WHAT AM I MISSING!

I seriously want to get all of my stuff out of boxes and set up a layout but I think it would be smart to plan it all out on paper first so that I have something to go by and to make sure it is going to "work"I found software provided by Atlas called RightTrack Freeware V5.0, so I have been using this to plot everything out. I want to make a siding like there are in real life but I cant seem to get it.I simply can not get the siding to run paralel to the main line. It either ends up Slowly creeping toward the mainline, or slowly away... How would I make it run paralel. I don't want to use flex track but If I have to I guess I will. I know the type of switch matters so I have Atlas Snap switches. I also have all sorts of track even if I dont have it, it is in the software library so that I can plan it out and order it from atlas. Would someone please tell me what I am doing wrong. The closest thing I have found is a #6 switch, 6"straight, 1

1/4" Straight, 1/3 18" raidus, 1 1/2" straight. but the siding ends up creeping toward the main line! GRRRR PLEASE HELP!
Reply to
Rosco
Loading thread data ...

=>I seriously want to get all of my stuff out of boxes and set up a layout =>but I think it would be smart to plan it all out on paper first so that I =>have something to go by and to make sure it is going to "work"I found =>software provided by Atlas called RightTrack Freeware V5.0, so I have been =>using this to plot everything out. I want to make a siding like there are =>in real life but I cant seem to get it.I simply can not get the siding to =>run paralel to the main line. It either ends up Slowly creeping toward the =>mainline, or slowly away... How would I make it run paralel. I don't want =>to use flex track but If I have to I guess I will. I know the type of =>switch matters so I have Atlas Snap switches. I also have all sorts of =>track even if I dont have it, it is in the software library so that I can =>plan it out and order it from atlas. Would someone please tell me what I am =>doing wrong. The closest thing I have found is a #6 switch, 6"straight, 1 =>1/4" Straight, 1/3 18" raidus, 1 1/2" straight. but the siding ends up =>creeping toward the main line! GRRRR PLEASE HELP!

Now you know why Atlas trackplanning software is "free." :-) When connecting a a turnout to the side track, you must put in a short curve - but I've found that even though Atlas short curve pieces are supposed to fit, they don't. Hence the "creep." This is true when you actually join the track sections together, BTW - there's just enough error to add up. It's only because you can fudge the joints a bit that the whole mess works out.

My advice: Don't sweat it - as long as the track plan is pretty close, you can build it. If you know where the beginnings and ends of the sidings are, you needn't worry about the track in between - you can even --gasp!-- draw it by hand on the printout. What's more, the plan and realitty will be somewhat different things, no matter how careful you are in drawing it up. I don't know of anyone who has built a layout as planned - there are always adjustmenst, sometimes major ones, as the building progresses. The plan is a guide, really.

That being said, some dimensions hsould be drawn as accurately as you can. They are: a) curve radius, and end of curves where the straight joins the curve b) track spacing (2" on straight, more on snap-track curves) c) distance of track from edge of layout - I recoemmnd 4" minimum d) distance of track from trackside objects such as buildings - I recommen

2-1/2" minimum

If your track has been in boxes for some time, it's probably brass track - if so, junk it, it will oxidise rapidly and you will have to clean it just about every time you want to run trains. (It's no longer made by Atlas, anyhow).

As for sectional track: avoid it if at all possible. Flex track, despite the fiddly business of joints of curves, is much better. But once you've mastered curved joints, flex is a breeze. The joints in snap-track are sources of mysterious electrical gaps, they increase the probability that wheels will pick rails, etc. If you do use it, make sure that all rail joiners fit tightly, that there are no gaps between the ends of the rails, and so on. BTW, even if the software tells you what pieces you need, you will find that you need others. Track comes in packages of 4 or 6 pieces, so order enough to cover. A few extra pieces won't cost you enough to worry about - and you may find that a stub-siding to an industry can be extended. If necessary, you can cut an extra piece to fit exactly.

Turnouts: Snaptrack turnouts are built to match the curves, but they don't fit exactly on the straight side - you will need an assortment of straight bits to make things come out right. I recommmend #6 turnouts instead - much more reliable operation, and better looking, too.

I _strongly_ recommend that you buy one of the books that describes building a complete layout. Such a book will answer questions you don't realise you should ask. It will also give you lots of good ideas about overall layout design.

Have fun!

Wolf Kirchmeir ................................. If you didn't want to go to Chicago, why did you get on this train? (Garrison Keillor)

Reply to
Wolf Kirchmeir

I found the secret to RightTrack is to use the "Special/Shape Flex Track/Curve (Left or Right)" drop-down menu. Use Radius=36"/Angle=12.5 for the CustomLine #4 turnout or Radius=48"/Angle=9.5 for CustomLine #6 turnouts. This will create a preformed track that will allow a parallel track.

Bryan

Reply to
sbryan

Your problem is that the numbered frog turnouts don't match up to any of the curve pieces of the snaptrack sections. Thus you end up with an off angle siding. The only solution (other than using snapswitches) is to use the flex track pieces to get the curves right. Once you start using the flex track in the longer sections, you will see that there is less trouble with joints in the track losing contact as there will be a lot less of them and that will drop your frustration with the layout having dead sections of track a lot better.

-- Bob May Losing weight is easy! If you ever want to lose weight, eat and drink less. Works evevery time it is tried!

Reply to
Bob May

Free is not always bad! Check out my track plan on my web site. All done with the Atlas software and a pait progam.

Just use the flex track to get things to line up.

Reply to
Kelly

Sorry, forgot to add my web site

Reply to
Kelly

I haven't used RightTrack, but isn't the diverging route of a Snap switch just 20° of an 18" radius curve? Wouldn't two 10° curved sections (a 1/3 18" radius track section is 10°) get things parallel again?

Reply to
Mark Mathu

The snaptrak turnout is a 2/3 of a standard 18" radius piece and comes with a 1/3 section to finish the curved section if so desired.

-- Bob May Losing weight is easy! If you ever want to lose weight, eat and drink less. Works evevery time it is tried!

Reply to
Bob May

So why use the flex track to get things to line up? Why not just two 10° pieces?

Reply to
Mark Mathu

The right track software has other turnouts besides the snap track ones.

If you use the #6 turnouts it is just simpler to use a length of flex track in the software

Reply to
Kelly

=3D>>>> Just use the flex track to get things to line up. =3D>>>

=3D>>> I haven't used RightTrack, but isn't the diverging route of a Sna= p switch =3D>>> just 20=B0 of an 18" radius curve? Wouldn't two 10=B0 curved sec= tions (a 1/3 =3D>18" =3D>>> radius track section is 10=B0) get things parallel again? =3D>>

=3D>> The snaptrak turnout is a 2/3 of a standard 18" radius piece and c= omes with =3D>> a 1/3 section to finish the curved section if so desired. =3D>

=3D>

=3D>So why use the flex track to get things to line up? Why not just tw= o 10=B0 =3D>pieces?

The problem lies in the way the track pieces are rendered on-screen and = in the computer. The computer operates with limited decimal places, but the=

dimensions of a curved track piece are perforce irrational numbers (on account if pi), hence the 10 degr. track piece cannot be exactly represe= nted

-- there is a built in, unavoidable error.

Moreover, the location of the track piece on screen (or in the "vector s= pace" if you are using a CAD) has a built in error - again, because of the limitations of representing irrational numbers w/ fixed decimal places. = In industrial applications, 64-bit arithmetic is used to reduce the errors = below the precision of measurement.

I suspect that Atlas's program uses simple 8 bit FP arithmetic (--> max = of about 3 decimal places). Thus the errors accumulate of the screen and printout -- but, as I said before, don't sweat it. Just locate the curve= s and turnouts, and draw the straight in by hand on the printout - it will be = more than precise enough to build from. And use "snap to grid" to ensure that= the pieces are lined up with the measured points, rather than with each othe= r.

Wolf Kirchmeir ................................. If you didn't want to go to Chicago, why did you get on this train? (Garrison Keillor)

Reply to
Wolf Kirchmeir

96 bit characteristics were being used 30+ years ago. And even then there were a number of abritrary precision packages available.
8 bit floating point!? 16 maybe, more likely 32.

Paul

Reply to
Paul Newhouse

Reply to
Max Coynes

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.