Brian,

I can't offer any specific answers to your question, I have an Allegheny, but as you know my curves are big......

But, I will share another basic principal of railroad physics often missed about steam locos and curves.

The most important factor for steam locos and curves is the rigid wheel base of the drivers. This controls minimum radius more than any other factor.

So, for example, the Allegheny has a very short rigid driver wheel base for such a large loco, only 11.83' for each three axle set. This makes the loco very nimble for its size. Especially considering the double swivel design of the model.

The Allegheny will work better and look better on any given curve than, for example, the C&O 4-8-2. Why?, because the Mountain has a rigid driver wheel base of 18.25'.

Here on the ATLANTIC CENTRAL, even with 36" radius curves, I limit steam loco rigid wheel bases to 21', with a few USRA 2-10-2 lights being the longest rigid wheel base. 

Examples of locos prohibited due to length:

SP GS4 4-8-4 - 21.5' rigid wheel base

B&O S1 2-10-2 - 22.30' rigid wheel base

Here is an example of how driver diameter effects rigid driver wheel base:

USRA 2-8-8-2, 57" drivers, 15.5' rigid wheel base

Compared to that GS4 above, 80" drivers, 21.5' rigid wgheel base

Eastern prototypes tended to have smaller drivers and were more suited to sharper curves.

A story from the prototype: The B&O has two main routes west, one directly west from Baltimore to Cincinnati, it is very curvy, and has many up and down grades. The primary large steam on this line were the 2-8-8-0's and the EM-1 2-8-8-4's, rigid wheelbase, 16.50'. Longest locos commonly used on that run, T-3 Mountains, 18.25' driver wheel base.

The other goes north thru Pittsburg to Chicago, it has longer, straighter, fewer (but just as steep) grades, and much broader curves. The primary steam on this line were the famous S1 2-10-2's, 22.30' driver wheel base.

At one point someone thought it would be a good idea to try an S1 on the line to Cincinnati - they put the loco on its side in a curve.........and put another on the ground on a different try........twice was enough.

This is exactly why the LIMA built the Allegheny as an 2-6-6-6, because they wanted larger drivers, they knew they would get better performance with a short wheel base, so this just put the max axle load on 6 drive axles rather than spread over eight like a Big Boy. Same is true of the N&W Class A.

While the C&O had ok success with the LIMA 2-10-4's built in the 1930, notice how they did not build more, but rather opted later for the Allegheny and the 2-8-4's - because the 2-10-4's were hard on the track, and hard on themselves in the curves.........

Way before these modern locos the C&O had a fleet of nearly 300      2-6-6-2's, not much more powerful than a 4-8-2. Why have the complexity of articulation for only that much power? Because they were more nimble, rigid wheel bases all around 10'. Worked so well they build the last ten in 1949......

I have lots of these on the ATLANTIC CENTRAL as well:

All of these prototype facts apply to your models as well.

So you can have big locos, on smaller curves, just pick the nimble ones......

Sheldon