2mm Found in the Attic!! - FS160 Switching Layout.


Western Thunderer
I was digging around looking for something recently and opened a drawer full of American N scale stock. Now I remember collecting all this stock some years ago. I wanted to get into DCC and I didn't have anything running in S scale at the time, so I opted to get a DCC fitted N scale switcher and some sectional track to try things out. This was all made a bit easier since I had found Model Junction at Slough which just happened to be on my way home from my business at Denham to Yate at the weekends. So the inital purchases turned into regular visits, with bits of stock being accumulated - as they do. :)

With all this stock I opted to build a layout to accommodate it - another switching layout since I didn't have room for anything larger. Seeing all the stock in the drawer reminded me of the planned layout and I remembered doing some work on it - I think I actually posted some pictures in a thread on the old RMWEB now lost after their recent update. So I visited the attic, found the two baseboards and this :-


...and details of what I did came floating back. :) I sort of remembered how it went together but I dug out my old PC and found the original Templot file dated 2008, printed it out and placed the track parts on it.


Here's a diagram of the layout from Templot.

The two boards are 3' 6" long and the found trackwork is destined for the left hand board above. The section in the green rectangle on the right hand board is a sector plate with a connection to the long straight hidden sidding at the top of the diagram. I think I remember that most of the track on the right hand board has been laid but I can't confirm that until I dig the boards out of the attic.

Here are a couple of close-ups of parts of the track.



The rail is 40thou wire rail from the 2mm Association soldered to copper-clad sleepers and timbers from the same source. It's all constructed with continuous check rails since the track is supposed to be inset in roadways and cobbles.

Sharp-eyed viewers may have noted that the clearances seem to be a bit tight for N scale. The track is actually laid to FS160 standards which is a finer set of standards designed by European modellers. Basically they have taken the 2mm Association standards and narrowed the gauge from 9.42mm to 9mm. The web site for the standards is here.

fiNe-scale or FS160

Here's a comparison of the wheels...


...with the standard N scale wheels on the left and the FS160 wheels on the right. I machine the wheels myself using Intermountain products with metal wheels. The standard wheels and axles on RTR products are all plastic and don't really machine all that easily. The wheels on the left are the Intermountain wheels before machining to the finished size on the right. It's a fairly straightforward process taking a skim off the front and rear to get the tyre and flange width down to standard. The flange depth happens to come out pretty well spot on after these operations.

The wheels on the first locomotive were treated in a similar way although I think I remember that I got spare wheels on axles from Atlas to use so that the locomotives could be restored to normal N scale standard if I were to sell them on.

I suspect this will be another iron in the fire along with the S scale layout and the layout in the garden. :):)



Western Thunderer
Has the old RMWeb really all gone? :(

I believe so. I remember Andy York admitting that the archive had not survived the latest update. There was a bit of a thread on the matter. I think I also remember that it was more of a problem of interfacing the old archive with the new software. There's progress for you. :( It's a great pity since there was a lot of very good stuff in the early days of RMWEB.



Mid-Western Thunderer
A nice simple trackplan, just to try out the scale, I see. :)

I started my US outline modelling with N gauge; at the time US N models were so much better than UK outline, as it was with HO-vs-OO as well. Must admit RTR UK outline has done some fair catching up in recent years.
I'm not sure I could return to such small models now after some years in O. As for such incredible hand-built track formations in that scale - wow!! Just - wow!! :thumbs: :bowdown: :bowdown: :bowdown:


Western Thunderer
I started my US outline modelling with N gauge; at the time US N models were so much better than UK outline, as it was with HO-vs-OO as well. Must admit RTR UK outline has done some fair catching up in recent years.

This will be my second American N scale layout. My first was started in the 1960s when N was in its infancy. The first locos and rolling stock available were either European or American outline and when the UK manufacturers eventually got onto the N scale bandwagon, a lot of the European and American outline stock was traded in. That's when I started buying second hand from Bill Eaglesham (to become M.G.Sharp) under the Highlandman's Umbrella in Glasgow. I had a working layout in one of the rooms in my house during the 1970s but the layout didn't survive our move to the Bristol area in the early 1980s. I remember selling the stock to an adjacent layout owner when exhibiting my S scale layout at the London exhibition in the 1980s. :)



Western Thunderer
I’ve done some experiments with FS160 myself and I’ve found that metal wheels don’t actually need machining. Most metal rolling stock wheels have one wheel pressed onto the axle while the other is in a plastic insulator. If the insulated wheel is pulled out slightly on the axle to the correct b2b, things work pretty well. It’s slightly off center, but as long as one is consistent within a truck, it’s hasn't caused me any problems. Diesels are even easier, since the vast majority are split axles with a plastic geared muff in the middle, just pull the wheels out until they’re in gauge. A lot of locomotives need this done to them even if they’re not going to run on finescale track.


Western Thunderer
I haven't tried standard unmodified N scale wheels - I opted to machine to FS160 standards since it was quite an easy option. Also, with US RTR N scale stock, the wheels and axles are a single plastic moulding, so adjusting the placement of wheels on the axle is not possible. But I have to agree that seeing the difference of the wheels on US diesel locos and bogie freight stock is almost impossible. :) However, I know that the FiNe-scale people who developed the standard have used it to great effect on steam locomotive models.



Western Thunderer
Thimgs have moved on a bit since July. My eagle-eyed club secretary noticed this thread and immediately lined me up to exhibit the layout at Thornbury at the end of next year - I didn't get much option to refuse. :) So I dug the two baseboards out of the loft, swept the cobwebs off and found that they were all in good condition with all the track laid on the right hand board (https://www.westernthunder.co.uk/index.php?attachments/fine160-001-jpg.108440/)

I had started work on the left hand baseboard, covering the board with 1/16" cork sheet and starting on work to fit point operating units under the turnout switches. But I suspect that I ran into the problem of how I was going to situate the POUs under several closely situated turnouts, as I noted the problem fairly quickly on my first inspection of the board. So I considered using the same method that I had used in my S scale layout, using tube in wire on the surface of the board with the point motors at the rear. The main problem with this method is that there is a single hidden siding track at the rear of the board and the point motors would have to be inboard of this track.

I then set about designing something which could be shoe-horned into the space available. Some years ago I had experimented with using servos as point motors and I had a few bits and pieces around. So I have designed a motor to, hopefully, do the job.

First, the CNC mill was pressed into service...


This is the base for the motor made from 40mm x 15mm aluminium angle from Screwfix.

Various other bits and pieces were got together...


The completed base is on the right with the hole for the servo at the top, and the two slots in the middle are for the micro switch. The servo - a Tower SG90 - is in the middle with its hacked down horn and the micro-switch is extreme left. The black item between the servo and the micro-switch is the operating arm which is 3D printed - I think that is Version 6. :) All the assorted screws are along the top - and I mean assorted. :) The screws on the left came with the micro switches and I'm not too sure what they are, but I think they are a small Imperial size, the screws to hold the servo, top centre, are 2mm metric and the pivot screw top right is 8BA. I can see me being cursed if someone ever strips one down in the future. But I'm now in the business of using what I've got. :)

Here is the unit assembled showing the operating arm at both extremities. It is designed for the servo to use its maximum 180 degree movement. I've read enough comments about servos kicking to their extremities on startup and damaging linkages so I opted to use the the servo to its extremities.


And here's the unit mounted on the baseboard. The woodscrews are in slots and provide adjustment. I will also have a clevis on the end of the operating wire with a screw adjustment to give finer adjustment. There are also five holes on the operating arm to give a selection of throws although the one nearest the pivot, with a throw of 1mm, will probably be the one used for FS160. The tube and wire to be used is shown - from Model Signal Engineering. This is of small enough diameter to be pretty well level with the top of the cork sheet and will be covered very easily by the road surface, which will be made from card with laser cut cobblework on top.


Here's the situation at the rear of the board, with just enough room to allow for a single fiddle track at the back. I might actually look at using a shallow "U" shaped cassette system which will prevent stock being accidentally knocked off the back of the board. I think there are another twelve point motors to be mounted along the edge.

You can just see the other board propped up in the background.

Last edited:


Western Thunderer
Things have moved on a bit in the last week or so. A bit of time has been spent getting a lot of things together to work on the FS160 track - like suitable soldering tips. I think I remember that the soldering iron I used for the track-making is long gone - a small one which I used for wiring at work which must have disappeared when I closed down my business. So I had a look at what I had now - a Weller with a range of bits which were all a bit heavy - and my temperature controlled Xytronics which had one heavy chisel bit and a very fine pointed bit which weren't really suitable. So I got some in between sizes of bit for the Xytronics and went about one of them with a file. :)

...and the end result after filing...

...with a sharp 90-ish degree chisel point and the copper centre of the bit is just starting to show. It will get tinned again by dunking it in Carrs 188 solder paint when up to heat. The copper will erode during use so an occasional dressing with a file will be necessary, but that's not a problem if it means that I get a nice smooth, sharp point on the tip.

I had a good think about how I was going to "nail" the track down. Since I was going to be experimenting with a few parts of the point operation I was looking for something that would allow track to be lifted and relaid. So I opted to lay PCB strips at several positions which are screwed to the baseboard and to which the rails would be soldered.

The PCB strips are 4mm scale sleeper size from MSE and No 0 brass countersunk woodscrews hold them down.

The next "fight" was establishing the rail ends at the baseboard join.

There were various trial fits to establish rail positions and finally the ends of the rails were trimmed flush with the board end with the Xuron cutters. I had to lift one set of rail ends on the other board since I couldn't reconcile the rail ends from the crossing of the turnout. I can't remember how I got into that situation eleven years ago. :) I also laid down a strip of PCB beside the PCB at the baseboard join so that I could tie the rail ends in position without having to solder the rails to the existing baseboard end strip of PCB. When everything is settled, the rail ends will be soldered to the end strip.

The rails ends are now lined up and soldered down to the auxiliary strip. 2mm Association members might just recognise a much modified Association roller gauge which has been cut down the middle with the cut edges faced off, a hole was then drilled and tapped 6BA through both parts and the two parts were adjusted to 9mm gauge and lock-nutted. The other small gauge is home built on the milling machine and it gauges both check rails.


A view of the board with all the track tacked down to the PCB strips.


...and an overall vew of both boards showing the track already laid on the other board.

The layout is actually 6' 8" long. I can't remember exactly why that length was chosen but I must have had a good reason at the time. :) What I do remember was building it using B&Q's cheap 4mm plywood of eleven years ago which was pretty poor quality, and using the glue and tape method to hold everything together. The boards have certainly survived very well above the insulation in the loft so they would have been subjected to plenty of temperature variation for many years.

The next job is to go over all the pointwork and replace the tiebars. I think I could make a better job of them today and I also see that I have to adjust how some of the point blades are mounted.



Western Thunderer
The next matter to deal with were the tiebars on the turnouts. I had noticed that one or two of the fixings had come adrift over the years and would need dealing with. I opted to change the method I had used to hinge the connection between tiebar and switch blade. The problem with the original method was that I used a pin through the tiebar from underneath which was soldered to the inner side of the blade. But the rail is just 1mm high and the FS160 flange depth is 0.5mm so there was just a bare 0.5mm of pin to solder to the rail, so not much area for a strong solder joint. Also the existing tiebars were intended for under board operation by spring wire and I now intended to use tube in wire on the baseboard surface.

So I opted to try another method which could give better results. I spent a couple of weeks trying a few different things, but came up with the following method.

The first job was to manufacture the new tiebars and the CNC mill came into use again - may as well use it if I've got it. :)


...and the end result


The tiebar is a 2mm Association PCB sleeper.


The next job was to make a small jig to help with the fitting of the hinge pins to the tiebar.


The hinge pins are made from 0.45mm brass wire. I opted to go with the wire after a bit of experimentation with pins which didn't work out.


The pins, with their top parts tinned, are placed in the jig.


...then the switch is placed on the jig with one switch rail butted up to its hinge pin...


...then the pin is soldered to the blade. This gives a good 4 - 5mm of solder joint length.


When the other hinge pin is soldered to its blade then the assembly is turned over...


...the pin ends bent over and trimmed down with side cutters...


...then the bend are flattened with light taps from my toffee hammer - one of my first tools from a Hobbies set in the early 50s. :)


The view from topsided when it's all finished. When the point motors are installed, I'll finish off all the blades in position with scrapers and needle files.

The problem with using various types of pin was feeding them through the tiebar and bending them over to a sharp right angle. Getting the bend close to the tiebar tended to put stresses on the tiebar at its weakest point at the pin hole. Finally I opted to make the right angle bend first and the way I could do that was to use plain wire. I had thought about using small washers soldered on to the wire under the tiebar but I had a go at bending the wires over and using a few gentle taps to flatten the corner and that worked. :)

I got all fourteen turnouts on the baseboard finished over the last couple of days so I'll now get ahead with the tube in wire installation then the servo point motors. With a bit of luck I might have them all installed by the year end and than I can start wiring and testing.

Last edited:


Western Thunderer
Other things seemed to get in the way over Cnristmas. :):):):)

But I have now got a good few of the servos installed with the tube-in-wire.



The cork underlay is beginning to look like the woodworms have been visiting. :):)


The plastic tube and nickel silver wire from MSE is installed by cutting away the cork sheet to provide a channel for the tube. The tube diameter is just under the thickness of the 1/16" cork sheet. I have used hot melt glue to hold the sheet down. This stuff is pretty well uncontrollable when trying to get a neat effect. :) I eventually found that if I applied a small dollop on the tube then allowed it to start setting for a few seconds, then pressed my steel rule flat onto the dollop it flattened it to the height of the cork sheet and the rule didn't stick to the glue - most times. :) The hot glue doesn't make a permanent joint with the MDF and the tube, so any problems are easily cleaned off to try again. But there's enough adhesion to keep the tube in place. The small guides on the tube ends are made from brass tube soldered to small lengths of the 4mm sleeper PCB. They also lift the tube and wire end to match the height of the operating arm.

I was going to fit clevises on the wire ends but opted just to bend the wire at right angles as a simpler solution both at the servo end and the tiebar end. An advantage of this is that the wire bends allow a small amount of springiness which comes in handy when setting up the tiebar throw.

At the moment I have come to a halt since I am waiting on more servos coming. I thought I might have got them by the weekend just past, but it looks as though it might be tomorrow at best. At around 80p per SG90 servo off Ebay, it makes all the time spent on doing the servo mounts worth while. :):)

Having the track removable in chunks has made this job a lot easier and I think it will help a lot when I get things running and I might have to do some adjustments.



Western Thunderer
Just a short update to show that something is still happening. :):):) I've had a bad attack of sciatica for most of the week - a hangover from my work many years ago when I carried a heavy tape recorder on my shoulder.


All the servo motors are installed and all the pointwork is laid again and is working. Work on the layout now stops until I've got the control gear for the servos built. I could operate each turnout individually using the servo tester but I would much rather have all the switches working when I'm doign the track testing.

I'm using a relatively new bus control system from MERG called Ezybus. This will allow me to operate all the pointwork using a four wire bus - two wires for the bus signals and two for the power feeds to the input boards at the control panel and the output boards feeding the servos. This means that I can have the control panel free to move between the front or the rear. At home I would prefer to work the layout from the front, but at the exhibition(s) it might be better to consider working from the rear.

The Ezybus system uses Arduino boards which I've never worked with before, so it could be interesting. :):)



Western Thunderer
Work has started on the electronics.


The input boards are the three on the left with the chip modules fitted. The main board is the blue Arduino UNO in the centre and the board above has an LCD screen and buttons to provide a user interface to the system. The output boards are to the right with all the various sockets and pins soldered in and the Arduino Nano chips, which fit on these boards, are the three modules on the extreme right. The Arduino chips have all had their programs uploaded to them. I've just got a few components to add to the output boards and get a suitable power supply and cable to hook everything up for testing.

I also had to sort out a small control panel which would be portable to fit either front or back of the layout. I spent a bit of time devising a representation of the track plan - my attempt at an Underground style map. :)


The result is in the centre above and the dimensions are 235mm by 110mm. I opted to use sub-miniature toggle switches and I had some 3mm Perspex sheet lying around to use as the front plate. The drawing for the track diagram was used to generate a file for the CNC machine to cut out the Perspex to size and to drill the holes for the switches. I also had to work out how to hold the drawing behing the Perspex since any form of adhesive would probably show. So I opted to cut a backing from 30 thou Plastikard sheet - on the left above. I couldn't go much thicker because of the size of the miniature switches.

I also found out the problem of working with Perspex sheet which has been around for a few years - the protective covering sticks on very firmly. I had two unsuccesful goes at getting it off the sheet since my efforts at removal had marked the surface. I did a bit of digging around on the Internet and found someone who recommended Eucalyptus Oil to help removal. I got some from my local Boots and tried it on the third attempt, and it worked quite well, allowing the covering to be scraped off fairly easily. I also made up a scraper from a scrap of the Perspex which helped to avoid marking the surfaces.


The sandwich was clamped together and the switches were fitted which then held the whole lot together...


...and the underside. Some of the switches are single pole and the switches for crossovers are double pole. The EzyBus system is quite basic and only allows a switch to control one servo, so no driving two servos from a single pole switch. There are one or two ways around this but I thought that the simplest way would be to use double pole switches for crossovers.

The next job will be to make an enclosure to hold the panel and the input modules.

Last edited:


Western Thunderer
And a bit further with the electrics - the panel now has its box


The panel face is recessed into the box by the depth of the switch levers to provide protection for them. They are a bit fragile and I have broken one already before the panel was fitted in the box.


And the underside showing the two input modules mounted on CNC milled mounts and screwed to the Perspex backing board. The Veroboard part in the centre is a tag board using pins soldered to tracks with all the switch poles coupled to the pins. I opted for this setup so that I could easily change things around without having to alter the connections to the actual switches. I was using a quite fine hook-up wire from the switches and its insulation did not like heat so it was quite a fight to make good connections and plenty of Hellerman sleeving was used to make it look half neat. :) The connections from the tag board to the inputs on the modules will be done with a slightly heavier hookup wire with more accommodating insulation, :) and these hookups will be done when the control panel is connected up to the control module. The connecting cable still has to be fitted and that will come in at the top of the box to feed the network and power connections at the top of the modules.

At the same time I've gone over all the trackwork and located the occasional sleeper where I hadn't cut the copper - that took a while to do. :) I thought I had cut all the copper necessry while building the trackwork, but apparently not. :headbang:

the next job will be to connect everything up and see what happens. :)

The next matter to deal with were the tiebars on the turnouts (...)
Really interesting your small jig for placing & soldering the pins to the blades.
I am using the same method (but with a plastic tiebar), never ever thought about a jig (I have always soldered the pins by eye and with my hands),
great hint!


Western Thunderer
Any more progress Jim?

I'm afraid not at the moment. I'm waiting on my hip's recuperation advancing a bit more before I can get back to the work on the baseboards. The problem is when you need both hands to lift or work with something and one hand is normally required to hang on to something for balance. :):) The next job was going to be fitting servos to the other baseboard and that was going to require a bit of woodwork and two free hands. It will probably be near the end of the year before I can contemplate re-starting.



Western Thunderer

I'm much taken with your servo mounts, they seem like a very solid job. I have been pondering using servos for a while, and it seems to me that the full 180 throw gives a further benefit - provided you don't drive the servo until it's stalled, you can have a quite stiff spring wire (Tortoise style) which will give a decent force at the point blades, without having to maintain a high torque on the motor, as the reaction forces will be towards the pivot of the servo actuator. Servos don't like being stalled, so I guess this will both reduce current demand and reduce failure levels.

thanks for posting those pix, and hope the recuperation makes good progress

michael mott

Western Thunderer
Thanks for the update Jim, Mobility does seem to become a big issue as we move along the path of life. I too thought that your servo rigs looked very nice.



Western Thunderer
It is exactly a year from when I stopped working on this layout because of my broken hip, and the layout and bits have lain pretty well untouched since then. It had been lined up for the Thornbury exhibition last November and work had been going on, flat out, to get it into a reasonable condition for that date, but that stopped due to my hip and the requirement to exhibit due to other pandemic matters. But with the good results from the present lock down and the vaccination operation, people have started looking at the possibility of returning to some sort of normality later in the year and the Thornbury club are considering running their exhibition this November if things go well. So the layout is back on track to appear there this year - so about eight months to go. :)

The first thing was to get the Ezybus bus control installed and working. This system was designed by Davy Dick of MERG and looked to be a relatively simple system which had enough facilities to let me consider having the control board on a lead which would allow me to operate from front or back of the layout. So the first job to start with was getting the cable attached to the control box.


...and here's the box with its new cable. The cable is cut from a Cat5 Ethernet cable as suggested by Davy Dick and the connector is a 4 pin XLR plug - much beloved by professional sound people. These connectors are pretty well bullet proof and they come in three to five pin varieties.


The cable was attached to the two input devices and the fine yellow wiring between the central patch card and the farther input module is the start of my hooking up the control panel switches to the module. Wiring is au naturelle at the moment until its all finished and tested. :):)

Over on the main baseboard, the control module and one of the output modules have been installed. This was the last work done before my enforced stoppage.


The control module is closest and the first output module is on the other side of the central patch panel. They are all mounted on a piece of perspex which I had hanging around and I've CNC cut fitted bases for the printed circuit boards of the modules as I did for the ones in the control box. The odd bit of wood just beyond the output module is there to protect the farthest servo drive - two operating arms have been broken already by hands catching on it. When the scenery and backscene are installed, I'll make a more sophisticated guard. :):)


A closer view of the main control module. The black buttons provide the controls for the system. You can set the range of movement and the speed of every servo and it is stored in the control module memory. The module also is the central point for the two-wire bus system - the I2C system. The green wire is the data line and the blue wire is the clock line. Red and black wires are + and - 12V. The system pretty well looks after itself and the only input from the operator is to adjust parameters for servos when setting up. The Arduino Uno module, which is the master controller, is on the blue circuit board under the display board.


This is the small socket panel to accommodate the two XLR four-pin female panel mounts - one for the control box lead and the other for the feed to the output module on the second baseboard. The barrel socket on the right is for the 12V DC power.


The is the first output module. The eight servo outputs are front right with five servo leads already plugged in. The Arduino Nano is to the top and this controls the servo movements.

And it works. :):):) I had a few dodgy moments when first connected up, but that was all down to my finger trouble. Once I had tracked the problems down it did what it said on the tin. The next job will be to install the second output module on this baseboard and get all the servos on the board hooked up and checked out. After that I have to install servos and an output module in the second baseboard. After that I can get the track wired up and check out all the trackwork.

I find the Easybus sustem more than suitable for my needs. It is limited on one way in that a control signal on Pin 1 on an input module can only control Pin 1 on its matching output module. So setting up which servo output is controlled by which switch depends on connecting that switch to a specific input on an input module. That might be a problem on a larger layout with a lot of turnouts and signals where situating output modules to control equipment and keep servo leads at a reasonable length may be a consideration. With my simple setup one input module will cover one baseboard with two output modules and the second one will cover the second baseboard with one output module. I think it is accepted that Easybus is a simple system and MERG has another solution for more complex situations.

Last edited: