Atlas C55 Turnout Failures Increase

[ednadolski]

Seems to me that when things reach this point, it may be better in the log run to start with hand-built turnouts.

BTW another reason to consider going with Code 40 rail, when you are freed up from the constraints of commercial offerings. 

Ed

[Mark W]

Seems to me that when things reach this point, it may be better in the log run to start with hand-built turnouts.

Trouble is, the initial time investment is huge compared to RTR.  I think the best option is hybrid 'optimized' RTR.  That eliminates maybe 80% of the time and gets you almost 100% of the benefits of hand built. 

[Sokramiketes]

Trouble is, the initial time investment is huge compared to RTR.  I think the best option is hybrid 'optimized' RTR.  That eliminates maybe 80% of the time and gets you almost 100% of the benefits of hand built.

I've also re-built a couple of Atlas turnouts in place with the PCB tie insert and continuous point rail method.  I'm not sure about an initial time investment in hand laying being "huge", but modelers should expect to throw out their first turnout or two (at about 2 hours each).  Then, turnouts 3 and on should take about 45 minutes a piece and should be usable.

I easily spent 45 minutes carving out ties and ballast and forming new point rails once the turnout is in place.  So for me, I also decided hand laid was a better option if I was going to be spending that same amount of time on a RTR option. 

I suppose if you just wanted to rebuild an Atlas with continuous point rails on the bench, that 45 minute cycle time could decrease a bit.  But then I'm still stuck with frogs that turn yellow when the plating wears off. 

I'm surprised China can't crank out handlaid style turnouts from jigs for a decent price!

[MichaelWinicki]

I've been able to repair all of my failed Atlas code 55 turnouts right in place by simply dropping a little solder on the brass plate that rests under the spot where the point meets the rail that goes to the frog.  A 5 minute fix at most.  I've had about 16 failures/repairs out of 120 switches since 2008.

If I had to spend a half-hour "improving" the switch prior to putting it in place I would have gone the hand-laid route.

[mmagliaro]

Having done this a lot, I think it's simply a matter of a wide variety of how many will fail for
different people.   

For me, it was not just the hinges, so soldering that spot would not be the
only "in place" fix I would have to make.

I think it took me well over 30 minutes to prepare each turnout.  More like 90.
Solder-jumpering all the potential failure points takes time.  But I also had to file and
polish the point rails on pretty much every one because they are too narrow.  I think this
is something that afflicts me more than a lot of you guys because I run a lot of steam.   The rails are too narrow
in gauge, but 4-axle diesels seem to be able to make it through okay.  But an x-8-x steam loco?  Forget it.

[robert3985]

Since I started building my own code 55 and code 40 turnouts before there were any RTR offerings to be had, I just continued the trend for several decades.

I learned to build 'em with only Gordy Odegard's MR article on how he did it, then after building a couple like he did, I developed my own techniques...without any jigs or fixtures...just printed templates.

As far as a "huge" expenditure of time is concerned...it doesn't exist.  Of course there is an expenditure of time, but building turnouts is not an uber-complicated task, or are there any extraordinary skills involved, and like those who do it know, it only takes two or three to get the "hang" of it.

The main problem I had with the Atlas turnouts, which I was initially very happy to see, was after I looked at 'em...I thought "Something's not right here....they look odd."  And "odd" they are.  Each one of the three is considerably shorter between the frog and the closure point toes than turnouts built to prototype proportions.  An Atlas code 55 #7 is only a couple of ties longer than an ME #6, and an Atlas #10 is waaaaay shorter than it should be.  This virtually negates much of the benefit of going with larger turnouts, as their effective diverging radius is also considerably smaller than if they were proportioned correctly.

Now that these other problems keep cropping up, I am even more happy that every turnout I've used on my own modules and layout has been made in my own factory.

Additionally, because there is very little difference between the total turnout lengths of the Atlas #7 and the ME #6, and the effective diverging radius of the Atlas #7 is much smaller than it would be if proportioned correctly, I see no logical reason to not go with ME#6's, which will not have the dead frog/dead closure rail problems.  They also won't have the plating wearing off the frog either since the ME frog is cast nickel silver and the guard rails and closure points are also Nickel Silver rail.  They might either have problems when brand new such as gauge narrowness (if they do, send 'em back and get replacements), or develop problems as ALL RTR turnouts do occasionally, but, at least the vast majority of problems the ME turnouts will have can be fixed promptly, without soldering in wires or replacing plastic ties with PCB ties.  On several modules I built for customers, I used ME #6's and used Neolube on the closure point heels/hinges.  Those modules are still running and the owners have never reported any dead closure point rails.  That's a really quick fix.

[Mark W]

As far as a "huge" expenditure of time is concerned...it doesn't exist. 

Where did you get a hold of your time-stopping-device?  I agree, it's not complicated.  But it absolutely takes a lot of time.  Some people here are talking between 120-300 turnouts on their layout.  Averaged at about 45 minutes per turnout, that's the equivalent of two to five and a half WEEKS at 40 hours/week doing nothing but building turnouts.  A modest 50 turnouts is just about one 40 hour week, 8 hours a day.  It would only take about 10 minutes to open that many turnout packages. 

[mmagliaro]

Where did you get a hold of your time-stopping-device?  I agree, it's not complicated.  But it absolutely takes a lot of time.  Some people here are talking between 120-300 turnouts on their layout.  Averaged at about 45 minutes per turnout, that's the equivalent of two to five and a half WEEKS at 40 hours/week doing nothing but building turnouts.  A modest 50 turnouts is just about one 40 hour week, 8 hours a day.  It would only take about 10 minutes to open that many turnout packages.

But unless you are extraordinarily lucky, you can't just open the package and install a ready-made turnout.
I think that's the point some of us are making here - that in the end, when you account for the time to prepare a
turnout in advance so it doesn't suffer failures, or to fix them in-place on the layout later if they fail, you are worse off
than if you had hand-made the turnouts. 

I certainly averaged at least 45 minutes per turnout using Atlas ready-made units.   Maybe people think that's weird, but
nearly every one of them suffers from this (photos below).  I don't know how people just install these things
and don't have cars and engines climbing up out of the points.   I know, I know.  People say it all the time, "I just
install them and they work."  But not me.    I find it takes a lot of time to file and smooth these
out so they are in gauge, and that's just one problem they have.

[ednadolski]

Some people here are talking between 120-300 turnouts on their layout.  Averaged at about 45 minutes per turnout, that's the equivalent of two to five and a half WEEKS at 40 hours/week doing nothing but building turnouts.  A modest 50 turnouts is just about one 40 hour week, 8 hours a day.  It would only take about 10 minutes to open that many turnout packages.

This is 5.5 weeks out of how many years to build a layout with 120-300 turnouts?  Plus figure on the time to find, troubleshoot, and fix all the balky ones in a large config.  I have zero expertise with large layouts myself, but I would think that as the number of turnouts increases, the long-term reliability becomes even more valuable.

Most folks should be pretty proficient at construction by the first dozen or so turnouts.  And OTS turnouts often need tweaking out of the package, i.e. filing points and trimming to fit the track configuration.  They sometimes have other issues such as overly-generous tolerances. Not to mention of course the cost difference, which will certainly add up on a larger layout.

Ed

[rschaffter]

I've found with Atlas C55 switches that often electrical contact to the frog will be lost when soldering to the lug to power the frog.  I now put a c-clamo on the frog before soldering...

[leikec]

I have a #7 Atlas turnout that doesn't work worth a darn, and the turnout isn't installed on a layout, I might try that one piece point modification.

Jeff C

[Kev1340]

Wow! I never realised there could be so many problems with RTR turnouts! Makes me glad I've always stuck with Peco electrofrog.

I've got a couple hundred on my main layout (code 80) and over half are thirty years old, have been reused on several layouts, survived house moves etc, and the only maintenance is cleaning contacts, I don't even bother to wire feeds to the frogs! The code 55s on my newer N&W layout are equally bullet prof, despite being on their second layout.

As for 'plating' wearing off due to track cleaning ... Peco only use solid nickle silver, no issues there, but it sounds like you should invest in an HF track cleaner and forget manual cleaning.

Cheers,

Kev

[mmagliaro]

Wow! I never realised there could be so many problems with RTR turnouts! Makes me glad I've always stuck with Peco electrofrog.

I've got a couple hundred on my main layout (code 80) and over half are thirty years old, have been reused on several layouts, survived house moves etc, and the only maintenance is cleaning contacts, I don't even bother to wire feeds to the frogs! The code 55s on my newer N&W layout are equally bullet prof, despite being on their second layout.

As for 'plating' wearing off due to track cleaning ... Peco only use solid nickle silver, no issues there, but it sounds like you should invest in an HF track cleaner and forget manual cleaning.

Cheers,

Kev

Peco have always been rock-solid units for me.  Except for the issue with the code 80 insulfrog ones where the guard rails are set a little too far toward the center, allowing correctly-gauged wheelsets to pick the frog (a problem that has been well-documented a million times and can be fixed with a shim in the guard rail or other methods), I never had a trouble with them.   My only grumble (and it's a common one, not just from me) is the big ties and wide tie spacing, even
their code 55 track.  If they had made a North American prototype style code 55 track with correct ties,
I would have gone Peco in an instant.

[sp org div]

I just replace the point and closure rails completely as one solid rail and reinforce the frog lead as necessary.



This could be done on a laid/ballasted turnout quite easily.  The hardest part would be replacing a few ties with PC boards, but a tool like Robert made would make it pretty straight forward.

Been there done that.  Found a high failure rate on the frog rails over here (due to that "preloaded cast in plastic" electrical joint they used)... Where is the mechanical bond?  Then the throw-bar arrangement is pretty pathetic....  sorry Atlas (love ya anyhow).
Good looking fix Mark!  I've only done that on the ME turnouts... so, good to know the, "slide the rail in" technique works on the Atlas turnouts also.
I like that custom under rail scraper you got there Robert!  I am going to steal that...   

[timgill]

My switching layout uses some older #5's and #7's that I had on hand. When installing a couple of them, I did have to retrofit jumpers from the stock to point rails because of failed jumper tabs.

I also recently noticed gauge issues on several of the turnouts between the frog and guard rails, which causes an unsightly "jump" as a locomotive passes through. After a good bit of filing it was fixed; but I haven't gotten around to doing all 6 turnouts yet. That, combined with the possible future failure of more point rail jumpers in the future, is really causing me to consider just handlaying new TO's and swapping them into place; some of them are already ballasted and it would be a pain-in-the-you-know-what.

I think Mark's idea to replace the point rails completely with some PCB ties is an excellent compromise, and I may try that first. I'll be watching this thread for any other bright ideas...