More prob w/ drawing layout on foam

[kasskaboose]

I printed out layout built on Atlas RTS 8.  The software enabled me to measure the location of turnous for the around-the-shelf layout.  Why doesn't the software mimic real life?  Whenever I connect 24" radius templates, they don't align to where I marked the turnout.  I tried using some flex track measured to 24" and that also ends in a different area.  How do I go from the blueprint to real-life? 

This is quite annoying! 

TIA!

Lee 

 

[dehusman]

Are you using sectiona track or flex track?

If you are using sectional track, then just assemble the sections, trace around them or mark the track centers, put down cork roadbed and then put the sectional track back on the cork.

Dave H.

[gandydancer19]

I think that it is unrealistic to expect any track planning software to allow you to print out the plan full size and make it all work without any adjustments.

What I typically do in tight areas is just print out the plan on one page. Draw 6 inch grid lines on the paper plan to scale, then draw 6 inch grid lines on the actual layout lightly, but dark enough that you can still see them. Then you should be able to see where things should be positioned, not exactly, but close. To me, a 5x12 or smaller layout is a tight area. But even on a larger layout like I am building now, when laying out a yard or small switching area, I lay it out with the items I plan on using so I can make sure it all fits correctly.

If you are working on foam, some T-pins can be used to hold the track and turnouts in place, but still will allow you to move things easily. Once everything is lined up, mark the track center line. Then you should be able to lay the split roadbed knowing that the track will connect properly.

There is a lot of trial and error involved to getting the track positioned correctly on any layout. The trick is to fasten it in place enough so you can see if it's going to work, without putting it down permanently first, so you can make changes if required.

[Packers#1]

Well, i can't help you there, but might I suggest you tinker around with it some. See what you can doi to make them align. I doubt that anything drawn on paper can become 110% replicated in the physical world. Don't forget: Some stuff only looks good on paper. I'd try to mess around with it.

[joe-daddy]

I did my layout in 3rdplanit version n-1.  It allows you to printout the layout as templates.  I taped the dozens of pages together and laid them down on the roadbed. Worded well, but it was a lot of work.  Only to find out that the layout as designed was nice an complicated, but totally implausible to build scenery much less operate.  Scrapped the whole durn layout and started over.

I abandoned train cads and layout my track on the deck, decide what and where I want it, mark the centers with sharpies and proceeded to track laying.  I find simple is better.  Quite satisfied with the results.

I can't imagine for a minute that your current conundrum is the only one you have or will have with the cads.

Just my 2 cents. 

Joe Daddy 

 

[tomikawaTT]

"No battle plan ever survives first contact with the enemy."

Likewise, no small-scale track plan ever survives the first contact between track and roadbed.

My personal solution is to temporarily lay the track on the subgrade, shifting as necessary to adjust alignment.  Care has to be taken to maintain the designed radius for each curve.  Turnouts might have to have some of the stright rail trimmed off if the diverging side, but will actually be the components that determine exactly where the whole track pattern will finally settle down.

Chuck (modeling Central Japan in September, 1964)

[fwright]

I printed out layout built on Atlas RTS 8.  The software enabled me to measure the location of turnous for the around-the-shelf layout.  Why doesn't the software mimic real life?  Whenever I connect 24" radius templates, they don't align to where I marked the turnout.  I tried using some flex track measured to 24" and that also ends in a different area.  How do I go from the blueprint to real-life? 

This is quite annoying! 

Lee

Others have answered your question about going from blueprint to real-life - you don't without changes.  Now let me explain some of the reasons why this is so...

• How accurate is the software model of the track geometry?  In Atlas's case, they have been manufacturing track for decades, at a variety of plants.  Recently, much of their track has been manufactured in China.  Do you think the track has remained identical through all these changes, or at least the same to .01 inches?  I don't, especially when they are on their 4th redesign of their Custom-Line turnouts.
• Speaking of turnouts, what is the correct frog angle for an Atlas Custom Line #6?  For years Atlas published it as 10 degrees, and used 10 degrees in their track planning books.  This also fits in with the use of the 10 degree section of 18" radius track.  NMRA RP specs a #6 as 9 degrees 32 minutes.  Some claim Atlas actually used 9.5 degrees, which would fit with Atlas manufacture of a matching 19 degree crossing.  I would have real difficulty measuring an actual turnout to better than 1/4 degree precision.  Which figure does the RTS software use?
• Is the Atlas #4 turnout frog angle actually 12.5 degrees?  Atlas planning books and the 25 degree crossing suggest it is.  But who has actually measured to know?
• What tolerances did you allow the software when drawing the plan?  I'm guessing, but I believe the software default permits mismatches of 1/4" and 2 degrees in the software model to connect.  This tolerance is adjustable in the software.
• What gap between rails at the track section joints does the software assume?  Did you consistently put your track or templates together with that same gap at every rail joint?
• How accurately does the software translate the model to your printer?  Display screens and printers have very different resolutions.  And translation of a complex and precise mathematical model to either will never be perfect.
• Manufacturing tolerances are another source of error.  I've never measured Atlas track over the years to check for differences.  But I do know from measuring that Lionel O27 straight track has varied in length from 8.75" to 9.00" from one piece to another.  That's enough to cause a kink in a small oval.  Similarly, some non-Atlas HO sectional track from a train set was enough off in the curves that it was impossible to lay a circle or oval without kinks.
• Digitization and measurement errors occur when entering the data in a software model.  How accurately was the piece measured, and to what precision was the digitization of the measurement performed?  This is very likely the smallest of any of the possible errors listed.

And I have to ask - did you measure the 24" radius to the track centerline?  If not, that's a more than 1/4" error every 90 degrees of turn.  I don't know about you, but given the springiness of Atlas flex track, there is no way I could maintain a consistent, exactly 24" radius from end to end.  The best I can do is draw a 24" radius and then try to lay the track on that centerline.  But there will be small deviations from 24" radius, particularly near the end of the track section unless I exercise real care.

Hope this helps with understanding the nature of the beast.  These are just some of the reasons we have to do as-built drawings in the projects I've managed after construction, instead of just relying on the initial plans.  If we just use initial plans, the building/system can't be modified because the initial plans are too far from reality.

Fred W

[tomikawaTT]

• How accurate is the software model of the track geometry?  In Atlas's case, they have been manufacturing track for decades, at a variety of plants.  Recently, much of their track has been manufactured in China.  Do you think the track has remained identical through all these changes, or at least the same to .01 inches?  I don't, especially when they are on their 4th redesign of their Custom-Line turnouts.
• Speaking of turnouts, (precise geometry discussion snipped)  But who has actually measured to know?
• Manufacturing tolerances are another source of error.  I've never measured Atlas track over the years to check for differences.  But I do know from measuring that Lionel O27 straight track has varied in length from 8.75" to 9.00" from one piece to another.  That's enough to cause a kink in a small oval.  Similarly, some non-Atlas HO sectional track from a train set was enough off in the curves that it was impossible to lay a circle or oval without kinks.

And I have to ask - did you measure the 24" radius to the track centerline?  If not, that's a more than 1/4" error every 90 degrees of turn.  I don't know about you, but given the springiness of Atlas flex track, there is no way I could maintain a consistent, exactly 24" radius from end to end.  The best I can do is draw a 24" radius and then try to lay the track on that centerline.  But there will be small deviations from 24" radius, particularly near the end of the track section unless I exercise real care.

Hope this helps with understanding the nature of the beast.  These are just some of the reasons we have to do as-built drawings in the projects I've managed after construction, instead of just relying on the initial plans.  If we just use initial plans, the building/system can't be modified because the initial plans are too far from reality.

Fred W

The long discussion about real versus theoretical turnout dimensions are the best argument I can think of for building your own specialwork.  You don't have to second-guess the manufacturer if you ARE the manufacturer.

The best way I have found to lay Atlas flex to a precise radius is three part:

1. Using a full-size trammel or template, mark the outer tie line on the roadbed.  My 610mm (24") radius trammel actually marks a 624mm curve, 610mm radius + 1/2 the 28mm tie length.  That way you have a control point at every tie, not just at the widely spaced track nail holes.
2. Pre-bend the track, with the non-sliding rail to the outside.  It may take squeezing it down to about 4" radius, but it is possible to get a length of flex to 'relax' at or very close to 24" radius.
3. Stagger the rail joints.  Don't clip the inner rail, slide it into the next length of flex.  When rail joint location is established, trim the ties to accept a rail joiner.  Do your final rail length adjusting on tangent track - or the next electrical gap, whichever comes sooner.

Fred made another good point.  Once you've established where track is really located, go back and develop an 'as built' drawing.  It will be invaluable if you ever want to add to or change the layout.

Chuck (modeling Central Japan in September, 1964 - Atlas flex, hand-laid specialwork)

[johncolley]

I do mine directly on the foam lightly with a ball point pen or fine point sharpie. Do it lightly so as not to dig grooves in the foam, but just enough pressure to get the ink to flow. Once all the locating and center points are drawn I put a finishing nail in the foam and use string for a compass, just hold the pen vertical as you swing the arcs. Once you get the tangents and arcs in place you can lay the actual turnouts in place and trace the centerlines by marking between the ties. Also mark the center at both ends of the throw rod. after you pick up the turnout  connect the two ends of the throw rod across the track center to know where to make your hole, if needed.. The neat thing about the above procedure is you can scribble across the lines and relocate anything you want. I have even done different color ink ball points for revising the plan. try it a few times and it goes very fast! jc5729 John Colley, Port Townsend, WA

[jrbernier]

  I have measured the Atlas Customline turnouts and they appear to be 12.5 degrees and 9.5 degrees.  The 12.5, 19.0, & 25.0 degree crossing appear to confirm that.  I have used the RTS software and have had no problems translating from CAD to layout.  I have only 'printed' a couple areas for some 'test fitting' and it works pretty well.  What I have noticed is that one can seem to 'cram' tighter track arrangements into a space that what RTS will allow!  I suspect somewhere along the line I may have 'cheated' on some of my flex-track curves.  I have only done this for some industry tracks....

Jim