Fixing points that cause derailments

[Sheffie]

Symptom:

Trains derail when running over the points. because they aren't going in the right direction. Either the points are set to straight, and some of the wheels try to turn out, or the points are set for the turn out, and some of the wheels go straight on. (This post is an example of the former problem.)

 

Diagnosis:

The point blades are not quite touching the track. This should be visible under magnification (see first image). Some train wheels may get pushed downwards by the end of the point blade (and go straight on), but on others, the flange will either ride over the end of the point blade and slip upwards onto the outside, or simply push the point blade down, away from the curved rail.  

 

Solution:

In a nutshell, we are going to provide more room for that upper blade so that it can move further up, to the point where wheel flanges can't get "above" it.

 

The first step is to mark the curved rail just past the end of the blade. I've used a drop of black acrylic paint. Rust pens or other technologies should work equally well.

 

Now that the rail is marked, it can be removed and filed.

Long nose pliers should be able to pull the rail out using smooth steady pressure.

A small flat file is needed for this job, as is some form of magnification.

 

The goal is to produce a sharp notch on the right side, that tails away smoothly to the left.

This allows the point blade to sit higher up, above where the wheel flanges will be running.

The black spot, not visible here, is just to the right of the filed-away area.

 

Final tip:

The rail is far easier to re-insert into the various plastic and metal clips (the latter being far more demanding) if you first file the end of the rail to a point. I wasn't able to capture this in a photo, but the goal is to make it look like an arrowhead with a sharp point and edges at +/-45º. This will then thread the needle much more easily.

 

You should be able to reassemble the points using the pressure you can exert with your bare hands, holding the rail between thumb and forefinger. If you find yourself using the workbench to push up on the end of the rail, stop. Look more closely, and try to adjust the position of whatever the end of the rail is touching.

Edited March 15, 2020 by Sheffie

[Sheffie]

Result:

There's still a noticeable gap. but it's much less severe than it was.

Ideally the wheel flanges shouldn't quite touch the rails, anyway, so the gap doesn't need to be zero.

The important thing is, the trains are no longer getting derailed.

 

Conclusion:

Points that are causing derailments can be fixed, with less than an hour's work, assuming you have a few tools.

You don't need to be ultra-precise to make this work.

Just take your time, and you can improve your points' performance to the point where derailments hardly ever happen.

Edited March 15, 2020 by Sheffie

[cteno4]

Excellent point tune sheffie! This is the classic fix for Kato #4 Points. Definitely worth the effort if you have any problem points as it does work!

 

kudos and thanks for the nice documentation!

 

cheers

 

jeff

[gavino200]

Sheffie, how has this held up over time? Did the technique completely eliminate derailments? Have you discovered any further tips in the time since you made this guide? I'm going to give it a try myself soon.

[Sheffie]

It’s been a good solution to derailment for trains approaching points from the sharp end. I’ve found that it’s more or less eliminated derailments. Perhaps not for all scenarios or for all trains. On occasion I’ve found it necessary to add a short straight between a curve and a point, to give the rolling stock a chance to straighten up before hitting the blades of the points. But most of the trouble is solved by applying this fix—maybe it takes two or three tries to get it right, but it generally does a good job. 

[Kamome]

This seems to be a well documented issue with the Kato #4 turnouts. I did exactly the same to my points thanks to the video from Mike Fifer about 7 years ago.  Definitely makes them more reliable, even bullet trains at speed seem to stay on the correct path.

 

 

[gavino200]

Thanks guys. I'll have a go at it today. I have a small automation practice layout that has four Kato #4 turnouts. I'm going to try out this technique on those. In the past I've found this problem irritating enough to consider switching to Peco track. However, I have really  quite a lot of Kato track at this stage and it would be a shame to waste it. 

[cteno4]

I havnt done one in like 10 years, but it did work well for me. Engineering a small straight before the blades is the other safety thing as Sheffie notes. The straight even goes for #6 points as well, especially with shinkansens at speed. From what I could tell I could see a bit of swaying in some cars when they come off the double viaduct curves at speed and that would at times detail around the blade point or even bump platform ends and cause derailments. Adding 128 between point and curve and pulling the platform back a half inch or so from usual Kato platform spacing solved most all of the derailments at the points in the station.

 

I’ve heard from many that the #4 tuning has worked for them.

 

jeff

[gavino200]

5 minutes ago, cteno4 said:

 From what I could tell I could see a bit of swaying in some cars when they come off the double viaduct curves at speed and that would at times detail around the blade point or even bump platform ends and cause derailments.

 

Do you think this is related to the super-elevation? I'm not the biggest fan of super-elevation. It looks nice but the effect is subtle. Mainly it drastically reduces my ability to make custom curves. If I replaced them with side by side regular curve track, I could then cut the viaduct sections to whatever circumference I liked. Of course, I couldn't do anything about the diameter. 

[cteno4]

This was on non super elevated track. I too am not a fan of the super elevated, especially with the planning issues it creates and the effect to my eye is a bit too hard for the short radius of double viaduct track. The old flat track is getting harder to find on the secondary market…

 

jeff

[gavino200]

Thanks guys, for your information and guidance. I also used the Peco turnout as a guide and tried to copy it. Peco notch theirs on both sides, so I did too. The #4s on the practice layout are now giving no trouble. I have a #6 that is giving trouble to my EH200. I think the #6s could benefit from the same treatment. But they're more complicated to disassemble. If I release the rail I'll have to solder it back in place. So I think I'll work on those another day.

 

This is the Peco and the Kato (before filing) side by side.

 

[cteno4]

Yeah it can’t hurt doing both sides and also #6 could stand it for the cranky trains and just the odd situations as point blades are usually the derailers!

 

jeff

[Martijn Meerts]

It's also a good idea to look at the series of videos by Fast Tracks that show how to build a turnout using their fixtures. Even if you don't build your own turnouts, they give you a good idea of the critical points in a turnout, and what you can do to minimise derailments. The first one can be found at