Jump to content

Posted Image It's time to vote in the British Model Railway Awards! Click here.

* * * * * 3 votes

Sprat & Winkle couplings

Posted by Mikkel , in Rolling stock, Miscellaneous, Couplings 05 November 2009 · 10,878 views

couplings stock
Posted Image
Shunter George "Bulldog" Mullins critically eyes stock fitted with Sprat & Winkle couplings.
His shunter's pole is legendary among shunters for being rather crude!

The following notes on Sprat & Winkle couplings seemed to generate some interest when first posted over on gwr.org.uk, so perhaps they are of of use to someone here also. I have taken the opportunity to take some new and better photos for illustration.


Although the latest RTR offerings have helped enhance the looks of the RTR tension-lock coupling considerably, I still find them a bit too bulky and not quite reliable. They also do not offer the opportunity of "delayed action" uncoupling, which allows you to propel stock forward after uncoupling.

Looking for an alternative, I have taken to the fairly well-known Sprat & Winkle coupling, which - although a compromise in some respects - has proved quite reliable and fairly easy to fit. I find the delayed-action feature of these couplings simple and effective, and a plus for me is that they allow cosmetic 3-links to be retained.

Posted Image
Hook and bar. It could be argued that it is no less obtrusive than the modern tension-lock coupling.
But I find it less bulky and with more functionality.

One-hook operation

The Sprat & Winkle couplings are available in 2,3,4 and 7mm scale versions. As I model in 4mm my choice was between either the standard 4mm version or the "finescale" version. The latter is in fact intended for 3mm modellers but works fine for 4mm (including OO), as long as your curves are not too severe ( ie less than 4' radius according to MSE). This is fortunate because the standard version is a bit on the large side for my liking, and so I have opted for the finescale/3mm version.

In fact, even the finescale version is a bit more prominent than I would personally have wished for, especially when uncoupled. To minimize the visual impact I therefore fit a coupling hook to one end only, adding just the loop at the other end. This obviously requires stock to be facing in a particular direction when placed on the track, but on my layouts (and I think many others) this isn't really a problem. The absence of a coupling at one end also facilitates the fitting process (since you only have to fit one hook per wagon) and means I can add a prototypical (but cosmetic) coupling hook here instead, enhancing appearances a bit.

Posted Image
Coupled up using the "one-hook" approach

Mounting the couplings

The coupling hook features a square "paddle" at one end, which works as a counterweight beneath the wagon or coach body. The MSE website has an instruction sheet for fitting the couplings, and details on various extra parts not described here (including custom-made mounting plates). The instructions suggest two possible ways of mounting the hook: An "Upper" method in which the coupling hook is inserted through the headstocks (ie the "buffer beam" of the wagon), and a "Lower" method in which the hook rests immediately below the headstocks, hinged to the wagon floor with wire bent to the shape of a paper staple.

It is necessary to standardize on one of these two methods, and in principle I prefer the latter, which also comes recommended in the instructions: This requires only minor modification to the wagon or coach body, and is also - in my opinion - rather less fiddly. That said, I have made two minor modifications to this approach:

* Firstly, I replace the curled-up wire included in the pack with straight brass wire from Alan Gibson. I find that this makes it far easier to craft the wire-staple needed for fitting the paddle. The staple is then fitted to a section of square plastic rod mounted on the wagon floor. The plastic rod is not always necessary - it depends on the distance between the floor and the lower edge of the headstocks.

* Secondly, I find that the "Lower" method of mounting the coupling can sometimes give problems in ensuring that the coupling hook is fully horisontal: Exactly because it is underhung, the hook may come to rest at a slight upward angle against the bar of the loop on some wagons, which is neither aesthetically pleasing nor good for operation. I don't think it's just me, as I have heard others mention this issue also. My solution is rather crude I suppose, but effective: I simply open out a slight slot in the wagon just above the coupling hook, thereby allowing it to move freely to a full horisontal position against the loop. This may not be to everyone's taste, but the slot is really quite unnoticeable and can always be padded over with a filler if the coupling is removed.

Posted Image
The "Paddle", anchored with a wire "staple" to a supporting section of plastic rod


Uncoupling is by means of magnets located beneath the track, nested into the track base. The magnets attract the 3-links, thus tilting the hook downwards. When moving back up, the hook comes to rest in a position which allows the wagon to be propelled forward and left where you want it in the siding. Hence the "delayed-action" concept. The following photos illustrate the four main steps of this process:

Posted Image
1. Wagons are propelled in fully coupled condition

Posted Image
2. Coupling hook drops down as it is attracted by a magnet beneath the tracks

Posted Image
3. As wagons are propelled forward the coupling hook moves back up, but does not fully engage the bar

Posted Image
4. The uncoupled wagon is left where desired, and the rest of train is drawn backwards

For me this works well, with one important modification: Because I use only one coupling hook, the very powerful magnets occasionally uncouple the stock even when they are not supposed to - ie when the stock is passing slowly by. This happens even with a good layer of ballast above the magnets, and attempts with a sliver of Plastikard above the magnet doesn't help much either.

Again, I resort to cave-man technology for the solution: I simply break the magnets in half, thereby reducing the overall magnetic field. I say "break" because cutting will get you nowhere with these magnets - they need to be broken in two by holding the magnet with one pair of pliers and breaking downwards with another pair. Crude stuff, but it works.

Finally, I should perhaps emphasize that I have no affiliation with the manufacturers, and that these are the experiments of a novice: I do not have experience with the other non-RTR types of couplings available (see jim s-w's blog for an interesting entry on the Dingham coupling).

Edit: For a discussion of fitting the couplings to locos, see the comments to this entry.
  • Like x 4
  • Informative/Useful x 1

More or less the same method of fitting I use on my wagon stock. The only difference is that I omitted the bar on the hook end and fitted sprung buffers on this end only to allow more realistic wagon contact. This coupling system is simple and reliable to fit and use.
Interesting Richard. I did initially consider leaving off the bar at the hook-end, but found that the hook would then be prone to tilt upwards (rather than remaining horizontal) . But maybe you have found a workaround to that?
Steve with the Hat
Nov 08 2009 10:52
Thanks for the detailed explanation of this system Mikkel. As a novice all these different systems can mean little unless you actually see them illustrated. Cheers again and have 5 stars ;)
Thanks Steve. Yes there's nothing like a simple photo sequence to help demystify things in this hobby. For too long the modelling press ignored this - although in recent years at least some of the mags have caught on to this, it seems.

More or less the same method of fitting I use on my wagon stock. The only difference is that I omitted the bar on the hook end and fitted sprung buffers on this end only to allow more realistic wagon contact. This coupling system is simple and reliable to fit and use.

Richard has now posted photos and notes on his use of S&Ws on his own blog here.
Isambard Kingdom Brunel
Nov 23 2009 22:03

Interesting Richard. I did initially consider leaving off the bar at the hook-end, but found that the hook would then be prone to tilt upwards (rather than remaining horizontal) . But maybe you have found a workaround to that?

If the loco has a bar at each end, How do you manage shunting? If the loco pulls a train into a terminus, it then runs round the goods wagons. The loco can then push wagons, but, cannot pull the brake van off the train or remove the train after unloading.?

Or am I missing something extremely fundamental here???

Not completely sure what you mean here, but the loco does need to have a hook at one end for doing run-rounds etc. It doesn't strictly need to have a bar at the hook-end though, as the hook-end of the loco would be facing a wagon with only a bar at that end.

This would be a lot easier to show than say Posted Image
Isambard Kingdom Brunel
Nov 24 2009 16:36


Not completely sure what you mean here, but the loco does need to have a hook at one end for doing run-rounds etc. It doesn't strictly need to have a bar at the hook-end though, as the hook-end of the loco would be facing a wagon with only a bar at that end.

This would be a lot easier to show than say Posted Image

Thanks, this is the problem I have had trouble understanding. I read on a previous thread(RMWEB Mk3 version 2 build 3 ;) ) that locos have a bar at either end. Which as far as I was concerned made running round a train impossible, you could push but, not pull. The reply about the bar at each end was a reply to a question of how do you fit a S & W hook under a loco, there does not seem to be enough room.

If I had a hook at each end of all other rolling stock and kept coaches in rakes with hook at either end then Running round and shunting would be possible.

You know I quite liked the 3 link coupling on my last layout, it was quite simple to fit and the only downside was seeing the end of the shunters pole and not derailing all the stock when uncoupling. :(

I must move on though as on this layout I will not be able to easily reach to uncouple.

Do you have a photo of a hook fitted to a GWR Loco? I shall be using 14XX 74XX, 22XX and Dean Goods primarily.
IKB you are quite right to point out that it can be tricky to find the space to fit a hook under locos. It's a definite drawback of the "one-hook approach" that I should have mentioned. With a two-hook approach it's not a problem because, as you say, the loco can just have the bars only and no hook.

So far I've been able to overcome it by finding different ways of fitting the hooks according to the particular loco, but it is a bit of a trial and error approach. I'll post a photo of a loco tomorrow.
Right then, below is the hook fitted to the underside of my 850 Class. There is a fair amount of room for the paddle in this case because of the large bunker, but I did have some trouble fitting it as the screw for the chassis is directly beneath the paddle. I had to build up a mounting block for the coupling from layers of Platiskard cut to shape (just visible to the front left of the paddle), which turned out to be enough to hold it in place. This shot also shows an alternative way of securing the paddle, using track pins cut in half. I am actually beginning to prefer this method as it is a bit quicker to do. The brass plate has nothing to do with the coupling btw, it's just there to level the chassis.

Posted Image

Below is the same 850 with the hook seen from the outside. Not too pretty to look at, but it isn't that noticeable during operation I think. You can see how I've had to dispense with my normal principle of fitting the couplings and have had to make it underhung. The coupling still works fine with the other stock in this case, but I have one other loco where I'll have to fit the larger 4mm version instead, since the currently fitted 3mm version is a bit too erratic (it doesn't quite have enough "reach" as it is underhung).

Posted Image

These various workarounds are of course only relevant if you are using the one-hook approach. With two hooks on each item of rolling stock you can just fit bars to both ends of the loco. However I still prefer this method as it is very time-saving having only to fit one coupling per vehicle, even if it requires some fiddling around with the locos. Significantly it also avoids the visual impact of having a hook at both ends, which I find a bit jarring to the eye.
I am a frequent visitor to The Farthings and an admirer of your creativity.
Can I first join with the other poster in thanking you for a clear account of using Sprat and Winkle couplings.
I have recently converted a number of my N gauge wagons to S&W. I have gone for the business each end, with just a loop on either end of my shunter.
The hooks unfortunately look much more overscale ( in N thats nothing new!) but visually still preferable in my opinion to the Arnold Rapido style of knuckleduster, and they do seem to work - well, at about 90% of the time, and then you have to adapt a particular shunting motion to get them to release!
The problem with 2mm comes more in clearance to fit the base and paddle beneath the end of the wagon, particularly so on shorter wheelbase vehicles.
The compromise for me is in the distance between wagons which is slightly more than with the stock coupling, though with the decided advantage of being able to shunt a yard without the Hand of God appearing all the time.
That's interesting Chris, I had been wondering how the S&Ws worked in N. I can see how clearance would be a particular problem there, as always I'm really impressed what you 2mm/N modellers can achieve with what must be quite fiddly work sometimes.

Agree that the S&Ws aren't quite as discrete as some of the other couplings available, but for me their rugged simplicity outweighs this. I've recently started following Richard's approach (above) and omitted the bar at the hook end. It's not as aestecthically pleasing because the hook somehow becomes more obvious, but it does speed up the fitting process, which in my case is a priority.
I should have commented here ages ago, but thanks for posting these - I paid close attention while building and experimenting with my first S&W couplings a couple of months ago. I'll have to give the track pins a go - that looks like a great idea.

Hi Will, glad it's been of use. The track-pin-in-plastic-block is the only method I use now, much quicker and less fiddly, I find.
I have just come to this after a general Google search on S&W delayed action - I couldn't visualise how it would work. Many thanks for such a useful summary, Mikkel.
Thanks Al, yes this also happens to me all the time. I go off looking for something in Google and end up back on Rmweb :)

A bit late to this thread, but a big thanks from me too for the simple explanation of how to fit and use the couplings.


Do you find the bars override each other when pushing several wagons at once, if the 1st pushed wagon has been uncoupled ?

Hi Stu, glad if this is of use.


No, I haven't really experienced that problem. I can see why you ask though, because looking at the photos above the bars touch, but not the buffers. That is misleading though. Those photos are quite old from when I first experimented with the couplings, and you can see that on the van I hadn't fitted the bar quite flush with the buffers. That's not quite as it should be. As long as the bars are same height and aligned with the buffer faces, there shouldn't be a problem.

Welcome to Farthing!

Attached Image: farthing2.jpg


This blog chronicles the building of "The Farthing layouts", a series of small OO layouts that portray different sections of a GWR junction station in Edwardian days.


Intro and concept
How to eat an elephant
Design principles
State of play


Gallery (1900-1904)
Four o'clock blues, ca. 1902
What really happened in the Cuban...
The honourable slipper boy (Part 1)
The honourable slipper boy (Part 2)
The honourable slipper boy (Part 3)


Gallery (1904-08)
The trials of Mr Bull
A most implausible arrival
A parcel for Mr Ahern
Blue skies and horse traffic
The Remains of the Day
Motley crew

Edwardian daydreams


Gallery (1914)
All in a day's work, Part 1
All in a day's work, Part 2
All in a day's work, Part 3
All in a day's work, Part 4


Out of period
Undecided sky (1867)
The sleeping giant (1887)
Bunker first (1927)
Fitted fish and piles (1947)


Once Upon a Time in the West
Summer silliness
The unbearable lightness...
Across the years
The Sidelight Job
Painting coach panels

Traverser testing


Low-tech pre-grouping stock

Short trains for short layouts
Short trains with a twist
Hand-me-down coaches
Low-tech coach restoration (1)
Low-tech coach restoration (2)
Low-tech coach restoration (3)
Low-tech coach restoration (4)
Low-tech coach restoration (5)


Sprat & Winkle couplings
3 plank Open in GWR red
Outside Framed 8 Ton Van

In the red: GWR 1900s wagon liveries
In loving memory...
Scratchbuilt GWR one-plank wagon (1)
Scratchbuilt GWR one-plank wagon (2)
MSWJR 3-plank dropside
LSWR 10 ton sliding door van
SDJR Road Van
LSWR stone wagon
Fake news and wagon sheets
Same but different: 1900s wagons


GWR 1854 Saddle Tank (1)
GWR 1854 Saddle Tank (2)
Shiny domes and safety valve covers
Backdating the Oxford Dean Goods (1)


GWR large flat dray
Ratkin & Son horse-drawn wagon
Kit-bashed GWR light dray
GWR horse-drawn trolley
GWR 5-ton horse-drawn wagon
Parcels van and coal trolley


Fun with crates
Barrels, baskets, bales
Small crates and tea chests


Porters and Barrows
Andrew Stadden 4mm figures
Backdated Monty's figures
Footplate crew
HO figures for an OO layout
Lesser known whitemetal figures


C+L underlay and Carr's ballast
Experiments with C+L track
Comparing track
Messing about with track panels
Laying track on "The depot"


Constructing the Branch Bay
First bite: "The bay"
Simple structures for "The bay"
Platform trolleys and barrows
Signs, posters and adverts
Six lessons learnt


Constructing the Goods Depot
Second bite: "The depot"
Shunting Puzzle
Sketches of The depot
Soft body, hard shell
Kit-bashed roof structure
Dry Run
Dusting off the cobwebs
Playing with mirrors
Mezzanine floor
Progress on "The depot"
4mm slate roofing
The treachery of images

A roof for "The depot"

A tall bird from Paddington
Cranes for the depot
Shoulders of giants
Flight of the bumblebee


Constructing the Old Yard
Third bite: "The sidings"
Wagon propulsion
Progress on "The sidings"
Rising from slumber
The Biscuit Shed
A shed and a lock-up
Agricultural merchant's warehouse
Mr Gorbachev, tear down this wall
Stops, levers, plates, gauge, wall
Lamps and Lamplighters


Constructing the Stables
GWR Park Royal stable block
GWR stables - an overview


Railway modelling and Art
Moving Pictures
Season's greetings


Layout ideas
A flexible layout
Kicking back in Gloucester


Pre-grouping livery clippings
Journey to Didcot
Detail hunting at Didcot
Here's looking at you
The mists of time (and all that)
My friend the operating chair
Ready-to-plonk freight
GWR Modelling website


RMweb Workbench
Flickr photostream

Recent Comments