Blog Archives

Kankool trackwork – Part III …

Hi all,

Following on from my previous post on the tiebar installation, I have been carrying out testing of the new tiebars and turnouts in situ.

As outlined in a previous post, the wire-in-tube (WIT) hardware were mounted on timber bases directly underneath the spline at each turnout location.

The wire from the turnout back to the lever frame was temporarily connected to the frame using brass screw ferrules from terminal blocks.

wire-in-tube termination at lever frame


During initial testing, I found that length of the vertical drive shaft attached to the brass block had too much flex in it to work the turnout reliably.  Due to the thickness of the spline at around 25mm, the resultant drive shaft length was around 30mm.  The image below shows the vertical drive shaft and brass block. (Note – image shows the revised drive shaft length)

drive shaft


It was also found that the length of tube attached to the underside of the tiebar was too long.  So this was reduced to around 3mm in length.

new tiebar throw tube length


To reduce the drive shaft length, I realised that a reduction in spline thickness would be necessary at each turnout location.

After carrying out some tests with different drive shaft lengths, I found that a length of around 15 to 17mm operated much more reliably.  I decided to reduce the spline thickness to 10mm.  This was removed quite easily with my multi tool and saw blade.  At one of the other locations, I had to use a router.

reduced spline thickness


Once the spline was cut away at each location, the previously made timber mounts for the WIT hardware were re-attached to the spline by gluing in place.


I am quite happy with the result, so the next stage will be to paint the turnouts and fix in place.

I have made a short video showing the turnout operating from the lever frame.



More foam roadbed & CV track installation …

Hi all,

I have completed installation of the foam roadbed all the way from Chilcott’s Ck almost to Ardglen.  I have also laid quite a bit more CV track base from Chilcott’s Ck through Kankool to just before the first peninsula.

The process was the same as outlined here when I did the Temple Court section.

Under where the turnouts are located at Kankool, I laid 3mm cork down instead of the foam.  This was done to make it easier cutting out a hole for the future throwbar rather than trying to cut it out of the soft foam.  In hindsight, I probably only needed a small section of cork directly under the toe of each turnout rather than under the whole turnout.  That is what I will do on future turnouts.

cork laid under turnouts

cork laid under turnouts


Taking on board an idea from Andrew, I also commenced installing strips of the high density yellow foam to the sides of the spline.  The idea here is to lay continuous foam strips along the spline to not only allow for shaping the trackbed, ballast profile and drainage, but to also make it easier to install the main scenery foam later on, rather than trying to cut many small irregularly shaped foam pieces to try and fit against the spline.

scenery foam strips attached to spline



New vee crossings & more turnout progress …

Hi all,

Following on from this post back in August 2015 where I outlined the move from Proto-87 standards to NMRA Fine:HO, the new Code 70 vee crossings arrived back in November courtesy of Keiran Ryan.

The picture below show the two new vee crossing etches, 1 in 6 on the left, 1 in 8 on the right.

new vee crossings


The picture below shows a completed crossing prior to being removed from the fret.

completed crossing


After I then assembled a few of the crossings, I proceeded to remove the P87 ones from the two turnout assemblies for Kankool and retrofit the new ones.  The 1 in 8 is on the left, the 1 in 6 on the right.

1 in 8 crossing1 in 6 crossing









Below is a shot of the completed turnout complex at the Werris Ck end of Kankool.  All that is left to do here is to fit the throwbars and paint.

Kankool turnouts


Before I could start laying the foam roadbed, I had to make cut-outs in the spline at the throwbar locations for each turnout at Kankool.  I used a multi-tool with a small saw blade held vertically.

throwbar hole

throwbar hole

throwbar hole



Spline now 80% complete …

Hi all,

Over the last few months, I have completed all spline work within the Ardglen area, including the ballast sidings, and also to a point about halfway between Ardglen and Pangela.

This now means I have completed construction of 80% of total spline, which equates to 292 feet (88.5 metres) with just 74 feet (22.5 metres) of spline left to build through Pangela and to connect up to the helix.

The two images below are views of Ardglen from either end, showing the storage and goods sidings.

Ardglen spline (tunnel end)

Ardglen spline looking towards tunnel











The next image is a shot further down the grade from Ardglen, looking back towards the ballast sidings coming off the quarry road.

Ardglen spline with ballast sidings


The next image shows the shunting neck at the end of the quarry road.

Ardglen spline with quarry road shunting neck


The next image shows a general view of the section above Temple Court.  I have yet to finalise setting the grade in this section.  Temple Court has been covered with drop sheets to protect the scenery from dust and dirt whilst working above.


The next series of images show the spline from Ardglen tunnel through to a point about halfway to Pangela.


Layout progress update …

Hi all,

Apologies for the lack of updates since early Dec last year.  I have been busy with some scenery in the Temple Court section as well as some more benchwork and spline construction.

Temple Court scenery

I’ve made a start on some static grass application but have not done much since Christmas.  It’s been tricky getting the look and colours right, but I think it will look OK as it progresses.

Some pics below of the progress so far.

Temple Court scenery

Temple Court scenery

Temple Court scenery

I still haven’t done any more with the rocky outcrop, but you can see lots of plaster castings sitting around for inspiration.


Lower level backscene, upper deck benchwork and spline

I also started installing the backscene from staging through Kankool, including the entry from staging onto the Chilcott’s Ck bridge scene.

Backscene at Chilcott's Ck bridge scene

Backscene & upper deck benchwork at Kankool

Backscene & upper deck benchwork at Kankool

The last picture above shows the new supports installed for the upper deck above Kankool as well for backscene support.  As mentioned previously, I am using 3mm MDF as the backscene and this will be painted with the sky colour.

I had always planned to work on the lower deck scenery and trackwork from Chilcott’s Ck bridge to Ardglen first before the upper deck, mainly due to thinking I would have access issues on the lower deck for scenery work if I built the upper deck benchwork.  In the last few weeks I have since decided to progress with upper deck benchwork, spline and scenery from Ardglen through to Pangela as the next stage.

This all came about after I had started to think about getting the bridge in place and laying track from staging through Kankool.  But I realised that this may not be such a good idea as once the bridge was installed, and I had scenery in place, I would be working above all this with the ever present danger of things falling on the scene below.

So, the decision was made to continue with upper deck benchwork above Kankool and around to Pangela as well with spline construction from Ardglen to Pangela.

As I had already built the benchwork for Ardglen and the section above Chilcott’s Ck to Kankool some time ago (see previous post), I started to lay out the spline in this area.

Ardglen spline construction

Upper deck spline

As mentioned above, the plan is to continue with upper deck benchwork all the way from Kankool through Pangela and to build the spline from Ardglen to Pangela, then start on basic scenery formation.  Once all the messy stuff with plaster is done on the upper deck, I’ll probably then look at getting the Chilcott’s Ck bridge installed and continuing with scenery through Kankool and beyond towards Ardglen.

Cheers for now.

A revised ‘new location’ – Temple Court …

Hi all,

Well things have been pretty quiet since my post in late August.  During September, myself, Jodi and Connor went to the United Kingdom for a holiday.  We all had a great time but were glad to be home after being away for a month.

Work on the layout was pretty much non existent since we returned in early October, but a few weeks ago Andrew and I cut up some masonite sheets for more supplies of spline strips, and this spurred me on to complete the section of spline from the bottom of the helix round to the Down storage yard.

Since my post way back in January this year about incorporating an additional crossing loop in this section, I have since changed my mind again and decided to just make it a scene from the area around Temple Court, which is between Pangela and Murrurundi.  The main reason for not going ahead with Wingen, is that trains just wouldn’t have been in the right context for that location (I know – picky, picky! Winking smile ).  If an UP train was being banked in the front all the way from Willow Tree, the bank engines, which were sometimes attached “in the shafts”, normally would have been detached at Murrurundi, and therefore a train with bank engines attached would have looked out of place at Wingen, which is further south from Murrurundi.

I had come across a couple of great pictures showing a big granite rocky outcrop at Temple Court that was just begging to be modelled.

4538+4512+4497+48138 with Up No.606 freight, make their way past

a granite outcrop near Temple Court on 19 May 1982.  The 44 and 48 are the bank engines.

Photo by Mick Morahan

4854+44224+4483+44211 with an Up wheat near Temple Court in 1980.

Photo by Chris Nelson

I have also laid down the foam roadbed in this section in preparation for commencement of laying down the Central Valley tie strips.  The photo below shows the foam in place on top of the spline.

foam roadbed in place

I also installed a few more sections of the PECO flextrack from where I finished off when tracklaying in the helix.  See previous post.


There will be one more length of flextrack installed than in the photo above, which will be close to where this section of track will emerge into the new Temple Court scene.  The CV track will then be laid all the way back to the staging yards.

Over the next few weeks, I plan to carry out some tests on painting the CV track in readiness for laying.  I will keep you all updated as that progresses.


The Helix Part V …

Hi all,

Well, I have reached another milestone with the completion of the helix.  When I say “completion”, it’s not 100% complete as I plan to put some ‘guard sides’ up on each of the layers to afford some protection to prevent trains from taking a plunge if they derail, but I plan to do this later on.  But otherwise, the main structure is finished.

Following on from the last post, the next stage was to set the height of the first (bottom) layer to the correct height as this would then set the grade for the whole helix.

My initial attempt at this didn’t go so well, as I think I miscalculated the height of each sequential riser, eight in all.  The reason I say this, is when the rest of the layers were lifted up into position, something didn’t look right.  Some segments of the helix looked steeper than others.

helix erected with first attempt at 1st layer risers

The picture above shows the helix layers lifted up and spaced with temporary blocks at the correct height.  This picture was taken with the first layer risers that were incorrect.  To lift each layer up, it was just a matter of carefully lifting the whole thing and placing the temporary blocks in place at each of the eight support locations around the circumference.  This was repeated for each layer.

After some later discussion with my consultant (Andrew), and after some more measurements and calculations, we came to the conclusion that I had indeed stuffed up the initial height calculations.

Some new risers were then cut, and fixed in position.

first layer riser blocks cut

one of the risers fixed in place

The pictures above shows the new first layer risers all cut and installed.  These were glued to the L girders as well as the ply roadbed.  After these were fitted, the whole helix looked much better and with even grades.

I had originally planned to make notched risers to support the helix at the eight points, but could not work out a way to accurately cut these notches.  After some discussion again with Andrew, I decided to still use the 2×1 as the risers on the inside and outside, but just glue blocks in place to support the roadbed layers.  See the pictures below.

helix layer support blocks being glued in place

a completed helix support riser

The blocks were cut from some of the timber I had used to make the temporary spacer blocks.  They were cut to about 15mm thick so as to not encroach too much onto the track and to allow enough room for the longest piece of rollingstock to pass through with clearance.  The blocks are only glued to the risers and not the plywood.  This should allow for any slight expansion and contraction in the timber.

Just to finish off, below are a couple of shots of the finished structure.  I have also fitted the transition piece of ply roadbed from the top layer to where the spline will be attached at the 70 inch level.  I am surprised at how rigid the whole structure is now.

completed helix   completed helix

Where you can see the red clamp, I have placed a temporary riser to support the transition section where the spline will attach.

At some stage, I want to temporarily connect a throttle of some sort to test run a loco up the helix.

Cheers for now.

The Helix Part III …

Hi all,

Since the last post, there’s been more movement on the helix build.

My package of PECO Code 75 flex-track from the UK arrived on Monday (30th).  Before anyone starts on about why I didn’t buy it locally, I’ll tell you why.  I got fifty 914mm (3ft) lengths delivered from the UK for $173.75.  A quote I got from Brunel Hobbies in Melbourne was $176.60 (posted to Port Macquarie) for 25 lengths – double the price!!  Why should I buy locally?  If importers in this country feel the need to put such huge mark-ups on items sourced from overseas, is it any wonder they are complaining about internet sales taking away their business.  I won’t say any more on the subject.

Anyway, back to the topic at hand.

Before track laying could commence, I had to build part of the first section of spline that would exit from the helix and continue on towards the storage yards.  So, I did the usual thing of marking out where it would go using measurements from the 3rd Planit drawing.  I also made a short transition piece of helix roadbed for the spline to attach to.  I decided to only make a short section of spline so I could later remove it to rebate the top where it meets the helix.

short section of spline exiting helix

Whilst the spline was being made, I mulled over how to start laying the flex-track.  It’s funny to think that I have never really laid any flex-track in all my years of modelling.  Anyway, first step was to mark a centreline on the helix roadbed.  I then spent some time thinking of the best way to lay the track.  As I would require joins on curves, I didn’t really want to just ‘butt’ join the sections as this generally does not allow for a smooth curve through the join.  So I decided to stagger the joins.  With PECO track, both rails are able to slide through the moulded chairs, so I pulled one rail back about nine sleepers, and cut away a couple of sleepers where the join would be.  As I didn’t have any PECO rail joiners, I did the next best thing – a short piece of 30 thou brass wire soldered to the outside of each rail.  Two lengths of track were joined together whilst straight.

rail pulled back for staggered join

two sections joined with staggered joint

Some more time was spent deciding on the best adhesive to use to fix the track down.  I wanted to find something that would glue the plastic sleepers OK, be a strong bond, be fast setting but allow time to finalise placement.  I didn’t want to be waiting around for hours whilst glue dried.  After heading to Bunnings and looking in their adhesive section, I finally decided on a Selleys product called Aquadhere Quickset.  It sets in about 5 minutes but fully cures in around 30 mins, is water-based and dries clear.

I also had to think about how I was going to hold the track in position in the curve whilst the glue dried.  I decided to use small panel pins.  This PECO track does not have any holes in the sleepers, so I drilled every tenth sleeper to make a neat fit for the panel pins.

Next, the track was placed on the helix roadbed and formed roughly into the correct curve.  I managed to find a Tracksetta 36 inch radius template which will do the job nicely.

So basically, it was just a matter of placing the Tracksetta between the rails, ensuring the track was on the centreline and using the same small drill as before, drill a hole into the roadbed.  The small panel pins were then pushed through the sleepers into the roadbed.

tracksetta in place with panel pins holding the track

I also found out that my plan of using two lengths of track pre-soldered together was not the best move, as there was a bit too much pulling of the rail through the sleepers to adjust things on the curve as I went.  For the next and subsequent layers, I’ll lay just the three foot lengths in turn.  It’s much easier to pull the rail through the chairs in three feet rather than six feet.

first layer of helix track tacked down

The section of spline was removed from the benchwork and set up in a jig to allow for the rebate to be made to allow for splicing to the helix roadbed ply.

spline section in the jig for rebating

The picture below shows the spline now back in situ and glued to the helix roadbed.  The spline was also set at the correct grade which at around 1.8%, is slightly steeper than the visible section of the layout but should still be OK.  This is the same grade as the helix.

spline attached to the helix roadbed

I decided to only fix in place a short section of track on the spline merging with the helix track, so I cut a strip of the foam tape roadbed and glued it to the spline.  The foam was cut to the width of the spline, but to allow it to follow the curve, some cuts were made about 75% of the width to allow it to curve and remain flat.  The cuts are just visible in the image below.

foam roadbed in place on spline

Another length of track was cut to splice into the section already tacked down on the helix.  Some extra support for the spline was also installed here.

track ready for attaching to spline

Today (6th) I glued the first layer of flex track down on the helix and the spline.  For the spline section, I used some black caulking compound as a trial, as I was unsure how the Aquadhere Quickset would fix the plastic sleepers to the foam.  A lot of articles I have seen about laying the Central Valley tie strips to foam roadbed mentions using caulk.  The Aquahere Quickset was used on the helix sections though.  Using a thin metal spatula, I carefully levered up the track a small amount, still making sure it was held by the nails.  The glue was then dribbled over the sleepers, and using the spatula again, smeared underneath the sleepers and over the timber.  The track was then pushed back down firmly.  No weights were used to keep the track down as the nails were a firm fit in the holes in the sleepers, and seemed to hold OK.

gluing down the flextrack

Within half an hour, the glue had set enough for me to be confident the track was held in position.  I then decided to start the next layer of helix roadbed.  Some small blocks were cut to hold the new layer above the first whilst being glued and clamped.

second layer of helix roadbed being glued in position

With the number of clamps I have, I can only do about three segments at a time, but it should proceed reasonably quickly.  Once another roadbed layer is complete, the tracklaying process starts again for that layer and so on until I end up with five layers of helix roadbed with track attached sitting on top of one another.  The next stage will be to notch out the support risers, fix these in place, and lift the levels to the correct positions but that will be described in a later post.

I did also think about soldering the track feeder wires on prior to fixing the track down, but I think I can do this later.

Cheers for now.

Peninsula benchwork continues …

Hi all,

As I am still waiting on a parcel of track to arrive from the UK, work on the helix has come to a standstill.  In the meantime, I have been continuing on with some benchwork on the peninsula.  As the spline has been approaching the teardrop on the peninsula, the height of the risers has been getting quite high, which I found was resulting in a slightly unstable roadbed foundation, so I decided to build a second layer of joists to strengthen up the construction which in turn will shorten the risers.

This second layer of joists has also filled some of the ‘void’ between the spline and the bottom joists.  This will save a lot of scenery ‘filling’ for no reason.  Plus I didn’t really need so much depth below the spline.  Refer to the pictures below.

section of teardrop showing new layer of joists

the other side of the teardrop with new joists

The spline has also been set at the correct grade right up to the end of the teardrop.  The photo below shows just how much the track will climb from the bridge at Chilcotts Creek (the gap in the spline on the right) to the teardrop.  At this point, there is approximately a 13 inch (330mm) difference in height.  The distance to this point is around 81 feet (24 metres).

showing the grade from the bridge to the teardrop

The roadbed around the peninsula  is now starting to get to a good height.  I can just visualise the trains climbing this grade now!!

That’s all for now.  Hopefully the next update will be on the next stage of helix construction.

Cheers for now.

Setting the grade …

In this post, I will attempt to explain how I have gone about setting the grade for the spline from the bridge at Chilcott’s Creek, through Kankool and currently halfway to the teardrop.  The same process will continue right through to Ardglen where it levels out halfway through the tunnel.  Apologies for the length of this post, but I thought that putting it all in one post would be easier than splitting it.

Following on from the last post where I explained how I marked the 42” datum level around the layout using the water level, the next thing I did was to flick a chalk line on the walls against the marks made using the water level.  I now had my datum “zero” level around the walls to Kankool.

When it came to making the marks along the peninsula benchwork, I soon realised that where the 42” level was, was about halfway up the joists sitting on the main girders.  This initially proved tricky to get a true datum level, so I decided to mark out the 70” top track level on the peninsula walls using stringlines.  I now basically had two datum levels from where to mark out riser heights.

In all the articles I have seen about spline roadbed, they always say to fix the risers in place first where you want the roadbed.  I couldn’t see how this would work for two reasons :-

  1. I didn’t know exactly the route the roadbed would take.  That’s the beauty of the spline – it’s adjustable.
  2. How do you set the riser heights, and therefore the grade, without having the spline in place to check for dips or bumps?

So, hence the reason for me building the spline flat on temporary joists, then when the time comes, fix in place the proper joists at set intervals, and then the process of setting the grade can commence.

The beauty of the 3rdPlanIt program is that you can set a grade and have it display the track height (elevation) at any location along the track.

layout dwg showing joists

The screen capture above is from the layout drawing in 3rdPlanIt and shows the locations of the joists along the route.  The lines actually are the centrelines of the risers rather than the joists, but there is only a difference of about 20mm.

Once the grade has been set in 3rdPlanIt, a text marker can be placed along the track that shows the elevation at that particular point.

layout dwg showing elevations

The screen capture above now shows the elevation markers at the joists.  The values shown are based on a zero elevation at floor level, so because my datum level is 42”, I just took the difference between the two.  So now I had my roadbed heights, I could now mark these at each riser location.  Using a large square and spirit level, I drew vertical lines on the walls that lined up with where the risers would attach to the joists.  The roadbed levels were then made around the walls by just measuring up from the 42” datum line and making a mark.

When it came to making these marks along the peninsula walls, I just reversed the process by taking the difference between the roadbed elevation and the 70” datum and measured down the wall ‘stud’.

Now that I had all the elevations marked out, it came time to cut up some timber to use for risers.  Under the single track areas I used some 2×1 DAR and under the double track areas at Kankool, some 3×1.

I initially started to attach the risers to the joists first with the aim of then attaching the roadbed to the top of the riser later, but this proved to be not the right way to do it.  I instead decided to glue the riser to the underside of the spline first, using Liquid Nails, and when set, I could then fix the riser in position.  This would then also allow for slight twisting of the spline to ensure it is level.

attaching the riser to the spline

The photo above shows a riser being glued and clamped to the spline.  Note the riser is only temporarily clamped to the joist at this stage.

Now, a short side note.  Up until early this week, I had fixed the roadbed/risers in position all the way through Kankool and half way down the first peninsular.  The grade was looking good, then for some reason, when I was looking at the overall layout drawing, I realised that when I first drew the plan, I hadn’t allowed for the peninsula wall on the teardrop section, and consequently discovered I hadn’t allowed for enough scene depth behind the track at both Pangela on the top level and the corresponding track underneath it.  AARRGGH!!!!  So after about two hours redrawing this section of the layout and pushing the track out further to give around 400mm scene depth, I got it looking better.  I then realised that I would have to recheck the grade through this section.  When I did, it had changed slightly, which now meant all the roadbed I had previously fixed in place was about 7 mm high at each riser. *@#$%!!!!!!

Also, during this time when fixing some risers along the first peninsula, I came across another issue.  I think I must have used a cheap glue when fixing the riser to the underside of the spline, because I had to twist a couple of risers quite a bit to get the spline level at that point, and it was here I noticed that the glue was letting go.  So a solution had to be found.  I went through a number of ideas, but I think this will solve the problem.

strapping to reinforce riser glue

I had noticed that the problem lay generally on the inside of curves where I had to twist the spline to get it level.  So a quick trip to the hardware store to get some hoop iron and small wood screws.  Short sections were then cut and bent to shape for each location and screwed to the spline and riser.  I may have to rethink the riser to spline attachment process, but for now I have changed to a known brand of adhesive, namely Max Bond.  I may have to use the method described here and nail the spline to the riser to compliment the glue.  See pages 20 and 21.

So, after that issue was sorted out, I printed out another layout drawing showing the revised elevation heights and proceeded to remark the heights and drop the risers down.  In the meantime I have also added more joists on the teardrop and started laying out more spline along the other side of this peninsula.

Now, back to the elevation setting process.

With the spline sitting on temporary risers, clamped roughly at the right height, I had to devise a way of accurately getting the riser at the right height.  I ended up using a steel right angle square held against the wall at each location with it’s bottom edge lined up with the height mark, and either raised or lowered the riser until it touched the square.  The riser was then clamped in position and screwed to the joist, after checking for level (see below).  A small piece of scrap masonite was placed on top of the spline to allow for the foam roadbed, as all roadbed heights referred back to this datum ie 42”.

square in place setting the riser height

the elevation mark on the wall

The two images above show the square in position against the wall with the masonite packer, and a close-up of the elevation mark on the wall (the green line) and the 42” datum (the blue chalk line).  The red line was the previous elevation mark prior to the track changes on the peninsula mentioned above.  At this point in the process, before fixing the riser, a spirit level was placed on top of the spline to check for level.

levelling the spline

Well, that’s about where things are up to at the moment.  I hope it all makes sense, as sometimes I find it difficult to put into words how I do something.  As mentioned above, I have started to install joists further along the other side of the teardrop peninsula so I can resume spline construction here.

Over the last week I have also got my hands on some more photos of the area.  I posted a message on RailPage Australia asking for photos of the signals and structures at Kankool, Ardglen and Pangela.  I eventually heard from a couple of blokes who are ex train drivers, and are both sourcing more stuff for me.  Some I have already received from one are real beauties.  If anyone else out there has any photos of the area, please let me know.

Cheers for now.

%d bloggers like this: