Model Railways On-Line - Littlehempston - Part 8
Littlehempston Part 8 - Laying the Track By Paul Plowman The building of Littlehempston has reached the stage where all of the baseboards are complete, the trackbed has been prepared and a sub-base of cork has been laid. It is time now to begin laying the track. Photo 1: A draughtsman's "railway curve" being used to check track alignment. The layout was initially designed on computer using Turbocad software. The first task was to mark out the track alignment using coordinates obtained from the computer design. As the layout is rectangular in shape with sides at right angles, the edges of the baseboards could be used as a reference. Significant points in the design are measured on the computer to the baseboard edges and then redrawn full size on the layout. This process, at best is only accurate to about 5mm. The method I adopted was to set out the straights first and then draw in the curves using “railway curves”. Sometimes they are referred to as “draughtsman’s curves” but they did originate in railway drawing offices and “railway curves” is their correct name even when used for other purposes such as highway design, see photo 1. The long curves at the ends of the layout have a radius of 5 feet on the inside track (Down Line). Because the curves are so long the use of “railway curves” is impracticable. To overcome the problem I made my own curve from mdf board. It provides a 90-degree arc of 5ft radius, see photo 2. Photo 2: The author with his homemade, 90 degree, “railway curve” made from MDF board. At an early stage in the building of the layout I decided to use track supplied by the P4Track Company. Two types of track are available, 95lb Bullhead and modern 113A Flat Bottom with Vertical switches and crossings. In choosing Bullhead I have had to make a compromise. The BH turnout kits are based on LNER designs whereas my layout represents a former Great Western main line. Whilst most modellers cannot tell the difference and even a professional permanent way engineer would need the use of a scale rule, the kits are indeed faithful models of LNER prototypes. For example a GWR B8 turnout was 1ft 4.5in longer than an LNER one. Probably the most obvious difference was the use of two bolt chairs by the GWR instead of three screws by most others. It would really take someone with a very sharp eye to spot the difference between an LNER turnout and a later BR equivalent, so my story is that the switches and crossings at Littlehempston were renewed in BR days. Fishplates Fishplates are ‘H’ shape and made of ABS plastic, see photo 3. They are incredibly small and need to be handled with tweezers. The fishbolts are part of the moulding with square nuts on one side and round bolt heads on the other. They are so small that I need to use a magnifying glass to see which side has the nuts. The instructions that come with the point kits say that the nuts should be on the outside of the rails. I cannot verify that this was LNER practice but when I worked on track maintenance on the Southern Region the practice was to turn the bolts around each year. Fishplates were greased every year and the work was required to be completed before the onset of warm weather. So that engineers and supervisors could identify which joints had been dealt with staff were required to reverse the bolts when they greased the threads. One year all the nuts would be on the inside and the following year they would all be on the outside. During the early part of the year there would be a mixture as work progressed. So for true authenticity one must find out which way round the nuts were orientated during the year represented by our model railway and install the fishplates accordingly (just joking!). Photo 3: Close-up of a rail joint. Note the square bolt heads on the fishplate. A fishplate has been placed against the rail to show the ‘H’ shape. It wasn’t until this joint was seen through a macro lens that it was realised just how wide the rail gaps were. The rail positions have now been adjusted. I am finding that the ABS fishplates are very fragile and I am having a large number of breakages. They have very little strength and unlike the Peco style of rail joiner do not hold the rails in vertical alignment. I am finding that the thickness of the glue used to fix the track needs to be consistent. I am using Selleys Kwik Fix, which appears to be identical to Evostic in the UK. Pressing the track down close to the joints can result in the fishplates breaking. The technique I have adopted is not to stick down the last two sleepers at the ends of lengths of flexitrack. When the glue has hardened I run my fingers along the rail heads to feel which side of the joint is higher. If one side is higher than the other I raise the lower side by sliding craft knife blades under the sleepers and pressing them in just far enough to raise the rails to be level. Care must be taken not to lift the rails too high or the fishplates will break. Finally I run SuperGlue in under the sleepers either side of the joint. The craft knife blades are removed the following day. Droppers Since the fishplates are made of ABS plastic every joint is insulated. It is always good practice to attach a dropper to every rail and not to rely on rail joiners for electrical conductivity, but in this case there is no choice. I have always advocated soldering droppers to the underside of rails before the track is glued down. The result is much neater than having wires soldered to the sides of rails afterwards. However, with everything now being much finer I have to take greater care with soldering. Both the rail and the wire need to be carefully tinned beforehand because I am using steel rail. Another problem, which came to light, was difficulty handling the track panel when fixing down if the dropper wire was too stiff. Because the soldered joint is very small on the underside of fine bullhead rail it can easily break if not handled carefully and thick wire creates strain. I found hook-up wire from Dick Smith’s to be most suitable and it is available in a range of colours if required. Now, of course, thin wire has a higher resistance than thick wire. Initially I had thought of providing two droppers for each rail. However, only one locomotive is likely to be drawing current through a dropper at any one time. Also the resistance of a wire is proportional to its length so if the dropper is connected to heavier wire immediately under the baseboard its effect can be kept to a minimum. Rail Length The maximum length of rolled rail available to the railways of Great Britain before the 1990’s was 60ft. For a 4mm scale layout to be authentic therefore the maximum length of any rail should be 240mm. There are two advantages to adopting a scale length. Firstly it indirectly increases the number of droppers thus reducing electrical resistance. Secondly short lengths of track are easier to handle when fixing down. The race against glue hardening to get the track in the right place is easier with only a short length to concentrate on. Tips for Laying Track Always use a mirror to take a sight along the rails when gluing down. It is easier than trying to place one’s head on the baseboard to eyeball the track. Before gluing down have a dry run and place drapers pins into the trackbed to hold the sleepers in alignment. Place pins against the ends of sleepers, not through them. Then when the glue is applied the panel of track can be dropped into the right place using the pins to locate it. I use two pins at each end plus two near the middle of the panel. If the rails won’t take up the desired alignment I might add further pins as necessary. Hidden Sidings The hidden sidings are an area where we try to economise on the cost of track. My layout plan allows for an eventual ten hidden sidings with two reversing crossovers, making a total of 20 points. Clearly it would be desireable to find a low cost method of construction. The most obvious method would be to use copper clad sleepers with rail soldered in place. However, I have not had that much hands on experience of track construction and P4 requires finer standards than Finescale 00 or EM. Another problem we need to be aware of is the interfacing of two types of track with different rail heights (measured from top of rail to underside of sleeper) and possibly with different rail profiles at a location, difficult to access, in a tunnel or under a bridge at the entrances to the hidden sidings. On balance I decided to continue with point kits and flexi track from the P4Track Co. in the hidden sidings, thus building the entire layout with just one type of track. In making this choice I also decided that I would concede to simpler methods of construction in the hidden sidings where authentic appearance was not essential. The track is being pinned down rather than glued, the maximum rail length available of 0.5m will be used, Peco metal fishplates will be used where possible rather than finescale ABS plastic fishplates and some cosmetic chair parts will be omitted. Pinning the track in place makes alterations and additions easier at a later date. Present Progress At the time of writing (September 2009) the head of steel has progressed about two thirds of the way around the continuous run.
The building of Littlehempston has reached the stage where all of the baseboards are complete, the trackbed has been prepared and a sub-base of cork has been laid. It is time now to begin laying the track.
Photo 1: A draughtsman's "railway curve" being used to check track alignment.
The layout was initially designed on computer using Turbocad software. The first task was to mark out the track alignment using coordinates obtained from the computer design. As the layout is rectangular in shape with sides at right angles, the edges of the baseboards could be used as a reference. Significant points in the design are measured on the computer to the baseboard edges and then redrawn full size on the layout. This process, at best is only accurate to about 5mm. The method I adopted was to set out the straights first and then draw in the curves using “railway curves”. Sometimes they are referred to as “draughtsman’s curves” but they did originate in railway drawing offices and “railway curves” is their correct name even when used for other purposes such as highway design, see photo 1. The long curves at the ends of the layout have a radius of 5 feet on the inside track (Down Line). Because the curves are so long the use of “railway curves” is impracticable. To overcome the problem I made my own curve from mdf board. It provides a 90-degree arc of 5ft radius, see photo 2.
Photo 2: The author with his homemade, 90 degree, “railway curve” made from MDF board.
At an early stage in the building of the layout I decided to use track supplied by the P4Track Company. Two types of track are available, 95lb Bullhead and modern 113A Flat Bottom with Vertical switches and crossings. In choosing Bullhead I have had to make a compromise. The BH turnout kits are based on LNER designs whereas my layout represents a former Great Western main line. Whilst most modellers cannot tell the difference and even a professional permanent way engineer would need the use of a scale rule, the kits are indeed faithful models of LNER prototypes. For example a GWR B8 turnout was 1ft 4.5in longer than an LNER one. Probably the most obvious difference was the use of two bolt chairs by the GWR instead of three screws by most others. It would really take someone with a very sharp eye to spot the difference between an LNER turnout and a later BR equivalent, so my story is that the switches and crossings at Littlehempston were renewed in BR days.
Fishplates
Fishplates are ‘H’ shape and made of ABS plastic, see photo 3. They are incredibly small and need to be handled with tweezers. The fishbolts are part of the moulding with square nuts on one side and round bolt heads on the other. They are so small that I need to use a magnifying glass to see which side has the nuts. The instructions that come with the point kits say that the nuts should be on the outside of the rails. I cannot verify that this was LNER practice but when I worked on track maintenance on the Southern Region the practice was to turn the bolts around each year. Fishplates were greased every year and the work was required to be completed before the onset of warm weather. So that engineers and supervisors could identify which joints had been dealt with staff were required to reverse the bolts when they greased the threads. One year all the nuts would be on the inside and the following year they would all be on the outside. During the early part of the year there would be a mixture as work progressed. So for true authenticity one must find out which way round the nuts were orientated during the year represented by our model railway and install the fishplates accordingly (just joking!).
Photo 3: Close-up of a rail joint. Note the square bolt heads on the fishplate. A fishplate has been placed against the rail to show the ‘H’ shape. It wasn’t until this joint was seen through a macro lens that it was realised just how wide the rail gaps were. The rail positions have now been adjusted.
I am finding that the ABS fishplates are very fragile and I am having a large number of breakages. They have very little strength and unlike the Peco style of rail joiner do not hold the rails in vertical alignment. I am finding that the thickness of the glue used to fix the track needs to be consistent. I am using Selleys Kwik Fix, which appears to be identical to Evostic in the UK. Pressing the track down close to the joints can result in the fishplates breaking. The technique I have adopted is not to stick down the last two sleepers at the ends of lengths of flexitrack. When the glue has hardened I run my fingers along the rail heads to feel which side of the joint is higher. If one side is higher than the other I raise the lower side by sliding craft knife blades under the sleepers and pressing them in just far enough to raise the rails to be level. Care must be taken not to lift the rails too high or the fishplates will break. Finally I run SuperGlue in under the sleepers either side of the joint. The craft knife blades are removed the following day.
Droppers
Since the fishplates are made of ABS plastic every joint is insulated. It is always good practice to attach a dropper to every rail and not to rely on rail joiners for electrical conductivity, but in this case there is no choice. I have always advocated soldering droppers to the underside of rails before the track is glued down. The result is much neater than having wires soldered to the sides of rails afterwards. However, with everything now being much finer I have to take greater care with soldering. Both the rail and the wire need to be carefully tinned beforehand because I am using steel rail. Another problem, which came to light, was difficulty handling the track panel when fixing down if the dropper wire was too stiff. Because the soldered joint is very small on the underside of fine bullhead rail it can easily break if not handled carefully and thick wire creates strain. I found hook-up wire from Dick Smith’s to be most suitable and it is available in a range of colours if required.
Now, of course, thin wire has a higher resistance than thick wire. Initially I had thought of providing two droppers for each rail. However, only one locomotive is likely to be drawing current through a dropper at any one time. Also the resistance of a wire is proportional to its length so if the dropper is connected to heavier wire immediately under the baseboard its effect can be kept to a minimum.
Rail Length
The maximum length of rolled rail available to the railways of Great Britain before the 1990’s was 60ft. For a 4mm scale layout to be authentic therefore the maximum length of any rail should be 240mm. There are two advantages to adopting a scale length. Firstly it indirectly increases the number of droppers thus reducing electrical resistance. Secondly short lengths of track are easier to handle when fixing down. The race against glue hardening to get the track in the right place is easier with only a short length to concentrate on.
Tips for Laying Track
Always use a mirror to take a sight along the rails when gluing down. It is easier than trying to place one’s head on the baseboard to eyeball the track.
Before gluing down have a dry run and place drapers pins into the trackbed to hold the sleepers in alignment. Place pins against the ends of sleepers, not through them. Then when the glue is applied the panel of track can be dropped into the right place using the pins to locate it. I use two pins at each end plus two near the middle of the panel. If the rails won’t take up the desired alignment I might add further pins as necessary.
Hidden Sidings
The hidden sidings are an area where we try to economise on the cost of track. My layout plan allows for an eventual ten hidden sidings with two reversing crossovers, making a total of 20 points. Clearly it would be desireable to find a low cost method of construction. The most obvious method would be to use copper clad sleepers with rail soldered in place. However, I have not had that much hands on experience of track construction and P4 requires finer standards than Finescale 00 or EM.
Another problem we need to be aware of is the interfacing of two types of track with different rail heights (measured from top of rail to underside of sleeper) and possibly with different rail profiles at a location, difficult to access, in a tunnel or under a bridge at the entrances to the hidden sidings.
On balance I decided to continue with point kits and flexi track from the P4Track Co. in the hidden sidings, thus building the entire layout with just one type of track. In making this choice I also decided that I would concede to simpler methods of construction in the hidden sidings where authentic appearance was not essential. The track is being pinned down rather than glued, the maximum rail length available of 0.5m will be used, Peco metal fishplates will be used where possible rather than finescale ABS plastic fishplates and some cosmetic chair parts will be omitted.
Pinning the track in place makes alterations and additions easier at a later date.
Present Progress
At the time of writing (September 2009) the head of steel has progressed about two thirds of the way around the continuous run.