I can't speak for the origin of the 'makin's' of the 9000s, but I doubt that UP had 2-8-8-0 Mallets with 67" drivers. The 4-12-2s were acquired to replace Mallets, to get the average freight train speed up.
During the second decade of the 20th century the AT&SF dismantled some Mallets of questionable parentage and converted them into reasonably successful 2-cylinder simple locomotives. I believe, but cannot prove, that the rear engines kept the (shortened) Mallet boilers and their original cylinders, while the front engines got new cylinders (of more reasonable proportions) and new boilers.
A lot of weight issues can be addressed by tweaking the springing and equalization, and later locos could be riding on heavier rail (and thus have higher permissable axle loadings.) As for curves, the writeup in Kalmbach's Cyclopedia - Volume 1, Steam Locomotives notes that one series had to be moved from the OSL to the UP's transcon main because of problems with the OSL's curves. Another series (possibly those with blind #4 drivers) remained on the OSL until it was dieselized.
Only one type of 4-12-2 was built, the Union Pacific Railroad's 9000 class locomotives, 88 of which were built by ALCO between 1926 and 1930. These locomotives were used to increase the speed of freight trains in flat country, and were fairly successful, but were maintenance nightmares, largely because of their use of an inaccessible third cylinder driving a cranked second driving axle between the frames. There was no inaccessible valve gear to worry about, however. ALCO had obtained permission to use the so-called conjugated valve gear invented by Nigel Gresley. This system used two hinged rods connected to the outer cylinders to operate the inner cylinder. The 9000 class locomotives were the largest ever to use Gresley gear.
Between 1934 and 1940 eight of the first fifteen locos had their Gresley gear removed and were converted to a "double Walschearts" valve gear which utilized a double eccentric (return) crank and second link on the right side (similar to the gear Baldwin used on its 3-cylinder experimental compound 4-10-2 #60000), which operated the valve for the inside cylinder. Union Pacific referred to this system as the "third link." The 4-12-2's constructed from 1928 utilized roller bearings in the Gresely lever bearings, thus none of these engines were converted. The pre-1928 engines not converted received the roller bearing levers in 1940, and no further conversions were made.
During the design phase "blind" (flangeless) third and fourth axle were provided in order to improve curve handling, but the use of ALCO's lateral motion devices on the first and sixth axles (which allowed the axles to slide up to two inches to the side) made this unnecessary. They were the longest rigid frame locomotives in North America, and until 1934, the longest in the world. The trailing truck carried the same axle load as the drivers, which was unusual.
There has been some debate as to whether the first driving axle of the 4-12-2 was cranked to provide clearance for the main rod connected to the second axle. It has been shown that there was no such crank on the first axle. Instead, the spacing between the first and second axles was increased by 18 inches to provide the necessary clearance.
One example survives, at the Railway and Locomotive Historical Society's museum at the Los Angeles County Fairplex in Pomona, California.
Something that I have been curious about with UP's 9000 class, is normal side rod locomotives have the drivers "quartered" at 90 degrees so that the piston thrusts are evenly timed. Do the 9000's have their drivers set at 60 degrees ( I think my math is right on the 60 degrees), to keep the piston thrusts balanced? or how was the center cylinder timed/balanced in the valving sequence, if not?
I know that on articulateds each engine set was independently quartered as if a separate engine, and that synchronizing between engines was not attempted, as they would go in and out of sync due to wheel slip, usually with the front engine set slipping more easily.
Thanks for any info,
May your flanges always stay BETWEEN the rails
sarges wrote:how on earth could you take an articulated loco and turn it into a rigid wheelbase loco.
Lessee-- weren't the first RDG 3000-class 2-10-2s built using modified boilers from 2-8+8-0s? And you remember the RDG 4-8-4s used boilers from 2-8-0s. Re-using old boilers probably wasn't that uncommon-- wasn't it Worley that said SFe considered using ex-N&W 2-8+8-2 boilers to build 4-8-4s?
Felton, I found the following on web site: http://www.steamlocomotive.com/3cylinder/"I have been told that the designers had to keep the quartering at equal 120 degree angles partially for centrifugal mass balancing reasons. However, if the inside cylinder is inclined at say 7°, it would normally be necessary to set the cranks at 120°-127°-113° to preserve the torque distribution. Apparently, this was not necessary for the 9000s. The #2 driver has an enormous inside steel crank/counterbalance assembly that has to spin in concert with and balanced to the outside main rods. (One hundred and twenty degrees is the quartering angle listed in the book "Union Pacific Type, Volume 1" and confirmed by one of the co-authors, John Bush.) It is suspected that one of the reasons the Baldwin #60000 used a 90°-135°-135° configuration had to do with it being a compound locomotive, with the center cylinder receiving high-pressure boiler steam, and the outer cylinders receiving the low-pressure steam from the center cylinder. And according to John Bush, the locomotive had "a very interesting" outside valve gear design."
The SP used 135 - 135 - 90 degree spacing with a cranked front axle, so I'm totally confused.
sarges wrote:...how on earth could you take an articulated loco and turn it into a rigid wheelbase loco... mick
Actually, it's really not that hard to take boiler off an articulated chassis, and install it on a rigid chassis of sufficient length and strength.
IIRC, the only place the boiler is actually connected to the chassis is at the cylinders.
so I'm totally confused.
Confused as to why the drawings haven't surfaced to put an end to this question.
BigJim wrote:I found the following on web site: http://www.steamlocomotive.com/3cylinder/"I have been told that the designers had to keep the quartering at equal 120 degree angles partially for centrifugal mass balancing reasons. However, if the inside cylinder is inclined at say 7°, it would normally be necessary to set the cranks at 120°-127°-113° to preserve the torque distribution. Apparently, this was not necessary for the 9000s."
BigJim wrote:The SP used 135 - 135 - 90 degree spacing with a cranked front axle, so I'm totally confused.
JanOlov wrote:[quoting from Wikipedia?]There has been some debate as to whether the first driving axle of the 4-12-2 was cranked to provide clearance for the main rod connected to the second axle. It has been shown that there was no such crank on the first axle.
It seems the first and second driving axles were 88 inches apart and the inside rod was 113 inches long. If I remember right that the inside cylinder tilt was 9 1/2 degrees, and if we assume the axis of the center cylinder intersected the axis of the crank axle, then that leaves 10.45 inches between the centerline of the inside rod and the centerline of the first axle. So how fat is the rod, and what's the axle diameter?
(We can't take it for granted that the inside cylinder axis intersects the axle axis. After all, the outside cylinders don't.)