Atlas Turnout Goblins


GNMT76

Active Member
I have seven Atlas code 83 Snap Switch turnouts (item no. 542, basically a #4) on my layout, all powered by Tortoises. They've been but moderately used for only about four years now. Over the past few months, three of the turnouts have begun misbehaving by derailing all my locos (four diesels and one steam) on nearly every occasion - but only when entering the diverging route. When switched to the mainline route - or when the locos are exiting the diverging route - they operate perfectly and cause no derailments. A fourth turnout, strangely enough, derails the locos about 10% of the time when they are exiting from an adjoining mainline track. But perhaps that's another story.

I've checked everywhere there is to check on the turnouts, the adjoining trackage (also Atlas code 83 snap track and flextrack) and the locos' wheels and trucks with the NMRA gauge. All appear to be properly and proportionally adjusted. I also see no kinks or other track or loco anomalies. And when the Tortoise is activated, both point rails snap smartly and firmly against their respective stock rails.

A close eyeball inspection of the problematic turnouts, however, shows that both trucks of the locos jump in the same two places on the outside point rail (again, only when entering the diverging track): first, one tie beyond the throwbar and again exactly half way across the point rail. That one outside point rail is also not precisely aligned at the pivot point with its other half (as is the inside point rail); it leans inwards just a tad and wiggles slightly at the touch of a finger - and it has a slight, but visible, bend along its entire length. The inside point rail is perfectly straight and firmly in place. Additionally, the ends of both point rails (which are sharp and not at all blunted) long ago worked themselves loose from the throwbar. I used to push them gently back in until they "clicked" in place in the throwbar, but they'd always pop out again. More recently, I used CA to secure them in place, a remedy that seems to have worked. Mostly.

All this tells me there's simply a quality issue with the design and manufacture of Atlas turnouts. I even replaced one problematic turnout with a brand new, identical one last week, only to find that the locos continue to jump at the same two places on the same point rail and immediately derail. Rarely will the locos continue ahead, but always with an audible "complaint" before re-righting themselves on the track. The new turnout also has the same point rail misalignment and wiggling issues as the older one (you'd think developing such a problem would take time!). Oddly, both turnouts also have a noticeable vertical bend in the middle, which according to a review on the M.B. Klein site I recently read, is a chronic problem with Atlas turnouts.

Given all this, is there a way I can fix and eliminate this problem for good? Am I missing something? Or, as I'm starting to think, should I dump Atlas turnouts and go with a better quality make? I'm thinking Peco, Walthers or Micro Engineering. Are there any compatibility issues with those three makes and Atlas tracks? Building new turnouts by hand is not something I wish to get into, however. So it's either fix or replace.

I look forward to reading your potential fixes and recommendations on quality of design, build and operation of other brands.
 
Last edited by a moderator:
I don't really have much to say other than I prefer Atlas Custom Line Turnouts over Atlas' "Snap Switches". The Snap Switches dead plastic frog and the fact they are a tight radius for locos to negotiate is why I don't care for them. You could try filing the points and juncture of the points and closure rails to get them to better align with each other. Every turnout on my layout has had the tips of the points filed to a rounder and sharper tip. If you haven't done this, this should be priority number one on your "to do" list. Sorry I could not be of better help!
 
I've never had snap switches but did have the regular ones and still use some Custom Lines.
Sounds like you've checked things over pretty good, unless I missed it did you look at the rivets to see if they've worn lose?
That's one thing that will let the points rock and can be tightened with the switch removed from the layout.
The throw bar pins can also get worn out and not hold the points tight against the stock rail.
 
I have no personal experience with snap switches. The only Atlas turnouts I have are 4 #6 customline turnouts in my hidden staging area. No problems at all in over 20 years with them. I have seen some snap switches years ago on a friends layout and the didn't impress me. They seemed sloppy and the points on them seemed loose.
 
I used Atlas Custom Turnouts on my current layout and haven't had any problems. On other layouts I used Code 70 Walthers and Shinohara and also had no problems.

Are you running 4 or 6 axle locomotives? The diverting route on your Snap Switches is designed for a 18" radius and many locomotives have trouble making the transition. Like Chet I've found Snap Switches to be loose and not have a long useful life.

Thanks.

Greg
 
Last edited by a moderator:
it has a slight, but visible, bend along its entire length.
That is to be expected to match the outside diverging rail.

Given all this, is there a way I can fix and eliminate this problem for good? Am I missing something? Or, as I'm starting to think, should I dump Atlas turnouts and go with a better quality make? I'm thinking Peco, Walthers or Micro Engineering. Are there any compatibility issues with those three makes and Atlas tracks?
First just a point in general. The snap switch is made to work with 18" radius curves (hence the noticeably curved point rail) and is closer to a #3 than it is to a #4. I did the exact calculation once but that info is not at my finger tips. For some reason when Atlas took their line to the code-83 rails they did not continue the hollow rivet design of the code 100 turnouts. This is one thing that leads to the wobbly point rail, so in my opinion that is what needs to be fixed. I have three ideas NONE of which have I tried. These are being thrown out for consideration and are things I would try if I had this situation.
1. Replace the pivot pin with a 00-90 screw. Drill up from the bottom, tap it, and use a washer to get more vertical stability and better electrical contact with the closure rail.
2. Reinforce the pivot pin. I was thinking of soldering something to the outside portion of the rail across the pivot. I was thinking of something like bell wire that could flex a tiny bit as the rail moves back an forth. Just a piece at the very top might be enough to help stabilize it. The issue is, of course, that eventually it will stress out and break too. I thought of soldering wick, but unless one is really good at soldering this could easily result in a soldered solid point.
3. Remove the pivot pin. Press a railjoiner onto the closer rail and then just put the point rail into that. Let the point "pivot" in the slop of the joiner. I do this all the time with scratch built turnouts, BUT they have the point ends of the point rails soldered together. So I don't know how well this would work with a free and independent point rail.

Don't know if that will help or not. Once again emphasizing I have not tried any of these as I don't use snap switches - they are just thoughts.

And as for compatibility, I have never used the other brands of code-83 so cannot comment.

As the others, I also have been using Atlas code 100 custom line turnouts for over 50 years with only 1 real issue. Code 83 custom lines for about half that long.
 
Last edited by a moderator:
Have you checked the gauge with the NMRA track gauge?
When the point rail gets worn, the rail head will tend to be a bit lower than the solid rail head. That's where you will get a derailment.
 
Problem #1.) You are using Atlas Snapswitch turnouts. I would say that about 1/2 of the turnout on my layout are snapswitch's and about 3/4 of them give me problems. Granted, I run code 100, but still... Their quality is really not that great. The few custom lines I have work without issue. The Peco turn outs work without issue. My scratch built turns seem to work with nearly no issues. So I would say that problems that I have are the snapswitch's themselves. They CAN be tuned to work better, though, like Iron Horseman suggests. I have done similar things to my turn outs with moderate success. Still, there are a few that are pegged for replacement.

Are you running 4 or 6 axle locomotives?
This is a great question. The nice, everybody-but-me-runs-these, 6 axle locos will fight you a great deal on Snapswitch turnouts. That 18"r curve is not a friend to any SD loco. So what do you run?

Your eyeball exam will be the best tool in figuring out whats going on. Try to look for that exact moment when the wheels start to lift and stop the loco right there. Take your time and examine exactly whats going on. It may not be the leading wheels that cause the problem but rather the trailing wheels that get stuck and won't allow the truck to turn - or something like that.

Let us know what you find! Good luck.
 
This thread has caused me to consider replacing my one Snap Switch with something better! This is going to happen!
 
Certainly makes one think twice about ever using these turnouts, particularly brands that utilize 'stamp sheet metal rails'.
 
Gentlemen,

Thanks to all for your suggestions and comments.

The local hobby shop tech gave the questionable turnout a micro inspection, finding that the point rails are indeed properly aligned. He then slightly twisted the beveled end of each point rail outward and, with a small screwdriver, gave the two connecting rivets (or whatever they are) a gentle push from underneath, hoping to tighten them a bit. Back on the tracks, however, the locos continued to jump at the same two spots as before and then derail. The tech also commented that Atlas' design of its turnouts is decades old (specifically the beveled point rail ends, which are absent in Peco, Micro Engineering and Walthers) and may not be compatible with the newer design of loco and rolling stock wheels today. All my locos (4-axle) are only four years old, so that's a big red flag. We're thinking that the bevels themselves may be the culprit.

I may play with bending the beveled ends some more, but at this point I have little hope for a resurrection. You comments regarding the quality of Atlas turnouts is noted and also confirms what I've been observing. I'll probably replace them with Peco #5 or Walthers #4 turnouts. What's your experience and degree of satisfaction with those two brands or others you've used?
 
I haven't been following this, and haven't even read your first post...please forgive me....but I do know how to fix turnout problems, and I have made my own turnouts for years. No matter who puts the thing on the layout, or who crafts it, the points should be filed to a long taper. I never attempt to file a bevel. I place a clean removed plastic tie, or a matchstick, between the stock and point rails, and then use a needle file to firmly, but carefully begin a series of long sweeps from frog to point tip, attempting to leave a taper of about 2". Even if the taper doesn't need to be that long, the point rail must be close to paper-thin by the time it ends over the throwbar. It must also be securely and tightly held against the inner face of the stock rail.

If your wheel sets are rather reliably picking points, and you are certain the wheels are NMRA RP 25 compliant AND in gauge, two things are most likely: the points are not sharp enough and they are not kept from wobbling by the hinge mechanism. One other unlikely even could be that the point's tip is a bit too high, causing the lead wheels to bump upon contact and possibly slide one way or the other, causing the first part of the derailment.

Have you gauged carefully with a flange gauge all along the flange path...right from must before the throwbar, through the points, along the curved divergent points rails and beyond to include the closure rails....and then through the guards and frog? If a pioint rail could stand to be widened, or it 'wow'd too much (has too much of a bow), you probably have a correctable gauge problem.
 
I haven't been following this, and haven't even read your first post...please forgive me....but I do know how to fix turnout problems, and I have made my own turnouts for years. No matter who puts the thing on the layout, or who crafts it, the points should be filed to a long taper. I never attempt to file a bevel. I place a clean removed plastic tie, or a matchstick, between the stock and point rails, and then use a needle file to firmly, but carefully begin a series of long sweeps from frog to point tip, attempting to leave a taper of about 2". Even if the taper doesn't need to be that long, the point rail must be close to paper-thin by the time it ends over the throwbar. It must also be securely and tightly held against the inner face of the stock rail.

If your wheel sets are rather reliably picking points, and you are certain the wheels are NMRA RP 25 compliant AND in gauge, two things are most likely: the points are not sharp enough and they are not kept from wobbling by the hinge mechanism. One other unlikely even could be that the point's tip is a bit too high, causing the lead wheels to bump upon contact and possibly slide one way or the other, causing the first part of the derailment.

Have you gauged carefully with a flange gauge all along the flange path...right from must before the throwbar, through the points, along the curved divergent points rails and beyond to include the closure rails....and then through the guards and frog? If a pioint rail could stand to be widened, or it 'wow'd too much (has too much of a bow), you probably have a correctable gauge problem.

Crandell,

All your suggestions - except for the last one about the flangeways gauge - are covered in my original posting. I just tried that one and find that point tabs and the flangeway tabs all fit as you describe, all the way from the beveled point ends through the guard rails and the frog. I believe my local tech guy made this same check the other day, but I'm going back tomorrow for a second look-see.

What are your thoughts on the local tech guy's suggestion that the beveled ends of the point rails (Atlas' decades-old design) may be the culprits, in that those bevels are not compatible with today's newer wheel designs? All my locos are only four years old. And, this turnout is brand new, purchased only last week.
 
Last edited by a moderator:
I can only think of one way in which they might not be compatible, and that would be if the angle from vertical exceeds the angle of the flange away from the plane representing the back surface of the wheel. Obviously this isn't a depth problem, such as one might have with older flanges in a filled frog. If it is the bevel, it can only be the angle, and this is what I bore in mind when I stated that you really need a taper....like a looooooonnnngg and very thin flat-tipped screwdriver blade. Mine, whether hand-laid or commercially produced, all end up with a thin long blade by the time I am finished with them. In fact, I could be accused of nearly compromising the integrity of what little metal is left. Fortunately, I have yet to experience a bent and irretrievable blade tip on my points, say somehow crunched by a locomotive.
 
I can say Peco turnouts are a cut above, and I've had no real issues with Walthers to speak of.
I replaced all the older Atlas switches on the previous layout with Micro Engineering ones... I now have a stack of thirty broken ME switches that I'll take apart and use as scenic elements. I'd stay clear of those.
 
All my experience with Atlas turnouts is with Code 100. Over the years, I have used both SnapTrack and Custom Line, and have experienced some problems with them all. The rivets that form the pivots for the points can loosen up over time. Generally it isn't so much the rivets as the plastic ties compressing or wearing. IF that is the problem, then removing the turnout and flattening the rivets slightly (you have to be real careful not to overdo it or the point won't pivot), may be the solution. Another possible problem is if the angle that the point makes with the little tab to which the rivet attaches may not be 90 degrees. In that case you might have to bend the point slightly in the vertical plane. With the Custom Mark IV Code 100 turnouts, I found in some instances that the diverging route point was NOT curved properly to be in gage with the diverging stock rail! I had to bend the diverging point to properly gage. The problem there was that the side of the point that contacts the straight stock rail did not have the base filed away enough (or the straight stock rail did not have the foot filed away enough to clear the bottom of the point!). I do use both Snap Track and Custom Line Mk IV turnouts on my layout, simply because the switch machines do not require under-the-table space. This makes for easier access, even when there is space for a below-surface switch machine, and to save wear and tear on 75+ year old body parts! I do use Shinohara and Peco turnouts, and have zero problems with them. Shinohara do require switch machines or manual ground throws. Peco can either be manually operated or with switch machines. However, I have had very poor operation with their switch machines, and have used some older Kemtron and other twin-coil machines, most mounted above-ground, and disguised by buildings, bushes, or even piles of ties, leaving clearances for the actuating rods. Recently, in one instance, where a Snap Switch began causing derailments and stalls on my main line, where the main run was on the diverging route, and because the straight route lead to a dead siding that disappeared into a tunnel, I simply eliminated the turnout altogether and made a dummy connection that appears to be a highspeed turnout that doesn't operate at all. Good luck!
 
I have been following this thread with great interest. I used Atlas snap remote switches on my layout # 540 and #541 when I started three years ago.. and I mean...alot of them. The problem of my locomotives (4 axle) going over the snap switches just started recently. Up until that started... my problems have been with the switch mechanism getting bound up and not fully engaging when I press the remote. I made the mistake of taking one apart to try and fix the problem... which became a tangled mess. This most recent issue with the locomotives is that the locomotive will continue on the straight through without traveling onto the diverging route when thrown (I hope that explanation makes sense). The locomotive does not leave the tracks.. it simply does not go onto the diverging track. Which like in the case of trailrider, the diverging track is the mainline. I am going to try the filing and working with the rivets as mentioned here.

My question is.... when funds allow, I would like to replace the Atlas remote snap switches with something else. I am hoping that I can find a turnout make and size to where I can simply remove the Atlas remote switch from the layout and install a different brand. Can anyone tell me what would be the best turnout brand and size to do this with so that I do not have to tear up and realign alot of track around the turnout being replaced.
 
I would like to replace the Atlas remote snap switches with something else. I am hoping that I can find a turnout make and size to where I can simply remove the Atlas remote switch from the layout and install a different brand. Can anyone tell me what would be the best turnout brand and size to do this with so that I do not have to tear up and realign alot of track around the turnout being replaced.
Never mind - Just as I finished the post, I realized you were talking code-83 which made it all irrelevant.

For those who might be interested in this question for Code-100 track -->

Without going to the used market, the only direct replacement that I know of is Life-Like as the link below. I've never used nor know anyone who has used these but just the brand name makes me wary.https://www.hobbylinc.com/life-like...nickel-silver-model-train-track-ho-scale-3002

One could buy a Bachmann EZ-track turnout and remove it from the plastic roadbed. That would be a very expensive option for a direct replacement. I do know several people who have used EZ-track (Spacemouse, a former member of this forum, for one) and they were not happy with the turnouts in the long run.

On the other hand one can sometimes find new-old-stock on the used market. In which case both AHM and Model Power made a direct replacement for these turnouts. I did use some of the AHM way back in grade school so I don't remember much about them. To the best of my knowledge they are no longer in my collection, so that might say something about them.

The perhaps not a direct replacement but one that might work is Marklin's small turnout (one would want the two rail K-line not the three rail C-line). The specs say the turnout branch is 22 degrees 30minutes. A Snap-switch is 20 degrees. They come in a package of 2. Once again I've never tried them so I am just answering the question by relaying information, not making a recommendation.
https://www.hobbylinc.com/marklin-k...ho-scale-nickel-silver-model-train-track-2265

Having said that, your best would be to start planning on biting the bullet and go with the Peco short turnouts.
 
Last edited by a moderator:



Back
Top