Vertical curve radius?

[aleopardstail]

horizontal curve radius are I think reasonably well understood, e.g. 15" = doable but don't, 18" = doable but may look daft, 21"-24" looks ok etc

pondering vertical curves, e.g. the rough radius when going from level track to a slope, basically if say going from level to a 1:50 gradient roughly how long do you leave it before you are at 1:50 as the slope gradually curves to that level?

guessing this is partly down to the length of rollingstock wheelbases (too short and you will get derailments as wheels lift)

any thoughts?

planning 1:53 and 1:86 gradients with maybe 6" to transition? and 12" for a "hump" from a track going up at 1:50 to descending at 1:86

wondering if that sounds roughly reasonable before I start making cutting diagrams for wood?

[aleopardstail]

Done a bit more reading on this, this has suggested allowing roughly 10" per 2% gradient change, I have eased slightly to 12" as I think its tidier

http://www.aleopardstail.com/vertical%20curves.png

roughly illustrated thus (MS Paint for the win!)

showing a 2% (1:50) transition from level and a "hump" (could be inverted to be a dip)

the net effect is that if you know you have say 12' of length for your gradient, when working it out deduct 6" at each end, then divide the desired height change by that distance to get the gradient allowing for the ends

apparently someone (someone other than me) did the mathematics behind this and this should mean that rigidly fixed wheels with no flex don't see the flanges go above rail height, noted that exceptionally long rigid wheelbases may need longer and shorter (e.g. diesel bogies) can handle a shorter transition.

in crude terms that can be managed also by having say a 12" length of track without track pins and letting it find its natural curve

[Dad-1]

The drag from a curve can easily cause wheel slip.
A curve on a climb can become impossible depending on the locomotive
and stock you want to lift up the gradient. I recall a club layout where
3 coaches was impossible, 2 with a good traction loco would just do.

One guy I knew fitted a magnetic under track 'aid' from someone like
DCC concepts - can't recall exactly who it was, but he moved from Dorset
to Scotland before it was tested.

Geoff T.

[aleopardstail]

The drag from a curve can easily cause wheel slip.
A curve on a climb can become impossible depending on the locomotive
and stock you want to lift up the gradient. I recall a club layout where
3 coaches was impossible, 2 with a good traction loco would just do.

One guy I knew fitted a magnetic under track 'aid' from someone like
DCC concepts - can't recall exactly who it was, but he moved from Dorset
to Scotland before it was tested.

Geoff T.

this is something of a concern here, I will have curves on a gradient, 1:50 with 24" minimum. only testing will really tell how much of a problem it is - not planning massive trains though in the past have seen that drag effect on level track cause some problems (e.g. an older tender drive hornby A1 that could just about get round an 18" radius with eight coaches, but only with a good run up and at full speed - crawling by the end of it.

thinking here more about the vertical curve element than the horizontal one, can have a similar effect though of reducing the number of driven axles contacting the rails

[centenary]

The drag from a curve can easily cause wheel slip.
A curve on a climb can become impossible depending on the locomotive
and stock you want to lift up the gradient. I recall a club layout where
3 coaches was impossible, 2 with a good traction loco would just do.

One guy I knew fitted a magnetic under track 'aid' from someone like
DCC concepts - can't recall exactly who it was, but he moved from Dorset
to Scotland before it was tested.

Geoff T.

Charlie at Chadwick used the DCC Concepts magnet system for some of his locos to get up his helixes which arent steep. He concluded it was worth it for those few locos that needed help but found in a bit of hassle fitting the magnets to the locos. Obviously the plates had to be planned and laid before the track.

[aleopardstail]

Have seen that video, very well done and explained. trains here will be shorter, and also the curve is generally 80-100 degrees with straight sections between so ideally never have all the train on a curved incline as you would with a helix.

have had enough in the past with derailments due to bow string effect, never mind lack of traction.

did have a garage plan (sadly garage not useful without many £££ of work) that had a sort of helix, though going all the way around the room only one turn to drop to a reverse loop with the fiddle yard location actually on the "helix". worked out there was about twice as much hidden track proposed as scenic track

plan here is build the lower level, this has some 1:50 bits at the start of an incline through a junction with a 1:80 odd drop back to the datum again. its enough to test it out. moving to a 5" vertical spacing (3" for trains then 2" for servos etc) if stuff struggles will re-design the upper deck to ease it but for now this makes the upper deck a lot easier to design

[Bigmet]

Done a bit more reading on this, this has suggested allowing roughly 10" per 2% gradient change, I have eased slightly to 12" as I think its tidier...

Allow as much as possible, I'll not have anything more than 1% change per foot. (I have ten coupled rigid chassis, long bogie frame diesels, and carriage stock with close coupling mechanisms to contend with; and then there is the trend to reduced flange depth in RTR OO as well, increasingly finding 0.6mm flange depth.)

...in crude terms that can be managed also by having say a 12" length of track without track pins and letting it find its natural curve.

Inadvisable as even a modest 500g loco will flatten the track for a good 6" and then face a vertical 'kink'. Creating the natural vertical curve with a thin ply strip so the track is fully supported throughout is way superior.

[aleopardstail]

Done a bit more reading on this, this has suggested allowing roughly 10" per 2% gradient change, I have eased slightly to 12" as I think its tidier...

Allow as much as possible, I'll not have anything more than 1% change per foot. (I have ten coupled rigid chassis, long bogie frame diesels, and carriage stock with close coupling mechanisms to contend with; and then there is the trend to reduced flange depth in RTR OO as well, increasingly finding 0.6mm flange depth.)

...in crude terms that can be managed also by having say a 12" length of track without track pins and letting it find its natural curve.

Inadvisable as even a modest 500g loco will flatten the track for a good 6" and then face a vertical 'kink'. Creating the natural vertical curve with a thin ply strip so the track is fully supported throughout is way superior.

sounds good to me, there is space to easy it out a bit, its technically something like 1:52 currently so a bit of easing should work, means the main gradient is a tad steeper but easy enough to arrange