g. the UK).
>*
>* A second advantage was that three and four cylinder locomotives were
>* much better balanced than two cylinder locomotives and hence gave a
>* much smaller dynamic blow to the track at speed. This allowed a
>* heavier locomotive to be built without needing to rebuild the track
>* or bridges.
>
>more weight, so I don;t think you'd have a linear increase in "tractive
>effort"
First, tractive effort was calculated without reference to weight, so
adding a third cylinder would increase the *calculated* tractive effort
proportionally. However, the *usable* tractive effort was limited to
approximately 25% of the adhesive weight of the locomotive, so a three
cylinder locomotive had to be heavier than the equivalent two cylinder
locomotive. This weight increase would be achieved in several ways:
* fitting a larger and more powerful boiler (which was needed
anyway because of the extra steam demand of the third cylinder).
* transfering weight from the leading and trailing trucks to the
driving wheels.
In addition, the 25% could be shaved to 28% or even 33%. The more even
turning impulses of the three cylinder design would mean that the
locomotive was less likely to slip than a two cylinder locomotive.
Implicit in your comment, of course, is the realisation that the
maximum weight of the locomotive is fixed by the weight the track
and bridges can support. But some countries, notable the UK, had
such a restrictive loading gauge that the cylinder dimensions were
more restricted than the weight of the locomotive. This was helped
by the realisation in the early twenties that the 'hammer' blow of
a two cylinder locomotive could mean than the instaneous weight on
a wheel when running at speed could be double the static weight. Since
three cylinder locomotive eliminated this 'hammer' blow, civil engineers
would accept a greater axle load from a three cylinder locomotive than
a two cylinder locomotive.
All this meant that it was easy to use the extra power that the third
cylinder provided in some countries. In others, where the cylinder
dimensions were not so restricted, it was harder to use the extra
power. The US was such a country, and three cylinder locomotives were
rare there.
>...but the engine would also use more coal and water( because of
>the extra cylinder)
The same argument could be made about the simple mallets which had
four cylinders. Generally, railways discovered that the cost per ton
mile reduced as more powerful locomotives were introduced.
>..so was the gain worth it...form maintenance, and
>fuel costs reative to availability, repairs and fuel cost?
In the US? Generally no - and few three cylinder locomotives were built.
In the UK the answer is more complicated. Between the wars, the answer
was yes and most large locomotives built in that period had three or
four cylinders. In the post war era, the railways clearly decided that
the answer was no. Wages had increased, hence simpler designs were
preferred. Improved boiler making techiques (particularly welding)
meant that increased power could be achieved by increasing boiler
pressure without a prohibitive increase in weight.
andrew waugh
Christopher A. Lee
>
>The 3 cylinder Baldwin at the Franklin Institute in Philadelphia is in fact
>a compound. It was an experimental locomotive, and had other interesting
>features as well, like a water tube boiler. Most other US 3 cylinder
>locomotives post-dated the time when compounding was in vogue, and were
>simple expansion engines. As you suggest, the crankpins were at 120 degree
>intervals on most if not all of the later designs. You commented that the
>valve gear for the center cylinder was driven by the cranked axle. This
>was true for a few designs (notably the Reading 3 cylinder Atlantics and
>a 10-wheeler, of 1909-1912, which used Joy valve gear operating from
>the center cylinder connecting rod to operate the center cylinder valve
>gear). However, most other 3 cylinder locomotives used a variation of the
>Gresley approach, where the valve motions for the center cylinder were
>derived from the motions of the 2 outside cylinder valve gears, which in
Which he got from Harry Holcroft.
Gersley's original 3-cylinder 2-8-2 used heavily inclined cylinders and an
extremely convoluted method to derive the valve motion for the third
cylinder. He'd done this to get increased cylinder power withing the
confines of the British clearences (high station platforms restrict the
size of outside cylinders) but he found that the engine started better
under load and ran smoother.
Holcroft was an ex-GWR (very senior) draughtsman who went to work for
Maunsell at Ashford, who had worked out the math for a simple derived
motion during earlier hospitalisation, as a theoretical exercise.
On most railways, the general principals were the responsibility of the
chief mecahnical engineer who gave his name to the design, like
Churchward, Drummond, Gresley or Maunsell.
But the detail work was done by a team of draughtsmen and checked by
people who added their signatures all the way up to the top.
So Holcroft was actually responsible for several GWR designs, including
the last 0-6-0 saddle tanks (which were actually developed from earlier
West Cornwall Railway engines supplied by an outside builder). Also the
very successful Churchward 2-6-0s.
Stanier was another another of Churchwards impressive staff at this time.
Gresley described his extremely complex derived motion on his first
3-cylinder engine in the Engineer, and Holcroft wrote a letter of
constructive criticism suggesting a better way, which was published.
The result of this was that Holcroft was invited to address the
institute of locomotive engineers (in England a locomotive engineer is
somebody who builds engines, not the driver). This was the professional
body whose membership included all the great names of locomotive design.
Gresley wasn't at this lecture, nor a subsequent one to the design team
at Doncaster but the upshot was a meeting between Holcroft and Gresley.
Gresley told Holcroft that his simple solution was the answer to all his
problems with the complicated design and that thereafter he would only
build 3-cylinder engines.
Shortly afterwards the LNER approached the SECR to ask whether they could
talk to Holcroft about a move to Doncaster (they were Gentlemen in those
days).
Maunsell said no, and that he was going to build 3 cylinder engines as well.
Holcroft had been heavily involved in the design of some of the SECR's
2-cylinder engines and these were variants.
Holcroft's 3-cylinder engines took the motion from the valve rod at the rear
of the cylinders, using a rod passing outside the valve to the front of
the engine.
Gresley simply extended a rod from the front of the piston valve to do
the same thing.
In engineering terms this is not so sound because it requires an extra
steam seal and the length of the rod varies due to the temperature of
the stream in the valve. This error is magnified by the lever effect
when it drives the other cylinder, but I'm sure the design took this into
account.
The uneven loads from the conjugation levers put a heavy strain
on the vertical bearings so they needed careful maintenance. But
this wasn't a problem until wartime when they couldn't give them
the level of maintenance they required.
Also one of the biggest differences between British and US use was that
the levers and their bearings were covered by the continuous extended
running plate in England and protected from the gritty ash and char emptied
from the smokebox, whereas the levers and bearings were exposed in the US
engines. This accellerated the wear and hence slop at the bearings.
Originally the motion was called Holcroft-Gresley after the designer and
the biggest user.
The LNER were probably the biggest user of derived motion for the third
cyliner anywhere.
But by this time most express passenger engines in England used three
cylinders anyway because of the smoother running and reduced hammer blow.
The exception was the GWR who used 4 cylinders with two sets of inside
Walschaerts valve gear, deriving the motion for the outside cylinders;
and the Stanier Pacifics on the LMS which derived the motion for the
two inside cylinders from the outside motion.
But most of the 3-cylinder engines except on the LNER used three separate
sets of Walschaerts valve motion. The inside cylinder and motion need
to be inclined so unless the outside cylinders are equally inclined
it's not quite so smooth running. Which gets away from Stevenson's
discovery that the cylinders should be as horizontal as possible
(the Rocket lost its steeply angled cylinders -in a rebulid pretty soon
after it was built). The Holcroft/Gresley approach was an answer to this.
The interesting thing is that according to Strapac's book on the Union
Pacific 3-cylinder engines, the guys from UP and the manufacture came to
England to visit Gresley. He says that while they were there, Gresley
took them to "Mavsell" (a typo for Maunsell?). The only reason I can
think of, would be to meet the originator the derived motion, who was
one of Maunsell's senior assistants.
Although Holcroft doesn't mention it in his memoires.
You don't introduce important foreign visitors to a grunt without going
through his boss. Especially when the boss previously refused permission to
offer him a job.
Tobias Benjamin Koehler
: * I have heard reference to some steam locomitve having three
: * cylinders. What exactly is a three cylinder steam locomotive? Where was
: * the third cylinder located? What was the advatage to having the third
: * cylinder?
: The third cylinder was located inbetween the two(you see) under the
: boiler in the middle of the frame. Advantages...more horsepower...but
: maybe not necessarily more tractive effort...ther is one t the Franklin
: Museum in Phila. and I think one out in California at the museum( a SP or
: UP engine) The one in Phila was the demonstrator ....form Baldwin I
: think...oh well
The axles of three cylinder locomotives are not straight, hence
they are more expensive to build, and lubricating/maintaining
the inner drive is more difficult. There is an angle of 120
degrees between the 3 drives (and not 90 degrees of a two- or
four-cylinder locomotive). Examples in Germany are 01.10 and 44
(of these classes, several are still working as museum
locomotives), there are many more.
I believe they have been less popular in the USA as the wider
loading gauge allowed to build bigger cylinders and reach the
same power with a cheaper two-cylinder locomotive.
toby
--
tobias benjamin köhler ,-/o"O`--.._ _/(_
_,-o'.|o 0 'O o O`o--'. e\
s_...@ira.uka.de (`o-..___..--''o:,-' )o /._" O "o 0 o : ._>
``--o___o..o.'' :'.O\_ ```--.\o .' `--
<<la propriété, `-`.,) \`.o`._
c'est le vol>> - p j proudhon pic: felix lee `-`-.,)
Darryl Van Nort
>In article <D58Ku...@nntpa.cb.att.com> f...@hogpb.ho.att.com (-F.KERFOOT) writes:
>>
>>The 3 cylinder Baldwin at the Franklin Institute in Philadelphia is in fact
>>a compound. It was an experimental locomotive, and had other interesting
>>features as well, like a water tube boiler. Most other US 3 cylinder
>>locomotives post-dated the time when compounding was in vogue, and were
>>simple expansion engines. As you suggest, the crankpins were at 120 degree
>>intervals on most if not all of the later designs. You commented that the
>>valve gear for the center cylinder was driven by the cranked axle. This
>>was true for a few designs (notably the Reading 3 cylinder Atlantics and
>>a 10-wheeler, of 1909-1912, which used Joy valve gear operating from
>>the center cylinder connecting rod to operate the center cylinder valve
>>gear). However, most other 3 cylinder locomotives used a variation of the
>>Gresley approach, where the valve motions for the center cylinder were
I have an image posted of a 3-cylinder U.S. chooch...New Haven 3559, a 4-8-2.
You can see it in my WWW pages under "The Roundhouse" "Mountains."
http://www.mcs.net/~devanno/index.html
will get you into "The Roundhouse" selecting "View Images" will get you into
loco classes, selecting "Mountains" will get you into a group of images of 4-8-2s.
Netscape or the OS/2 Warp browser are probably the best to view these with, as
the pages use in-line JPGs.
Nick Leverton
>In the case of compound locos, the third cylinder was running off exhaust steam
>from the first two, so it wasn't all that costly from a fuel/water standpoint.
That turned out not to be the ideal configuration for a three cylinder
compound. The infamous Webb compounds, which we discussed last year,
used this arrangement; Even with two quite small high pressure
cylinders, though, the middle low pressure cylinder rapidly became
constrained in diameter by the distance between the frames; and hence
"strangled" the throughput of steam.
Johnson on the Midland Railway turned the idea around : he had a single
high pressure cylinder feeding two low pressure cylinders. This avoided
the problem of steam throughput at high speeds. To maintain tractive
effort at starting, he used a special regulator (US=throttle) valve
(patented by one Smith, whose details I forget) which upon first opening
admitted high pressure steam to all three cylinders. When it was opened
to the position known as "second valve", the regulator cut off the
direct supply of steam to the outside cylinders and diverted the inside
cylinder's exhaust to them. Thereafter this mode was maintained until
the regulator was closed and re-opened.
This was of course a relatively complex arrangement, but as the Midland
took care of their loco's it worked well. Johnson's five loco's were the
precursors of the Smith/Deeley Compounds, which in turn were propagated
by Fowler of the Midland and later the LMS and which remained in service
until near the end of steam. Johnson's loco's were handsome late
Victorian designs; Deeley rebuilt them into the somewhat plainer form in
which the first, Midland Railway no. 1000, survives today. But at least
we do still have a Midland three cylinder compound to drool over !
Nick Leverton
Wes Barris
rlpe...@cacd.rockwell.com (ROBERT L PERKINS) wrote:
* I have heard reference to some steam locomitve having three
* cylinders. What exactly is a three cylinder steam locomotive? Where was
* the third cylinder located? What was the advatage to having the third
* cylinder?
See http://www.arc.umn.edu/~wes/philly/3cylinders.html for information
on a few surviving 3 cylinder steam locomotives.
--
o O O O O o o o o o o o o . . . . . . .
__________ ___ ___ __ _
/_______|--`--|_ _|--=-==---|_ _|--,_|__|_ |-+-------`____________,
| ||_| / - | / - \ | || ________ | |
|_|___/------------|----__---------|------|||__|__|__| |SOUTHERN PACIFIC LINES|
|_|____________/-------{__}---------------{__}_______|=|_________ ________|
/_(o)=(o) ( )( )( )( ) [__] ( )( )( )( ) [__]--(o) (o)(o)(o)-----(o)(o)(o)
===============================================================================
Wes Barris Web: http://www.arc.umn.edu/~wes/steam.html
Minnesota Supercomputer Center, Inc. PH: (612) 626-8090 Email: we...@msc.edu
Roger Mitchell
moved per ton of coal, the most efficient steam locos that the UP
had. Many lasted until the end of steam on the UP. There were a
couple of different valve gears used on them, the Gresley and a
variation of the Walschaerts gear with 2 gear on the engineer's
side and one on the fireman's side.
Roger Mitchell
FCMRy
David Bromage
>There was also the (D)58 class of 3 cylinder steam locos, built as a
>modification to the (D)57 design during the years following WW2. They were
>a dismal failure in service because of continuous trouble with the tuning
>of the mechanics controlling the valve gear for the 3rd cylinder. These
>locos used a partially geared, partially levered setup to operate the
>valve gear for the 3rd cylinder instead of the 'conjugating' gear as
>fitted to the (D)57's.
I have read that the gears had to be retimed after only a few hundred miles
of running.
>I don't have info on whether any other Australian systems has 3 cylinder
>steam locos.
VR had the S class 4-6-2 and H class 4-8-4.
Chers
David
A R BREEN
Nick Leverton <n...@rb.icl.co.uk> wrote:
>In article <3jql37$p...@news1.mcs.com>, Darryl Van Nort <dev...@mcs.com> wrote:
>>In the case of compound locos, the third cylinder was running off exhaust steam
>>from the first two, so it wasn't all that costly from a fuel/water standpoint.
>
>That turned out not to be the ideal configuration for a three cylinder
>compound. The infamous Webb compounds, which we discussed last year,
>used this arrangement; Even with two quite small high pressure
>cylinders, though, the middle low pressure cylinder rapidly became
>constrained in diameter by the distance between the frames; and hence
>"strangled" the throughput of steam.
>
>Johnson on the Midland Railway turned the idea around : he had a single
>high pressure cylinder feeding two low pressure cylinders. This avoided
>the problem of steam throughput at high speeds. To maintain tractive
>effort at starting, he used a special regulator (US=throttle) valve
>(patented by one Smith, whose details I forget) which upon first opening
>admitted high pressure steam to all three cylinders. When it was opened
>to the position known as "second valve", the regulator cut off the
>direct supply of steam to the outside cylinders and diverted the inside
>cylinder's exhaust to them. Thereafter this mode was maintained until
>the regulator was closed and re-opened.
Fred Smith, chief draughtsman of the North Eastern Railway in the
early 1900s..
The NER had been playing with compounding for some years in the
1880-1900 period, generally using the Von Borries system (2
cylinders, both inside the frames. Inside Stephenson valve gear
working via rocking levers to outside steam chests - Aerolite
in the NRM at York is the last surviving NER Von Borries Compound,
tho' the 4-4-0 no. 1629 was built as one). There proved to be
very little advantage in this system & under Smith's tutelage
the new CME, Wilson Worsdell, went over to high pressure simples
(2-cyl, inside).
Smith however also peristed with compounding, developing the 3-cyl
system subsequently used by the Midland.. However there were problems
with the prototype at high power output (the NER tended to trash
its engines) as the steam pipes could not be made large enough
to pass steam from the single (inside) HP cylinder to the two
(outside) LP cylinders. So not ideal & the prototype remained
unique (on the NER anyway).
Now.... about 1900 Worsdell visited America & was impressed by
the large Atlantics (4-4-2s) running on (I think) the Pennslyvania
system. When he got back he designed a series of large 2-cyl
simple atlantics (the V class). Powerful for their day but
coal eaters.
Smith was ill during the design of the V's.... when he returned
he was outraged that a design had been produced in his absence
and insisted on the building of 4 atlantics to *his* design.
These had 4 cylinders, 2 HP inside & 2 LP outside. 2 sets of
Stephenson gear worked the 4 cylinders via rocking levers.
All four cylinders drove the leading coupled axle - so the
design was clearly different from the De Glehn compounds
of the same period.
This type (the 4CC class) were very successful - a great
improvement on the V's but were not repeated as Smith died
just after they were introduced. Worsdell retired around the
same time & the new CME, Vincent Raven, prefered 3-cyl simple
expansion designs (3 sets of Stephenson gear, all inside).
The atlantics built to his designs (Z class) were *very* good -
superheating made them every bit the equal of the 4CCs. The Z;s
had all 3 cylinders in line, driving cranks set at 120 degrees
on the leading coupled axle.
AFIK Raven didn't think the compound design was worth developing
even though it promised *very* high power outputs because his
aim was electrification of the NER main line between York
and Newcastle... this didn't happen (until 1992)
The 4CCs were still active well into the 1930s.
"Without us, chickens would devastate the countryside..."
*****************************************************************
* Andy Breen Adran Ffiseg/Physics Department *
* Grwp EISCAT group Prifysgol Cymru Aberystwyth *
*****************************************************************
* PC-A pay me: they don't have to share my opinions. Yet. *
*****************************************************************
RICHARD ALAN HUSS
[good stuff about the Midland compounds]
> Johnson's loco's were handsome late
> Victorian designs; Deeley rebuilt them into the somewhat plainer form in
> which the first, Midland Railway no. 1000, survives today. But at least
> we do still have a Midland three cylinder compound to drool over !
>
> Nick Leverton
We also still have a Great Northern Railway (Ireland) V class No. 85
(a 4-4-0 compound working on the same Smith's system) in working
order, on the main line :-)
Sky-blue 4-4-0s for ever!
Richard A Huss (ra...@aber.ac.uk)
+------------------------------------------------+
|Talyllyn Railway Company |
|Wharf Station, Tywyn, Gwynedd, Wales. LL36 9EY |
|Phone: (01654) 710994 |
|Fax: (01654) 711755 |
+------------------------------------------------+
| The World's First Preserved Railway |
| |
| WWW Timetable & info available on: |
| http://www.dcs.aber.ac.uk/~rah94/talyllyn.html |
+------------------------------------------------+
Donald Nelson -FT-~
Let's not forget all those wonderful <Lima> Shay and Willamette Shay locomotives --
The class B and C all had 3 cylinders connected to the gear train!
--
ALL OPINIONS EXPRESSED ARE MY OWN AND NOT NECESSARILY THOSE OF INTEL CORP.
DON NELSON
Carol & Pete Jobusch
>
>Robert Perkins writes:
>
>> I have heard reference to some steam locomotive having three
>> cylinders. What exactly is a three cylinder steam locomotive? Where
>> was the third cylinder located? What was the advantage to having the
>> third cylinder?
>of the drivers demonstrates this admirably.
>
>Jobst Brandt <jbr...@hpl.hp.com>
Glad you mentioned Shay locomotives ... most of them had three
cylinders, and not inside the frame!
The German class 44 and 45 were three cylinder locomotives, and got
spread around Europe after the war (i.e. a.k.a. SCNF 150X). Having
been built in 1926-1944 these would be relatively late designs
compared to the US, where I think the three cylinder designs were built
mostly before 1930. The Aster model of the BR44 sounds like the
valve timing is really messed up when it runs with six valve events
instead of the "normal" four, but is a smooth runner and really
pulls!
A R BREEN
-F.KERFOOT <f...@hogpb.ho.att.com> wrote:
>.................... As you suggest, the crankpins were at 120 degree
>intervals on most if not all of the later designs. You commented that the
>valve gear for the center cylinder was driven by the cranked axle. This
>was true for a few designs (notably the Reading 3 cylinder Atlantics and
>a 10-wheeler, of 1909-1912, which used Joy valve gear operating from
>the center cylinder connecting rod to operate the center cylinder valve
>gear). However, most other 3 cylinder locomotives used a variation of the
>Gresley approach, where the valve motions for the center cylinder were
>derived from the motions of the 2 outside cylinder valve gears, which in
>turn were developed by a traditional valve gear such as Walchaerts or
>Baker.
This might have been true in the US but in Europe (inc. Britain,
wrth gwrs..) most 3-cylinder steam locomotives used 3 independent
sets of valve gear..
Certainly the Midland Railway's 3-cyl. compounds (a successful
& long lived type, using the Smith/Deeley system) used three
sets of Stephenson gear, all inside. So did the various North-Eastern
Railway designs (4-4-2, 4-6-0, 0-8-0) - but of course what else
course you fit at Darlington? The NER 0-8-0s were still hard at
work around Sunderland in the later 1960s.. I remember them as
being much quieter & smoother in their running than the 2-cylinder
WD and BR 2-10-0s (which were ~30 years younder, at the least).
The Southern Railway's Schools 4-4-0s had 3 sets of Walchaerts gear,
2 outside and one inside, as did the earlier Southern moguls and
2-6-4 tanks (the ill-fated Rivers) - with a couple of exceptions
which briefly used the Holcroft 2:1 derived gear. This was replaced
with 3 sets of Walschaerts after Gresley claimed patent infringement.
The Royal Scots, Patriots and Jubilees of the LM&S all used 3 sets
of Walschearts (2 out, 1 in). I'm not honestly sure what the Webb
compounds used 2 sets of Joy gear & a slip eccentric perhaps?
The Pickersgill 4-6-0s built for the Caledonian in 1919 started
with a derived gear of Pickergill's own confection (which didn't
work) and finished with 2 sets of Walschearts gear and 1 of Stephenson,
all inside (I think). This didn't work either.
The L&NER were the only big british users of derived valve gear
on 3-cyl engines (almost everyone used rocker gear for 4-cyl
designs) but even there it wasn't universal.. Gresley built
a number of 4-4-0s (Hunts?) in the 1930s which had 3 sets
of Capriotti (sp) gear driving poppet valves. The same arrangement
was used in some of the big P2 2-8-2s. Thompson (Gresley's
successor) rebuilt quite a lot of Gresley engines with 3 sets
of Walschearts .. all his own three cyl engines had 3 sets of gear,
as did those of Peppercorn (Thompson's successor).
So in the UK the jury definitely remained unconvinced on the
merits of derived valve gear...
I believe that in the rest of europe 3 sets of Walschearts
was the usual system, but I'm not able to quote examples in
the same way.
-F.KERFOOT
>...
>Now.... about 1900 Worsdell visited America & was impressed by
>the large Atlantics (4-4-2s) running on (I think) the Pennslyvania
>system. When he got back he designed a series of large 2-cyl
>simple atlantics (the V class). Powerful for their day but
>coal eaters.
In 1900, it was the Atlantics of the Philadelphia and Reading (later
the Reading Co.) which were the rage. These were Vauclain compound,
Wooten boilered camelbacks, built by Baldwin. They were regularly beating
the Pennsy's D16sb's (4-4-0) on the two line's parallel racetracks between
Phila. and Atlantic City, NJ. The Pennsy. originally responded with
a virtual copy of the P&R engines, designated the E1. These were the
Pennsy's first, and only, camelbacks, and were the the ancestors's
of the Pennsy's famous E6 Altantics, which eventually (1914) were
arguably the finest American Atlantics ever built. However, in 1900,
it was the P&R leading the show, and the P&R stayed in the contest
through the first decade or so of the century.
Frank Kerfoot
f...@hogpa.att.com
John Wilson
: Frank Kerfoot
: f...@hogpa.att.com
This brings back some memories. When I was in elementary school I
took ham radio classes at the Franklin Institute Saturday
mornings and then spent the rest of the day hanging around the
Institute. Got to know the 60000 very well. As far as I know, the
valve gear for this engine is unique. The middle cylinder valves
are operated by a third Walschaerts link, which is driven from
the crosshead of one of the outside cylinders. I think it was the
engineer's side crosshead, but I'm not sure. The third link is
coaxial with the engineer's side link. A rocker shaft transmits
the motion from the third link into the innards of the engine
where you can't see without getting under in a pit, which the
Franklin Institute doesn't have (too bad). Must have been a
marvel to see this machine go by!
The engine was way ahead of its time in other ways, too. For
instance; water-tube firebox, thermic syphons, 350psi steam
pressure (about 100-150 psi higher than common practice at the
time), and other technological whizzbangs that I don't remember.
73,
JohnW (K3KXJ)
A R Hook
: Robert Perkins writes:
: Jobst Brandt <jbr...@hpl.hp.com>
For what it's worth, don't forget that the World Steam Record holder,
Mallard, is a 3-cylynder loco. The A4 pacifics do sound odd to anyone
weaned on GWR 2- & 4-cylynder locos. I always understood that this
was due to the conjugated lever set-up which activates the valve of
the inside cylynder.
Andrew. (Andrew...@bris.ac.uk)
slj...@delphi.com
were not a major success.
Sohnson
Bill Donald
|>
|> ------snip--------
|>
|> This might have been true in the US but in Europe (inc. Britain,
|> wrth gwrs..) most 3-cylinder steam locomotives used 3 independent
|> sets of valve gear..
|> Certainly the Midland Railway's 3-cyl. compounds (a successful
|> & long lived type, using the Smith/Deeley system) used three
|> sets of Stephenson gear, all inside. So did the various North-Eastern
|> Railway designs (4-4-2, 4-6-0, 0-8-0) - but of course what else
|> course you fit at Darlington? The NER 0-8-0s were still hard at
|> work around Sunderland in the later 1960s.. I remember them as
|> being much quieter & smoother in their running than the 2-cylinder
|> WD and BR 2-10-0s (which were ~30 years younder, at the least).
|>
|>
Sorry Andy, but you're way out of bed on some of your statements here. The ex-NER
0-8-0 "around Sunderland in the later 1960s" were NOT 3-cylinder locomotives -
they were the 2-cylinder Q6 class. You are confusing them with the later Q7 class
which were withdrawn by the end of 1962/3.
Anyone who is interested in the locomotive history of the LNER and its
predecessors should purchase the 11 volume "Locomotives of the LNER" book set,
published by the RCTS. This is the primary reference source for the subject.
I think you'll also find any BR steam running shed fitter will give you some
rather pungent views on the maintenance of THREE inside sets of valve gear.
Happily, Gresley saw the light and avoided this layout - but even he is not above
criticism when you consider his middle big-end design on the A3s and A4s.
BTW, the BR 2-10-0s WERE 3-cylinders.
--
Bill Donald don...@ecid.cig.mot.com
Documentation Group Tel: (44) 017 9350 0088
Motorola ECID FAX: (44) 017 9354 1226
Swindon, SN5 8YQ Home email: w...@farnbell.demon.co.uk
Wilts. UK G7ENQ
A R BREEN
Bill Donald <don...@merlin.ecid.cig.mot.com> wrote:
>Sorry Andy, but you're way out of bed on some of your statements here. The ex-NER
>0-8-0 "around Sunderland in the later 1960s" were NOT 3-cylinder locomotives -
>they were the 2-cylinder Q6 class. You are confusing them with the later Q7 class
>which were withdrawn by the end of 1962/3.
Apologies on this one.. I checked the RCTS book when I got home and f'sure
the Q7s had gone by the time I was old enough to be taken out. Must have been
th T2s (Q6) I remember.
>
>I think you'll also find any BR steam running shed fitter will give you some
>rather pungent views on the maintenance of THREE inside sets of valve gear.
>Happily, Gresley saw the light and avoided this layout - but even he is not above
>criticism when you consider his middle big-end design on the A3s and A4s.
I never said that 3 sets of inside gear were easy to maintain - only that
some successful and long-lived designs used this approach. I certainly
wouldn't say that multiple sets of inside valve gear were a good idea: if
nothing else the L&Y 4-6-0s of the early years of this century proved that
(4 sets of Joy gear, all inside....). Not a howling success.
>
>BTW, the BR 2-10-0s WERE 3-cylinders.
Not Evening Star when I looked at her from underneath in York a few
years ago... Nuttin' in the middle there.
All BR standard types were 2-cylinder (with the sole exception of
Duke of Gloucester): I *think* the 9Fs used the Britannia cylinders
but I wouldn't swear to it.
Thanks for prompting the T2 memories though..
Michael Ingram
>>There was also the (D)58 class of 3 cylinder steam locos, built as a
>>modification to the (D)57 design during the years following WW2. They were
>>a dismal failure in service because of continuous trouble with the tuning
>>of the mechanics controlling the valve gear for the 3rd cylinder. These
>>locos used a partially geared, partially levered setup to operate the
>>valve gear for the 3rd cylinder instead of the 'conjugating' gear as
>>fitted to the (D)57's.
>I have read that the gears had to be retimed after only a few hundred miles
>of running.
Apparently the spur gears on the main shaft that replaced the conjugating
arm (as used on the earlier 57 class) were the major cause of timing
problems with the middle cylinder. Rapid wear of the tooth surfaces and
the consequent slack that developed in the gear meshing were probably the
major factors.
>>I don't have info on whether any other Australian systems has 3 cylinder
>>steam locos.
>VR had the S class 4-6-2 and H class 4-8-4.
Speaking of the H-class, how many were actually built? Was H-220 the only
one to ever be constructed? What sort of arrangement did it (they?) use
for controlling steam flow to the third cylinder?
Mike.
David Bromage
>In <3k2k9n$9...@harbinger.cc.monash.edu.au> bro...@mdw079.cc.monash.edu.au (David Bromage) writes:
>>VR had the S class 4-6-2 and H class 4-8-4.
>Speaking of the H-class, how many were actually built? Was H-220 the only
>one to ever be constructed? What sort of arrangement did it (they?) use
>for controlling steam flow to the third cylinder?
Only H220 was completed, but two incomplete locomotives (which would have
been H221 and H222) sat in Newport Workshops until the mid 1950s.
The S and H classes had the same arrangement for working the third
cylinder. The actual working can be described simply as a single lever,
working in a 2:1 rocking arrangement, actuated the piston valve of the
middle cylinder. This was the most successful 3 cylinder arrangement ever
devised and never gave any trouble. The S class could run express
passenger trains for 13 days straight without giving any trouble. One day
each fortnight was needed to wash out the boiler. H220 also ran freights,
and the occasional passenger train, with a similar good record (except
when burning Wonthaggi coal).
Cheers
David
Barry Fitch
>>Speaking of the H-class, how many were actually built? Was H-220 the only
>>one to ever be constructed? What sort of arrangement did it (they?) use
>>for controlling steam flow to the third cylinder?
>Only H220 was completed, but two incomplete locomotives (which would have
>been H221 and H222) sat in Newport Workshops until the mid 1950s.
What happened to H-221 and H-222? Were they sold to scrappers or kept as
spares for H-220? How long did H-220 keep running before it was retired?
This might sound naive, but I believe that H-220 has been preserved, so is
it ever planned to restore it to operating condition?
>The S and H classes had the same arrangement for working the third
>cylinder. The actual working can be described simply as a single lever,
>working in a 2:1 rocking arrangement, actuated the piston valve of the
>middle cylinder. This was the most successful 3 cylinder arrangement ever
>devised and never gave any trouble.
Same setup as on the NSW (D)57's. I don't understand why the alternative
method used on the (D)58's was trialled given the fact that (D)57's had
been in service since the 1920's and been extremely successful at what
they were built for without any of the serious hassles the (D)58's had
during their brief lifespan.
>H220 also ran freights, and the occasional passenger train, with a
>similar good record (except when burning Wonthaggi coal).
That reminds me of something a few old steam drivers have told me on
enthusiast trips about using Thirroul coal (known as "Bull's Wool")
instead of coal from Metropolitan Colliery (Helensburgh). These old
drivers reckoned that Thirroul coal was the worst for steaming qualities
because it produced hugely copious quantities of ash which would impede
air flow through the bottom of the firebox and would require cleaning the
ashpan at much more regular intervals than normal. Probably clogged up the
spark arrestors in the smokebox too.
BF.
--
Barry Fitch (bar...@ppt.orb.apana.org.au).
Modelling modern era Australian prototype railways.
David Bromage
>In <3oieda$k...@harbinger.cc.monash.edu.au> bro...@mdw078.cc.monash.edu.au (David Bromage) writes:
>>>Speaking of the H-class, how many were actually built? Was H-220 the only
>>>one to ever be constructed? What sort of arrangement did it (they?) use
>>>for controlling steam flow to the third cylinder?
>>Only H220 was completed, but two incomplete locomotives (which would have
>>been H221 and H222) sat in Newport Workshops until the mid 1950s.
>What happened to H-221 and H-222? Were they sold to scrappers or kept as
>spares for H-220? How long did H-220 keep running before it was retired?
>This might sound naive, but I believe that H-220 has been preserved, so is
>it ever planned to restore it to operating condition?
The partly coompleted H220 and H221 were cut up, although the odd few bits
and pieces were kept as spared for H220.
H220 was withdrawn from service in April 1958 and was held in dry storage
at Newport Workshops for several years before being places in the ARHS
museum at North Williamstown.
H220 was inspected by Steamrail in 1991 and found to be capable of being
restored to service. However, at that time it was decided that such a
costly project could not be justified due to the lack of turning
facilities which could take such a large locomotive. H220 requires a
triangle or a 90' turntable. The only places available wehe the traingles
in Melbourne, Geelong and Warracknabeal and the turntable at Ararat.
Bearing in mind that H220 is capable of being converted to standard gauge,
this may not be the end of the story.
>>The S and H classes had the same arrangement for working the third
>>cylinder. The actual working can be described simply as a single lever,
>>working in a 2:1 rocking arrangement, actuated the piston valve of the
>>middle cylinder. This was the most successful 3 cylinder arrangement ever
>>devised and never gave any trouble.
>Same setup as on the NSW (D)57's. I don't understand why the alternative
Not exactly the same as the D57, but very similar.
Cheers
David