Powerful new curve radius insights for any scale

And one that gives a far more relevant meaning than the old "sharp, conventional, and broad" curve descriptions.  Those were good for their time, when flextrack was in its infancy, and many layouts featured bowl-of-spaghetti track configurations.  Since the dominant freight car in those days was the venerable 40-footer, it seems that they were on the right track.  However, as freight cars grew, so did the minimums needed for reliable operation.  Passenger cars were always a special needs population as far as our model empires were concerned.  A couple of the railroads I operate on have passenger operations, with most (if not all) of their switching taking place on straight track.

I look forward to more articles like this, Joe.  It's an exciting time be be a model railroader.

Joe,

Since I am currently immersed in planning our layout this is very timely information.  Thanks for a great article!

Armed with these simple rules and the insights I gained from Tim's article on S-curves I feel that much closer to being "qualified" to plan my layout.

Thanks again!  As I expected the first issue has blown me away.  I brought my wife and son over to the computer on a number of occaisions today when they heard me exclaim, WOW!  The whole family has enjoyed looking through issue # 1.  I'm the only one in the family that even turns a page in the print mag I get regularly.  That says something for the power of rich interactive media.

I had read about the ratio method before but seeing the photos illustrating through example was very nice. I think too often guidelines are given without providing "real" examples of why the guideline is in place.

Chris

Last year the Australian Model Railway Association (AMRA) adopted a minim radius standard based on a series of formulae. For bogie vehicles the AMRA standard used 3x the length as the recommended minimum radius.  An excellent article that confirms the assumptions used in the AMRA  standard are valid.

Terry Flynn.

I found the article on projecting curve radi very informative.  Very interesting to know you can actually compute radi based on your car length.  Good info to have.  Thank you for a great article.

Terry in FLorida

I have to thank you for your time and effort that you put into this great article,your description is just plain "EXCELLENT" and very clear photography.

I have also taken some of the plain-jane Locos and converted them to take the decoders with sound. Had good luck with them.

Thanks again.

James.

Wow!  Another outstanding article that really "shows" what the standards define.  Timely too, as I am planning a major expansion of my European layout this year.

David Martin, Ph.D.

Best article in the issue IMO.  I would like to see a similar approach taken with easement curves.  Would be a good follow up to this I think

Agreed. I always see suggestions for the "lay out a cubic spiral". However I have yet to find any instructions for actually doing so. Apparently it's so darned blatantly obvious to everyone else but me. :(

Apparently the "stick" method is where you threaten someone else to do this for you. :D

-
S

Joe Most excellent work on your part. Just as I have supected for years thank you for doing the leg work and the math. It is a very good way to guesstamate how a plan will operate. Another formatable tool to add to the tool box of model railroading.

This is indeed a very timely article. Without even knowing it I've used this method before, when planning layouts, just never knew what it was called. IIRC they were discussing this a long while back on the LDSig.

CJ

Joe,

A great article!  A follow-up article on easements and superelevation would be equally helpful with a particular focus on the length of easements, the length of the transition from level to superelevated track, and the amount of superelevation that is appropriate.

Joe - I cannot believe how simple that seems to be - but I can understand it and thanks to you for making it easy to understand, as usual.

The magazine is a success for me right off the bat, and I had to download it in a marina deep in Guatemala. Thanks for keeping me occupied with trains instead of the work of my sailboat.

Cheers, Jake

This article cannot be applied directly to what track curves/radii a steam engine can be expected to navigate, i.e. the engine without it's tender. With engines such as 0-6-0, 2-8-0, 2-8-4  and 4-6-4, one is not concerned about coupling but with what curves the series of 6 or 8 fixed wheels will navigate.

Could there be a similar kind of calculation that would apply to curves various steam engines can be expected to navigate?

Malcolm Horsnell (Modelling the Toronto Hamilton & Buffalo Railway)

I'd expect you will want to measure the length of the rigid wheelbase of any steamer and then probably go for a 4x factor to have minimum issues with binding, etc on the curve.

Of course it gets more interesting with things like articulated locos, since they may track well on 4x of the rigid wheelbase, but they will look like heck.

So for looks, the length of the equipment guideline still applies. 3.5x of the loco's actual length will look decent from inside the curve, 4x will look decent from outside the curve - and at 5x the loco will couple just like on straight track.

Locos, because of the mechanism tied to the wheels, are something of a special case. You should always test out your locos on various curves before  you commit yourself to that 20 year dream layout project only to find your favorite loco won't run worth a darn on that minimum radius!

IIRC they were discussing this a long while back on the LDSig.

The results of that collaboration are still up on the LDSIG Wiki (also maintained by our host Joe F) and are a good compact reference to complement the visual article in MRH.
Curve radius rule-of-thumb

I felt that was the best article in the issue as well.

Nice work, Bone Daddy!

How many of the drivers are "blind" - where they don't have any flanges? The more blind flanges, the tighter the radius (and probably the more ridiculous looking, but there you go).

 As I understand it, even prototype locomotives had blind drivers.

These guidelines are going to come in very handy in planning "aesthetic" vs. "reliable but tight" curves on a layout.

For my predominantly 40' cars, I can use the 5x (37.5") radius for the visible (& switching) sections, and stick to 2.5x or 3x (18.75" or  22.5") radii in the secondary (& hidden) areas. Should I ever want to realistically run 85' heavyweight passenger equipment, I know I should be planning on 63.75" (4x) radius or greater.

I'd better plan on a larger house purchase; I'm going to need 12' of width just to turn a passenger train around :-)

Solved my planning issues ... I will replan my project for apprpopriate curves ... doing Amtrak N-scale passenger sercice between Seattle, Edmonds, and Everett. Cars will be +6" so I'll look at 5.0, 4.0 and 3.0 for my staging/holding pen. Don't need to switch/decouple on curves, but trains may be double ended to simplify operations.

Great article--leads me to your supporting adveretisers.

Bill, Edmonds

Hello Joe,

firstly, a very interesting article with clear explanations and useful pictures.

Second, is that so simple? When I ended reading your article I thought: wow it's totally straightforward I should have got to the same results If only I were smarter than I am.

A really nice way to establish the behaviour of cars on a given curve before buying the first piece of equipment and in addition totally scale-free.

Third, it came perfectly on time for me since I'm starting to plan my first layout and I'm eagerly looking for as many info as possible. And this article is going to be a reference for me.

Regards, Ricky4208.

P.S. as many other have written it would be nice do follow up with other articles about easements, superelevations and all that is related to what you can to in order to predict beforehand how you equipment will perform.

This article makes me feel a lot better about some things in my track plan. Salmo BC may yet have passenger service!

Thought the article on curve radius guidelines was quite eye opening!

I'm in the process of designing a new HO layout and am severely limited by an 8' room depth (20' long). I'm modeling the C&NW in and around Chicago and one of my "givens" is running bi-level 85' commuter cars. As much as I'd love to have 40" radius curves that ain't gonna happen!

So, here's my thinking - since these cars would only be viewed from inside the curve (it's going to be an around-the-walls layout) could I minimize the downside of a smaller radius by increasing the layout height?

My thinking is the lower the layout height the more pronounced the car overhang would be perceived.

Make sense or not?!

Chuck

Chuck:

Yes, that's a great application of the curve radius guidelines, plus the value of high benchwork. You can drop to about 3.5x and view the curve from only the inside at about armpit height and your commuter cars should look reasonably okay. However, as I illustrate, they may need coaxing to couple, and any diaphrams may not mate perfectly.

I find expressing curve radius as a ratio of rolling stock length suddenly causes the whole radius question to make much more sense. And if you relate it to how the rolling stock looks on the curve, and how well things couple on a curve - you're now considering all the proper variables that need to be considered when deciding on a minimum radius.

And your point of getting the track level higher also works as a visual trick to allow sharper curves, especially when viewed from the inside.

Well, here it is, over 4 months since the MRH article (Jan. 2009) on radii as a ratio of car length.  I find that I am automatically thinking in those terms as I design and redesign on paper and on plywood.  If the radius is not at least 3X the expected car length (and aiming for over 4X), I'm asking myself why I should lay out the track this way, rather that try a different solution.  Thanks once again, MRH.

You're welcome, Don.

I find the ratio of equipment length approach really anchors the whole radius question in reality much better and keeps you from cheating so much. Or if you do cheat, you at least know what you're giving up (easy coupling, looks less real, ... )

I hope MRH can do more articles like this that really make concepts practical.

Now we just need a similar "ratio" method for determining min frog number..

There is what looks good.. what runs well.. and what at a bare minimum will run. I personally think that many folks use too high a frog number for their space.. I have never actually seen anything to back up assertions that using a #6 is more reliable then a #5. At least, not an apples to apples comparison where a repeatable set of tests were performed on turnouts otherwise indentical.

I did read an article a while back, a long while back, which I THINK was a John Armstrong article that showed how a #5 was plenty big in comparison to any reasonable model railroad min radius. My understanding is that a N scale #5 has a radius equivelent of 24inches.. which is huge.

Its been my admittedly limited experience that the way you arrange and tune the turnouts has more impact then the size. Don't get me wrong.. watching a train glide through a #8 is wonderful to watch.. but are you really serving your layout operationally by taking the space? Maybe.. but I want to see some real world results..

Chris

I'm not normally a fan of Wikipedia, but I found this article on railroad switches informative. This speaks of a higher numbered turnout (ie a shallower diverging route angle) based on speed, and not necessarily car length. The shallower the angle, the higher the through speed.

In modeling, I'm not sure how relevant speed is. I think your article suggestion is a good one; we do see a difference in turnouts with longer cars at our scale when it comes to coupling performance and aesthetics.

I personally settled on 5's instead of 4's per Lynn's book. On the mainline, I'm running sixes, in order to do what you specify - save space without hindering performance too much.

One dynamic not covered in the curve radius guidelines is the transition zone from straight to curve.

In this transition zone, the prototype often does a spiral easement, meaning the straight track slowly gets "curvier" until the curve reaches its full sharpness. This eases the equipment into the curve, and makes it track better, makes it less likely that loads will get thrown around inside the cars, avoids throwing passengers or crew around, and so on.

This curve transition zone also makes the perfect place to go from level to superelevated on curves.

By definition, a turnout is always a transition zone (going from straight to curved), so using a broader radius turnout than your minimum radius creates a sort of easement - which is more prototypical. Using larger radius equivalents for by going with a larger frog number actually makes all kinds of sense when you consider this need to have easements on your curves if you want to model prototype track alignment more properly.

When you do include easements, broader frog numbers, and superelevation (if it's appropriate for your prototype), the track just looks more real and it feels "right". Many people won't be able to put their finger on why your track looks so good - but instinctively it will resonate with the eye as more real.

So you can't just blindly apply the curve guidelines in this transition zone. By adding correct easements to your curves, you'll end up offsetting them maybe half an inch in HO, so the easement doesn't dramatically alter your track placement - but the visual impact of proper easements and larger frogs on turnouts is significant and not to be underestimated.

One reason why I love spline roaded so much is it adds these easements automatically. And using #6, #8, or even #10 turnouts adds these easements to your turnout locations automatically. I also believe using layout planning software like 3rd PlanIt gives you these easements automatically in your curves if you don't use sectional track.

I do use easements and stubbornly believe EVERYONE should.. and per John Armstrong's "Track Planning for Realistic Operation" book I treat the curved leg of a turnout as an easement.

Anyway.. your point is well taken that a longer turnout will create a better easement. I have a more pragmatic reason for trying to stay with one turnout size.. the cost of the fast track jigs. I still hope someday I can afford one. If I discover I can get good results without then I will start using more a variety.. I really should work on that but there are so many other things to do..

The shallower the angle, the higher the through speed.

I was just reading something about that somewhere. I think it was in John Armstrong's book "The Railroad: What it is, What it Does" which had a nice chart showing turnout and curvature vs speed. I will have to double check.. but its the last book I read that had stuff like that in it..

Chris

btw- XtrkCad also does easments. Unfortunately XtrkCad does not let you set a fixed length of easement. Its proportional to the change based on a constant. I think it was Andy Sperandeo, not sure, wrote a nice piece a while back explaining why for model railroad purposes the length of the easement should be relative to the length of your rolling stock rather then the degree change in curvature.. So... I would like to be able to specify "6 inches" say. and have that be constant. I get around it.. but I have to fuss with it a bit at times.