I have read many posts from newbies trying to build their first layout with grades that are too aggressive. Add me to that group. So now I am redesigning after ripping up most of the track laid. My question is as follows-I plan 36"curves on with 2.1% grades and will be running 85’ passenger cars,is there a way to get an idea of the "true"grade? I mean that with the curve on the grade the grade is really more than 2.1% and I want make sure I can really run the passenger cars. I do not want to rip up tracks a second time.
Any curve, grade or not, will add to the pull needed. If you are trying to figure your grade on a curve, lay a piece of string along the track, measure the string and the height difference at the two ends of the string. Compute yur grade.
Suggestion, don’t completely secure your risers or your track (just a couple of track nails) and run your trains up and down to see how they do.
In the 60’s John Allen came up with a ‘rule-of-thumb’ that the effective grade should be 32/R greater than the actual grade. So your planned 36 inch radius cuve on a 2.1% grade is roughly equivalent to a 3% grade on straight track (2.1 + 0.89). About 2 years ago the folks at the Layout Design Special Interest Group forum on Yahoo had quite a discussion (over 200 posts) on how to use physics or experiments to develop a better rule. The take-away is that it would not be easy to come up with an exact solution and that 32/R should be a conservative estimate, especially with today’s more free-rolling trucks.
The best advice is to test your own equipment. This is especially critical if you are trying to run steam locomotives without a helper. Model locomotives vary widely in pulling power. Many model steamers are underpowered compared to their prototypes and many model diesels are overpowered. Folks who run diesels tend not to worry about the problem, they just add more units.
My railway ‘engineering’ handbook indicates that one should add .05% ‘compensation’ for each degree of curavture. A 36" radius curve is about a 22 degree curve. So, add 22X.05 - about 1.1 percent compensation for just the curve. Of course, this is based on steel wheels on steel rail. Our ‘softer’ n/s wheels/rail may add even more drag. In my case, I have a 2.7% grade with a 22" radius(40 degrees) curve at the top of the grade - that works out to about 4.7% with compensation - no wonder I need a second unit on my 15 car trains!
thanks for the info, very helpful. when I see pictures of layouts, especially of a helix, it seems to be a very aggressive grade. am I correct in assuming a helix must be a very low grade to work? pics I see seem to look otherwise. again don’t make fun of me I’m a newbie
Grade in a helix depends on radius and height between turns. For a well-designed 36" radius helix in HO with 4" railhead-to-railhead elevation from turn-to-turn, the grade of the helix itself is less than 2% (around 1.75%), Adding in the effect of the curve increases the effective grade to around 2.7% (if one uses the 32/R first approximation). This would be generally workable for most, depending on era, equipment, train length, etc. Using thinner roadbed material, one could reduce this grade a bit.
Prototype railroad friction estimates are not directly applicable to the model because the physics and forces involved are completely different for our lightweight equipment and tight radii.
Many hobby newcomers plan helixes that are too tight in radius and thus too steep. In addition, there is a danger of “stringlining” longer trains across the center of these tight helixes due to the build-up of oblique forces caused by the interaction of grade, friction, and the drawbar pull of the locomotives.
So generally, yes, broader radii (and thus, lower grades) work better in a helix.
Some folks design their layouts hoping that a model railroad helix will suspend the laws of physics. Sadly, this is not the case. Note also that some commercially available built-up helixes are likely too tight in radius to work with longer trains – and sometimes this has been compounded through mis-labeling grades. Mathematica Emptor (buyer, do the math)