Let's talk radii

[SP-Wolf]

Out on my mine line - I am using a min of 21"R. Same for going into and out of the yard. My switch complex is a bit different. I went with what ever radius would fit a particular industry. (I have no idea what size they are). For me, it doesn't matter.

Thanks,
Wolf

[crrcoal]

Something I didn't see mentioned yet is to plan for easements when using smaller radii.  They won't help with the visual, but will definitely aid in operation.

Jason

Doodling on xTrackcad for my layout I came up with 16-15-13-13-13-13-15-16  for the main curve......

[John]

I have one area thats 15" (Wheeling Jct) and it's signaled for restricted speed (15mph) .. 89" go through there ok, but don't go fast .. the rest is 22" or 24"

[Santa Fe Guy]

Ed, I would definitely add some super-elevation on the curves. I have it on my main visible curves and seeing the trains lean in looks damn cool. [ Guests cannot view attachments ] As a side issue I added drainage ditches beside all of my mainline tracks to make the mainline ballast stand out even more.
Rod.

[pdx1955]

I'm running 17/19" inches on my "mainline" double track and it seems to work fine and looks good as well. A number of the curves are at least double that though. The larger cosmetic curves 30"+ look really good on a mainline if you have the right sight angles set up.

Peter

[basementcalling]

A sharp thread on a broad subject.   

[coldriver]

I'd suggest going with 57" radius curves, like this one.  Unfortunately this was the only one I could fit on the layout...   My mainline minimum is 22".  I had a peninsula-ender at 19" that I ripped out and replaced with a 22" while sacrificing 6" of aisle space.  Yes, it did make a large difference (to me anyway!). 

IMG_1339 by Dean Ferris, on Flickr

[bnsfdash8]

I'm stuck with using at max 17" radius curves. My modern equipement sometimes looks a little goofy going around it but I don't have any issues with any of it. Whenever I can upgrade from a HCD layout I'll stick with 20"+ curves.

[robert3985]

Lots of good information here, and I probably don't have a lot to add to what's already been said. 

However, let's see if I can sum it up.

(1) 18" bare minimum mainline radius.  Do your best to keep it in areas you can't see too well (cuts, fiddle yards, tunnels, etc.) or on branchlines.  It will function just fine, but suffer a bit in the appearance dept.

(2) 16" bare minimum industrial/commercial trackage.  This is restricted to smaller-wheelbase motive power...just like the prototypes do it.

(3) 24" minimum preferred mainline radius.  This is the MINIMUM...if possible, make other curves as broad as you can.  I have several mainline curves with radii exceeding 100", and as John Armstrong knew, they are very photogenic.

(4) Superelevate your mainline curves.  This is a really great way to make running and photographing your trains much more realistic.  Be cautioned to not do it too much...I consider Kato's superelevated Unitrack to be WAY TOO MUCH...about half their angle would have been much better looking.

(5) Use easements...especially on smaller radii mainline trackage.  This really makes your trains glide smoothly from straight track to your fixed radius curves.  It DOES take up a bit more room, but operationally and visually it really makes a huge difference...especially with longer cars and engines.

(6) Get a good layout planning CAD program.  The one you have now isn't fitting the bill!   

I think that about covers it.

In addition, I have several large curves on my layout that are nothing but spiral easements, with a short fixed radius connecting them in the middle at the apexes of their smallest radii.  Trains really look good going through these spiral tracks, especially noticeable being the passenger trains, with car ends not wagging around anywhere on the curves or transitioning to straight trackage.

If anybody isn't already aware of it, you should be advised that Atlas 55 turnouts have a much sharper effective diverging track radius than equivalent prototype turnouts do because of the disproportional shortness of the closure rails on all of their turnouts.  This means their #7 turnout has an effective diverging radius equivalent of a properly proportioned #5.5 (approximately) turnout.  The Atlas 55 #10's diverging radius is approximately what a properly proportioned #8 would be. 

On the other hand, Micro Engineering #6's ARE properly proportioned and their effective diverging track radius is slightly larger than the Atlas #7's.

IMO, if you're concerned about what radii you're using on your layout, you should be aware of the eccentricities of the Atlas 55 turnout diverging track effective radius.

Photo (1) Big Blow on spiral radius, superelevated mainline trackage at Echo Curve heading east out of Echo to Wahsatch:


Photo (2) UP Livestock Dispatch leaving the signal at Echo Yard.  Engines are on 71" radius trackage, and cars in the distance are rolling on a 106" radius curve:


Cheerio!
Bob Gilmore

[coldriver]

Photo (1) Big Blow on spiral radius, superelevated mainline trackage at Echo Curve heading east out of Echo to Wahsatch:


Photo (2) UP Livestock Dispatch leaving the signal at Echo Yard.  Engines are on 71" radius trackage, and cars in the distance are rolling on a 106" radius curve:


Cheerio!
Bob Gilmore

Wow Bob, those are some of the best looking model railroad curves I've ever seen!  All hail N-scale!

[nkalanaga]

A 10 degree curve is sharp for a modern mainline, but there are quite a few out there.  In N scale, it scales to just over 43 inches.  Anything over 36 would look "right" in scenery where sharp curves would be found.

A prototype GP28-2 will take a 140 foot radius - by itself.  Coupled to a 50 ft car, the limit is 302 feet, or about 22.5 inches in N scale, according to the EMD operators manual.

As others have said, all of your curves don't HAVE to be the minimum radius.  If you have room, widen any that you can.  Even the prototypes don't use the same curve everywhere.  Curves don't even have to be circular.  The radius can change within a curve, as long as none of it is below the minimum.  And, yes, easements do help, where there's room for them.

[wm3798]

Consider the following prototypical considerations:

You're modeling York PA in a heavily industrialized area, so tighter curves than the California desert should be expected.

You're modeling the Northern Central, the second line out of Baltimore only to the 1827 route of the B&O.  Despite that early start, York was there before the railroad, so tighter curves than Montana should be expected.

You're modeling the Atlantic Piedmont region, where older rail lines followed narrow water courses through undulating countryside, so tighter curves should be expected.  Okay, not as tight as say the M&P, but you're in the same neighborhood, so you're not building a race track.

Some modeling considerations:
You're modeling primarily yard operations, so the main visual consideration should be how the trains look approaching and departing the yard.  How they get to staging or to the return loop are minor issues visually, but should be accounted for operationally.

Longer cars will be in play, the NC was a key route for Truck Trains during the PRR and PC eras, so 89' flats and auto racks should be in your fleet.  Make sure your curves play nicely with them operationally to get them in and out of staging, and don't worry so much about the camera angles.  Again, my layout handled this traffic just fine on 15" r even in the helix.



Note that the outer loop was 16.5 r and designated for the up hill run to reduce the effective grade slightly.

Lee

[jagged ben]

Train length effects whether you can run certain equipment on tight curves and steep grades.  Or at least, it affects where it can be in a train.   If you're trying to run a 10ft or longer train up a 16.5" radius and 2.5% grade.... you'd better not have 89' flats with body mounts up front. 

I think in general it will hold true that if your space is tight enough to warrant smaller curves then you're also going to end up with shorter trains, so that mitigates the issue somewhat for layouts with less space.   The pitfalls are for the guys with a lot of space who want to take advantage of it and really run long trains.  You've got to take advantage correctly or you'll be disappointed when you (or your buddy) says 'oh come on, let's see how long a train you can run'.  Which will happen.   

[crrcoal]

So photo #1: what is the radius here we are looking at? Also what is the width of the layout here?

Your overall post is sticky material imho. 

Photo (1) Big Blow on spiral radius, superelevated mainline trackage at Echo Curve heading east out of Echo to Wahsatch:



Cheerio!
Bob Gilmore

[wcfn100]

(4) Superelevate your mainline curves.  This is a really great way to make running and photographing your trains much more realistic.  Be cautioned to not do it too much...I consider Kato's superelevated Unitrack to be WAY TOO MUCH...about half their angle would have been much better looking.

(5) Use easements...especially on smaller radii mainline trackage.  This really makes your trains glide smoothly from straight track to your fixed radius curves.  It DOES take up a bit more room, but operationally and visually it really makes a huge difference...especially with longer cars and engines.

Maybe because it's just obvious, but it never really gets mentioned that super elevation is directly tied to easements.   They start and end together i.e. where the SE reaches it's highest point is where the easement becomes tangent with the curve.

Or another way to say it, if you have super elevation, they should be coupled with easements.

Jason