A Question About Garratt-Type Locomotives...

Sorry to take so long to get back to this...
The model of an "American" Garratt that Komachi posted a link to at the beginning of this string is a (fictional) Rio Grande narrow gauge "Double Mikado" (2-8-2+2-8-2; phrases like "Double Mikado" are I think often used to describe Garratt wheel arrangements), with the running gear of two D&RGW narrow-gauge Mikados. Given the sorts of applications the Garratt-type locomotive had on other continents, thisa is perhaps about as plausible "prototype" as could be chosen for an American Garratt: big standard-gauge Garratt's for U.S. mainlines were proposed (and fantasized about), but this is closer to the scale of Garratt's actually built. A few comments:

Boiler:
Garratts tended to have-- and this was one of the selling-points of the design (I'll try to discuss boilers in a later post)-- short, fat, boilers. If there is anything "wrong" with the model, then, it's the boiler: too skinny, too long, (too much like the boilers on "conventional" Rio Grande narrow-gauge locomotives). At a guess, to kit-bash a plausible Garratt for the D&RGW, you'd want the running gear used here and the boiler from, say, a standard-gauge Pacific, the boiler further shortened by taking segments out of the smoke-box and the middle boiler-course. (The tube-length-- length of fire-tubes between the firebox and the smokebox-- of boilers on large Garratts was often under 14 feet, whereas conventional North American locomotives often had tube-lengths of over 19 feet, and sometimes will over 20 feet.)

Wheel-arrangement:
4-8-2+2-8-4 seems to have been a more popular wheel arrangement than 2-8-2+2-8-2 for Garratts-- maybe an effort to spread the weight of a large locomotive over more axles for lightly-built track, maybe because railways in the British Commonwealth didn't trust 2-wheel leading trucks as much as U.S. railways did. ??? And, of course, the over-all ratio of driving to unpowered axles on a Garratt doesn't look so bad, in comparison to U.S.-style "conventional" steam locomotives if you count the Tender axles of the conventional locomotive!
That said, there were numerous classes of "Double Mikado," built for many railways over a significant period of time: no reason to think the Grande wouldn't have gone for this type!
(Over all, Garratts were built in 0-4-0+0-4-0, 0-6-0+0-6-0, 2-4-0+0-4-2, 2-4-2+2-4-2, Double Mogul, Double Consolidation, Double Prairie, Double Pacific, Double Mountain, and Double Northern wheel arrangements... and a "Mallet-Garratt" 2-6-6-2+2-6-6-2 was proposed, apparently seriously.)

Size:
The one Rio Grande narrow-gauge Mike that I have weights for (480-class, K36) was roughly 145,000 lbs on the drivers, 185,000 lbs total engine weight. Most narrow gauge Garratt classes seem to have had lower axle loadings than this, though South Africa and the then-British colonies of East Africa had very big power on 3'6" and meter-gauge track. (There was a proposal, late in the steam era, for a 3'6" gauge East African Garratt which would have been a large locomotive even by U.S. standard-gauge standards: a Double Northern, with a maximum axle loading of about 58,000 pounds, and a boiler with a maximum diameter of 99 inches!) So a Garratt for Rio Grande narrow gauge would have been right in the middle of Garratt size ranges. (One narrow-gauge Double Mikado class, a bit lighter than a "Double K-36" but approaching it, was built under British War Department auspices in 1943 for use somewhere in Africa. So, imagine a strategic reason for building new power for Colorado narrow gauge during World War II and you can justify a very modern D&RGW Garratt.)

Builder:
Alco, I believe, held licenses to the Garratt patents for North America. They also had patents for all-welded boilers...

Good luck, Komachi, if you go ahead and model such a creature!

At length, this thread has discussed the pros and cons of the Garratt style steam locomotive. Yet, in all my research (which, admittedly, has been FAR from extensive) I have yet to come across either a clear description or a diagram/picture depicting the layout of the steam passages from the boiler frame to the two steam engines.

Can anyone explain/provide pictures of how the steam was provided to the engines, and the exhaust piped back to the smoke-box.

Thank you.

The answer appears to be through a lot of long pipes with both spherical and sliding packed joints.

Weiner says "The exhaust steam from both engines is led to two concentric exhaust pipes of the same sectional area.
The live steam pipes have a swiveling joint the centre of which is co-incident with the pivot of each bogie. the exhaust steam passes through similar joints before entering the smokebox. the rear joint is prolonged by a sliding tube."

This drawing - http://i894.photobucket.com/albums/ac14 ... ner190.jpg
shows long, mostly horizontal runs of both live steam and exhaust piping running hither and thither, with a few different joints visible in the exhaust line below the boiler barrel. This little engine doesnt appear to have a superheater, but almost all Garretts did have them.

The LNER crews who ran U-1 2395 reported among other problems (2395 was not a popular engine among its crews), that wait times between runs would result in these long pipes filling with condensate, which caused delays starting.

Could you provide a higher-res scan of that diagram? Its kinda hard to make out the intricate detail.

Also, does anyone have any construction photos of any garratt?

As it was described, I get the impression that the steam pipes are used as the king-pin on a diesel's truck assembly?

Could such a pipe be reasonably expected to handle the stresses that such an engine would encounter in mainline freight service on a US railroad?

Here is the 300 dpi scan of the drawing requested.

I dont think the steam pipes are ever load bearing parts of the actual bogie joint, rather they pass through the center of the king pin which is itself a larger diameter tube.

jgallaway81 wrote: Could such a pipe be reasonably expected to handle the stresses that such an engine would encounter in mainline freight service on a US railroad?

It should be remembered that some really large Garratts were built . LNER 2391 had a rated tractive effort (85%) of 73,000lb

The largest steam engine ever built for use in Europe was a Garratt , engine Y-01 built by Beyer-Peacock for the Soviet Union in 1933, 1524mm gauge.
The top of the stack is 17 feet above the rail!

There's a nice comment in the Durrant book about the Ya-01: the very generous Russian loading gauge allowed tall smokestack and dome, so people used to the way large steam locomotives have short stacks and domeless or near-domeless boilers (think NYC Niagara, or the South African 26 class) may not appreciate how big the thing really was from the photo!

The Garratts in South Africa, East Africa and Australia were also very large and powerful locomotives capable of handling quite a bit of stress hauling heavy freight trains in mainline service. In Kenya it's uphill all the way from sea level to Nairobi at 1,700 m (5,500 ft), a distance of around 487 km (302 miles), finally topping out at Timboroa summit at 2,785 m (9,136 ft). Kenya's Class 59 Garratt had a tractive effort of 83,350 lb (at 85%), while South Africa's GL Garratt was 78,650 lb (at 75%). Although these locomotives are narrow gauge, either 1 metre in East Africa or 1.065 metre (3' 6" Cape Gauge) in South Africa, they are built to a loading gauge which is bigger than many standard gauge railways and they are really huge.

A Class 59 is still operational in Kenya, and a couple of GMAMs (big like a GL but not quite so powerful) are still operational in South Africa.

One other point worth mentioning, Mallets were almost universally compound engines, whereas (almost) all Garretts were single expansion. That distinction suggests Mallets' better thermal efficiency would benefit larger more advanced railroads where economies of scale would overcome the (implied) additional maintenance cost of the more complex machinery, and conversely would never pay off for the smaller, lower margin roads.

But note that most US railroads simpled their Mallets starting in the 1920s.

4-8-2+2-8-4 seems to have been a more popular wheel arrangement than 2-8-2+2-8-2 for Garratts-- maybe an effort to spread the weight of a large locomotive over more axles for lightly-built track, maybe because railways in the British Commonwealth didn't trust 2-wheel leading trucks as much as U.S. railways did.

I'll assume it's an issue of track quality. Well into the 20th century, no pilot wheels was considered acceptable in Britain and Europe for road freight engines. Then look at, for example, the many 4-8-2s built for African countries. These often didn't have high drivers. They weren't specifically passenger or fast freight power as they were in North America, but all-purpose engines. I gather the 4-wheel pilot truck was needed on the rougher track of these countries. I suspect this is also why their switchers tended to be tank engines with pilot and trailing wheels rather than tender 0-x-0s (except the South African 0-8-0s, some of the very few examples of purpose-built switchers of this configuration outside North America).

Here is the Garratt from Thailand:




Unfortunately out of operation.

And here is one in best conditions:

There are couple of pictures of South African GMAM no 4079 in action in 2008 at http://www.friendsoftherail.com/phpBB2/ ... hilit=4079

The Royal Siam 457 Class is interesting in having a third tank, a well tank, under the boiler;
and in having been built by Henschel in Germany, rather than by B-P.

There are a few pictures of a large Garratt in action in South Africa yesterday posted at http://www.friendsoftherail.com/phpBB2/ ... 137#p25137 and http://www.friendsoftherail.com/phpBB2/ ... 140#p25140. I have fired this loco in recent years - a magnificent beast.