Slope of cuttings & embankments

[rodent279]

What is the typical slope of a railway cutting or embankment? I guess would vary with the soil type, but is there a typical figure?

[Dagworth]

There is, I remember seeing diagrams online somewhere, I’ll have a dig and see if I can find a link. 
 

Andi

[br2975]

14 minutes ago, rodent279 said:

What is the typical slope of a railway cutting or embankment? I guess would vary with the soil type, but is there a typical figure?

 About 1 in 1.5,  I believe.

.

The drawing below is for 4mm scale dimensions.

Edited January 24, 2022 by br2975

[rab]

I think we've had this before as you say its all down

to the embankment/cutting material.

 

Try googling 'natural angle of repose'

[Darius43]

7 minutes ago, rab said:

I think we've had this before as you say its all down

to the embankment/cutting material.

 

Try googling 'natural angle of repose'

What he said.

 

 

Coulsdon railway cuttings on the London to Brighton line in Surrey by DigPeter on Flickr.

 

These slopes are a lot steeper that 1 in 1.5.

 

The big advantage of modelling cuttings like this is that you don’t need to worry about slope stability or slip failures.

 

Cheers

 

Darius

Edited January 24, 2022 by Darius43

[Chris M]

A typical model railway embankment is too steep.

[Trog]

Older cuttings and embankments also tend to have steeper slopes than more recent ones, and as stated above cuttings through stiffer materials such as chalk also tend to be steeper sided. One access point into Tring Cutting in the 1980's consisted of a length of rope tied to a fence post, so you had something to hold onto while climbing in or out of the cutting.

[didcot]

Hand dug cuttings were more open and less steep. Parts of the Didcot, Newbury and Southampton were dug with a steam shovel which gave almost vertical sides. They were prone to collapse and had to be opened up.

[jim.snowdon]

1 in 1.5 is as good a rule of thumb for slopes in earth, but for clay soils it needs to be significantly flatter than. Cuttings in rock can be significantly steeper, but even that depends on the type of rock. Cutting sides in hard, sound rock can be nearly vertical, but with anything friable or prone to weathering, such as chalk, the slope can still be 1 in <1, as evidenced by the picture of the cuttings south of Coulsdon that feature in Darius43's post earlier. Even then, that cutting has not been without its problems in recent times and parts of it were the subject of some quite major slope regrading some years ago.

 

[Compound2632]

I believe it's the case that Victorian civil engineers took 1 in 1.5 as their usual working figure, except through harder materials such as chalk. They didn't have as good an understanding of the stability of different types of materials as modern civil engineers, so in some cases this has turned out to be unwise, for example, the continual difficulties experienced with the cutting at the north end of Harbury tunnel, between Banbury and Leamington.

[Nearholmer]

15 minutes ago, jim.snowdon said:

Even then, that cutting has not been without its problems in recent times and parts of it were the subject of some quite major slope regrading some years ago.

 

Its been falling down since it was dug.

[eastglosmog]

Brunel cut Sonning cutting at 1 vertical to 2 horizontal (26.5degrees to the horizontal).  It famously collapsed a few years later!  As a general rule, clay slopes would later be cut at 1 in 3.  Rock cuttings could be near vertical, and many of them today shed rock blocks.

Quote

 

 

[Butler Henderson]

Down to what is being cut through and how stable it is, take the near vertical cutting at Nottingham Victoria

https://www.pinterest.co.uk/pin/563020390894995314/

[TheQ]

Yep, see how the line is kept away from the high Side. 200ft IIRC on the left. The cliff face was there before the railway cutting on the Kyle line

[Compound2632]

Well, yes, a rock cutting is one thing, but the OP is asking for what is typical, rather than extreme; also mentions soil. So the canonical 1 in 1.5 is, I think, what they're looking for. But I think in a model one can get away with something a little steeper, especially if viewed sideways-on. (In much the same way as the narrowness of 00 is less prominent when viewed side-on.)

[Miss Prism]

Note the '1 in 1.5' is the recommended steepest.

 

[eastglosmog]

See attached report by the Transport and Road Research Lab regarding what gradients have been found to stand up over the years - this may be work on motorway embankments and cuttings, but the same physics apply to railway works.

TRL R199 - Survey of slope condition on motorway earthworks in England and Wales.pdf

Edited September 16, 2022 by eastglosmog
Restore file

[Compound2632]

17 hours ago, eastglosmog said:

the same physics apply to railway works.

 

But not the same knowledge, is the point I will continue to press, the vast majority of railway works having been carried out in the 19th century, well before such studies were available. So the fact that most cuttings and embankments have stood the test of time is more down to the experience and empirical judgement of the civil engineers of the day.

[The Stationmaster]

1 hour ago, Compound2632 said:

 

But not the same knowledge, is the point I will continue to press, the vast majority of railway works having been carried out in the 19th century, well before such studies were available. So the fact that most cuttings and embankments have stood the test of time is more down to the experience and empirical judgement of the civil engineers of the day.

Although they sometimes got it wrong.   -But basically exactly as you say - the engineers arrived at their decisions regarding the slope of embankments and cuttings through experience and judgement and were pretty good at it..  

 

The interesting thing is how over the years slip sites have been seen on lines which were constructed relatively late.  For example one of the worst slips in recent times on the Western has been near Brinkworth on the Badminton cut-off - a section of route which didn't open until 1903 -and there are other places where slips have been experienced on that section of route.  And there have been some dodgy bank movements, with the need for major improvements, on the Stert and Westbury which was opened in 1900.  Whether the GWR was alone in getting it 'not quite right' on later works I don't know - these are just examples that I'm  aware of on sections of route that I'm familiar with.  But it does suggest that even as late as the start of the 20th century the science of embankment and cutting grading was still not fully understood in relation to varying soil types although I know that in some cases slips in cuttings have been a consequence of things done, or not done, on land beyond the railway fence.

[rodent279]

What a fascinating subject!

 

Does anyone know to what standards new works are built? For example, the Selby diversion, Penmanshiel tunnel diversion, HS1 etc?

[kitpw]

On 24/01/2022 at 21:46, Compound2632 said:

I believe it's the case that Victorian civil engineers took 1 in 1.5 as their usual working figure

Tring Cutting (the famous Bourne engraving is probably familar to modellers) is in chalk to a depth of about 40 feet and when excavated in 1835 had slopes each side at 1:1 (since altered and now much steeper sided, presumably with slope stabilisation). By contrast, New Cross cutting is through London clays, the first 50' of depth through brown clay and then blue clay down to the Woolwich & Reading beds at a depth of about 68'. Both side slopes are 2:1 (2 along, 1 up). It is not uncommon to find berms (effectively a shelf) at a change in soil type but these weren't provided at New Cross.  The familiar rule of thumb of 1:1.5 is a useful guide for modelling although in any depth of cut, it still produces quite a width if the whole cutting is modelled for a double track route. I'd look at a few pictures from the web and rough out some 3D sketches in cardboard to see what slopes look right for the width and depth which is wanted, starting with 1:1.5.

 

Kit PW

Swan Hill - https://www.rmweb.co.uk/community/index.php?/blogs/blog/2502-swan-hill/

[eastglosmog]

2 hours ago, Compound2632 said:

 

But not the same knowledge, is the point I will continue to press, the vast majority of railway works having been carried out in the 19th century, well before such studies were available. So the fact that most cuttings and embankments have stood the test of time is more down to the experience and empirical judgement of the civil engineers of the day.

But the paper I referenced is based on what has been found to work - it quotes slopes for which only 10% have failed - hence my statement about the physics being the same.  Gravity has not changed since the 19th Century!  With reference to the slips that Stationmaster has noted, soil slopes tend to get weaker with age due to weathering and stress relief, resulting in failures some time after construction (e.g. Sonning).   Another case of the GWR getting it badly wrong c. 1906 is the Cheltenham to Stratford line around WInchcombe, as the GWRS have found to their cost.  Much earlier, the OWW got it badly wrong in several places, Aston Magna cutting (resulted in the sharpest curve on the OWW to avoid it) and the cuttings approaching Chipping Campden tunnel.  The GWR also inherited a few instabilities from the Banbury and Cheltenham Direct, the cuttings at Hook Norton failed repeatedly over the years (and were the final excuse to close the Hook Norton section).  A bit of searching will reveal other railways also had trouble, not just those taken over by the GWR!

Regarding modern railway cuttings, they are designed on the basis of extensive strength testing, computer modelling of failure modes and a sizeable allowance for unknown variables.

[Compound2632]

25 minutes ago, eastglosmog said:

But the paper I referenced is based on what has been found to work - it quotes slopes for which only 10% have failed - hence my statement about the physics being the same. 

 

Not disputing that, merely pointing out that it's hindsight; such knowledge was not available to the civil engineers building the lines we have. That there has been a 1 in 10 failure rate shows the limitations of the rules of thumb they were obliged to employ. In most cases they had no appreciation of the geology they were going to encounter - in fact, it was their work that made geology possible!

Edited January 26, 2022 by Compound2632

[rodent279]

1 hour ago, eastglosmog said:

Regarding modern railway cuttings, they are designed on the basis of extensive strength testing, computer modelling of failure modes and a sizeable allowance for unknown variables

So in other words, each case is designed according to the local conditions, there is no rule of thumb applied any more?

Edited January 26, 2022 by rodent279

[laurenceb]

I remember reading a photo caption that said that the embankments & cuttings on the GCR London extension were of a shallower angle because of Board of Trade rules. How accurate this statement was I know not.