Union Pacific FEF-3 4-8-4 Northern type steam locomotive, by Smokebox

[FourEightFour]

I really want to open her up on the NEC. She was good for 110 mph.

I have heard even better. According to UP documents, the 837 was hauling a test train on the south-central district pulling an equal consist to the City of Los Angeles. She was rolling along at 110 mph, with only 8 pounds of back pressure. The UP official did not allow the engine to go faster, but she could of considering the max back pressure for this engine is around the neighborhood of 10-12 psi.

I will be opening her up on Tennessee Pass myself.

[_o_OOOO_oo-Kanawha]

Looks good, mr Rennie. You are from the UK, right? Have you visited the real FEF?

One questing concerning realism, do you think it is possible to animate the reverser position into the valve gear movement? Not the cut off setting per se, but rather the forwards, neutral and reverse position of the Walschaert's gear?


Over at Elvas Tower, some very fine new steam locomotives are being built for MSTS as well. Well worth taking a look at the new WP steam locomotives.

Steam is not dead, Steam lives!!!

[mrennie]

Since the FEF does not have a hydrostatic lubricator, you never have to worry about turning the lubrication on owing to the fact all lubrication is provided via two Nathan DV-7 36 pint mechanical lubricators. You would need to fill these every 100 or so miles to keep them full(*I think, never messed with the big'ns*). These lubricate the pumps, the valves/main piston, and a few other areas not worth mentioning.

I'll be animating the lubricator levers connected to the valve gear.

[mrennie]

Looks good, mr Rennie. You are from the UK, right? Have you visited the real FEF?

That's right, I'm an Englishman, born accidentally in Wales, moved to England, lived for many years in Scotland from age 9, then in England again, and 21 years ago I came to Spain (not the wisest decision of my life). I feel a bit ashamed to have to admit that despite having family in the US (my sister lives not far from Truckee), I've never been to the USA myself. To model the FEF, I'm relying on drawings, photos, videos and anything that my good friends here can send my way.

One questing concerning realism, do you think it is possible to animate the reverser position into the valve gear movement? Not the cut off setting per se, but rather the forwards, neutral and reverse position of the Walschaert's gear?

I looked at this when I was doing the Connie but it's highly impractical (virtually impossible). To get a smooth animation of the running gear and valve gear for a given position of the reverser, I have to split each 360 degree revolution into 40 steps (it's just like cartoon animation, where things move a small distance between each animation frame). In each step, I have to position each moving part (and there are many) manually. The automatic animation plug-ins simply can't handle it, except for the wheel rotation (where I place four frames, for each 90 degree rotation and let 3DC fill in the gaps), so I have to do it manually. This is very tricky and sometimes involves changing the position of the group centres to overcome a floating-point bug in 3DC that causes inaccuracies leading to jumps in the animation. It can take a couple of days to complete the animation for just one position of the reverser. Now, to do that for every position of the reverser (which is also split into 40 frames), it would take months. The result would be a set of 40 animation timelines, one per reverser position. But then there's another problem. The animation of the wheels + running gear + valve gear are done together, but the animation of the reverser cannot be included in the same animation - it has to be done separately. Then the problem is how to enable the correct animation from that set of 40, when the reverser position changes, and how to transition between them smoothly, without the animation starting from the beginning (this would be seen as the wheels instantly jumping to the start position each time the reverser is moved, which would look awful).

Something I might experiment with is to just have three animation timelines, for full reverse, neutral and fully forward, but only if I solve the problem of how to set dynamically the start frame for the animations, to avoid the wheels instantly rotating to frame #1 in the timelines.

[_o_OOOO_oo-Kanawha]

Thanks for your eleboration. I thought it could only be done by compound animation, which obvious from your reply, isn't possible in RW.

With Walschaert's valve gear, it would add a nice visual touch. On the FEF, that big and bulky frame surrounding the reverse quadrant blocks most of the view anyway.
That reverse quadrant hanger looks just like with Baker valve gear, so it is probably a standard ALCo casting.

UP never fitted lightweight roller bearing rods to its FEF's?

That was probably because of the fact that Walschaert's valve gear still had sliding surfaces susceptible to wear and intolerances thereoff.
It looked ever so modern on those late 40's steam locomotives.



I believe the Q2 was the only freight locomotive to have roller bearing rods, but perhaps some NYC 4-8-2's got them when they were rebuild with Scullin disc drivers.

[mrennie]

I believe the Q2 was the only freight locomotive to have roller bearing rods, but perhaps some NYC 4-8-2's got them when they were rebuild with Scullin disc drivers.

Timken fitted their roller bearings to a stock ALCO 4-8-4, officially TRBX 1111 but known informally as the "Four Aces", in 1929, for trials and as a demonstrator. Two Northerns on the DL&W got them some time later, in the 1930s.

[_o_OOOO_oo-Kanawha]

Four Aces only had roller bearings on all axles, drive gear still had solid bearings IIRC.
UP 8444 still seems to have the tapered side rods that initally were also fitted to Santa Fe's 2900 class war babies.

Roller bearing side rods and needle bearing valve gear linkage was only possible with rigid cast engine beds, a later innovation.
Some roads later replaced the tandem side rods linking #2 and #3 drivers with singles.
In the ATSF 2929 restoration website there are many photographs showing the assembly of the roller bearing rods and the fitting to the drivers.


The previously unheard of close tolerances necessary for the lightweight drive gear were only possible with those one piece engine beds with integrally cast cylinders.
These were foundry masterpieces that are probably impossible to reproduce today as the necessary plants and skills are no longer there.
Norfolk and Western's splendid J-class 4-8-4 are another example of thoroughly modern US passenger steam locomotives.

She does look a little old fashioned to me because of the Alligator crosshead guides and spoked drivers.

[UP3985]

The later N&W A Class 2-6-6-4s had them along with the ATSF 2900 4-8-4s.

[mrennie]

Four Aces only had roller bearings on all axles, drive gear still had solid bearings IIRC.
UP 8444 still seems to have the tapered side rods that initally were also fitted to Santa Fe's 2900 class war babies.

You're absolutely right. UP 844 does indeed have tapered side rods. I haven't modelled the tapering though because it's hardly noticeable and it saves a few polygons if I omit it. All the other dimensions are spot on.

[_o_OOOO_oo-Kanawha]

Four Aces only had roller bearings on all axles, drive gear still had solid bearings IIRC.
UP 8444 still seems to have the tapered side rods that initally were also fitted to Santa Fe's 2900 class war babies.

You're absolutely right. UP 844 does indeed have tapered side rods. I haven't modelled the tapering though because it's hardly noticeable and it saves a few polygons if I omit it. All the other dimensions are spot on.

Mike it is, right? I don't know about not modeling the tapered side rods, they are an essential and distinguishing feature. Unless you texture them very dark and dirty so they won't stand out from likewise dark and dirty wheels. I do like the spic and span shining rods in the top photo, but shining rods perhaps don't look good in the game. The fluting, which is milled/chafed out on the prototype, can be simulated by the texture. A few polygons don't matter that much, do they?

Here, by comparison are the model rods of the MSTS WP GS64, currently under construction at Elvas Tower:


I am hunting for more good pics of light weight drive gear, from the streamlined art deco NYC J3a "Super Hudsons" which looked great with their Scullin disc drivers, C&O's "Poppets" ( also 4-6-4's) and C&O's Greenbrier 4-8-4's, the big Santa Fe 3765 and 2900 class 4-8-4's and more.

[mrennie]

Mike it is, right? I don't know about not modeling the tapered side rods, they are an essential and distinguishing feature. Unless you texture them very dark and dirty so they won't stand out from likewise dark and dirty wheels. I do like the spic and span shining rods in the top photo, but shining rods perhaps don't look good in the game. The fluting, which is milled/chafed out on the prototype, can be simulated by the texture. A few polygons don't matter that much, do they?

Yes, it's Mike

I was referring to the tapering of the ends of the side rods as seen from above (perhaps we're not talking about the same thing).

Here's how the rods look at the moment (simulating the shiny rods you see on UP 844 today):

2013-04-20_00001.jpg

2013-04-22_00002.jpg

2013-04-27_00001.jpg

[_o_OOOO_oo-Kanawha]

Sorry Mike, I think we misunderstood each other. So UP 844's rods were also tapered in width. I think that is an detail that can be omitted as nobody wants to stick his head between a locomotive's drive gear.

Superb modeling you have done so far! That looks very much real!

Here is some more pics of other engines, no doubt the same pictures can be found with the FEF-3
A modellers nightmare for sure in this perspective/view, a close up of C&O's Baker valve gear, power reverse, lubricator (?) and other appliances.

Notice the tandem rods linking drivers #2 (main) and #3.

Oiling around, giving you some idea of the sheer size of a main rod bearing.

With your experience in modeling the FEF-3, there are many other big 4-8-4's people would gladly buy.

[FourEightFour]

UP never fitted lightweight roller bearing rods to its FEF's?

Rods themselves are of the I section design which provides the most favorable strength to weight ratio. The side rods have no knuckle joints on the front and rear sections. The front side rod joined the intermediate rod between the latter's jaws. A brass bushing on the front rod operated on a steel unit of the intermediate rod. The back rod was the same as the front and were all channeled to reduce weight, the sections being narrower. The rod design of this engine was one of the most noted advancements in design, considering these engines were mostly designed in 1936.

[mrennie]

Rods themselves are of the I section design which provides the most favorable strength to weight ratio. The side rods have no knuckle joints on the front and rear sections. The front side rod joined the intermediate rod between the latter's jaws. A brass bushing on the front rod operated on a steel unit of the intermediate rod. The back rod was the same as the front and were all channeled to reduce weight, the sections being narrower. The rod design of this engine was one of the most noted advancements in design, considering these engines were mostly designed in 1936.

This series of screenshots illustrates it quite well:

2013-05-02_00004.jpg

2013-05-02_00003.jpg

2013-05-02_00002.jpg

2013-05-02_00001.jpg

[bpetit]

Man you will raise the bar so high RSC wouldn't be able to top you for a while.