DERAILING, GARGRAVES TRACK

It is hard to determine from your photos. Could be the track, could be the engine is binding and trucks are not swiveling or maybe you are clipping the scenery.

Is it 072 or greater or less than 072? Can you see the engine derail? Have you tried other engines?

I have the same question is Don.  Also in the third photo the track appears kinked.  The large engines like the rails flat with wide smooth curves.

Hard to tell from the pictures.  I would suggest you run the offending locomotives through the location as slowly as possible while you are there watching what happens.  If you are lucky you will get to see what is taking place and can figure out if it is trackwork or the locomotive.  

The one curve on the right side of the first picture looks tighter than O72. Also, have you leveled across the track, parallel to the ties? The track would be best to be level or slightly downward sloping toward the inside of the curve, but should never slope toward the outside of the curve. You might also inspect all track joints to see if any rails opened up at the bottom and pinch them closed.

George

Jerry, is the track Gargraves flex trak that was bent to 072, sectional curved track, or a combination of flex and sectional curved?

If it is flex track, how did you bend it and lay it down?

A suggestion:  buy a piece of Gargraves 072 sectional track; then, flip it upside down and run it over the flex track you've laid.  Doing so will give you a good idea of any problem areas.

Good luck.

At first glance I thought this thread stated "detailing Gargraves track"

Are you having problems with the Gargraves track itself or is it at the switches? I have a couple of MTH engines that don't like Gargraves switches so I went to Ross Custom switches and no more derailing at the switch. Another thing to look at is the area where the Gargraves track and tubular track meeting smooth? If not that can cause derailments as well.

Lee Fritz

Dollars to Donuts, it's that kinked area where the curve becomes almost straight.  Try reforming the curve to eliminate the straightened section.  Hopefully the deviation from a true circle isn't there for a reason. It's not surprising that a Premier Triplex requires nothing but the best track work, given the complexity of 3 interconnected drivetrains.

Bruce

A slight kink can cause lead wheels to climb out and derail.  We built a bending board jig setup to get smooth curves, and in tricky places lay a template we have for curvatures alongside when fixing to roadbed.  On the few 072 curves we cheat them into about 074 by swing wide around the curve.  We have one spot the huge engines will not track, regardless of how often we have tried to fix it, relay it, re-bend it,... finally just put up a sign "No Class 1 beyond this point" ... they are restricted to the 080+ trackage, of which most of it is thankfully.  Also, a couple of locomotives, like the UP Veranda turbine from MTH have a binding problem with the trucks, which will not swing to a full 072.

 

Looking at the first photo, take a GOOD look at the shims under the track.  What are you using for shims?     It looks like you have varied thickness shims spaced too far apart.  Do yourself a favor, take all the shims out and use a good cork roadbed under the track.  Do NOT screw the track down yet.  Get a good 3 ft. metal level and check the track for 'level' the entire length.  The Triplex is a heavy engine and needs a level roadbbed.  Go to Hobby Lobby, Michael's, or a craft store and  purchase a package of Jumbo Craft Sticks to use as shims.  Shim as necessary but do not leave large gaps.

Once level, use #4 cabinet screws to secure the track and roadbed.  Remember to drill thru the wooden ties or they will split.  Those screws are double thread and work fine for this job, they are also black.

bruce

Sounds like an NMRA gauging tool might be handy here too. Making one isn't to hard either. This kinda stuff bites the small gauges running steam with flex a lot. If it went under gauge while bending, it doesn't stand a chance. Being slightly over would be better. Being at the locos minimum dia. isn't helping either.

Using Kerrigan's experience, I think you have two choices, increase the curve size with Gargraves or use a flat top rail track, like Atlas or ScaleTrax.

I get approximately 19" from the end of the outside rail to the end of the inside outer rail on a section of GG O72. Measure from the first driver flange to the rear of the 8th driver flange on the opposite side. I think you'll find it's longer than 19". The last set under the tender articulates, so those under the main boiler are the concern. The O80 GG curve is almost 21".

I would guess that the first 8 drivers are binding and pushing over the round rail.

If you only have space for O72, use a part of a circle of MTH RealTrax for a test. It has a flat top rail and is inexpensive or a half circle of ScaleTrax. The edge of the flat rail may stop the runover from the binding.

That's one long engine!

It looks like the track will flex vertically when a heavy steam engine passes the large space between the shims.

How thick is the plywood you're using to support the track in picture #1 it looks like the ply might even flex

when you first enter the 0-72 curve 

Tom N

 

The skirt between the engine and tender on a Triplex must be flipped up or it will derail on curves.

Superelevation - which I have on my layout, and did it 20+ years ago - can be a real problem if not done carefully. The shim/space/shim arrangement does indeed look suspect, and you are apparently trying to run a truly enormous and heavy loco (Triplex) over it. 

Cumulative tolerances issues - in the equipment and the track/roadbed - by definition, add up. Stabilize (and reduce?) that superelevation. I used long pieces of scrap vinyl flooring - cut in curves and stacked where needed on the outside of the curve/ties - to do my superelevation years ago. It supports the entire curve - no gaps. Someone mentioned a wire; same result, I suppose. The vinyl is tough and easy to work; disappears under the ballast.

rattler21 posted:

My first thought is if the shims are for super elevation, you may consider removing them and place continuous wire beneath the ties under the outer rail.  Second, if the third picture shows hand bent rail with a slight S in it, articulated engines may not perform properly in that area.  You are trying to force it to do something beyond its design capabilities.  You may run into the same problem with jury rigged cross overs without a short piece of straight track between the track switches.   John in Lansing, ILL

The method we used to super-elevate the curves, a small wire under the outside ties.  Works great!

Thanks again guys, preciate your help.

Jerry

I just recently ran through similar issues with my layout.  The problem with Gargraves track is even the standard pieces are a bit flexible.  Flexible enough to cause irregular geometry where curve sections meet.  I have an 054 loop on my layout.  When I first built it, I had numerous issues with quite a number of engines and even passenger cars derailing.  The worst of it was around the Ross Switches I was using.  Steve from Ross actually sent me a replacement for one of the switches which turned out not to be the problem.  I use RR Track and there is a measuring tool in it.  I used the measuring tool to discover that my half circles were just a bit egg shaped.  The entire loop was about 3/8" off spec (too wide) and that was enough to cause the issue.   Once that was fixed, I dropped my MTH Reading T-1 on the rails (it requires 054) and found that some of the curve to curve joints were just slightly out causing the pilot truck to hop a bit.  I am talking 1/8" tolerances.  None of this was visible to my eye.

While the super elevation could be your issue, I would double check the track joints too.  From the pics above, it appears that at least one joint is kinked (first photo to the right by the last vertical piece of homosote).

You may want to consider Ross Roadbed.  I plan to add it to my layout in the future.  The way it's constructed, it will eliminate any issues with kinked joints.

Tony

I have 2 Reading T-1's from MTH and can run them both on 042 Gargraves track & 042 Ross Switches. My MTH Reading T-1's don't like Gargraves switches. Maybe your curve sections have an unsmooth part to them or are bent or flex track has been bent too many times. I have had derailments with some of my engines and had to replace a section of Gargraves track because the rail got bent.

Lee Fritz

phillyreading posted:

I have 2 Reading T-1's from MTH and can run them both on 042 Gargraves track & 042 Ross Switches. My MTH Reading T-1's don't like Gargraves switches. Maybe your curve sections have an unsmooth part to them or are bent or flex track has been bent too many times. I have had derailments with some of my engines and had to replace a section of Gargraves track because the rail got bent.

Lee Fritz

Interesting.  I didn't try them on 042 because they were spec'd for 054.  My '91 LTI T-1 runs on 042.  My layout has no flex track.  The only 2 switches on this loop are Ross 054/072 switches.  The initial problem was the approach to the switch.  The derailments happened when the train went through the 054 turnout.  The 072 through was fine.  Because the switches are located in an 054 radius turn, the angles didn't look wrong but the measurement told a different story.  The Lionel T-1 didn't have an issue with any of the loop once the switch entries were corrected but the MTH T-1's front truck would hop and sometimes derail on several curve joints.  A bit of tweaking has everything running nice.

Tony

Great replies, best, Jerry 

Tony, my MTH T-1's are MTH # 20-3130-1 & 20-3250-1, cab # 2124(Reading Lines) & 2102(Blue Mountain & Reading). The MTH boxes mention using 042  curves. My MTH T-1 cab # 2124 likes to throw it's traction tires a lot. Both are full PS-2 equipped.

Lee Fritz

Two big things you can do for Gargraves track are painting "rust" on the rail sides, which reduces the appearance of the height of the rails, and ballasting. The Ballast pile, since it's capped at the tops of the ties, completely conceals the thickness. What we did initially was use playground sand as a base fill, then ballast on top of that as the ballast pile was kinda thick. Now, we use #16 "silver" construction sand straight out of the bag (100 pounds for about $10-12). It's easily painted if one chooses.

AGHRMatt posted:

Two big things you can do for Gargraves track are painting "rust" on the rail sides, which reduces the appearance of the height of the rails, and ballasting...

Matt, it's "derailing", not "detailing". I made the same mistake(as did others)!

ADCX Rob posted:

AGHRMatt posted:

Two big things you can do for Gargraves track are painting "rust" on the rail sides, which reduces the appearance of the height of the rails, and ballasting...

Matt, it's "derailing", not "detailing". I made the same mistake(as did others)!

Yes, but good information anyway.

I started my 3-rail layout in the 80s when the only realistic track was Gargraves. The switches were powered from below with NJ and Toshiba machines and a rotating linkage I designed.  Yes, derailing was a problem, especially in reverse. This was rectified by using a Dremel cone grinder to slightly open a cavity in the rail where the switch rail made contact, just enough to make the switch rail flush, like Lionel tubular switches.  In some cases slight frog adjustments were made. These early Gargraves also had short guard rails.  I replaced them with longer guard rails of Gargraves rail itself, inserted and glued into tie slots cut with the Dremel.  I also used this technique for guard rails on bridges.  This made for more realism and smooth operation.  Yeah it's a little work but worth the effort.  My layout, by the way, was donated and is now situated in a restored baggage car at the Strasburg VA Museum, and was featured in the July 2015 issue of Classic Toy Trains.

Thanksfor your help. 

Jerry