I haven’t been on the forum too long, but I did notice a fair number of questions about the impact of grades on locomotive hauling capacity. Is everyone familiar with this information?
Also, this allows the influence of curved track radius to be accounted for:
Have there been any more recent studies or data developed?
Marty
Yeah, and anyone have a similar chart for N?
From what I know, no new studies have been done. I don’t know how much actual experimentation went into John Allen’s grade compensation for curved track - he wasn’t know for using large radius curves. Also, these figures were developed before modern free rolling trucks and wheel profiles were developed. I suspect that the impact of curves is less today down to the radius where the wheels must slide instead of roll to accommodate the different distances each wheel travels. At that radius (somewhere in the neighborhood of 20" for HO, I believe) I would expect a sharp increase in friction (increase in grade compensation factor) as radius decreases.
I suspect, but have no evidence, that the grade compensation factor curve is as valid for N as for HO in the '60s if one substitutes N equivalent radii. This is because rolling friction does not scale well. As the scale decreases rolling friction decreases only slightly, if at all. The rolling friction of an N car is nearly the same as an HO car. So using the older HO friction values for N would seem more appropriate.
The same is true of the grade effects on train length. On the prototype, train weight is almost the only factor in determining the effects of grades. The difference between the rolling friction of 20 loaded cars compared to an equivalent weight train of empties is not nearly as much as on a model train. The more free rolling the trucks, the closer we come to the prototype impact of grades curves. Since rolling friction does not vary with grades, and rolling friction is a greater proportion of the total load in our models, the impact of grades on our models is actually less than on the prototype.
I’d love to quantify some of the above, but it would take lots of controlled experimentation. Most folks simply test their individual situation to see if performance is satisfactory. If performance isn’t as desire