Gradient profiles of Heritage Railways - List in Post 1

I suspect that the gradient profiles of most ex BR lines are those from the date they were built. The exceptions there will be the likes of Swanage and the GWSR where the track was originally lifted on closure. As for the NYMR, Ploughman will be able to give chapter and verse as he has surveyed relayings in recent times. I understand that the steepest is 1 in 41 at Greenend but this is only a short stretch. There is a length of 15 chains around Darnholme at 1 in 42/43 which is also on a 13-15 chain curve. The length means that you have the whole train on it for a reasonable period.

 

I suspect the gradient profiles of the lines which were lifted are still the ones from when they were built, with exception of the likes of the Bluebell where one part has definitely been re-profiled

 

A few years back there was a short section at Darnholm of about 3 lengths that was at 1 in 18 now blended into the ruling grade either side.

 

They will be generally because the base formation is unlikely to have changed but tipping ballast over the years will alter the top and can make a difference over short lengths. Underbridges might also have been raised to give higher clearances.

 

I believe the Bluebell has been entirely resurveyed in the preservation era, not just the section north of Horsted Keynes - @Robin Moira White might know. There are some oddities on it that arise from subsidence of the formation over the years; for example the section of 1 in 75/134/60/75 south of the tunnel near Black Hut is due to reprofiling where the ground has subsided: originally it would have been a steady 1 in 75. Similarly I think the bit coming out of Horsted at 1 in 83 / 139 / 63 / 75 is due to historic subsidence around the area known as Fireslip. As a fireman you tend to think of that bit of line as just a steady 1 in 75 all the way, but you notice the subtleties more when driving, particularly driving unfitted trains.

The 1 in 55 at the top is also on a reasonable curve which adds to the drag. If you are on your game as a fireman and in tune with the driver, then the last pull just knocks the stuffing out of the fire and you slow down at the summit to about 15 - 20mph or so; that means the driver doesn't need to brake significantly on the following 1 in 60 descent which stops the water surging forward, when it has already dropped a big amount anyway on account of the gradient change.

Tom

 

The SVR has been resurveyed too, but the gradient profile is the GWR one!

 

In my professional life I’ve come across a double track railway (I won’t say which one!) which according to the official survey was rising in opposite directions across three stations. Assuming the lines at the same height at station A this would have resulted in something like a 2m height difference at station C. We asked them to check again and the profile has changed at least twice since. It’s much more sensible now. (The railway hasn’t moved, of course!)

Many years earlier I was on a site survey myself where we could clearly see that we were at the top of a hill but the official plan had us at the bottom of a hill. I’m pretty sure the signalling had been laid out incorrectly as a result of having the gradients back-to-front and I don’t think anyone ever changed it at the time (although the line has been resignalled since; I don’t know the current arrangements).

I’ve also come across survey data which had a repeating cycle of two different gradients - it was something like 1:190 and 1:215 alternating over an 18-20m cycle. I came to the conclusion that the track had settled slightly where the fishplates used to be in jointed track days and for any meaningful purposes could be considered a steady 1:200 (or whatever the average was), but it was pretty annoying having to work with pages of this data...

The point of all this is that this isn’t an exact science and there is certainly room for errors in the official data.

Sent from my iPhone using Tapatalk

 

The Blubell location map and gradient profile I produced - oh so long ago now -

http://www.bluebell-railway.co.uk/bluebell/map_grad.html

were corrected at the time for some variations from the BR originals. I can’t speak for whether they need updating now. For photographic / public purposes, I dare say the variations do not matter much but I would guess that the infrastructure professionals of most lines these days would have a clear idea.

Robin

 

There are likely to have been changes to the gradient in preservation with the line being relaid. It's also not beyond the realms of possibility that, like the discussion on the Welshpool & Llanfair, the GWR profile was 'as planned' rather than 'as built' . The SVR was supposedly built to a ruling gradient of 1:100 and the plan shows exactly that for the whole length of Highley Bank. However a significant deviation was added during construction due to land instability, which resulted in the S-curve and may well have changed the profile in doing so.

 

Mislaid my copy of NER Gradient profiles, but recall from it that the Author of the booklet was a BR Civil Engineer and he said that as a junior tech officer in the office he was tasked with resurveying many of the routes in the book.
I believe he found one that roughly matched the accepted profile, all the other lines were nowhere near the expected profile even allowing for subsidence or track renewals over the years.

 

No, the reverse curves on Highley were laid as opposed to the planned "straight and level" the deviation from that is largely down to subsidence.

 

I think the point is that if a railway is planned to rise, say, 30' in 1000 yards in a straight line, then that's a gradient of 1:100. If it ends up being built with an S bend making it 1100 yards between the same points, then it's 1:110, over essentially the same route.
I've always thought there must be a bit of tolerance in gradient markers anyway. Surely there must be the vertical equivalent of transition curves as the track changes from one gradient to another, so which end of the 'transition slope' is the marker placed?

 

That would come under the title of Vertical curves.
In general for 25 mph they can be ignored unless you have a severe gradient change.
A study of the Track design manual will give you the details.

The marker post will be placed at the point of change of gradient, determined by surveying.

 

Added Spa Valley to the list on page 1, following a post by @Jonnie in the Spa Valley thread.

Spa Valley Railway
https://www.spavalleyrailway.co.uk/shopimages/SVR MAP 2012 - portrait - colour - DPP.pdf

Tom

 

When I was working with such things a lot of the gradient information we used was taken from a set of diagrams, always referred to as ‘Sidestrips’, which I believe derived from LMS practice. As you might expect these diagrams were long and narrow, but the long dimension was used vertically. Each diagram had a representation of the track layout and a gradient profile drawn either side of a central line marked off for the mileposts, some also indicated the line's curvature. The text showing place names, milepost numbers and the individual gradient sections was applied horizontally (i.e. across the layout/gradient, etc.). As desks and tables are generally wider than they are deep it was normal to read these diagrams with the long axis in the horizontal position, which was OK as long as you had them the right way up. I well remember a senior engineer complaining that a particular signalling plan he was reviewing had all the rising gradients shown falling and vice versa, until it was pointed out to him that he had the sidestrip upside down.

 

North Norfolk Railway gradient profile

 

Thanks - added to the list on page 1.

Tom

 

The link on page 1 for the KWVR is broken, but there is a gradient profile in the Railway's own website at https://kwvr.co.uk/wp-content/uploads/2020/04/Gradient-Map.jpg

 

Hmm. Not SO flat, Norfolk....

Robin

 

Thanks, I’ll change that link when I get a chance.

Tom