A tale of three turnouts (2)–my approach to turnout building

While this story was about three turnouts, all were built in the same way. Here’s how I did it.

With the rails for the straight route in place, it’s time to establish the diverging route through a turnout.  The next rail laid will be that of the diverging route.

I use offsets to determine the routing of the curved closure rail through the turnout. Both NMRA and the prototype CN turnout data that I use give these offsets.  NMRA recommended practice RP-12x available for free download at


gives their standard turnout data for many model rail scales.

CN no. 7 turnout offset diagram

CN no. 7 turnout offset diagram.  From the point of the frog in this example, measure towards the switch points a scale 56′-1″ using a scale rule.  This is the location of the toe, or the tip of the switch points.  With this established, use a fine-tip marker pen to mark the location of the offsets and the toe of the switch.

SAFETY FIRST.   Wear eye protection.  DO NOT solder or use any power tool without wearing the appropriate safety glasses or goggles. 

Ask yourselfhow much do you value your eyesight? 

Curve some rail gently, then starting at the frog, measure carefully at each offset while you solder this curved closure rail in place.  Be as accurate as you can.  I use separate switch points, so the curved closure rail ends a scale 11′ from the toe (tip) of the switch points.

CN no. 7 turnout diagram detail of bend in stock rail

When laying turnouts by hand, it’s necessary to bend the stock rail of the diverging route to receive the switch point for the straight route.  This CN standard no. 7 turnout diagram shows how it’s done on the real railway.  Bending the diverging route’s stock rail ahead of the points in this fashion is a common practice on North American railroads;  UK railways use a “joggled” double bend, see–http://www.templot.com/martweb/gs_realtrack.htm

The stock rail is bent per the above diagram for the no. 7 turnouts that I am laying. This is followed by filing off the base of the rail for a scale ten feet or so at this bend to receive the switch point.  I place the bend of the stock rail about a scale nine inches from where the tip of the switch point will go.  Using an NMRA Standards Gauge to make sure that I am laying this rail down true to gauge, I solder the stock rail in place for the diverging route, gauging off the previously laid curved closure rail.  As I am using Code 88 wheelesets on my rolling stock, I lay as tight a gauge as possible through the turnout.


Point end of an HO scale CN no. 7 turnout, showing bend in stock rail to receive switch point for straight route.  The base of this point for the normal route is soldered to the PC board throw-bar for strength;  the point for the diverging route is soldered to a Detail Associates’ .010″ x .018″ brass wire spike-shaped pin which is inserted into a slot drilled into the throw-bar.  This allows the pin to move while still maintaining this point’s location relative to the stock rail.  A little tweaking is still needed, as this layout has been in storage and the point for the straight route has gotten bent a little.  

With only the switch points to be installed, I drill a 1/16″ diameter hole between the ties at the heel end of the point location.  In this hole goes a short piece of 1/16″ o.d. brass or copper tubing.  A longer piece of 1/32″ diameter brass wire is bent 90 degrees at one end for about 1/8″ to form a support leg for the base of the rail at the heel of the switch point to be soldered to, and a hinge pin for the point.  I tin rail base and that 1/8″ long wire leg, then drop the wire through the tube.  The bent part of the wire is held in place to solder the rail to it by a spare switch tie on top of the rails.  It’s then an easy matter to solder the rail to the wire, judging by eye the location of this rail to line it up with its mating closure rail.

Take out the tie and drop the point into place.  Move the point around and check for binding, adzing off a thin layer of wood from the top of any tie that it binds on.  Do the same for the other switch point.

Cut a PC board throw-bar to length.  Drill a hole in the middle of it to receive the wire from your choice of switch-throwing device–Blue Point, Cobalt, Tortoise, slide switch, etc….

Take out the switch points and drill three 1/8″ diameter holes between the scale 16′ long headblock ties that are at the toe of the switch, between the stock rails.  I use a Dremel-type drill for this, using slight side pressure to connect the holes to form a 3/8″ long slot between the ties for the throw wire between throw-bar and switch machine.  Take your time; it’s easy to break a drill bit when abusing it like this.  Alternatively, you can use a Dremel milling bit to connect the holes to form a slot.

Take the PC board throw-bar and be sure to gap it on both sides before sliding it in place between the headblock ties.  Replace the switch points, and solder the base of the point for the normal route through the turnout to the throw-bar.  Move the point back and forth, ensuring that the throw-bar does not bind.

The point for the diverging route is soldered to a piece of pre-tinned Detail Associates’ wire as in the above photo.  A piece of tie with a hole drilled in it holds the wire pin while soldering it to the base of the rail in the same manner as a track spike. Drop this pin/point assembly into the throw-bar and check for binding.  If this assembly does not bind, you can insert the throw wire into the throw bar and connect it to your choice of switch machine or hand throw device under the roadbed.  The throw wire should not protrude more than a few thou above the throw-bar, and must not protrude above the top of the rails.  Check for binding of the throw wire in the slot that you’d cut in the roadbed.


The completed turnout, built following CN no. 7 turnout diagrams and plans. Looking a bit rough after soldering the rails in place, a little cosmetic titivation is needed to improve its appearance.  Another layer of ballast, especially! 


A look at the frog and guardrails.  Following CN practice, the guardrails are the same 11′ length as the points of this turnout. The centre-point of the guardrails must be located at the point of the frog. The paper template under the turnout did not line up perfectly with the established track centrelines; I used it merely for tie spacing here.

And now, off to something new…


A tale of three turnouts (1)



Lindsay’s west crossover between No. 1 Siding, with the turnout to the Haliburton Subdivision in the foreground.  Turnouts are CN no. 7. 

After I had laid Lindsay’s yard tracks 3 through 6 using Code 55 rail, I spun the section of layout around 180 degrees and laid the Campbellford/Midland Sub. main track, No. 1 siding and no. 2 yard track using Code 70 rail, to represent the 100-pound rail that CN used on these tracks in my modelling era.

I’ve already described my method for laying ties and ballast down; so I won’t repeat it here. I used differing tie spacings in these tracks, with the main track using 18 ties per 33′ rail length, No. 1 Siding laid with 16 ties per 33′ rail, and yard track #2 using 14 ties per 33′ rail.   All are in accordance with CN 1939 standards for each track.

This spot in the yard contains a crossover between No. 1 siding and the main track of the Midland Sub.  There is a short straight section of track  between the crossover and the switch defining the start of the Haliburton Sub.  CN would likely have used Standard no. 10 turnouts, but I found that CN no. 7 turnouts work quite well for layout use.  The closure rail radius in a CN no. 7 turnout is far greater than the 30″ or 32″ minimum that I will use elsewhere on the layout, so a larger turnout to me is overkill and a waste of valuable layout space.

On a previous section of layout I found that CN no. 7 turnouts and crossovers are long enough to allow full-length passenger cars and steam loco’s to run through them reliably.  And a CN no. 7 turnout in HO scale is about an inch longer than an NMRA no. 6 turnout, and shorter than an NMRA no. 7. Quite a space saving for the modeller compared to NMRA no. 7 or 8 turnouts.

CN no. 7 turnout diagram0001

CN Standard no. 7 turnout, 1939. Note the “Diagram of Bend in Stock Rail” and “Offset Diagram”.

Taking out the CN 1939 Maintenance-Of-Way rulebook, I scanned the no. 7 turnout diagram in .pdf.  I was then able to enlarge the turnout plan to HO scale and print off copies.  Carefully sited to the established track tie end lines and and glued onto the roadbed with spray adhesive, these turnout plans became the reference for laying down ties. I also used pushpins to indicate where the point of frog was on each turnout, as the drawing would be obscured by ballast once the ties were laid down.  Measurement using an HO scale rule established the lead as 56′-1″ between switch point and frog point.

I intersperse Fast Tracks’ PC board ties with wood ties in turnouts.  Some shimming is necessary to get the PC board ties to the same height as the wood ones.   But when the result is a turnout that I may never have to adjust the gauge on again, the extra work to use PC board ties is worthwhile. 


Careful work is called for when aligning turnouts in a crossover to ensure that the rails in the connecting track will line up when track is laid.  “Sin in haste and repent at leisure.”  You’ll notice that tie lengths are called out on the drawing.  I pre-cut ties from scale 16′ lengths to lengths in increments of six inches from 9′ to 15′-6′.  These are stained using the Jack Work leather dye/alcohol method and glued in place per the tie lengths called for on the diagram.


Almost ready to lay rails down.  The blue-headed pushpins indicate frog points–simply measure scale 56′-1″ to the long headblock ties sited at the tip of the switch points to establish where the switch points will be.  Again, it’s worth taking the time to make sure that turnouts are laid out properly.  Plain track is laid after turnouts in a situation like this.  It’s far easier to adjust plain track slightly while laying to align it with turnouts, than to rebuild the turnouts to suit the plain track that they are connected to.  

The location of the point of the frog determines how track will flow through a turnout, so I lay the frog first. It used to be that one would spike the frog in place with spikes in every tie for stability, but Fast Tracks’ PC board ties have made building equally or more reliable track so easy now. Gauge the adjacent straight stock rail off the frog using an NMRA Standards gauge, then lay all straight rail that you can in both directions from the frog and the previously gauged straight rails. This includes the straight closure rail. Don’t forget to file the stock rail base off to receive the switch points.

An easy way to check if rail is straight is to simply sight down it.  Your eye will pick out out-of-line rail immediately.  Be sure that one rail is straight, and the NMRA Standards or three-point gauge that you use will help you lay the other rail straight as well.

Lindsay west crossovers in HO scale 001

Some rail has been soldered in place using regular 60/40 electrical solder.  I clean the PC board ties off using a sanding stick , sand the bottom of the rail using 400 grit paper, and immediately solder the rail in place.  I do not have to use rosin or acid flux as a result.  Frog assembly, switch points and guardrails are premade and placed in an envelope–each envelope contains one complete set for a turnout.

With the straight rails laid, we’ll look at laying the rest of the turnout in the next installment.

Thinking of the people who built our railways, there is a lot of truth in this song from the band Genesis.  My grandfather was a track maintainer at the Steel Company of Canada in Hamilton, Ontario for many years.  The railway cannot run without track maintainers.

“We followed the rail, we slept under the stars
Digging in darkness, and living with danger
Showing no fear of what lies up ahead
They’ll never see the likes of us again.”


Modelling Lindsay’s track scale (2)

I laid the scale and its approach tracks last on this side of the yard.  Being nearest to the aisle, I decided to solder the Code 55 rails of this track to Fast Tracks’ PC board ties rather than spike them to wood ties.  This makes it more difficult to damage the track via wayward elbows, etc.  Most likely it’s my wayward elbows that would be the culprit anyhow.

First, that picture of Hamilton’s scale to compare with my model of Lindsay’s.

CN scale house, Hamilton, 1993.  Steve Lucas photo.

CN scale house, Hamilton, 1993.  My photo.  Notice how the rails on the scale platform are higher than those in the rest of the yard–the substantial downgrade on the far side of the scalehouse is proof of this.  I will model this feature. The scalehouse sits on a concrete base and is elevated a bit above the scale platform.

Here’s my model along with a few notes–


Ties are being laid to gradually bring the height of the scale rails from the rail level in the yard tracks up to place the rails on the moulded scale platform from the Moffett Models kit.  Tie sanding to level them relative to each other, creosote coloured tie stain, and several layers of ballast are to follow.  Fast Tracks’ PC board ties are interspersed with wood ties having wood shims under them–the rails will be soldered to the PC board ties. The scale house is part of the Moffett Models kit, but I made a mould for and cast the “concrete” base under it. 

Lindsay yard and scale track 25-1-12 006

The ties having been ballasted, unweathered Code 55 rails were soldered to the Fast Tracks’ PC board ties and to flat-head brass machine screws driven through the resin scale deck of the Moffett Models kit.  Two of those screws do triple duty as screws to solder rail to, DCC feeders to the scale rails, and attaching the scale deck to the layout..  Prototypically, there is a gap between scale rails and those on the approach tracks.  The “switch points” either side of the scale on the approach tracks are expansion compensators to keep the scale rails from binding against the rails of the approach tracks.   On the prototype, normal rail expansion will result in the rail expanding downgrade toward the compensator, past the switch points, and thus prevent the rails’ binding against the scale rails.  These are a simple device for the local section forces to keep in good repair.  I modelled the rails on the scale deck as being about six inches higher than the rest of the yard. 


Lindsay yard and scale track 25-1-12 004

A view which I hope will show the difference in elevation between rails on the scale deck and the tracks in the rest of the yard.   There is still some scenic work to do.  Scale six-inch-wide or so “flashing” is to be added to both sides of the rails, touching the web of the rails to cover what would be the top of the weigh scale mechanism of the prototype.  The deck on track scales is tarred as well, roofing material commonly being used in the manner of that applied to a flat roof.  The layout edge will have to be brought out about half an inch to support the base of the scalehouse–though I’d like a little bit of ground behind it as well for the viewer to imagine that this is in a real location rather than on the edge of a layout.

Modelling Lindsay’s track scale

“And distant memories are buried in the past forever…”—–The Scorpions, Wind of Change.

The model rail hobby can bring those memories back in 3D.  Here’s how.

CN Lindsay scalehouse & scale pit foundation0001

What was left of Lindsay’s track scales in the fall of 1986.  This part of Lindsay’s Durham Street yard was torn up in the fall of 1964, leaving the scale house isolated until torn down circa 1967.  The scale house foundation is the concrete pad to the right; the foundation for the scale pit itself is faintly visible poking through the grass.  My photo.

Track scale stuff0001Sample (for illustration only) scale ticket from the CN General Weighing Instructions manual, 1970. The weight figures “117800” looking like they are typed in, were printed by metal type in the scale mechanism by the person weighing when the balance beam on the scale was balanced. This was the total (gross) weight of the car; subtracting the car’s empty (tare) weight of 43,800 pounds results in the shipper being billed for 74,000 pounds or 37 tons of freight at the applicable tariff rate.  As the car on this ticket is rated for a total rail weight or load limit of 133,200 pounds, it is not overloaded.  Keep in mind that a scale ticket is a legal document.  Ray Kennedy, on his Old Time Trains website, states that weighmen or yard foremen (yard conductors) had to take a sworn oath that they would favour neither the company nor the shipper when weighing cars! 

Lindsay, like many CN terminals, had a track scale for weighing cars travelling from the branchlines for furtherance out of Lindsay and also to establish an accurate empty car weight to be re-stencilled on them after being repaired.  By the time that I came on the scene to document what was left of CN’s Lindsay yard in the mid to late-1970’s, most of its railway facilities were gone. What I had left to work with was the foundation of the scale and scale house, in the above photo.

Lindsay0007 copyTo the left of this photo are cars awaiting weighing on Lindsay’s track scale, 1952.  The incline up to the scale is hardly noticeable, but very necessary.  From the collection of the late Al Paterson.

Years later I made the acquaintance of Keith Hansen.  His book, Last Trains from Lindsay, is an essential read for those interested in the railways of the Lindsay area. He supplied me with some 1965 CN Bridge and Building plans and diagrams for Lindsay and the Campbellford Subdivision.  One of the plans, dated 5 November, 1965,  contains a description of Lindsay’s scalehouse.  Notes indicate that it was fitted with a Fairbanks-Morse 100-ton scale. Yes, that Fairbanks-Morse who also built locomotives.

The floor plan and written description shows it to be very similar to CN scalehouses across the system, including those that stood at Hamilton, Brantford, Palmerston, etc.  Hamilton’s scalehouse I am rather familiar with, having worked in Hamilton for a number of years from when I hired on and also weighed few cars there.

CN scale house, Hamilton, 1993.  Steve Lucas photo.

CN scale house, Hamilton, 1993.  My photo. Note how the scalehouse is elevated above the scale deck on its own raised foundation.  This kept the scale mechanism under the rails and scalehouse above water, the yard being partly on landfill. The stacks may have been for a coal or oil stove, or maybe they simply served as vents for the scalehouse basement and scale pit.. There was a small three-pane window at the rear of the foundation. Only live rails are provided on this scale.

The track to the scale gained elevation as it approached the scale on most CN installations to raise the scale mechanism above the water table in an effort to keep it dry, and also to permit easy and accurate weighing of cars. You can see this track incline towards the scale in the above Lindsay photo if you look for it.  Here’s how cars were weighed at CN.  The description of poor weighing practice is almost folksy–

Track scale stuff0002Cars were switched onto the scale track by the yard crew.  They were allowed to run over and past the scale if necessary, as the scale mechanism, when locked, allowed normal traffic to pass over the live rails without injury to the scale mechanism.  Some CN scale installations such as at Brantford used live and dead rails and switches at either end of the live rails, but I’ve seen more CN track scales with just a pair of live rails.

When it came time to weigh cars, the crew would pull out the entire cut from the track. The first car was weighed, then kicked by the loco and uncoupled in motion to allow it to roll away from the scale with a yard helper (brakeman) working the handbrake to stop it where desired.  In this way, cars could be easily weighed uncoupled at one end.  After each car was weighed, it too was kicked towards the cars already weighed and secured with the necessary number of hand brakes.  Scale tickets were prepared as each car was weighed. When finished weighing, the yard crew could switch out the entire cut of cars by destination. The scale tickets were stapled to the waybills (more legal documents) for each car.

To verify the accuracy of track scales, the railways used scale test cars of a known weight.  In my 1956 modelling era, these were four-wheel cars with handbrakes only and roller bearings.  Scale test car weights were often multiples of 20,000 pounds, 60,000 pounds being common.  A through 1 1/4″ inside diameter heavy-wall AAR standard brake pipe ran under the car so that train air brake continuity could be obtained between loco and caboose, which these cars were hauled just ahead of in a train. The lack of air brakes reduced the change in weight due to cast iron brake shoe wear as time went on.  The use of roller bearings on these cars reduced fluctuation in weight due to journal bearing brass , babbitt, and journal wear, as well as the necessary addition of oil to plain bearing journal boxes from time to time.

CP scale test car 420926

CP scale test car in the Toronto, Hamilton and Buffalo Railway’s Aberdeen Yard in Hamilton, ON, 1983.  Roller bearing journal boxes are the same used on CP’s streamlined 2200-series coaches and baggage cars.  I wonder if CP just used spare passenger wheelsets for this car.  My photo, taken using Ilford XP-1 black and white film (if you are old enough to remember that stuff).  Stencilled on the car side are certification dates stating when inspected by the Canadian Department of Weights and Measures.

A method for checking the accuracy of track scales absent a scale test car to do so was to do “three-spot weighing”, where the three weights obtained had to concur.  A scale test car , if available, was ideal for this due to its short 8′ or so wheelbase.  Again, from CN’s instructions–

Track scale stuff0003

We’ll come back with how I modelled Lindsay’s track scale in the next installment.