Conventional Lionel slowing on curve

If it's tubular track, check for tight gauge at the problem area. Sometimes the rails get bent inwards from handling or being stepped on. Pic shows a simple way to make a serviceable track gauge with cardboard and paper punch. It will help spot any major deviations.

What do you mean by 32R? If you mean 32-inch-radius, that would be O-64, which is a pretty wide curve.

It's my understanding that conventional will slow down to some degree on any curve because of slight binding and lack of electronic cruise control to maintain speed. If it's only a slight reduction in speed it would be considered normal.

Ace posted:

If it's tubular track, check for tight gauge at the problem area. Sometimes the rails get bent inwards from handling or being stepped on. Pic shows a simple way to make a serviceable track gauge with cardboard and paper punch. It will help spot any major deviations.

 

Track is brand new gargrave sectional curve but I can check that. Thanks.

Matt Kirsch posted:

What do you mean by 32R? If you mean 32-inch-radius, that would be O-64, which is a pretty wide curve.

It's my understanding that conventional will slow down to some degree on any curve because of slight binding and lack of electronic cruise control to maintain speed. If it's only a slight reduction in speed it would be considered normal.

Sorry, my bad the radius would be 16".

As with the real RR's, a (non-superelevated) curve is the same as a grade - flange and wheel friction add to the apparent weight of the train, so it can slow down. Just good, garden-variety physics.

Variables that make this more - or less - apparent include loco gearing; most PW locos were geared fairly poorly (too "high"), so it doesn't take much to inhibit some of them.

D500 posted:

As with the real RR's, a (non-superelevated) curve is the same as a grade - flange and wheel friction add to the apparent weight of the train, so it can slow down. Just good, garden-variety physics.

Variables that make this more - or less - apparent include loco gearing; most PW locos were geared fairly poorly (too "high"), so it doesn't take much to inhibit some of them.

Good point. Wonder if I just shim that curve a little if it would help. Give it a try.

radar493 posted:

I have a 2018 that slows up significantly on a 32R curve. My 2037 same thing so basically two identical engines. Power feed exactly mid point of curve. Is it just the tightness of the curve? Or is it something else? My meter shows pretty even voltage around the layout.

Does the binding occur on all your GarGraves 32 curves, or  just in certain places? I run similar 2037 and 2026 locos, which are made to run on Lionel O31 curves. They shouldn't have significant binding on O31 curves if wheels and track have correct gauge. Check that the curves have a uniform radius throughout with no kinking at the rail joints. You might also check for too-wide wheel gauge on the locos. 

I have no direct experience with GarGraves track. Does GarGraves rail have a profile substantially different from tubular rail, which might create more drag for some wheels on sharper curves?

Ace posted:

radar493 posted:

I have a 2018 that slows up significantly on a 32R curve. My 2037 same thing so basically two identical engines. Power feed exactly mid point of curve. Is it just the tightness of the curve? Or is it something else? My meter shows pretty even voltage around the layout.

Does the binding occur on all your GarGraves 32 curves, or  just in certain places? I run similar 2037 and 2026 locos, which are made to run on Lionel O31 curves. They shouldn't have significant binding on O31 curves if wheels and track have correct gauge. Check that the curves have a uniform radius throughout with no kinking at the rail joints. You might also check for too-wide wheel gauge on the locos. 

I have no direct experience with GarGraves track. Does GarGraves rail have a profile substantially different from tubular rail, which might create more drag for some wheels on sharper curves?

This is the only 032 on the layout. It's gargrave sectional curve and is new. Not sure about it's profile vs tubular as it's the only track I've used in O gauge. I would think new sectional curves would be in guage. It definitely does slow up in the middle of the turn. The fact that the two engines are the same (2037/2018) it would seem to mean it's not the engines but track. 

Drive wheels are pressed on the axles. This means that one wheel has to slide going around a curve. This takes energy and slows the loco down. Another issue is that post war wheels are designed to run on post war track.  The wheel set gauge is measured at the intersection of the tread and the flange. But the flange has a 25 degree angle on it tapering inboard away from the rail.  On tubular rail the narrow point is in the middle of the tube shape, which is well below the top surface of the rail. This taper on the flange and tubular rail shape gives more clearance between the flange and the rail than will be available on rails with different shape heads.  

To reduce this effect, you could insulate the curves from the straight sections and put a resistor or diodes in series with power feed to the straight sections. This would give you a little voltage boost on the curves to help overcome the curve resistance. 

David Johnston posted:

... post war wheels are designed to run on post war track.  The wheel set gauge is measured at the intersection of the tread and the flange. But the flange has a 25 degree angle on it tapering inboard away from the rail.  On tubular rail the narrow point is in the middle of the tube shape, which is well below the top surface of the rail. This taper on the flange and tubular rail shape gives more clearance between the flange and the rail than will be available on rails with different shape heads...

Excellent point.  This raises the question whether sharp-radius GarGraves track is not as well suited for operation of older 3-rail steamers with six or more drivers, because of the different (more prototypical) rail profile. The steam locos have a longer fixed wheelbase which makes it more of an issue.

The original poster should still check his 2018 and 2037 locos to make sure the drivers aren't gauged too wide. Try pushing it through the curve and see if the wheels have some side-play in the track, not riding up on the flange faces.

Given the 2018 and 2037 were both sold as O-27 locomotives I'd think they can handle O27 and certainly O31 curves without binding. Given that it's both while I wouldn't rule out driver gauging or even sufficient thrust moment binding the side rods, but it seems the common mode is with the track. As has been suggested I start there.

Bogie

Wow this is interesting. If I run one of my relatively late model command engines conventionally with 6 drivers given the same voltage and it maintains it's speed through the curve would this establish that it's the difference of postwar vs gargrave track, correct?

if it is lionel 6 dash number  might help.

in any case older engines without some kind of speed control will slow down on curves because inside wheels turn at a different rate then outside wheels.

if you have ever seen an 18 wheel truck or a camper trailer with dual wheels make a sharp turn, those rear tires go through changes that one would think they might fall off the rim.

Flange bind. A big problem with toy trains, park trains and real trains. The tighter the curve the worse it is. When I ran the trains at the local park railroad (24" gauge") we greased the outside rail to aide in wheel/rail wear, flange bind and noise. Made a big difference when greased and not greased. Granted our curves weren't THAT bad but we had a few very sharp curves. 

I was also given a chance to operate a PCC trolley at the National Capital Trolley Museum. They have a 90* curve that nearly brought that PCC at 30MPH to 0! 

I wouldn't worry about any problems. That's physics! 

radar493 posted:

Wow this is interesting. If I run one of my relatively late model command engines conventionally with 6 drivers given the same voltage and it maintains it's speed through the curve would this establish that it's the difference of postwar vs gargrave track, correct?

The track might seem suspect since two similar locos have the same issue. But you need to verify that track gauge and wheel gauge are all normal, with no binding while going through the curve. Push each loco through the curve and feel for some side-play by turning the loco diagonally both ways, at different places in the curve. The flange faces should not be riding up on the inner face of the rail. Also make sure there is nothing else binding or dragging, like pilot or trailing wheels or tender drawbar or center-rail pickups hanging up. The curve should have a uniform radius with no kinking.

Then I would be looking at whether the flanges have more "scrubbing" on the rail in the curve, due to the combination of wheel and flange profiles. Are there any particularly rough finishes on the flange faces or corresponding rail surfaces? If so, they can be smoothed with very fine sandpaper.

On one old loco I found that side-play in the driver axles caused the wheels to drag on curves in one direction, because the wheel gears would contact the frame.

You just have to look closely and methodically at everything to figure out what is dragging or binding. If you are still stuck, show us some photos of the locos and curves.

Thanks guys. I will check out all these possibilities.