Is There a Different Take on Derailments at Turnouts?

[BradKT]

I use Bachmann EZ-track on my layout and I still have not licked the problem of derailments at turnouts.  I think that the most maddening thing about this issue is how randomly it happens.  The same car can cross over the same turnout without problem for 3 laps and derail on lap number 4.  One car can derail consistently at the same turnout every time.  The problem seems to happen much more with longer cars, such as 50' freight cars and 72' passenger cars.  It can happen with brand new RTR cars or with assembled car kits.  It can happen with engines...especially the second engine in a 2 engine consist...so the issue isn't the weight of the cars or engines.  It can happen in the middle of a long straight.  I even modified my layout so that the turnouts aren't anywhere near a curve and I have two problem #5 turnouts.  One of the worst on my layout is a remote crossover  turnout (single piece) where the car switches from one track to a second track.

I have watched the cars as they cross a turnout closely and have observed that it is usually one set of wheels in a truck (usually a rear one) that is not making complete contact with the rails...it hits the frog and bingo!...the car derails.

Then I had a thought and I want to know what the group thinks about it.  Has anyone has had any experience using spacers (Kadee or others) as a way to keep the trucks and wheel sets level on the bottom of the cars so the it maximizes the  contact of all wheels with the track...especially as they cross the frogs on the turnouts?   We all know that the truck needs a certain amount of play horizontally and vertically as the cars navigate around the track..but that also seems to allow a certain amount of vertical wobble so that one wheel may not be on squarely on the track as it crosses a turnout frog.  It may be raised just a little bit...and that seems to me exactly what is happening.  Does anyone have any thoughts about or experience in this area or am I totally off track in more ways than one?

I have replaced turnouts and that solved some some problems.  I have adjusted the screws that hold the trucks to car bodies.  I have added weights to cars.  I have even tried thin strips of plastic on the inside of turnout frogs to keep the wheels close to the track as they pass through the guide and over the frog.  I haven't messed around with the frogs, but nothing so far seems to work consistently and I have a hard time believing that Bachmann's turnouts are that bad.

If the turnout frogs sometime sits a bit higher than the rails, should they be filed down?  What part of the frog should be filed?  Is there a demo video anywhere on the net on how to do this?  I believe this to be the case on my problem switches because I notice the cars (and engines) visibly wobble as they cross a switch, one set of wheels loses contact with the rails and the derailment happens.  I have also confirmed this using that clear flat car that Micromark sells.  If so, will someone please walk me through how to do this?  How far should the frog be filed down?    

It has been my experience that the derailment issue is the main problem with HO scale...as opposed to O scale or S scale.  Do I have start over and rip out all of my Bachmann track?

There ain't nothing easy about this issue and it is totally ruining any positive experience that I have having.  I have spent an awful lot on money on this hobby and still can't be able to have the basic experience of sitting back and enjoying simply operating my trains.  I still have to hold my breath every time my trains go around the track.  It is not a question of if a car will derail...it is only a question of when and which car will be be this time?  

That is not the way that this is supposed to be working out.

[Ken G Price]

Have you checked the wheel gauge on the cars? If not you need to do so to eliminate or confirm that this is or is not a problem. Putting wheels in gauge is a lot less expensive then replacing switches. Also low flange thinner wheels ride better through turnouts them deep flange wheels.
If you do not have a NMRA gauge you really should get one.

[OldTimer]

I can certainly sympathize with your problem.  I don't use EZ Track, but not because of this issue....I prefer flex track for the look, but I've had a lot of experience with derailments and turnouts over the years.  In my experience, derailments are primarily caused by three things:
 
1.  wheels out of gauge and other wheel problems
2.  points that do not fit tight up against the stock rails
3.  uncoupling pins that hang too low

If you do not have an NMRA standards gauge, you need to get one.  Sometimes you can find them at you LHS, or you can order on-line directly from the NMRA.  You also need a Kadee coupler gauge. They are widely available.

Check the wheel gauge on every car.  Wheels that are out of gauge can sometimes be twisted on the axle and moved back into gauge.  Apply a tiny drop of ACC to fix the wheel in place.  Look for plastic wheels with "sharp" flanges and throw away any that you find.  Throw away trucks that "crab" when they go down the track unless turning one axle around fixes that problem.  Throw away trucks that have truck-mounted couplers.  It is a good idea to standardize your wheels when you find a brand that is reliably in gauge and has the RP-25 contour.  Doesn't have to be all done at once, but it is a goal.  The Kadee washers are used to adjust the height of body mounted couplers.  Putting a washer on a car will push the body up, not push the truck down.  Trucks should be adjusted so that on each car, one truck can only swivel and the other truck can swivel AND wobble up and down just a little

Take your NMRA gauge and go over your turnouts with a fine tooth comb.  The gauge comes with directions.  Fix any problems that you find.  Take a small flat file and file the upper corner off the points.  You should be able to run your finger along the inside edge of the stock rail and not feel the point.  Make sure that your switch machines are throwing the points far enough and then locking in place.  

Use your Kadee gauge to make sure that the uncoupling pins are not too low.  Kadee makes special pin adjusting pliars, or, if you're careful, you can do all the adjusting you need with small needle nose pliers.  

The problem with the #5 crossover is, in part, that it is a #5 crossover.  It introduces a sharp ess curve and must be negotiated slowly, preferably with shorter cars and 4-axle diesels.

I have had a zero derailments goal for every railroad I've built and I've been successful in reaching that goal.  You can too.  Good Luck!
OldTimer

[BradKT]

I have replaced all of my wheels on my rolling stock with the Walthers-Proto 33' metal wheels, so I really don't think that the wheels being out of gauge is the problem.  Plus, this problem happens at random to any car, including an RTR car.  I have dozens of cars and they are primarily made by Athearn, Atlas, Walthers, Accurail and some by Bachmann.  On some of them, I have replaced trucks and complete wheel sets.  I have used that little tool made by Micro-Mark on my plasic trucks to ensure that the wheels roll freely.  The same car that previously crossed a turnout one time can derail the next.  

Besides, how would one fix an E-Z track turnout if it is out of gauge?  Bachmann's rails on a piece EZ track fit in a fixed guide and it is impossible to get a rail back in the track it if comes out...that piece is ruined.

Now that I think some more about it, I am beginning to really think that the issue is the height of the plastic frog.  My engines (particularly the second one in a 2 engine consist) and cars visibly rise and/or wobble as they pass over it on almost EVERY Bachmann turnout.

How do I fix that?  Is it the top of the frog, the groove for the wheels or what?  If anyone has done do, please give me some specific instructions with details...and don't assume that I know anything.  A diagram would be most helpful.

[Jim Banner]

I am not sure what OldTimer uses to achieve zero derailments but I can tell you what works for me.  On my layout, every derailment requires a written report of which end of which car derailed, which cars it was coupled to at the time, exactly where the derailment occurred, and the type of movement (accelerating, decelerating, forward, reverse, approximate speed) was involved.  Then the report is attached to the car with an elastic band and the car is set aside.

Between operating sessions, the car goes through a check process -
- do the wheels spin freely?
- do the wheels wobble?
- are the wheel treads dirty?
- are the wheels in gauge on all axles?
- do the trucks rotate freely?
- does the car's weight match the NMRA recommendation?
- will the car run freely down a 3% grade without being nudged to start it?
- do the trucks have enough play to tilt properly side-to-side and fore-and-aft?
- are the couplers at the correct height?
- do the couplers return to centre properly and freely?
- are the coupler jaw pins in place?
- are the coupler trip pins the correct height?
- do the couplers uncouple  and recouple properly?
- does delayed uncoupling work properly?

When any faults are found, they are corrected and the car retested.  When there are no more faults, the car is returned to service.  Admittedly, many of the possible faults listed above are not likely to lead to derailments.  But when the car is off the track and on the bench, it is a good time to check everything.  Once the cars are fixed, the bad order reports are then filed against parts of the layout (the blocks from when it was wired for dc).  If an area has too many reports, then the track in that area is checks.  Since derailments almost always occur at turnouts, the checks are mostly of turnouts:
- are all the rails in gauge as measured with an NMRA wheel and track gauge?
- do the points throw freely and close properly every time?
- if the point rails are moved slightly away from the stock rails, will the switch machine or spring return them to the stock rails?
- are the frogs dirty from wheels shedding dirt as they hit the frogs?
- are the point hinges tight enough to avoid misalignments between the heels of the point rails and the frog rails?

If the turnouts pass all the easy tests above, then it is time to get more serious:
- are there any kinks, bumps or twists in the tracks leading to the turnout, or worse, in the turnout itself?  Kinks (horizontal deviations) are usually easy to see or can be checked with home made radius gauges.  Bumps (vertical deviations) and twists are harder too see.  Turnouts and the tracks leading up to them need to be dead flat (although not necessarily level) for perfect performance.  One way to check is to set a piece of flat, clear glass on the turnout and make sure all the rails touch it.  If you have problems telling which rails are touching the glass and which are not, a thin layer of grease applied to the tops of the rail heads, followed by putting the glass on straight down will usually help.

Years ago, an older model railroader put it this way:  Perfect cars can run satisfactorily on mediocre track and mediocre cars can run satisfactorily on perfect track, but it takes perfect cars running on perfect track to run perfectly.  I see perfection as something to strive for but something about impossible to achieve.  To me, having half a dozen guys running trains during a two or three hour operating session with zero derailments is a great goal but having one derailment or maybe even two during such a session is good enough.

Jim

[NarrowMinded]

I didn't read everyones response but I add a little more weight to cars that derail often, but I always check the cars first to make sure they are within spec, if your pulling long freights sometimes a car is just not heavy enough to stay down on the track, also check the couplers for up and down play, to much can cause the whisker to catch on the turnout, by itself it may look fine but under load it may be held lower at times

NM

[Jim Banner]

I have replaced all of my wheels on my rolling stock with the Walthers-Proto 33' metal wheels, so I really don't think that the wheels being out of gauge is the problem.

Without checking the gauge, you will never know whether the wheel sets are in gauge or not.

Besides, how would one fix an E-Z track turnout if it is out of gauge?

After determining whether the turnout is under or over gauge, you would widen it out with jeweler's files or narrow it with shims.  The areas involved are usually small.  If you have a friend doing stained glass work by the Tiffany process, bum a bit of self sticking copper foil from him/her for use as shims.

The same jeweler's files can be used to slightly round over the top inside corners of the points and/or sink them slightly into the stock rails.

If the frog sits higher than the stock rails, you can reduce the height of the frog using a sheet of 320 grit wet and dry sand paper on a sheet of glass.  Turn the turnout upside down and move the whole thing longitudinally over the paper, not cross-wise.  And check frequently to see how you are doing.  Needless to say you have to have the turnout separated from the rest of the track work to do this but you will need it separate anyway when you visually check to see if the frog is high or if the flangeways in the frog are too shallow.  Either can cause locomotives and/or cars to bump up as you mentioned but filing/sanding the wrong one will make the problem worse instead of better.

Brad, I cannot assume you know nothing.  I must, for example assume you have a working knowledge of English and can Google any unfamiliar terms.  What I can and do assume is that there will be times when the reader is left wondering just what the writer meant.  As often as not it is because the writer did not explain in enough detail.  One of the joys of writing on the internet is that there are no extra publishing costs for ink and paper if the writer wants to expand on what he is saying.  If he goes into a level of detail that starts boring some of the readers, well, nobody will complain if they speed read through those parts.  Another advantage of writing on the net is the ease of editing even after uploading whatever you have to say.  And along with that, the ease with which one can ask the writer to say a bit more.  So - if you need more, don't be afraid to ask.

Jim  

[BradKT]

Thanks guys.  You have given me some things to think about.  The rails are straight reading to the turnout...whether they are perfectly level and FLAT is something that I have to check out.  One of the worst offending switches is a single piece crossover section that is about 18" or so long.

I am NOT running long trains.  I gave that up.  This can happen with trains of 3 to 7 cars.

Jim Banner...a question.  If I run a small flat piece of glass (or clear plastic...that Micromark clear-bed flat car comes to mind) over the turnout, is it supposed to pass smoothly over EVERYTHING where the frog and turnout rails are located?

I also have noticed that on the Bachmann turnouts, the bottom of the groove (or channel) itself where the wheels pass through definitely rises.   I don't know if that is a problem areas or not.  I don't think so, but who knows.  The type of derailments are almost always of two types...(1) if the cars are running on a track from left to right, it is the right rear wheels that derail (most common) and if they are pulling another car, the front wheels of the second car then derail like they are being pulled off of the track; or (2) the rear wheels of the second engine in a two-engine consist derails.  If the switch is set to go straight, the wheels that derail always are taking the turnout as if someone has thrown the switch after the front wheels of the car passed the frog.

[jward]

since you mentioned that your cars are derailing at the FROG and not the POINTS, i have a different take on your problem. there should be NO WAY a properly guaged wheelset should EVER pick a frog.......

you mentioned that you used proto wheels. do these have a plastic axle by chance? i had a problem with warped axles on kadee wheelsets which had plastic axles. a warped axle will cause the wheels to wobble in and out of guage as they rotate. if you find wheelsets like that, replace them with something that has a more substantial axle such as jaybee, bachmann of nwsl wheelsets.....

assuming that your wheelsets are in guage, another place to look for problems is the guardrails. if these are spaced too far from their adjacent rail, they won't steer the wheel flanges away from the point, and that's the whole reason they are there.
checque their spaciing using the flangeway part of an nmra guage. if the flangeways are too wide, you can glue some thin styrene strip along the edge of the guard rail and shim it to the proper spacing. let the styrene dry overnight and you can trim it to the profile of the guard rail with a hobby knife. i did this with some older peco switches with wide flangeways at the guard rails, and it drastically improved their performance.

lastly, you'd mentioned that the bottoms of the flangeways seemed to rise at the frog. i don't think they are your issue, but there is a simple checque you can make on their depoth. take a spare truck, and roll it through the frog by hand. if you feel it bumping over the frog you have a problem. on a properly constructed frog, the wheels should roll smoothly. the wheels shouldn't drop into the flangeways as they pass through the frog, nor should they bump over anything.

[BradKT]

A couple of things to add.  I have had success with the crossover switch, but before I discuss that one, I want to tell you that I believe that my original hunch was right about using kadee or other spacers (I used one gray one and one red one) between the trucks and the body of my Con-cor 72' passenger cars.  The use of spacers does have the effect of keeping the trucks level and maximizing the contact between the wheels and the track.

I also added one ounce of weight to each end of the Con-cor passenger cars for a total of 2 ounces per car.

So long as the wheels remain in full contact with the track, they pass over the crossover switch with no problem.  I tried it dozens of times successfully.

I did some slight sanding and filing at the frog area and the combined effect seems to have fixed this particular switch.  While I am still having a problem with one Con-cor passenger car, I think that this is due to the truck itself and I am replacing it.

Does anyone know what would be a decent substitute for Con-cor passenger car trucks?  Their coupler is attached to the truck itself...as opposed to the body of the car.

Now with all of that said, I know that there are a few areas on my layout where the track is NOT absolutely level and this causes the trucks to sometimes slip slightly off the track, resulting in a guaranteed derailment at the switch.  It will be a pain in the butt, but I know that I have to fix that.

I haven't addressed the other switches, but I am fairly optimistic that it is a combination of all of this that will ultimately fix the problem.

[Ken G Price]

After I went and added the proper weight to each car the amount of derails dropped immensely. For me adding metal wheels are a big help for smooth operation. Then I could start to get at the switch and track problems. I have most of those fixed. 

As you have found there is a lot of engineering involved in getting the perfect layout. But the more you do it the more you learn, and the easier it gets.

[mf5117]

I found that on my #5 turnouts , switches I'm still bad with the terms . A few of them I perchased had positive camber at the frog ie" a hump and the 2" straights didn't match up correctly "difference in the rails" I changed them out got them off after a curve . Changed my layout up so many times . Turned them around from rights to left . I also found with the EZ track that coming into the #5's I couldn't have a run of 3" , 4.5" and a 2" to make up a diversion turnout ect so I had to lengthen my table and widen it so I could utilize long pieces 9" and 4.5" pieces .Although I thought I had made sure everything was flat it was just enough to cause me problems .

When running a string of rolling stock a mixture of bachmann ,athern cant remember the other cause I'm at the hospital my gf wrecked one of our Harley's . but anyway .I had some lighter cars in the string and noticed in some of the ways I coupled them on the string 3 box cars bachmann 40 ft a couple of athern flats then maybe a couple of coal cars hoppers . Its was like the cheaper rolling stock would come off the rails usually the front inside wheel sets it was like they were just held together by the couplers . I also found I had to run my sd40-2's in one direction due to they would derail running the other direction into the long part of the #5 9" then the frog .Its on this site as I asked so many questions .gauged couplers weighted rolling stock to nmra specs gauged the wheel sets you name it I did it . I finally took the #5's out of the mainline a bought a couple of the #6 crossovers and and added 3" straights to make up the difference on the inner and outter rad. I had to come up with same rolling stock strings and be Happy like Jim said with a couple of derailment every once in awhile if i ever do another HO layout it will be with flex track and code 100 track . for a complex layout I feel Ez track is not the way togo

[mabloodhound]

Sorry to hear about the Harley and hope your GF is OK.
You are learning all the tricks to create a good layout and before long you'll be on hear advising other newbies on how to fix something.
Keep plugging.

[BradKT]

for a complex layout I feel E-Z track is not the way to go.

I have to agree.  In a couple of places, if I can't fix those areas in the next couple of weeks, then I am ripping out those particular sections of E-Z track and will substitute Atlas track and cork roadbed.  Even though most of my track will be E-Z track, I just want this to work and if that is what it takes, that's just what it takes.  This is ridiculous.

[mf5117]

I got a good deal on a digital commander set and got the NS expantion set for Christmas as a gift .So I thought well EZ track would be the way to go .If I kept the simple oval with the one turnout then everything ran ok .But even said there was misalignment of curves and straights at the rail joiners and rails . I think I even said, were these pieces manufactured on late Friday or early Monday . That's how I came to have so much extra track and turnouts . I love all the 23 bachmann diesel locomotives I have and rolling stock . parden my Spanish, but the track sucks . I have less problems with my brass G scale track and switches . And I feel sometimes . Why did I go with the HO . Its alot of fun the modeling aspect everything that is available and the cost . But before I went head over heals I wish I would have done alittle more research and spent alittle more time learning the Hobby and instead of thinking I knew everything from the get go . But will work threw it as my gf has some grand kids and I can pass this down to them someday and teach them how to cuss "lol" had to say that . Anyways good luck great layout you have and Merry Christmas