Steam & Excursion > Vauclain's triple-expansion, double-hinged 2-8-8-8-2

Yesterday I was rooting around in old patents looking for ones related to Santa Fe's infamous "flexible boiler" 2-6-6-2's ( see: https://www.trainorders.com/discussion/read.php?11,1845249 , http://www.douglas-self.com/MUSEUM/LOCOLOCO/flexmallet/mallet.htm ) .  The pressure vessel itself, of course, was not actually hinged - the forward part of the boiler contained a feedwater heate, superheater, and a reheater to try to put more energy into the exhaust steam from the high pressure cylinders before feeding it to the lower pressure cylinders.  They were Baldwin "separable boilers" that were in these cases quite literally separated. 

I found William Leighty's double-ball-and-socket design as well as the accordion-pleat conjured up by Baldwin's Sam Vauclain.  Santa Fe had examples of each type:

https://patentimages.storage.googleapis.com/c3/7d/45/3c17d77413be70/US1074877.pdf
https://patentimages.storage.googleapis.com/d3/ff/f0/d917343d076833/US912923.pdf

The search also turned up one other relevant Vauclain patent on a much more ambitious bit of engineering, US Patent 976,014: a triple-expansion 2-8-8-8-2 featuring not one but two accordion-pleated boiler hinges.  Figures 1 and 2 from the patent are in the first attached image below, and the full patent can be read here:

https://patentimages.storage.googleapis.com/f3/bd/53/821082e76fa19f/US976014.pdf

Some time ago, the late Dudley Westler loaned me a drawing of a triple-expansion steam locomotive conjured up by the rather obscure Doble Steam Motors company in 1928, which I posted here:

https://www.trainorders.com/discussion/read.php?10,4057725

Looks like far less-obscure enterprises were also interested in the concept nearly two decades earlier.  Anyone know if Baldwin's triple-expansion design got any further toward realization than this patent?  Baldwin did build 2-8-8-8-2 "triplex" locomotives for Erie in 1914 and 1916 and a 2-8-8-8-4 for Virginian with the third set of drivers under the tender, but these were only double-expansion compounds:

https://en.wikipedia.org/wiki/Triplex_(locomotive)
https://www.steamlocomotive.com/locobase.php?country=USA&wheel=Triplex&railroad=err#3300

George Henderson's 1914 Baldwin patent for a "quadruplex" 2-8-8-8-8-2, shown in the second image below (US Patent 1,100,563), likewise is only double-expansion, and mercifully only has one accordion-pleated hinge in the boiler.

https://patentimages.storage.googleapis.com/bb/47/65/9cd083361cee79/US1100563.pdf

The August 1951 Trains Magazine has some Charles O. Egerton sketches interpretating Henderson's designs as might have been built for Santa Fe had the railroad not come to its senses after their disappointment with the 2-10-10-2's. One sketch on page 47 shows a 2-10-10-10-10-10-2 "quintuplex" (reprinted on p.377 of Iron Horses of the Santa Fe Trail).  I don't know how many times it supposedly expanded steam before sending it up one of three funnels.

---

---

When guessing about the possible performance of these  ‘nightmare designs’, several basic principles of loco thermodynamics are useful.

First is grate area.  There is a corresponding relationship of cylinder volumes to be filled and steam generating capacity.
The widest the grates could be is 10 ft...so take a guess as to the length of the grates..example #1.

Next, is the number of superheater return-bends exposed to the direct heat of the fire’s flame tips.
It is the flame tips that does the superheating...the cold water surrounding the boiler tubes very quickly cools-off the fire’s hot flue gasses...especially after traveling about the first 5 to 6 ft of tube length.  The gasses are divided into little individual streams when they enter the boiler tubes...there is more cold water surrounding the flues than can possibly be heated to any appreciable degree
by the small streams traveling in the tubes, & after only first few feet down inside them.

 In all these examples, so-called ‘superheaters’ are far removed from the direct flames of the fire...after traveling through the boiler’s heating section, there could never be much beneficial degree of superheating ...
The same realities also apply to so called ‘re-heaters’.
Its all thermodynamics B/S...

As far as compounding goes, the HP cylinders are actually pushing all of the pistons, downstream of the first high pressure cylinders....thus, there is high back-pressures on the primary pistons—- which subtracts from the differential pressures acting on the HP pistons, and reduces any tractive effort from the HP cylinders.

Another factor to consider: Starting.
On compound articulateds it’s common to use a ‘starting valve’...this device admits boiler steam into the receiver pipe feeding the LP cylinders.... However, because of the larger diameter of the LP pistons, full boiler pressure would cause uncontrollable driver spinning...thus, the ‘starting valve’ ( controlled by the engineer when needed) reduces the live steam oressure to a fixed value that approximates the steam pressure ( acting on the LP pistons) when running compound....  
So, on these concoctions, how do they get the heavy trains started?
That little grate can’t possibly supply or keep up with the huge steam demands of 6 or 8 cylinders —- when starting as “simple”....

What was the reality of the functioning of the so-called ‘reheaters’?   
The warm, smoky gasses from the earlier journeys of traveling through the cold water containers, the superheaters, etc.,
would never have enough residual heat remaining  to benefit  these ‘reheater’ volumes..
You don’t have to be a thermodynamic engineer to laugh out loud at these preposterous proposals.

Example no.1: does anyone realistically believe that a possible grate area of 60 sq.ft. can possibly generate enough steam to supply three sets ( 6 cylinders) of compound cylinders? & super heaters? & Reheaters? ....
Especially, if trying to fight up the grades, pulling the heavy freights,  prevalent in the eastern U.S.?

Anybody take a look at those tiny tenders...with 12-tons of coal and 8,000 gallons of water....how far are they going on THAT?
(All right, you tell me what the possible capacities are...!)

All of these multiple-framed compounds were abject failures and couldn’t compete with the high-HP designs that followed the first Berkshire locomotives, and that led to the successful further designs —- culminating in the ATSF’s wonderful 2-10-4s.
Study those 2-10-4 behemoths, if you want understand the limits of thermodynamic principles applied to recip-machinery...

In these drawings, you are looking at engineering’s preposterous, lying —- applied to sales pressures — in trying to generate revenue for, primarily, Baldwin.

There is no possible way that flue gasses, cooled-off by cold boiler water,  would get past the first stage of boiler capacity...
At least the last design, of the sketches above, does include a little bit more grate area..but not enough to make a material difference in the downstream flue-gas temperatures..

AND...Don’t even get me started on those ridiculous, ‘accordion bellows’ in the middle of the reheaters/boilers....
What happens when the tons of flyash, cinders, etc. solidly pack the ‘valleys’ between the bellows’ rings....then, whole boiler becomes rigid and the front engine don’t flex no more...
You tell me what the result is..with a solid, unflexible boiler unit... 
How do you keep them things cleaned out?
 Once it gets solid-packed and rigid..then what do you do?
 
​W.



Edited 3 time(s). Last edit at 09/11/19 07:39 by wcamp1472.

Evan, for shame! You have somehow missed the pioneering effort in accordion-boiler locomotives, the Gilderfluke Perfected Articulated, perhaps because it was patented in Lithuania, rather than the U.S. This image comes from a 1909 issue of Locomotive Design Gazette Illustrated. It was said to have inspired Santa Fe's later efforts, although the Gilderfluke prototype was only built to 1" scale, and for some reason, only ran backwards. Must have been a problem with the top angled nut janker. The bellows are clearly visible between the hard clank berf plates and the swiveling nodular drop spindles, but otherwise, I'll be darned if I know what's going on. Perhaps Wes knows.

E.O.

---

Santa Fe actually did have issues with cinders in the bellows, not to mention the other shortcomings. The non-hinged "Prairie Mallets", 69" drivered 2-6-6-2's from 1910, were fairly sucessful with some lasting 20 years on the West Texas plains hauling oil trains and other tonnage.

That triple would have stopped the diesel thing in it's tracks!! (Pardon me, I just hurt myself from laughing).
Seriously (and Wes is totally right) the builders knew the issues with both multiple expansion and length/
steam travel issues, I'm surprised they'd even waste their time on this type of concept.

Out

hogheaded Wrote:
-------------------------------------------------------
> Evan, for shame! You have somehow missed the
> pioneering effort in accordion-boiler locomotives,
> the Gilderfluke Perfected Articulated, perhaps
> because it was patented in Lithuania, rather than
> the U.S. This image comes from a 1909 issue of
> Locomotive Design Gazette Illustrated. It was said
> to have inspired Santa Fe's later efforts,
> although the Gilderfluke prototype was only built
> to 1" scale, and for some reason, only ran
> backwards. Must have been a problem with the top
> angled nut janker. The bellows are clearly visible
> between the hard clank berf plates and the
> swiveling nodular drop spindles, but otherwise,
> I'll be darned if I know what's going on. Perhaps
> Wes knows.
>
Let's see....   That would be, what, 42 sets of brasses and wedges to maintain, besides the bellows.

-LD

I don't know, the Triple Expansion Engine was pretty sucessful in marine applications.
Applying that to lets say an Eire Triplex might have worked if you can find a way to fit the very, very large low pressure cylinders into the constraints of a standard gauge locomotive frame.

Compounding Absolutely works....

Steam turbines, with their successively larger ‘stages’, are the clearest examples of effective ‘compounding’.
Marine and stationary turbines are engineered to squeeze every ounce of heat energy out of their steam-driven devices.

Interestingly, replacing steam as the heat source, with petroleum fuels — gave rise to the jet engine...a direct descendant of steam turbines.  I love to stare in amazement as contrails crawl across the skies....streaming behind today’s skyliners.
Direct descendants of attempts to perfect compounding principles..

It was inventor Parsons who built the first steam turbine...from then on it was engineering evolution to better and better improvements ....

W.

Evan_Werkema Wrote:
......... by the rather obscure Doble Steam Motors company.........

Well, not that obscure from a railfan standpoint:
 -  After Besler bought assets and [ some ] patents of the company, they went on to provide the propulsion unit for a New Haven steam railcar, and would have provided the drive units for the stillborn B&O W-1 "4-8-4".
 -  Doble himself provided the design basis for the well known Sentinel small switchers and railcars of the UK, many of which were exported
 -  He also worked with Henschel of Germany in the automotive/truck area, which, more importantly, provided the basis for the Reichbahn's experimental Class 19 "2-8-2" with individual axle drive, which was built in 1941, came to the US courtesy of the US Army after the war, and was scrapped at Ft Eustis.

Best regards, SZ
 

Of course compounding works, and yes quite well in marine applications but we're talking locomotives here. I can see
the logic involved if you consider that the steam used in single expansion may not be fully expanded by the end of stroke
and hence could be exhausted into a second set of cylinders. But as 1309 shows that's not really possible (sorry for that).
But at the same time and with a much lower amount of mechanical mass, correctly sized single expansion cylinders in
an articulated application are also effective and frequently more so.

Out 

MP4093 Wrote:

> Santa Fe actually did have issues with cinders in
> the bellows, not to mention the other
> shortcomings. The non-hinged "Prairie Mallets",
> 69" drivered 2-6-6-2's from 1910, were fairly
> sucessful with some lasting 20 years on the West
> Texas plains hauling oil trains and other tonnage.

Some of the flexible-boiler engines lingered on the roster for 18-19 years, with the last ones going to scrap in 1929.  How many of those years they were actually out pulling trains, I can't say.  The 3322 at least was still servicable in 1920 when it was tested against a new 3800-class 2-10-2 (and lost out in nearly every category).