Imaginary Locomotives

[The Johnster]

I'm mystified by Mallets and Garratts in general - the long path between the boiler and cylinders, between cylinders (in a compound Mallet), and the exhaust ought surely have been a disaster for thermal efficiency? But I must have missed something as they were evidently successful. What's going on? 

 

Forget about thermal efficiency, a largely irrelevant figure used in part of the hard sell of early diesel manufacturers' sales patter to infer that steam locomotives were more inefficient than they actually were, and to infer that diesels, with demonstrably lower figures, were more efficient than they actually were.  

 

It actually doesn't matter except to engineering purists; from a railway's point of view a locomotive that will pull the train load to time and cost less that the income the train generates after taking out the necessary overheads like empty stock and light engine working, track, signalling, labour, and administration costs, and repayment of capital loans, without destroying the track, is successful irrespective of how much of the heat goes up the chimney.  

 

Big American Malletts, admittedly not compounds but suffering the same drawbacks you describe, were very successful in the US in two particular applications; banking (or helper engine as they call it) work, and hauling very long heavt trains through long single track sections in order to save paths.  AFAIK only the Union Pacific Challenger class were used on fast passenger work.  I doubt thermal efficiency even occurred to their designers or builders.  The amount of head end power needed to replicate the performance of these thermally inefficient steam locos is revealing!

Mmm. That was a bit gruff!

Perhaps a bit billy goat gruff!

[The Johnster]

I was on a steam excursion once when KGV disproved your theory. Somewhere near Wrexham as I recall.

 

I seem to recall this being an altercation between the twin arch road overbridge at Ponthir, between Caerleon and Llantarnam Jc, and KGV's safety valve bonnet, which the bonnet lost.  The build up of track level resulting in the thickness of modern ballasting had not been taken into account, and the old route restrictions relied on.

[The Johnster]

I'm probably about to make a comment that will upset all the proper GWR fans, and shows my lack of knowledge on GWR at the same time; I thought all their 4-6-0s were pretty much the same (they look it!). Do Kings and Castles have bigger boilers then? I was aiming for something of a theoretical 1905ish vintage (hence Saint, small Prairie, and TGB tender), could a king/castle boilered version be its potential 1930s version maybe?

 

Go to your room, young man, and have a long hard think about what you've just said.

 

To summarise; the first GWR 4-6-0s were the 2 cylinder Saint and 4 cylinder Star classes introduced by Churchward.  Both had 4-4-2 versions in their development process, and used a new (then) boiler design called the Standard No.1, which also featured on the concurrent 28xx design.  Both 4-6-0s used 6'8" diameter driving wheels.  The No.1 boiler was used by Collett on the Hall and Grange 4-6-0s, the Hall being a version of the Saint with smaller, non GWR standard, 6' driving wheels and the Grange with 5'8" drivers, this time a GWR standard size and a very similar loco had been drawn out by Churchward but never built as there were sufficient 43xx 2-6-0s to do the work.  

 

The Castle is basically a Star with an enlarged boiler, the Standard No.7, and they first batch continued the 40xx number series, an indication of how different the Railway, if not it's publicity department, thought the locos were.  The King is a further enlargement of boiler, but with a requirement to clear under bridges for loading gauge purposes Collett needed to set the boiler lower, hence the need to reduce the driving wheel diameter to 6'2", another non standard size, so that it could sit lower between them.  The cylinders were larger as well, and the the front bogie had to have outside frames for the leading wheels to clear the inside ones.  6'2" drivers were later used by Hawksworth on his County 4-6-0s, which used another non-standard boiler, a domeless version of the Stanier 8F boilers for which Swindon was tooled up, having built this type during WW2.  

 

This leaves the Manor, which is a variation on the Grange but with a smaller boiler, Standard No.14.  This boiler was later used in a domed version for the BR Standard Class 4MT 4-6-0, 75xxxm designed at Swindon.  This boiler was reputed to be a bad steamer, and much work was done on draughting with both classes to improve it, eventually with some success in the case of the Manors.

 

GWR locos in general have a family resemblance, the result of the railway having a centralised and standardised locomotive policy throughout a very long period of time uninfluenced by grouping.  The safety valve bonnet, for example, used on all GWR locos from the Fire Fly class onwards, is designed by Daniel Gooch to project the sound of steam from lifting valves skywards away from those alongside the loco, a good idea as anyone who as spent time platform-ending on a steam railway will agree.  But no two are exactly the same; I concede that to the uninitiated the difference has to be looked for.  The Hall and Grange are particularly similar, but the Grange's footplate is set lower behind the cylinders where the Hall's is straight, and the boiler is pitched lower.

 

Putting a larger boiler on a loco is generally a good idea if you want to increase it's power, but in order to use the extra steam efficiently you may have to enlarge the cylinders or provide extra ones.  There is a top limit imposed by loading gauge and weight restrictions, and on very big locos by the capacity of a human fireman, which may lead to mechanical stoking or oil firing and a whole new set of problems.  A successful loco is usually one that is liked by it's drivers and firemen, and may not be the same thing as what the designer was thinking of when he was drawing it out.  

 

Drivers like locos that ride well and have plenty of power in reserve so that they can keep time easily and make it up if some is lost.  Firemen like this as well, as a steady footplate makes their labour less onerous and tiring, an important consideration on a long non-stop run.  But the fireman also appreciates a boiler that steams freely, i.e. will create steam faster than the driver can use it in most situations.  No loco can do this pulling hard up a gradient of course, but free steaming means that the situation can be quickly recovered once the top is reached.

 

This was not much of an issue in the 19th century, when locos were small and pulled light trains, boilers were set low and forward visibility was good, and boilers were well within the capacity of a human being to fire successfully.  Bigger 20th century locos were a different beast, and new issues presented themselves to designers; forward visibility was compromised by long, fat boilers, and the resulting higher centre of gravity played havoc with the ride quality.  The GWR's Churchward designs and their derivates generally fared well in this environment; efficient, free steaming boilers, tapered to lift steam away without smoke deflectors, and the taper enabling the centre of gravity to be kept as low as possible.  Even the 2 cylinder engines rode well by the standards of 4-6-0s generally, certainly compared to Black 5s or Royal Scots, and the 4 cylinder engines delivered the power smoothly.  

 

The railway can be accused of small c conservatism in the post Churchward era, but the Churchward type of loco worked well and efficiently.  Pacifics were thought unsuitable for a railway with steep banks in South Devon and South Wales, and the Britannias, when they appeared in 1953, did not fare well in these places.

Edited March 2, 2018 by The Johnster

[melmerby]

  The King is a further enlargement of boiler, but with a requirement to clear under bridges for loading gauge purposes Collett needed to set the boiler lower, hence the need to reduce the driving wheel diameter to 6'2", another non standard size, so that it could sit lower between them.  The cylinders were larger as well, and the the front bogie had to have outside frames for the leading wheels to clear the inside ones.  6'2" drives were later used by Hawksworth on his County 4-6-0s, which used another non-standard boiler, a domeless version of the Stanier 8F boilers for which Swindon was tooled up, having built this type during WW2.  

 

Sorry, but I think some of that is wrong

Kings had 6' 6" wheels and the originally drawn out version had standard 6' 8½" wheels as on the Castle

The as built loco had 6' 6" wheels to increase the tractive effort to over 40000lb which Collett had been instructed to exceed by Sir Felix Pole. With Castle wheels it would have been just over 39000lb

 

The inside cylinders would have fitted without the funny bogie if Collett had put a slight incline on them but that would not do on the GWR so they are horizontal, in line and would foul a Castle style bogie.

(Stanier managed with a standard bogie on the Princess which had generally the same front end layout)

Only the first 6 had 16¼" cylinders the rest had 16" the same as a Castle but would be bored out during service to 16¼"

 

The loco was inside the height loading gauge by more than some GWR locos and the extra 1¼" boiler centre line with 6' 8½" wheels could have been accomodated (just).

 

Info on these matters is freely available in various books. I have several which lay out the same general info.

 

Cheers

 

Keith

Edited March 2, 2018 by melmerby

[JimC]

Kings had 6' 6" wheels and the originally drawn out version had standard 6' 8½" wheels as on the Castle

The as built loco had 6' 6" wheels to increase the tractive effort to over 40000lb which Collett had been instruct to exceed, by Sir Felix Pole. With Castle wheels it was just over 39000lb

I fear that is incorrect, even though it is in a lot of books.

 

The smaller wheels were decided quite early in the design process according to Cook, who was closer than most, and of course its a fundamental aspect of the design. In order to test this out they turned down a Castle's tyres to that size, technically below scrapping thickness, but safe on new tyres for a limited period. According to Holcroft this was late in 1926. King drawings in the NRM index are all from 1927 or later. One of the earliest, from Jan 1927, is for the driving wheels.

 

The little tweak to get the TE over 40,000 was to take the cylinders out to 16.25in - first rebore in effect. 

 

Sir Felix Pole got the wheel size/cylinder size confused in his book, but he was writing when an old man decades after the event, I think he can be forgiven. 

 

With 16in cylinders and 6'8.5in drivers the TE would have been under 38,000.

 

-----------------------------

 

The pitch of the King boiler, BTW, was just 3/4in lower than that of the Great Bear, and the same as the 47s.

Edited March 2, 2018 by JimC

[melmerby]

I fear that is incorrect, even though it is in a lot of books.

 

The smaller wheels were decided quite early in the design process according to Cook, who was closer than most, and of course its a fundamental aspect of the design. In order to test this out they turned down a Castle's tyres to that size, technically below scrapping thickness, but safe on new tyres for a limited period. According to Holcroft this was late in 1926. King drawings in the NRM index are all from 1927 or later. One of the earliest, from Jan 1927, is for the driving wheels.

 

The little tweak to get the TE over 40,000 was to take the cylinders out to 16.25in - first rebore in effect. 

 

Sir Felix Pole got the wheel size/cylinder size confused in his book, but he was writing when an old man decades after the event, I think he can be forgiven. 

 

With 16in cylinders and 6'8.5in drivers the TE would have been under 38,000.

 

-----------------------------

 

The pitch of the King boiler, BTW, was just 3/4in lower than that of the Great Bear, and the same as the 47s.

Where does the early drawing of a King with 6' 8½" wheels fit into it? It must have been a serious consideration initially, else why do it?

 

Keith

 

BTW OS Nock quotes both sources and seems to give them both some creedence.

Edited March 2, 2018 by melmerby

[Budgie]

Outside valve gear for the rear truck. No space for the ashpan otherwise.

Couldn't you use inside underslung Caprotti valve-gear?

[rodent279]

No doubt that the Robinson 2-8-0 was a rugged thing, and seems to have been grudgingly admired even by GWR loco crews for it's toughness, but there are important differences between it an a 28xx. They locos are of a size and haulage capacity, and have similar sized driving wheels, but the 28xx were vacuum fitted, which the ROD (to use the GW vernacular) was not, and the 28xx were happy running at 50+mph, which the RODs were certainly not! Their best work on both the Great Central and the GW was as heavy mineral haulers, a job which requires brute strength but nothing much else; speeds are 25mph or lower. A 28xx must have been a nicer place to work than an ROD on similar work. A 28xx, being vacuum fitted, will be used on part or fully fitted freight work which the ROD cannot handle; they will in consequence be regarded as faster locos.

 

It is a little like comparing an Austerity 2-8-0 to a Stanier 8F, the loco it was based on. The 8F is more refined, rides much better, and is probably easier to fire, but the WD is designed to be cheaply and quickly built with readily available materials in a wartime economy. Nobody liked them much, but like the 04s, they were tough and robust.

 

As Ian has pointed out, the coupling rods on 42xx/5205/72xx tanks were jointed in order to cope with sharp curves, I believe with the Western Valley route from Ebbw Jc to Ebbw Vale's iron ore trains in mind. Even with this, the frames strained and the locos suffered from tank leakage, a problem on a loco with a barely sufficient water supply to start with. As I said, I am not a loco engineer, and do not know if the 2nd axle drive of these engines is a consequence of the jointed coupling rods; perhaps someone of greater erudition in the matter will comment?

The position of the driving axle, on the 2nd or 3rd axle, may be tied in with bending moments on the frames, in other words the tendency of the frames to twist with each stroke.

 

Edit: just thought that balancing will come into it as well. A longer connecting rod will create more rotational and fore/aft forces than a short one.

Edited March 3, 2018 by rodent279

[rockershovel]

re “Big 4 Garratts tending to look the same” isn’t that a sort-of function of their main characteristic - being of no real use, for reasons which were well known at the time? The LMS, LNER, and subsequently BR would demonstrate that it was quite possible to develop eight- and ten-coupled locomotives which could exceed available traffic requirements and operational constraints, why bother with articulated types? (Aside from the LMS using them for certain very specific types of traffic on lines with well-known constraints not present elsewhere)

 

Mallets are the same (including the simple-expansion types later developed in the US). US railroads are largely devoid of platforms, overbridges or other sorts of trackside furniture; even fences or handrails on bridges, in many cases. This leads to the characteristic US-style loco outline, of huge cylinders, very low to the ground. The particular conditions of the US, particularly long lengths of single track mainline placed a value on extremely large locomotives hauling extremely long, heavy trains. Mallets didn’t develop in the U.K. because their primary purpose didn’t really exist, and their required environment (specifically, clearance for the inevitable side-throw) didn’t exist either.

Edited March 3, 2018 by rockershovel

[rodent279]

How does one adapt a tender loco to be able to run tender first as well as a tank loco runs bunker first?

 

Is it a stability issue? Wear on the rear drivers? Danger of getting hit in the face by airborne coal?

Didn't European loco designers make tenders more suitable for tender first running by mounting them on bogies?

[rockershovel]

Digressing slightly, but further along the path already begun, it does occur to me that apart from occasional aberrations like the experimental 4-14-4 (really more an expression of the Soviet world-view, than a serious Engineering design), the Russians never really felt the need for the enormous US locomotive types. Russian locomotives were never SMALL, by any standards, but their enormous loading gauge isn’t fully utilised - the ubiquitous 2-10-0 and 2-10-2 types were in the 85-100 tonne range.

 

The Russian, and later Soviet railways served a comparable function of creating, and unifying a single nation across the vast distances of its own geography; but the Soviet “command economy” tended not to repeat the American pattern of hauling vast quantities of raw materials over great distances, preferring to take the manufacturing to the materials, including the workforce if required.

[runs as required]

I fear that is incorrect, even though it is in a lot of books.

 The little tweak to get the TE over 40,000 was to take the cylinders out to 16.25in - first rebore in effect. .

I've read too of the boring out of the Kings cylinders to achieve the 40000

[Regularity]

The 14 coupled experience in the Soviet Union demonstrated why articulated engines came about: such a long rigid wheelbase is fine, but only on straight lines.

 

Mallets were often used on “branchlines” in the Appalachian mountains, to provide brute force on steeply graded sinuous lines built to access the coal mines: nothing else could cope, and having several engines with several crews would have been a nightmare to coordinate when switching (shunting), as well as creating problems with drifting smoke, etc. On the plains, with level routes and long distances, more conventional locos were more than up to the task, as the Nickel Plate knew - one of their Berkshires (2-8-4) was an easy master of a long freight train running steadily at 55 mph, and outperformed diesels in this scenario. The NKP actually held off dieselisation for a few more years because of this, but eventually the economics of better fleet utilisation won through.

 

Of course, when diesels came along with their MU capabilities, they offered a much more flexible approach that only required a single crew, and with all the units operating together.

 

As well as a more generous loading gauge, American lines were also built to accommodate heavier axle loads with closer spaced ties and heavier rail, and the bigger engines were not the same problem as over here.

[runs as required]

Do I understand most N American Mallets were not actually true Mallet compounds? So more like Garretts for unlimited loading gauge height clearances.

dh

[rockershovel]

Do I understand most N American Mallets were not actually true Mallet compounds? So more like Garretts for unlimited loading gauge height clearances.

dh

Most later American “mallets” were, indeed, single-expansion types, but quite unlike Garratts.

[Regularity]

Do I understand most N American Mallets were not actually true Mallet compounds? So more like Garretts for unlimited loading gauge height clearances.

dh

  

Most later American “mallets” were, indeed, single-expansion types, but quite unlike Garratts.

Yes, not Mallets in the sense that M. Mallet originally intended, but otherwise the front frames were pivoted at their rear to the front of the rear, fixed, frames.

 

Definitely not a Garret.

[runs as required]

When you both say not like a Garratt, presumably you mean in appearance. What they did have in common was a 'double engine' supplied by one boiler (and one crew).

 

The Gorton version was more compact (it evolved first for narrower gauges) and enabled a larger diameter shorter boiler slung lower between the two articulated engine units. Garratts were usually double ended.

The US Mallet had a rigid cast steel frame with an enormous elongated high pitched boiler with its front bearing onto (and looking to slide across the front engine). They also appeared to be very much unii-directional but unlike the Garratt presumably adhesion was more constant than the Garratt where the water tank and bunker gave differing adhesion full or empty.

Did US railroads generally turn their engines on turntables or triangles?

 

dh

[burgundy]

If I have understood it correctly, a Mallet has a fixed rear engine and an articulated front engine (as described by Regularity above).

On a Garratt, both engines are articulated in relation to the boiler.

On a Meyer, both engines sit under the boiler and are articulated, connected by a Z link.

Hope this helps

Best wishes

Eric  

edited to correct spelling

Edited March 4, 2018 by burgundy

[Ohmisterporter]

Does it make sense to refer to Compound Mallet and Simple Mallet? I think the term articulated is too simplistic to describe a simple Mallet: there are several forms of articulated locomotives, as described in earlier postings.

Edited March 3, 2018 by Ohmisterporter

[asmay2002]

On a Garrett, both engines are articulated in relation to the boiler.

 

No a "Garrett" isn't articulated at all.

 

[Regularity]

The US Mallet had a rigid cast steel frame with an enormous elongated high pitched boiler with its front bearing onto (and looking to slide across the front engine). They also appeared to be very much unii-directional but unlike the Garratt presumably adhesion was more constant than the Garratt where the water tank and bunker gave differing adhesion full or empty.

 

Other than having simple rather than compound expansion, the US mallet was exactly like M. Mallet’s design. That’s why it’s called a Mallet.

But I seem to be repeating myself.

Did US railroads generally turn their engines on turntables or triangles?

 

Yes, but not on all branchlines. Edited March 3, 2018 by Regularity

[Hroth]

No a "Garrett" isn't articulated at all.

 

You might say it was inarticulate....

 

H,C&O....

[Compound2632]

Beyer-Garratt; Meyer; Mallett - all very confusing.

 

A Meyer seems to be somewhere between a single and a double Fairlie?

[Regularity]

You haven’t mentioned Forney tanks, or Pechot-Bourdon, or...

[Regularity]

You can start with Wikipedia, of course:

https://en.m.wikipedia.org/wiki/Articulated_locomotive

Edited March 3, 2018 by Regularity