hi everyone started running mainline for my northeastern coal train which amounts to two K-4’s pulling and one pushing the rolling stock and would like to get up another 6" to run the sides of the mountains for the return run but wanted a idea from everyone some grade% that everyone has used that is trouble free that would get up 6"and back down say with in 8’-10’ and what is the sharpest trouble free radius a person can get away with and still run a variety of cars
Nothing is trouble free in the hobby. 8’ to 10’ means around 5% grade or steeper. That is Shay country, not K-4 Country.
Having a helper and a lead engine on a heavy train that is travelling up or down in combination with curves is a real issue depending on a number of factors.
Assuming twin bay 50 ton hoppers you probably can get down to about 22" with the K4’s without too much trouble. Just try to avoid starting and stopping the grade on a curve and also try to give room to change from level to grade at both the bottom and the top at least one engine length long.
You will experience a limit to the number of rolling stock up the grade with or without helpers. You may need to double or even triple the hill.
To add to what Safety Valve said, keep in mind that including a curve in a grade increases the effective grade by the ratio: 32/Curve Radius.
For a 24” curve, that ratio is 1.3. If the tangent grade is 1.3%, then the eqivalent grade with a 24” curve is 2.6%. I’ve not faced it, but I’d imagine this can play havoc with steep grades and sharp curves.
Two show stoppers. One is the type of rolling stock and how well they can take being accordioned by two locos, one shoving and one pulling. My guess is that you will have problems from the get-go, especially on steep grades (greater than 3%), and doubly so if you add in curves tighter than about 30".
Secondly, unless you can find a way to synchronize your engines perfectly so that they pull and push in near-perfect concert, you will be pulling and compressing the couplers on models that are already going to have enough issues on the lesser demanding trackage.
Good luck.
Once again, we need to know what scale you’re using. Minimum radius is very dependant on the scale. A 15" radius curve is fine for N-scale; but tight in HO.
If you don’t mind a toy-like appearance, you can buy an Over-And-Under Trestle train set and put it on a 4x8 table, then run a train half as long as the track. I did this as a kid. (HO scale) It was fun to play with until some of the cars started falling off the elevated track. I would not recommend this for any rolling stock or locomotives you prize.
In real life, Madison Hill (Madison IN) had a 6+% grade in a straight line and required specially geared diesels, speed limits, special operating rules (engine must always be on downhill side of train), &c to run. I think this would make a fun branch line in a bigger layout, but I would not want this on my mainline, especially in an infinite loop.
Hello all, I have a question (surprise, surprise…)
I wasn’t aware of a formula for determining total grade. This is cool, and scary too… Are the curves really that harsh on the grade? I would expect you would multiply your grade by the result of the equation, not add them together; 1.33 x 1.3 = 1.73% total grade effect. A net increase of 33% on the grade from the curves. If done the way you show it, even a 32" radius (Quite a broad curve.) will add 1% to the total grade, that seems like a lot. Obviously, if my method were the way it were applied, then curves of > 32" radius wouldn’t effect the grade at all. In other words, the drag applied by a curve of at least a 32" radius would be no greater than that of tangent track. I would further observe that this is scale sensitive, these numbers are for HO only and the radius factor is different for other scales. Am I right or am I crazy? (That’s kind of a rhetorical question.)
Inquiring minds want to know…
I believe that it is a multiplier. When I was constructing my layout (small steam, mostly on grade); I found (thru experiments) that a 18" radius curve was equal to my 2% grade. By having a 18" curve on a 2% grade, I had an effective 4% grade. I wish that I had known of this formula then, could have saved some time and effort.
Hi Alan,
I guess the difference isn’t that noticeable when you’re calculating for a tight radius. If you’re adding the factor:
32/18 = 1.78 + 2.0% = 3.78% effective grade
-vs-
32/18 = 1.78 x 2.0% = 3.56% effective grade if you’re multiplying them.
But, if you’re figuring for broader curves then… adding:
32/30 = 1.07 + 2.0% = 3.07% final grade
-vs-
32/30 = 1.07 x 2.0% = 2.14% final grade.
That seems like a difference we can all sink our teeth into, yes?
If it’s a multiplier then a very broad curve would DECREASE the grade and that is counter intuitive.
32/48 = .67 * 1.78% = 1.19% That just doesn’t make sense.
Taking the limit, a tangent on a grade would be 32/infiinty = 0 0 * anything = no grade at all.
Addition seems more likely unless the formula is an approximation valid for a certain range of curvatures.
“Curiouser and curiouser”, said Alice
Is this 32/radius just for HO? It seems excessive in N scale…
I’ve just returned and seen the responses to my post ragarding the impact of curves on the effective gradient.
You all have raised good questions–the bad news is, I’m not an engineer and can’t answer them!
I got the information originally from an MR Information Station document posted November 2005, page 105. The source noted was John Allen, and the formula was presented (or I interpreted it; I’m not sure which) as a multiplier. I don’t have any more info than that–sorry. As to the scale, as Allen worked in HO, I assume (!) it applies to HO.
Rick Krall
I’m sure that someone in this very well read group has a copy of that document. Let’s hope that they see this post and check it out for us. [:)]
Karl
As you say, it makes no sense to reduce the grade when using a larger curvature. Like I stated in my earlier post:
My gut tells me that this doesn’t make sense either though. While a 32" radius curve (in HO) is a decently broad curve, it’s still a curve and should be adding some effective drag to your train.
But if this formula is additive, then IMO this factor becomes very important when designing your pike. I just can’t believe that a curve of 32" R. will add 1.0% to your grade. That’s serious!!! And furthermore, if this is the case, WHY WASN’T I TOLD???!!!
I’ve been reading MRR for years, surfing the web, and (of course) I own John Armstrong’s Track Planning For Realistic Operation. Well, I’m in awe that this is the first I’m hearing of how dramatically curves effect your grade. When planning our layouts (Like I’m doing now.) we’re told to keep our grades to < 3.0% with a maxium grade of 2.0-2.5% being preferred for reliable operation with decently long trains. (Unless we’re planni
Well, Grades are grades. Throw in a curve and they will get steeper.
I can take a 18 wheeler up a straight Mountain grade all day and have it stable from the bottom to the top. But using Babcock Ridge as an example near Altoona PA there is a two lane road that has a hairpin about 1/3 of the way up and when you come out of that you have effectively downshifted two gears and buried the torque on the engine lugging the load into a ever-increasing mid grade for the fight to the summit.
Raton Pass at the Colorado Border is absolutely engine breakers if there is insufficient power, traction and sand. Sand is king on that grade. You can stack all the engines until the durn thing moves but once it hits that Raton with all of the curves your power requirements increase even more.
Broad curves of 24" to 32 or more are always GOOD in a plan. But even a 20" curve on a 2% grade is not too bad. It’s when we put a 5% grade on a 18" curve and expect to run 20 cars and a 2-6-0 up it. Aint going to happen.
Model grades that bad makes it helper country. We all have several engines on hand. It would not be too difficult under today’s DCC to assign a helper and crew to assist heavy trains over difficult grades. Even a local can double a grade if necessary. That is where good sidings come in. Building good sidings into a plan on both ends of a grade gives the operator a fighting chance to deal with a grade in less time than it might take to notify the dispatcher and have a helper sent out.
Regarding the “Who Stole my Cheese” feeling when no one talks about effects of grade on a curve in Model Railroading to potential people considering joining the hobby; I think it’s hard to describe such a engineering intensive part of railroading without burying the hapless and now sleepy reader in tech babble and math forumlas. There are some who have written very good articles such as the old ABC’s of Operations from years ago (They should re run that) which had a excellent
Try these out … this question comes up all of the time on the forum …
I think trial and error was easier for me than the math. I am building a helix with 30’'radius and will try 3%. I was going to clamp the subroadbed to the risers and test it out. I will be running 4 sd40-2 with 20 cars up the grade. I tested it on a straight piece of track at 6% and it worked. If 3% does’nt work ill try 2.5%. Its not that hard to readjust. It seems sometimes its to hard to find the formula and make it work when you need it. Joe A.
One of the sad facts of circular helii is that the radius, grade, and separation between the tracks are locked together. You cannot change the grade of a helix of fixed radius without adversely impacting the separation between tracks. In most cases, the separation between track levels is a pre-determined minimum (pre-determined by construction method and vertical clearance of trains), and the radius and grade select each other.
In HO, 4" is a typical minimum vertical separation (6" allows most people to get a hand in to rerail rolling stock but produces too steep a grade). A 30" radius gives 188" of run or 2.1% uncompensated grade. Add another 1% of compensation for the curves (I remember the Allen formula being additive, but am willing to be proved wrong) and your helix is already at 3.1% effective. A smaller radius just makes things that much worse.
I don’t know how valid the Allen formula for grade compensation is. It was derived before the RP25 wheels and modern free-rolling trucks were common-place. And I have no information about how much testing went into the formula. According to my calcs with RP25 wheels, wheel “slippage” (the wheel slides instead of rolling) to compensate for the extra distance the outer wheel travels is likely a significant friction factor at smaller radii.
My calcs:
Difference in length of outer rai
Fred, I may be out to lunch on this one, but I have a strong sense that my steamers don’t do as well with traction on a curve…if there is superelevation involved and it is on a grade. As for level track, I think that traction should be fixed, just more friction as axles on free-playing locos designed for short radii get pressed tight to one side, or when flanges get pressed the same way. And for sure, if one doesn’t do justice to grade transition, 4-8-4 and larger locos really pay the price.
Superelevation works when you hammer thru a curve at 100 mph or similar.
A local at 8 mph likely wont benefit from it. In fact I would avoid superelevation. I dont mind a little bit made of several layers of regular masking tape under the center outer rail of a curve under the ballast. About two or three helps without making it obvious.
I was on Atlanta’s Speedway once some time ago, just to stand on that super-elevated section of track was nearly impossible. But I bet at 170 mph those cars will be “Nailed down” without fear of losing traction and over the side.