Selective Compression- how-to?

[kiwitram]

Hello everyone,

I'd like to pose a question which I believe afflicts many modellers who wish to represent a prototype area or site;

How does one selectively compress to fit the space available?

As some of you may know, I'm modelling Hastings station and yard prior to its rebuild into a conventional through station format in 1930/31. This also precludes the line's electrification, something which only implemented from Tonbridge down and between Hastings and Eastbourne around 1936.

From a Google Maps printout at a scale of 100 feet to 4.5 centimetres (100:4.5), the length of the site is 25cm long, and 9.5cm deep. I'm already struggling to calculate the real-scale distance (maths never was my strong point) but members of my model club suspect that this would scale out to 40 feet, clearly far too big for the space I have available. I suspect my calculations are incorrect anyway; map scaling is definitely one of my weak areas.

Now, my bedroom is 11'6'' square, but with furniture in the way and whatnot, I have far less space than this realistically.

My question here is, how do people compress the prototype to fit the space available? I'm thinking you should compress (or remove) around 40% of what there is (this can include sidings and shortening platforms) but maintain, as far as possible, the scenic break lengths.

Of course I could be totally wrong.

Does anyone have any other ideas? It's something I know I should do, but I'm struggling to see 'how'.

Alex

[flying scotsman123]

A good question, but rather complicated, with the real place and personal preferences all making it difficult to have a one size fits all approach. However, as you say, shortening platforms is a good start. The percentage of platform occupied by buildings is usually quite small, so the rest can be cut down to suit, along with train lengths.

The station throat is usually surprisingly long. A mixture of removing extra sidings and making use of three way points to at least preserve some of the sidings if scenically important might be helpful. If you've got a mesh of pointwork a few inches here and there can be shaved off by clipping the plain bits of rail on each end of each set of points. Using shorter radius points helps here as well.

Something I've done with mine is to introduce a slight curve along the length of the scene. This increases the platform lengths you can fit in and reduces the lengths needed "off scene" for curves to return to the other side. This works better for stations that already have a slight curve which you increase, but could also work for straight platforms too if you feel that's acceptable.

Some stations with overhead canopies etc. are suitable to be "halved" if that's acceptable, especially termini. Ie. you have one approach, the train disappears under the canopy in theory to the rest of the station but actually off scene/end of the layout. I seem to recall this approach was quite popular a while back.

Anyway, those are just a few thoughts, this hobby is all about making compromises, I'd need loads of space to model my station in it's entirety but I've managed to squish the key things in such that anyone who knows it will instantly recognise it, at which point they probably won't notice all the compromises made anyway.

[Bufferstop]

As you have discovered scale length can mean massive. Our little station, which never has more than a four car DMU stop at it, is on a former secondary mainline. So the platforms will take around 10 coaches except that about half their length is now buried beneath the creeping green stuff that seems to invade anything that remains stationary or unswept around here. I could certainly get a model in half the length, and then the open line between the end of the platforms and an overbridge could probably be done in a third or quarter of the space. If you need anymore compression, then the section with the buildings on could be squeezed by taking one section of window and door out of the building. GWR conveniently used modular designs for their brick built buildings with standard sized windows, doorways and intervening brickwork, you can either chop out a module or reduce the standard lengths. You have to choose which looks best. Then there's the bit between the footbridge and the road overbridge you could cut that down a bit but not too much or the sloping path up to the road would look too steep. Perhaps the road between the road bridge and the path sloping up to it could be made a little steeper hidden by the hedge and the time table boards, then the path wouldn't have to be so steep. Fortunately the side that most people approach from faces the other platform which only has a single little waiting room with acres of platform beyond it. That will cut down easily and Joe public won't suss what you've done but will still recognise the station. It's the compromises you make and the degree to which they alter the perception that decides whether or not you've captured the essence of the scene.

[b308]

If you can get hold of Freezer's Track Plans 2nd Edition book it has a short bit on compression. He took a plan of Exeter Central and then did two plans, the second smaller and compressed and talks you through what he did to keep the "essence" of the station in the smaller space. In the space you have you may find that compression and subsequent simplification will be pretty drastic as 11ft 6ins if not very long.

The answer is, of course, to go down a scale if that's possible!

[Emettman]

From a Google Maps printout at a scale of 100 feet to 4.5 centimetres (100:4.5), the length of the site is 25cm long, and 9.5cm deep.

Alex

Someone else check me?
Real size 100ft for every 4.5 cm on map. 4.5's in 25cm = 5.5 recurring. x100 for feet =556 feet. Divide by 1/76for OO = 7ft 4 in
Which looks desperately too short.

Has the map been printed out at 1:1? if not, its scale legend will be badly off.


Chris.

[mumbles]

The compromise is to shorten the trackplan but keep the essential parts required for operation, and keep the important structures.

This is effectively what I've done for my current and past garage layout. The guiding factor being the length of trains ( another compromise) and platforms. Track work is as close to prototype as RTR points allow with double slips and 3 ways to compress but keep the same routes, and as someone said, the station throat offers a good place to compress too. I've made the entrance to Sandgate on a curve (as it runs round the garage) which in real life it's straight.

[Emettman]

The relatively easy compressions are:
having platforms barely longer than the trains, having loops and crossovers that only just clear the reduced platform lengths, trimming the length of sidings.
Moving"end of scenic section" limits closer to the platfortms.

Then, using sharper angled points to reduce length of crossovers and other pointwork,
losing a siding out of a set.
Reducing the length of trains by a carriage or two (back to step 1!)

An alternate interesting way is available if the station is crossed by a high level road, or indeed station building.
This gives the possibility of modelling *half" a station with less compression, as beyond the scenic break tracks can curve and the necessary pointwork can be made as compact as possible.

Chris

[kiwitram]

From a Google Maps printout at a scale of 100 feet to 4.5 centimetres (100:4.5), the length of the site is 25cm long, and 9.5cm deep.

Alex

Someone else check me?
Real size 100ft for every 4.5 cm on map. 4.5's in 25cm = 5.5 recurring. x100 for feet =556 feet. Divide by 1/76for OO = 7ft 4 in
Which looks desperately too short.

Has the map been printed out at 1:1? if not, its scale legend will be badly off.


Chris.

I did feel the map printout felt off, and whenever I've tried to print out without adjusting the page (which I also did here) it's come out with entirely unrealistic scale lengths.

I suppose what I can take from all this is:

- shorten long stretches of plain track where I can
- Be clever with angles and curves (the station is on a curve anyway, so perhaps tighten this a little to fit more onto the board?)
- Rejig siding placement where necessary
- Shorten buildings
- Use scenic breaks to create perceptions of depth (Hastings is ideal as it has a road bridge at either end)

But I'm happy to hear people's ideas, and I hope this thread will help other modellers as well as myself- it *IS* a big problem that modellers sometimes face and it can be disheartening or confusing.

- Alex

[Mountain]

The problem is that either one adopts a branch line but runs it as if it is a busy main line, or one compresses lenghts. The difficulty is, if one runs an express locomotive it just does not look right with only three or four coaches in tow.
One idea is to model a modern intermediate station on a mainline as some of these were short so could only accommodate the local service trains. The expresses didnt stop.
The issue of a short mainline station rarely happened on the prototype but... There were many examples of secondary routes with shorter platforms but were unbelievably busy in summer seasons due to holiday traffic.

[Mountain]

If you ever get the book called "The Living Model Railway" by Robert Powell Henry, go for it as part of it covers the issues you have brought up and the answers. The book describes the issues 0 gauge modellers have and the nifty ways they use to overcome such issues and make their railways believable. Some of the layouts dont have much in the way of scenery and have centre rail pick ups but they have a more realistic feel to them then the most detailed layouts which dont have a model purpose to them.
It is a good read.

[NickH]

Using the NLS mapping website, I came up with an approximate length between over bridges at around 420m and the width at 180m. This would work out as 1,400 feet by 600 feet, which, at 4mm to the foot, roughly comes to 5.6m x 2.4m, or 19 feet by 8 feet.
The biggest problem with Hastings is its sheer width, with the way the platform lines are set at a very wide angle. Whilst the ideas suggested by others will have some impact, it won't be easy to maintain its characteristic shape in the space you have.
One possible way to compress the layout is to appreciate that the geometry of readily available pointwork is usually very different from the real thing. Creating the critical station throat layout in your chosen track points would be an initial minimum starter from which to develop your ideas, taking into account the potential for reducing platform and siding lengths, and possibly omitting sidings to get your best fit.

[kiwitram]

The problem is that either one adopts a branch line but runs it as if it is a busy main line, or one compresses lengths. The difficulty is, if one runs an express locomotive it just does not look right with only three or four coaches in tow.
One idea is to model a modern intermediate station on a mainline as some of these were short so could only accommodate the local service trains. The expresses didn't stop.
The issue of a short mainline station rarely happened on the prototype but... There were many examples of secondary routes with shorter platforms but were unbelievably busy in summer seasons due to holiday traffic.

Hastings is a strange mainline station, though, in that it most likely only saw around maybe seven or eight trains an hour;

Ashford-Brighton
Brighton- Ashford
Hastings-London (C.X. or C.S.)
London-Hastings (C.X. or C.S.)
Hastings- London (Vic.- this was probably less frequent than the other London services)
Hastings- Tonbridge/Tunbridge Wells (portioned with the Up London services)
Hastings-Eastbourne (could be tagged onto the Ashford-Brighton service)
Bexhill-Rye autotrain service (this probably shuttled between 10:00am and 4:00pm, and even then only on market days)

We also have the 'periodical' trains, such as the City Limited, which would run once or twice in the morning and again in the evening, and the Sunny South Special; this could probably be just a coach or two portioned from Eastbourne or at Ashford from the Kent Coast services (people tend to forget the SECR and later the Eastern division of the SR operated Kent SSS too).

Now, having had a look at the coach sets that are outlined here- http://www.semgonline.com/coach/sets.html - it seems most trains to Hastings were between two and ten coaches long; this, to me, is a bit too fluid, and obviously if I'm limited in space, it makes sense to me to compress the train sets, and if I use the following coach types (obviously these are 'mainline bogie' coaches, and so of course will be longer than older pre-grouping types) I can probably compress down to three or four coaches maximum:

60' Birdcage three-set
57' R0 Pullman car
59' R0 Maunsell (maybe only one or two due to my time period having an upper limit of 1930)
64' 8'' 'Continental' (the brakes of these were 65' 2 3/4'' long, which is just an awkward size and these were more seen on Kent Coast services anyway, so I doubt they would be included)
60' 'Hundred Seater' (again, perhaps less likely to appear due to their use on the London commuter belt; more likely to have been for shorter journeys, no further south than Tunbridge Wells, possibly)


Other coaches would include the LBSC bogie, coming in at 54', the same company's 48' six-wheel coaches, and various other smaller coaches between 26 and 50'. There will be a couple of Pullmans at 64' but they'd be running in sets and so could be excused, almost. Inter-regional stock on the Sunny South courtesy of the GNR and LNWR would come in at 48-50', also.

I'm sure that no coach set following this principle would exceed more than maybe 75/80cm, with another 20cm or so for the locomotive? Still quite long, but I'm sure I'd manage to work out a decent number of coaches that looked right.

Of course, this is what would work for my selected location, and I need to work around that. Other people may be able to rejig and compress their trains more sensibly than I have, if that's the correct way of saying it. I should also add that my layout will be set in winter and hence I gather fewer people would be out and about anyway, and so I shouuuuld be able to get away with the degree of compression I've suggested.

With the compression of the station itself, having read the post by NickH (thank you, by the way for that- I wish I'd been able to find a way of working that out myself! Shocking size, really!) it seems another major limiting factor is the angle of the old station platforms; this leads me to think that perhaps, if I reduce this angle further I'd still be able to fit both platforms in and a fair representation of the station building to still capture the character of it.

It wouldn't be as accurate as I'd like, but unfortunately, I simply do not have the space for a scale portrayal of the site (maybe one day, with another prototype) and so these compromises are a necessity.

This is another thing to add to the possibilities of selective compression, then:

Make what you want to represent fit your space available. It's all good and well taking things out and shortening, but sometimes things literally need a 'squish', in this case, the depth of the station footprint.

I hope this makes sense,

- Alex

[TimberSurf]

From a Google Maps printout at a scale of 100 feet to 4.5 centimetres (100:4.5),

The problem may be that your statement is incorrect,100 feet to 4.5 centimetres is not the same as 100:4.5! Plus printing and measuring a print is dependant on what scale you printed at! Make a link to the map and we may be able to discern were you are going wrong.

[mumbles]

Makes perfect sense. Its worth remembering that the look and feel of the layout are as important to creating the realism as size and scale IMO

I'm all aboard with your plans, sounds great

[kiwitram]

From a Google Maps printout at a scale of 100 feet to 4.5 centimetres (100:4.5),

The problem may be that your statement is incorrect,100 feet to 4.5 centimetres is not the same as 100:4.5! Plus printing and measuring a print is dependant on what scale you printed at! Make a link to the map and we may be able to discern were you are going wrong.

Timber,

I've attached a screenshot of the map, which is also available at https://www.google.co.uk/maps/search/Ha ... 767244,18z

When I printed it out I think it might have automatically adjusted to the A4 page, and so skewed the scale, but we'll find out what the scale is soon enough, I'm sure.

I hope this will prove an education for other modellers in the future- check your scaling and calculations properly!

- Alex