Trains.com

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Minimum radius for a real main line in HO-scale

25849 views
47 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    February 2017
  • 19 posts
Minimum radius for a real main line in HO-scale
Posted by CP Modeller on Wednesday, February 21, 2018 4:47 PM

I see there is a thread going on here regarding the minimum radius for model train equipment but I was wondering what the minimum radius would be if I was to model an actual minimum radius of a main line. I have heard in the past that it is. somewhere approaching 30' radius but that would seem to be an almost impossible curve to model. Is this correct or am I way off the target???

Regards  Colin  

  • Member since
    January 2010
  • From: Fruita, CO
  • 541 posts
Posted by slammin on Thursday, February 22, 2018 9:45 AM

Welcome to the forum Colin. I remember a thread that discussed how much room it would take to model a real curve. Prototype road measure their curves in degrees. In HO a 30" radius is not consider "big". Large home and club layouts have 48" and larger radii when modeling class 1 railroads. Most us don't have that kind of space and tend to model short lines and branchlines that look reasonable running 30" and smaller radii.

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Thursday, February 22, 2018 10:20 AM

 Real railroad use degrees, because unlike a model, you can't just sit above the earth adnd use a giant hand to draw out a radius. You have to use surveying instruments and measure how far you offset over a given distance. Common distance is 100 feet, so a 1 degree curve changes 1 foot for every 100 feet. This gets you the chord of the circle (in this case 100 feet) The radius of such a curve can be computed as Chord Length/2Sin(Degrees/2). Say you have a 4 degree curve, that's 100/2Sin(4/2), 100/2(.0349), 100/.0698, 1432 feet. In HO that's 16 feet. A 2 degree curve is about 32 feet. A 10 degree curve is about 6.6 feet in HO. You can work backwards and figure what degree curve a typical HO radius is, but the real world equivalent of 18" is pushing it even for trolleys. Luckily physics does not scale and we can get away with things in the model world that would never work in the real world, or we would be forced to work in Z scale or smaller.

                                     --Randy


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    February 2005
  • From: Vancouver Island, BC
  • 23,330 posts
Posted by selector on Thursday, February 22, 2018 2:10 PM

CP Modeller

I see there is a thread going on here regarding the minimum radius for model train equipment but I was wondering what the minimum radius would be if I was to model an actual minimum radius of a main line. I have heard in the past that it is. somewhere approaching 30' radius but that would seem to be an almost impossible curve to model. Is this correct or am I way off the target???

Regards  Colin  

 

I believe some early roads used 100' chains to run a cord to the next intersection point along the circumference of a section of a circle, that section comprising the needed curve to maintain grade.  They'd specify a degree of curvature, meaning the cord line had to angle from the tangent at its first reference point at so many degrees.  Degrees in the 6+ range meant the trains would be restricted to slower mainline speeds there.  Less than about 3 deg meant higher speeds, and less than 1.5 deg or so meant high limited speeds one would expect from passenger express trains.

http://trn.trains.com/railroads/ask-trains/2011/01/measuring-track-curvature

For a more robust answer:

https://mysite.du.edu/~jcalvert/railway/degcurv.htm

 

 

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Thursday, February 22, 2018 2:42 PM

I just measured the curve radius of the UP (formerly WP) mainline to Oakland at Sunol, CA.  I picked a near 180 degree turn just a bit east.  It's 1000'.  That's 11 1/2 feet in HO.

It is not, by the way, a high speed section of the railroad.

Ed

  • Member since
    February 2002
  • From: Mpls/St.Paul
  • 13,892 posts
Posted by wjstix on Thursday, February 22, 2018 3:04 PM

Randy's statistics sound good. I recall in one of John Armstrong's books he said that in HO I think a 30-something inch radius curve (forget exactly what) worked out to be about a 10-degree curve. 10-degrees would be about the tightest curve for a prototype mainline you'd find, and would limit trains to like 15-20 MPH.

BTW same goes for turnouts, our model passenger equipment will run nicely through No.6 turnouts, but the prototype would use No.20 or larger.

Stix
  • Member since
    October 2008
  • From: Calgary
  • 2,047 posts
Posted by cx500 on Thursday, February 22, 2018 4:11 PM

The magic formula is radius (in feet) = 50 divided by the sine of D/2 where D is the degree of curve.  A 12-degree curve (rare, but sometimes unavoidable) works out to be 5'-6" in HO scale.   Remember to run your trains very slowly through that sharp a curve thoughSmile.

John

  • Member since
    December 2001
  • From: Northern CA Bay Area
  • 4,387 posts
Posted by cuyama on Thursday, February 22, 2018 4:21 PM

wjstix
I recall in one of John Armstrong's books he said that in HO I think a 30-something inch radius curve (forget exactly what) worked out to be about a 10-degree curve.

A 10-degree curve in real life is over 570' in radius and thus about an 80" radius in HO. 

  • Member since
    September 2003
  • 21,669 posts
Posted by Overmod on Thursday, February 22, 2018 4:40 PM

Surprise

Real railroad curves are impossibly long for indoor layouts.  I wanted to simulate 100-mph mainline traffic and I did not have room on a 2-acre property, outdoors, in HO scale, to model the necessary 'prototype' radius.  As noted, you could fit something 45-55mph track speed into something that limited 

  • Member since
    July 2009
  • From: lavale, md
  • 4,678 posts
Posted by gregc on Thursday, February 22, 2018 5:49 PM

Here's a table for maximum-curvature/minimum-radius for Reading steam locomotives.

i've converted some to HO and listed the loco types

    6.9"   50'  OE1 boxcab

10.3" 75' A4 0-4-0 13.8" 100' B8 0-6-0 17.2" 125' B7 0-6-0
32.8" 238' D5 4-4-0 35.9" 260' E5 0-8-0 37.7" 273' I6 2-8-0
39.4" 286' I8 2-8-0 39.4" 286' M1 2-8-2
41.4" 300' D11 4-4-0 41.4" 300' I10 2-8-0 41.4" 300' Q1 2-6-4
45.9" 333' P3 4-4-2
49.4" 358' G1 4-6-0 49.4" 358' L10 4-6-0 49.4" 358' N1 2-8-8-2 49.4" 358' P7 4-4-2

greg - Philadelphia & Reading / Reading

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Thursday, February 22, 2018 6:05 PM

 Welp, can't even run a D5 on my layout, and I can totally forget about the I10.

Good news, Boxcab 50 and 51 (or 98 and 99, depending on what year you are talking about - they have to be 98 and 99 for me, as I have S-1 50 and 51) can take a tighter radius than the Roundhouse models can manage. I don't think the trucks on mine turn enough to do under 7" radius.

                           --Randy

 


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    December 2008
  • From: Heart of Georgia
  • 5,406 posts
Posted by Doughless on Friday, February 23, 2018 7:55 AM

cuyama

 

A 10-degree curve in real life is over 570' in radius and thus about an 80" radius in HO. 

 

Bumping this for its importance.  

With seemingly more vocal angst over prototype fidelity these days, its important to know that we suspend realism once we decide to build curves (among other things).  But we do the best we can.

- Douglas

  • Member since
    February 2002
  • From: Mpls/St.Paul
  • 13,892 posts
Posted by wjstix on Friday, February 23, 2018 10:09 AM

I've often thought that the comparatively sharp curves we have to use is another reason to run trains slower than the prototype - running a model train at 70 scale MPH on a curve that a real train would have to slow down to say 30 MPH doesn't seem to me to be as realistic as running the model train at 30 scale MPH is.

Stix
  • Member since
    July 2009
  • From: lavale, md
  • 4,678 posts
Posted by gregc on Friday, February 23, 2018 10:27 AM

i think there's a common misconception that it is the flanges that keep a train on the track.   They probably do for models, but not for the prototypes which have smaller flanges

see Dr Feynman's explanation 

https://www.youtube.com/watch?v=y7h4OtFDnYE

greg - Philadelphia & Reading / Reading

  • Member since
    February 2008
  • From: Potomac Yard
  • 2,767 posts
Posted by NittanyLion on Friday, February 23, 2018 12:19 PM

The tracks outside my office window are in the low 20s and long equipment looks like it is hanging over like an 89' flat on a 22" curve. Flanges scream on the garbage flats.

It would be a 139" curve in HO. 

  • Member since
    July 2009
  • From: lavale, md
  • 4,678 posts
Posted by gregc on Friday, February 23, 2018 3:00 PM

base on Dr Feynman's explanation, the minimum curve radius is related to ratio of the circumferences of the rails and the ratio of the wheel diameters due to taper.   The following drawing indicates that there's only about 0.2" of taper per wheel.   This corresponds to curve radius of 742'.   Anything less and one of the wheels has to turn more than the other.

greg - Philadelphia & Reading / Reading

  • Member since
    February 2002
  • From: Reading, PA
  • 30,002 posts
Posted by rrinker on Saturday, February 24, 2018 11:13 AM

Richard Hammond (you might know him from the original British Top Gear show) did another show on engineering marvels, and in one episode demonstrated very clearly that it's the wheel taper adn not the flange that keeps the train on the track. He laid out a J shaped section of track and rolled a cylinder down it, starting on the straight vertical leg of the J. Of course, it hit the curve and apart from a slight deflection, ran right off the table. Next he took a piece made of two cones glued together, fat part in the middle, tapers outward. An exaggerated set of railroad wheels. At any speed, until inertia overcame friction, it faithfully followed the curve in the track.

 Of course, no one compares to Fenyman when giving a lecture. There's a series of talks he gave at Cornell on various aspects of partical physics that are on YouTube and quite understandable. Had he been lecturing, I never would have had to drop the quantum physics class I took in college. And yet this incredibly brilliant guy sounds like a brawler from the heart of Brooklyn.

 The singing is the flanges momentarily hitting the rail - if the flanges actually rode continuously against the rail it would be more of one long continuous screech. The flange does help guide things back - if the curve is too sharp, the wheel will want to shift even more on the conic section which would push the wheelset completely off the rails, the inside wheel inside the rail, as the wheel tried to move to even a smaller diameter area past the face of the wheel, and the outside wheel outside the rail as it tried to climb to a larger diameter past the inside edge. But they don't usually rub continuously except on very sharp curves, and it's very obvious as an annoyingly loud squeel. These spots also require very slow speeds - the flange won't really be able to hold the wheels on the rail if enough force can be generated between the flange and the rail, it will lift right up. Poor trackwork makes it even more likely. We used to listen on the scanner some times at the train club, one of the members was an engineer on the PBNE when the Bethlehem plant of Bethlehem Steel was still operating, at least in some capacity. Seemed like every day something would go on the ground.

                             --Randy


Modeling the Reading Railroad in the 1950's

 

Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.

  • Member since
    June 2007
  • 8,892 posts
Posted by riogrande5761 on Monday, February 26, 2018 9:23 AM

slammin
In HO a 30" radius is not consider "big".

"Big" is a relative term.  John Armstrongs Track Plannining for Realistic Operation book lists 18" as sharp, 24" as conventional and 30" as broad.  For some, 30" would be considered a big radius, if they are very cramped for space. 

But times have changed and standards with it.   If I were to revise John Armstrongs terminology for today, I'd list 18" as very sharp, 24" as sharp, 30" as conventional and 42"+ as broad.  Very broad would be limited to what John Armstrong coined as scenic curves, a feature he recommened layout builders include to have a least one scene with a very broad curve.

While some have disputed my recollection, I did remember reading lots of MR magazines in the late 1970's and through out the 1980's and noticed a pervasive trend; that moderate to large layouts had standardized at 30" for their minimum.

Large home and club layouts have 48" and larger radii when modeling class 1 railroads. 

That may be true, but I've read a lot of articles showing large home layouts and the minimum was not 48" radius that I've seen.  That is unless you are talking about home layouts the size of a small barn or aircraft hanger.  Large layout minimums still seem to be less than 48" - again, I'm talking minimums which are often the lions share of the curves, baring scenic curves.  On my 10x18' layout, I did include a John Armstrong feature, a scenic curve of 56":

Most us don't have that kind of space and tend to model short lines and branchlines that look reasonable running 30" and smaller radii.

Yes, but even if we don't have a large layout space, it definitely does not limit us to branchlines.  My small layout was a railfan type layout with mainline running in the 10x18' room and minimum curves were 32" except in the yeard where a few spurs were 28 or 24 inches.

Rio Grande.  The Action Road  - Focus 1977-1983

  • Member since
    July 2009
  • From: lavale, md
  • 4,678 posts
Posted by gregc on Monday, February 26, 2018 9:58 AM

i tried to show in the chart for Reading locomotives that prototypes have restrictions on their locomotives.   Presumably they use this chart to determine which locomotives can run over various trackage.   My impression is that they also design locomotives to meet the restrictions of specific trackage

one example I've read about is the choice of the Reading T1 for the Freedom train on the east cost.   It could handle the relatively sharper curves on the trackage.

when we design our layouts, we make choices, this includes recognizing the limitations on what locomotives we can run, just like the prototypes

greg - Philadelphia & Reading / Reading

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Monday, February 26, 2018 10:58 AM

From the Free-mo Standards:

 

"S3.12 The minimum permitted curve radius on a through route of a Mainline Module is 42 inches.  This includes through track sidings and other tracks where through traffic will run."

The above is followed by a recommended practice:

"RP3.12.1 While the minimum permitted radius of curves on the through route of a Mainline module is 42 inches, 48 inch and larger curves are preferred."

 

Not in the Standards, at this time, is the internally highly recommended practice of using easements, both for operational and appearance reasons.

 

 

Ed

 

  • Member since
    December 2001
  • From: Northern CA Bay Area
  • 4,387 posts
Posted by cuyama on Monday, February 26, 2018 12:40 PM

slammin
Large home and club layouts have 48" and larger radii when modeling class 1 railroads.

riogrande5761
That may be true, but I've read a lot of articles showing large home layouts and the minimum was not 48" radius that I've seen.

Concur with riogrande. I've designed 200+ layouts for individuals, clubs, and commercial organizations. The largest HO multideck Class 1 mainline designs range from 1400 square feet to 3800 square feet. None of those large layouts has a greater minimum radius than 40”, and most use smaller radii. A handful of other layouts I’ve designed have used 48” minimum radius, but they are railfan layouts that made many compromises to attain those curves. And one is an HOn3 backwoods rambler.

The massive (5500+ square feet) Colorado Model Railroad Museum layout is one that I think uses 48” minimum radius -- but it’s an extreme outlier, in my experience. So I think that it’s incorrect and quite misleading to say that mainline modeling in HO demands 48” radius. Thousands of successful layouts over the decades prove otherwise.

Byron

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Monday, February 26, 2018 1:32 PM

gregc

base on Dr Feynman's explanation, the minimum curve radius is related to ratio of the circumferences of the rails and the ratio of the wheel diameters due to taper.   The following drawing indicates that there's only about 0.2" of taper per wheel.   This corresponds to curve radius of 742'.   Anything less and one of the wheels has to turn more than the other.

 

 

For the "slip point" radius* in model railroading, as opposed the the real world (above), I calculated the radii for HO scale 36" wheels with the 3 degree taper:

 

The NMRA standards include tolerances for track gage and wheel gage.

For track gaged narrow, and wheels gaged wide, the "slip point" radius is 191".

Fro track gaged wide, and wheels gaged narrow, the "slip point" radius is 60".

 

So it's pretty much impossible to not have wheel slip in typical HO curves.

 

Ed

 

*Slip point being where the curve is tight enough that the wonderful wheel radius differential for railroad wheels fails.

 

 

 

  • Member since
    August 2013
  • 3,006 posts
Posted by ACY Tom on Wednesday, February 28, 2018 1:30 PM

Minimum radius for a prototype NKP Berkshire works out to about 39-40" radius in HO, and that means the tightest curve allowable. Any Berk negotiating that curve would do so at restricted speed.  Typical HO passenger cars don't feel comfortable on anything less than about 30" radius. But look at the misalignment of the vestibules when they do it, and you can see that something in excess of 40" is more reasonable for the safety and comfort of the passenger.

In reality, few of us have the space to build our layouts with curves anything like those of the prototype, so we do what we can and what we must. In practice, bigger is always better when it comes to operation on curves. 

Tom 

  • Member since
    June 2007
  • 8,892 posts
Posted by riogrande5761 on Wednesday, February 28, 2018 2:39 PM

cuyama
Concur with riogrande. I've designed 200+ layouts for individuals, clubs, and commercial organizations. The largest HO multideck Class 1 mainline designs range from 1400 square feet to 3800 square feet. None of those large layouts has a greater minimum radius than 40”, and most use smaller radii. A handful of other layouts I’ve designed have used 48” minimum radius, but they are railfan layouts that made many compromises to attain those curves.

Yes, it's not unusual for moderate to large layouts to use minimum curves under40 inches.

I have a primary space of about 500 sq ft with an additional room 144 sq ft and just roughed in one possible layout sketch which uses 32-inch minimum radii but have been able to fit in larger curves in part of the track plan.  It it is fully walk-in but walkways narrow to 24" in some places.  I'm going to work up a 2nd track plan which makes use of a lift out gate and will allow more generous walkways.

Rio Grande.  The Action Road  - Focus 1977-1983

  • Member since
    January 2009
  • From: Maryland
  • 12,897 posts
Posted by ATLANTIC CENTRAL on Sunday, March 4, 2018 5:41 PM

In my experiance, 36" radius seems to be a very common minimum for those with larger layouts, 30" being the next "step down" for those with less space.

Club layouts and large home layouts in this region seem to trend around the 36" radius number, even 3-4 decades ago.

Paul Mallery was a big supporter of larger curves, more so than Armstrong, and lobbied hard in all his books for 48" radius for modeling class I lines in HO.

I have managed to stay comfortably at and above 36" for all my mainlines for many years now. My next layout will push that minimum to more like 40".

And yes, I too have been explaining to people for years that the flanges don't hold the train on the track, wheel tapper does...........

Sheldon

    

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Monday, March 5, 2018 2:54 PM

ATLANTIC CENTRAL

 

And yes, I too have been explaining to people for years that the flanges don't hold the train on the track, wheel tapper does...........

 

 

On the real ones.  Not on our model layouts.  Unless your curve radius is between 60" and 191", depending on variables.

We DO use the flanges on pretty much every curve we use.

Ed

  • Member since
    June 2007
  • 8,892 posts
Posted by riogrande5761 on Monday, March 5, 2018 3:45 PM

Since space is limited in my new basement, and I can't manage a 40" minimum, and the next step down is 30", I like to try to boost things a little because even at 30 inches, a couple of extra inches makes a difference in better operation, even if you don't see any real visible improvement.  So I'm going with 32" as my minimum but whereever possible, bumping it up to 36" in visible area's where I can manage it.  I've already drawn up a scale plan to rough in benchwork, walkways and main line.  In my case, I'm debating using a lift out for two tracks near the bottom of the stairs to be able to fit in wider walkways - leaning in that direction.

Rio Grande.  The Action Road  - Focus 1977-1983

  • Member since
    May 2004
  • 7,500 posts
Posted by 7j43k on Monday, March 5, 2018 7:18 PM

This layout:

 

 

does a fine job of having the sharp curves concealed, while having wide curves exposed.  If the concealed curves were 30" radius, I estimate the "showing" curves at, very roughly, 90".  I recommend considering this approach in track planning.

It's Michael Rose's Georgetown and Allen Mountain, which shows up from time to time in the Layout section of this forum.

 

Ed

  • Member since
    January 2009
  • From: Maryland
  • 12,897 posts
Posted by ATLANTIC CENTRAL on Monday, March 5, 2018 7:56 PM

7j43k

 

 
ATLANTIC CENTRAL

 

And yes, I too have been explaining to people for years that the flanges don't hold the train on the track, wheel tapper does...........

 

 

 

 

On the real ones.  Not on our model layouts.  Unless your curve radius is between 60" and 191", depending on variables.

We DO use the flanges on pretty much every curve we use.

Ed

 

Yes Ed, I was refering to the prototype discussion earlier in this thread.

I'm still amazed at the number of model train "enthusiasts" who have no understanding of how real trains work.

Sheldon

    

  • Member since
    October 2008
  • From: Calgary
  • 2,047 posts
Posted by cx500 on Monday, March 5, 2018 8:53 PM

Further to Sheldon's comment, it is not just the model side that has no understanding of how real trains work. It can be found fairly easily in the railfan hobby too.  In my experience even railroaders may know comparatively little beyond the narrow confines of their job or desk.

John

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Users Online

There are no community member online

Search the Community

ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
Model Railroader Newsletter See all
Sign up for our FREE e-newsletter and get model railroad news in your inbox!