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!

Laying Flextrack on curves.. or I can't believe I did that!

7310 views
19 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    November 2002
  • From: US
  • 4,648 posts
Laying Flextrack on curves.. or I can't believe I did that!
Posted by jacon12 on Friday, December 7, 2007 7:08 PM

It has been a while since I last laid Atlas Flextrack, especially on a curve.. so naturally I screwed up big time and have kinks in two places on one large 22 inch R. curve.  I couldn't for the life of me figure out what I'd done so wrong so I Googled for information and one of the hits was the Atlas website.  There it says that the 'sliding' rail must ALWAYS be on the inside of the curve.  Guess where I put mine.  Go on.. take a guess.

I've worked on these kinks for the last two or three days, off and on, and they're not getting any better, so I think it's rip out and redo time and try to get it right this time.  I've come across other information where the tracklayer advises tacking down the first section of track on the curve BUT leaving the last foot or so straight, then joining the next section together with joiners and soldering the joint.  Then proceed laying the track around the curve and doing it again with the next section if one is needed.  But, what about the ties?  Won't they get in the way and prevent the track from sliding in the ties as you bend it around the curve.

How do you lay flextrack on curves and eliminate kinks?

Jarrell

 HO Scale DCC Modeler of 1950, give or take 30 years.
  • Member since
    February 2002
  • 533 posts
Posted by CascadeBob on Friday, December 7, 2007 7:29 PM

I've used the method you describe in which the sliding rail is on the inside of the curve and each subsequent flex track section added to the curve is added to the free end that is still in the straight configuration.  Soldering the track joiners on those sections of track comprising the curve will allow you to pull each newly added section into the curved configuration without any kinks.  You'll need to remove the ties on either side of the joiners so they'll not interfer with this process.  I save these ties and after the track is in its final position, I trim off the "molded spike heads" and slip the ties under the rails to fill the gap.  Those ties that must be positioned directly under the rail joiners may need to have some of their top surface shaved off to allow them to fit without creating a bump in the rails.  These ties are then held in place with a small amount of super glue or other suitable adhesive until you've weathered and ballasted the track.

Hope this helps,

Bob

  • Member since
    February 2005
  • From: In the State of insanity!
  • 7,982 posts
Posted by pcarrell on Friday, December 7, 2007 7:49 PM

I always have the sliding rail on the outside.  Iput the moving rail on the outside because it has more play in it's base, and so if it's on the outside it would (very) slightly loosen the guage between the tracks. That would be desireable in that it would allow you some wiggle room for longer wheelbases. If you put it on the inside it would tighten the guage which could cause binding issues. This effect would be intensified the tighter radius you get.  It's very minimal to be sure, but you can see it with an NMRA guage.....at least in N scale (where tolerences are more crucial then the larger, more forgiving scales).

To actually lay it, I solder one end of the flex to the end of the straight section it comes off of, go about 6" more before I begin my curve, then add a gentle easement.  I lay the curve, securing it with latex caulk as I go until there's about 6" left.  Next I bend the rail into the place it's going to be and then trim the rails.  Then I let the track snap straight again, add a joiner, solder the next piece of track on, then start bending and securing the rails in place until I come to 6" from the end of that piece where I do it all again.

Now that I think of it, I suppose I could do away with the trimming and just stagger the joints.  It'd probably be stronger and resist kinking even more that way.

 

Hmmmmm................

Philip
  • Member since
    November 2002
  • From: "Steel, Steam and Thunder"Fort Wayne, Indiana
  • 1,177 posts
Posted by TheK4Kid on Friday, December 7, 2007 9:37 PM

Joe Fugate has a downloadable video on laying flex track available through the MR forum site. I downloaded it, followed his advice, and it went well. The sliding track on the inside. I solder two pieces of 36 inch flex track together first, then gradually bend them. Some guys do three pieces at a time, but that gave me problems, two at a time worked best for me.
I have  two 34 inch radius curves, and two 32 inch radius curves so far on my layout, both tracks running side by side, 2-1/2 inches apart .
I am working in HO with code 100 Atlas flex track.

  • Member since
    April 2003
  • From: Martinez, CA
  • 5,440 posts
Posted by markpierce on Friday, December 7, 2007 10:13 PM
 davidmbedard wrote:

Solder your joints BEFORE you lay it.  Draw a 22" centerline and stick to it.

David B

For curved track, you don't ever want to solder rail ends together until you have shortened the inside rail.  That's best done as you are laying down the track (or at least after the track has been formed to the desired radius) so the amount of rail to be cut off has been accurately measured.

Mark

  • Member since
    July 2006
  • From: Pocono Mts. of Pa
  • 196 posts
Posted by LNEFAN on Friday, December 7, 2007 10:20 PM
I'm right in the middle of replacing ties as described (though using Peco code 83 track). Though I have done it many times before, it remains a basically tedious and time consuming job. I have found however that my new cordless Dremel is great for grinding off the molded spike heads and a touch of clear adhesive caulk holds everything in alignment very well. It's now extremely difficult to detect that these are replacement ties. Ballast will perfectly complete the  disguise.
  • Member since
    February 2005
  • From: Vancouver Island, BC
  • 23,330 posts
Posted by selector on Friday, December 7, 2007 11:04 PM

Good folks, here is what you do.  Slide two pieces of flex track together, and solder the outer rails together.  That leaves two close inner rail ends that can slide, but have no retainer.  Good...exactly what you want.  Then, slide both inner rails one way or the other for no more than 3".  The ends are still tight, but displaced from the outer joinered meet by however many inches to the side you have placed them.  Solder the inner join while both rails are dead straight aligned...between two plastic spikes, natch.

Now adjust, by cutting, the rail ends on one end of this mated pair of flextrack segments to fit nicely with track already in position on the layout.  You can use joiners, spike them temporarily with nails on each side of both rails, solder 'em, whatever.  Now make the two-segment length  bend it to conform to your curved centreline.  Tack it down, and repeat with another length of twinned flextrack for as long as it takes to complete your trackplan until the next turnout or backstop.

The staggered join will move a bit to one side as you bend the joined segments of flex, but not much if you keep the soldered join within 3" of the outer one.  Staggering the joints lends much better strength to them.  The spike heads help in this regard, but so does the soldered joint.

If you ever find that you have had to gap along a curve, and that you just can't seem to get a nice kink-free curve as a result, force the rails on both sides of the gap toward the centre of the curve by using track nails driven into the roadbed. That typically means four nails: one nail for each rail end on each side of the gap, and they will all be driven on the outer side of the curve.  Press fairly hard on the rails down the radius line to clean up the outward leaning kink, and then drive your nail so that its head sits atop the foot of the rail.  Don't bend the rail outward by depressing the metal on the foot of the rail or by deforming any adjacent ties!! Check gauge!!

That is the only way, besides what I describe above, to get some semblance of a normal curve when you have a break between the rails along the curve.  Check gauge!!

  • Member since
    November 2002
  • From: US
  • 4,648 posts
Posted by jacon12 on Saturday, December 8, 2007 7:16 AM

 pcarrell said:

"Now that I think of it, I suppose I could do away with the trimming and just stagger the joints.  It'd probably be stronger and resist kinking even more that way"

 

I've also thought about staggering the joints but it seems like you'd have to remove even more ties because of that one staggered joint sliding back on the ties than you would if you cut the rails off square with each other.  I dunno, haven't tried it yet.

Jarrell

 

 

 pcarrell wrote:

I always have the sliding rail on the outside.  Iput the moving rail on the outside because it has more play in it's base, and so if it's on the outside it would (very) slightly loosen the guage between the tracks. That would be desireable in that it would allow you some wiggle room for longer wheelbases. If you put it on the inside it would tighten the guage which could cause binding issues. This effect would be intensified the tighter radius you get.  It's very minimal to be sure, but you can see it with an NMRA guage.....at least in N scale (where tolerences are more crucial then the larger, more forgiving scales).

To actually lay it, I solder one end of the flex to the end of the straight section it comes off of, go about 6" more before I begin my curve, then add a gentle easement.  I lay the curve, securing it with latex caulk as I go until there's about 6" left.  Next I bend the rail into the place it's going to be and then trim the rails.  Then I let the track snap straight again, add a joiner, solder the next piece of track on, then start bending and securing the rails in place until I come to 6" from the end of that piece where I do it all again.

Now that I think of it, I suppose I could do away with the trimming and just stagger the joints.  It'd probably be stronger and resist kinking even more that way.

 

Hmmmmm................

 HO Scale DCC Modeler of 1950, give or take 30 years.
  • Member since
    November 2002
  • From: US
  • 4,648 posts
Posted by jacon12 on Saturday, December 8, 2007 7:35 AM

Mornin' Crandell!

Questions!  Are you using track joiners in your method and approx how many ties need to be removed?  Or am I missing the point somewhere?

Jarrell

 

 selector wrote:

Good folks, here is what you do.  Slide two pieces of flex track together, and solder the outer rails together.  That leaves two close inner rail ends that can slide, but have no retainer.  Good...exactly what you want.  Then, slide both inner rails one way or the other for no more than 3".  The ends are still tight, but displaced from the outer joinered meet by however many inches to the side you have placed them.  Solder the inner join while both rails are dead straight aligned...between two plastic spikes, natch.

Now adjust, by cutting, the rail ends on one end of this mated pair of flextrack segments to fit nicely with track already in position on the layout.  You can use joiners, spike them temporarily with nails on each side of both rails, solder 'em, whatever.  Now make the two-segment length  bend it to conform to your curved centreline.  Tack it down, and repeat with another length of twinned flextrack for as long as it takes to complete your trackplan until the next turnout or backstop.

The staggered join will move a bit to one side as you bend the joined segments of flex, but not much if you keep the soldered join within 3" of the outer one.  Staggering the joints lends much better strength to them.  The spike heads help in this regard, but so does the soldered joint.

If you ever find that you have had to gap along a curve, and that you just can't seem to get a nice kink-free curve as a result, force the rails on both sides of the gap toward the centre of the curve by using track nails driven into the roadbed. That typically means four nails: one nail for each rail end on each side of the gap, and they will all be driven on the outer side of the curve.  Press fairly hard on the rails down the radius line to clean up the outward leaning kink, and then drive your nail so that its head sits atop the foot of the rail.  Don't bend the rail outward by depressing the metal on the foot of the rail or by deforming any adjacent ties!! Check gauge!!

That is the only way, besides what I describe above, to get some semblance of a normal curve when you have a break between the rails along the curve.  Check gauge!!

 HO Scale DCC Modeler of 1950, give or take 30 years.
  • Member since
    February 2002
  • From: PtTownsendWA
  • 1,445 posts
Posted by johncolley on Saturday, December 8, 2007 10:51 AM
Unfortunately that is bad advice. If you lay the sliding rail on the inside the guage will tighten up. However, if you lay the sliding rail to the outside of the curve, the guage will open up slightly, providing less friction resistance and easier tracking. jc5729 John Colley, Port Townsend, WA
jc5729
  • Member since
    November 2002
  • From: US
  • 4,648 posts
Posted by jacon12 on Saturday, December 8, 2007 11:22 AM

 johncolley wrote:
Unfortunately that is bad advice. If you lay the sliding rail on the inside the guage will tighten up. However, if you lay the sliding rail to the outside of the curve, the guage will open up slightly, providing less friction resistance and easier tracking. jc5729 John Colley, Port Townsend, WA

 

Well, now I am confused.  One would think that Atlas wouldnt give out bad information, but I've seen manufacturers do it before.

Jarrell

 HO Scale DCC Modeler of 1950, give or take 30 years.
  • Member since
    February 2005
  • From: Vancouver Island, BC
  • 23,330 posts
Posted by selector on Saturday, December 8, 2007 12:39 PM

There is some truth to what John says.  It stands to reason that since the spike heads retaining the sliding rails will permit more deviation from the gauge than will the ones retaining the fixed rail, as you form a tighter curve the sliding rail on the outside will want to place itself as wide, and therefore as close to the outer spikes as possible.  That may be desirable for very sharp curves where you are at the mins for your longer steamers.  It may be.  However, for more generous curves, I would follow Atlas advice.  In any case, one should be checking the gauge everywhere as one lays each section of track.

Jarrell, for your question to me, if the two sections joined as I suggested will make up much of, or all of, the entire curve, even with the inner displaced a couple of inches they will not move much as you bend the two sections because they'll be at the apex of the curve.  It is when such a staggered joint is near the limbs of a curve, at its ends, that the sliding rail does most of its movement, and for which it will need to be free to do so...hence the solder will have to be very cleanly filed, or done later...which tends to defeat the purpose of the stagger in the first place. 

The last two inches of free rail on flex will not bend the same as the rest of the rail length on a curve.  It just doesn't.  That is what causes the kinks when you try to merely lay them in place and curve them after joining them with two joiners.  That is why you must solder the two rails if possible, or at least stagger the inner rail joints to partially overcome this tendency for the meet to want to straighten somewhat for both mated track section ends.  When you keep one rail tucked into spike heads, it helps for force both of them to continue the curve that was commenced further back.

I use nails, staggers, solder....whatever it takes in situ to maintain accurate overlay on the drawn centreline for that curve.  Takes some thumb pressure and spiking often, but I get it.

  • Member since
    July 2004
  • From: Weymouth, Ma.
  • 5,199 posts
Posted by bogp40 on Saturday, December 8, 2007 1:21 PM

The guage differences wil vary with the brand of track used. Atlas both code 100 and 83 flexes and springs for the bend very easily. I have no experience w/ Peco, but the Walther's/ Shinahara is a bit tighter yet works easily. The ME flex track, although the most detailed is the most difficult. This actually has to be forced to shape and can kink easily.

The varying tollerences and tightness of the molded spikes for the sliding rail, will of coarse allow the looser, Atlas, to have the most guage change. Not a bad thing, you just need to figure if you would like the guage to tighten or loosen pending your needs and radius. I have never had a problem with Atlas, and it actually is the easiest to lay turns w/o any kinking. The Walther's/ Shinahara was selected by my club after examining and testing all other makes. We felt it to be the most workable and still have decent detail. The ME track is left for bridge track as this bridge flex is absolutely fantastic in appearance once weathered.

Modeling B&O- Chessie  Bob K.  www.ssmrc.org

  • Member since
    February 2005
  • From: In the State of insanity!
  • 7,982 posts
Posted by pcarrell on Saturday, December 8, 2007 2:12 PM

 johncolley wrote:
Unfortunately that is bad advice. If you lay the sliding rail on the inside the guage will tighten up. However, if you lay the sliding rail to the outside of the curve, the guage will open up slightly, providing less friction resistance and easier tracking. jc5729 John Colley, Port Townsend, WA

Thats exactly what I was saying!  You can prove it with an NMRA guage.

Philip
  • Member since
    November 2007
  • 105 posts
Posted by JulesB on Saturday, December 8, 2007 3:15 PM

With 24" radius curves I do the following.

I use track nails because I had a hard time finding spikes and I'm laying track on cork, spikes don't do well in cork I'm told.

After drawing my curves with easments I lay the cork down following the track line. Then looking at the underside of Atlas flextrack you can see pre drilled holes, but they are not all the way thru. I drill em through. Then at the center of the flextrack you will notice a spot where the plastic ties are not joined. I solder #20 AWG feeders to those spots, but I remove  two ties on each side. I use .015 dia solder (.020 is ok too), this enables a nice solder joint on the bottom of the rail. After the track is down I take a knife and cut a little slot in the cork to tuck the wire into. This really hides it! I solder my curved track joints before bending the track. By cutting away those ties in the center as explained above the center wire soldred points can move. where they were opposite one another they can separate by as much as an inch or more.

When soldering joiners on a curve I stagger the rail joints by at least an inch. I do it while it's straight, using a straight edge. I never have kinks. I use a track guage  as I nail down the track around the curve. I found that by leaving a .020" gap on straight sections I have no more problems with expansion or contraction. Live and learn. I drop feeders for every section. Joiners are no good for elecrical conductivity, at least not dependable.

 

JulesB

 

  • Member since
    November 2002
  • From: US
  • 4,648 posts
Posted by jacon12 on Saturday, December 8, 2007 9:12 PM

Ok, thanks guys for all the tips and suggestions!

Jarrell

 HO Scale DCC Modeler of 1950, give or take 30 years.
  • Member since
    November 2002
  • From: Colorado
  • 4,075 posts
Posted by fwright on Saturday, December 8, 2007 10:01 PM
 jacon12 wrote:

 johncolley wrote:
Unfortunately that is bad advice. If you lay the sliding rail on the inside the guage will tighten up. However, if you lay the sliding rail to the outside of the curve, the guage will open up slightly, providing less friction resistance and easier tracking. jc5729 John Colley, Port Townsend, WA

 

Well, now I am confused.  One would think that Atlas wouldnt give out bad information, but I've seen manufacturers do it before.

Jarrell

Actually, Atlas flex track tends to be already set at the wider tolerance for gauge from the factory (ME flex track is definitely gauged narrower, and both are within NMRA specs).  Add to this the fact that wheels are already gauged about 1.5" narrower than the track by NMRA specs, and all of a sudden there is quite a bit of slop.

The slop is not an issue for the overly wide code 110 wheel treads in HO, and the wider gauge does help you cram a too long locomotive through a too small radius.  But the more slop, the more small model steam engines, short truck wheel base diesels, and 4 wheel bobbers will "wallow" down the track.  It's likely that at the extreme (I haven't run the calcs to be sure) a prototype width wheel running on track at the widest possible NMRA gauge will drop between the rails.

I'm not taking sides - I'm just saying make your decision an informed one.

yours in tracking

Fred W 

  • Member since
    December 2005
  • From: East Granby, CT, USA
  • 505 posts
Posted by jim22 on Saturday, December 8, 2007 10:16 PM

I had a lot of trouble wher I had a transition from a turnout to a curve using flextrack.  Following conventional advice, I did NOT solder the flex to the turnout  If I had to do it again, I would put a  piece of 18 inch radius sectional track to make the curve and then transition to flex for basically straight sections.  If you mess up an 18 inch radius curve, part of the curve is tighter than 18 inches, which some rolling stock can have trouble with.

Jim 

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!