This Blog describes the ongoing development of a 16mm scale 45mm gauge garden railway situated in the North West of England, UK from 2004 to the present day.
This blog describes ongoing progress in the development of a G gauge Garden Railway from its inception to the present day.
NEW (29 June 2024)
NEWS
The blog has now had well over a million visitors. Many thanks to all those who have provided me with support, suggestions and feedback over the years.
When I became interested in building my own garden railway I spent a considerable amount of time (and money) on books, videos, DVDs and scouring the internet for information, ideas and inspiration. When I eventually started construction, I used some of the ideas I had discovered, but also experimented with my own approaches. This blog outlines how I have gone about constructing my own garden railway. My aim is to provide the sort of information I was looking for when I was getting started, and also to share what I've learned (or 'borrowed' from others). I've tried to include a few 'How I ........' postings interspersed with occasional 'Progress Reports'. I do not profess to be any kind of expert - what I offer here is an opportunity for you to metaphorically look over my shoulder to see how I have gone (and am going) about this fascinating hobby.
As this is a blog, the various posts are presented in reverse chronological order (ie the most recent first). To see a categorised list of contents, go to the Blog Contents Page.
If you are thinking about building your own garden railway, then why not join the 16mm Association or the G Scale Society - you'll get plenty more advice and opportunities to visit other peoples' garden railways. Alternatively, browse through theG Scale Central website - there's plenty more guidance here and an opportunity to sound out the views of others through the G Scale Central discussion forum.
The Blog
The advantages of blogging are that it is immediate and uncomplicated when creating and uploading information. The other, of course, is that with Blogger it is free. The major disadvantage is that I have minimal control over how the postings are presented. The blogging system adds the most recent information to the start of the blog, hence the postings appear in reverse chronological order (most recent first, oldest last). Whilst there is a list of postings on the right hand side, it's not particularly easy to see what is there. This introduction is an attempt to provide you with a contents list of the postings organised into categories so, hopefully, you see if what you are looking for is presented in this blog. To ensure that it always appears at the start of the blog, I update its content and set its presentation date into the future each time I add a new posting.
Wow! It's hard to believe that this is now my 100th Progress Report. When I first started this blog back in 2006, I honestly thought it would be something I might add posts to every six months or so, this will be my 369th post about my railway which means on average, I've posted in the region of 19 posts a year or between two and three posts per month! Over that time my blog pages have had around 1.25 million views, and I've received over 900 comments - which doesn't include spam messages which I have to delete on a daily basis. I've also posted over 230 YouTube videos about Peckforton Light Railway which to date have had over 830 000 views. Who'd have thought so many people interested in my little garden railway set up in a modest back garden in Cheshire?
Anyway, what has happened on the railway since my last update in August?
The most significant event has been the complete refurbishment of Beeston Market station and the Copper Mine, after I discovered some of the baseboard timbers had rotted away. In addition, there has been the general maintenance of locos and stock, motorising my semaphore signals, the continued painting of 3D printed figures, a newly acquired app for helping with freight management, hosting a visit from Zach Bond - who really enjoys the challenge of running his live steam locos on my track and, of course, a few running sessions.
Permanent Way
For a couple of years, I've noticed that the trackwork at Beeston Market and the Copper Mine had dipped in a couple of places - to the extent that I was having to prop up the underlying wooden baseboards with additional supports. The hump and dip in the track on the approach to Beeston Market can be seen at 14:40 in this video I recorded in September last year:
After crawling underneath, I realised that there were several places where wet rot had attacked the treated timbers and so I had little alternative but to lift the track and investigate further.
What was revealed presented me with a dilemma - should I just patch up the worst affected areas, or rip out all the baseboards and start again from scratch?
The same problem was found at the Copper Mine.
The baseboards had been in place for 17 years and so, I figured that if I
replaced them, I'd be at least 90 years old before they would need to
be done again. And so, with a spell of decent weather forecast, I
decided to rip all the boards up and start again.
A quick calculation and a trip to my local fencing supplier (one advantage of living in a rural farming area), and a few days later sufficient boards were delivered for the revamp.
I discovered some of the legs supporting the baseboards had also rotted and so these were replaced and a few additional legs added where I thought they might be needed.
Cross-members were then screwed into place .....
..... so the new boards could be fixed into place.
I had worked out, together with comments from some of my fellow forumites on the GardenRails.org forum, that the rot on the previous boards had mostly occurred where I had cut the treated boards. And so, I made sure that every cut end of the new 6" x 1" pressure treated timbers was given a liberal soaking with wood preservative.
At the Copper Mine, I decided to raise the sidings slightly above the adjacent main line, to help differentiate the two.
Once all the boards had been levelled and fixed into place, they were covered in good quality (15 year guaranteed) roofing felt (aka tar paper), making sure that all joints were sealed with bitumen based felt adhesive - again following suggestions from fellow forumites.
Tracklaying, then followed. I spent a couple of days, trying various combinations of pointwork until I was satisfied that the new layout would make shunting operations easier and/or more interesting.
At present, the track is unballasted and the platform is just a plank until I am completely sure that the new layout works well and there are no snags.
Something I was keen to install was a double slip - partly because I thought it might add a couple of options for shunting, and also because I find them fascinating.
So far, I've run a few test trains over the new set-up and had a couple of full operating sessions - so far, so good.
See - How I refurbished Beeston Market Station - pending
Rolling stock
No major developments here since the last update, though I have now given the de Winton loco a decent paint job and had an opportunity to run it in.
I must admit to being very pleased with this loco. Not only because I feel it looks good, but also because I made its mechanism from scratch, something which, over the years, I've tended to avoid as my modelling skills are insufficiently adept - until I acquired a 3D printer which, of course, can produce models down to 0.1mm accuracy.
The dozens of 3D figures which I received from Dean at DesignScanPrint3D after a scanning session with my fellow thespians at our local amateur dramatic society are still in the process of being painted.
I have to admit that progress is painfully slow. It takes me ages to paint each figure as, rather than slapping on a single layer of paint to each piece of flesh and each item of clothing, I go over each element several times to try and get the shading and the facial features looking reasonably realistic. With some figure, this process takes ages, as I find I have to redo them several times until I am satisfied.
The added complication is that I have two of each figure to paint - one for me and one for each volunteer, as a reward for being scanned.
Hopefully, by the next progress report, they will be finished.
Operation
Motorised semaphore signals
The most significant development here has been the motorisation of my semaphore signals, which I constructed in 2014 - See How I constructed 19 semaphore signals. After an initial attempt at motorising some of them using a Picaxe micro-controller(see - How I radio control my semaphore signals), I was reluctant to extend the process for the remaining fifteen signals as it was a bit cumbersome and would have required a fair amount of additional wiring together with devising a robust, weatherproof system of electrical connection for each signal.
When I spotted some tiny linear servos, together with a compact Arduino based processor and a tiny keyfob actuated receiver, all of which could be powered with a single li-ion cell, I experimented with producing a self-contained mechanism and control unit which would sit neatly beneath each signal.
It took a couple of months, but I now have all nineteen signals motorised without the need for any additional wiring. The next stage will be to provide solid bases for each signal around the garden so they can be quickly and reliably installed at the start of an operating session. In the meantime, I have deployed a few of them during a running session to test their reliability and effectiveness. For more information, see - How I radio control my semaphore signals - Part 2
A new app to support freight handling
As regular readers will know, I have used various versions of my computerised freight management program over the years to help me decide what wagons will be used in each goods or mixed train and where they will be sent. For example see, How I created a Psion version of my freight management program
However sometimes, if I am in a hurry or if I want to run some special goods trains, I turn to a different method to determine which wagons will go where. For example, in a recent Southwold-focused running session, I used a simple die-rolling technique to decide where to send the few items of Southwold Railway goods stock I had at my disposal.
In addition, from time to time, I have used a simple app on my phone to carry out a similar process. In the past, I used an app called Weighted Random Generator but when I accidentally deleted it from my phone I discovered it was no longer available. Fortunately, I discovered a similar app called Rand1which is equivalent and so now use that from time to time.
For more information see Issue 353 (January 2024) issue of the Garden Rail magazine pp18-22 or - Various teachniques for generating freight movements - pending
Running session
In addition to the test runs for the new layout at Beeston Market (see above), I also took the opportunity, during the spell of sunny weather at the start of April, to run a full session which I decided would have a Southwold Railway focus.
Over the years, I have slowly and steadily been adding items of Southwold stock to the railway, ranging from a couple of locos.....
.... to a rake of Cleminson six-wheeled coaches.....
.... and a few items of goods stock.
So I decided it was time to showcase these items in a Southwold-focused running session. As indicated above, I needed to change my approach to managing the destinations for the goods stock, but I found it an interesting change from my normal running sessions.
And finally, I hosted another visit from Zach Bond who brought with him his latest acquisition - a rather beautiful Accucraft Caledonia model which he had had repainted in Southwold Railway livery as it reminds him of a Southwold Sharpie.
We spent a very pleasant afternoon running the loco up and down the line alongside some of my own stock and catching up with the latest news on garden railway developments.
[Awaiting video - currently in the editing shop]
Conclusion
There are still several jobs on the todo list:
finishing the painting of the 3D figures,
ballasting the track at Beeston Market and the Copper Mine,
providing bases for the signals at each station;
... and I have got a few ideas for additional locos, particularly now I have developed the confidence and the skills to make my own powered chassis. But, for now, I am quite pleased with how the railway has progressed over the years.
Having recently constructed a range of small diesel locomotives from 3D printed parts, I was on the look-out for a reasonably priced sound card for them, preferably with a small footprint.
On AliExpress, there are several small "diesel simulation" soundcards mainly designed for use in radio-controlled lorries. For example:
So, for just over a fiver (inc. postage), it seemed like too good a bargain to miss.
As can be seen from the photo, the card is designed to connect between the receiver and the ESC (Electronic Speed Controller) in a radio-controlled vehicle using JST connectors. This enables the sound card to vary the sound in relation to the speed of the loco.
The beauty of Deltang (Rx65 etc) and Micron (MR601 and MR603) radio receiver/controllers is they combine the receiver and the ESC into one small, neat package. The disadvantage is that the above soundcard cannot be connected directly. Fortunately, Andy Rutter, the designer and constructor of Micron RC equipment, has anticipated this and can provide his receivers with one or more of the outputs equipped with a servo signal relative to the voltage output of the receiver's built-in ESC. You can specify this when ordering one of his receiver/controllers.
Alternatively, it is possible to re-program a Deltang or Micron receiver/controller to output this signal yourself. Here is how I reprogrammed a Micron MR603b receiver. The same process is used for the earlier MR603a and the later MR603c. However, if you are using an updated version (since MR603c), it is advisable to check the Micron website to find out id any changed have been made.
1. Identify the pad you want to use
The first job was to decide which pad I wanted to reprogram to output the servo signal needed by the soundcard. Any of them could be used but opted for P3 as it looked reasonably easy to solder a lead on to. Any pad can be re-programmed so, if you want to use P3 for something else, choose a different pad. Pads 5 and 6 are trickier to attach wires to as they are designed for the connection of a JST socket. You can choose to have a JST socket soldered on when you order the receiver - this would save you having to do some soldering on to the receiver yourself.
Interpreting this for the first time can be a bit tricky, but once you've done it a few times it becomes common-sense.
As (hopefully) you can see, the code we need is:
2, 3, 1, 1, 1
...... because
2 = Table 2 (for servo configuration)
3 = Pad 3
1 = Normal servo (the default at the time of writing)
1 = Radio control channel (Channel 1 controls the speed of the motor)
1 = Normal speed (to map the servo signal directly to the speed of the motor from Channel 1)
3. Putting the receiver into programming mode.
BTW, when programming a receiver, I usually connect it to a 6v or 9v battery pack, using electrical hook clips ...
... as they can easily be connected to the power pads on the receiver.
Firstly, I made sure the receiver was bound to a Deltang Tx20 transmitter - or a Micron Tx20 would also do the job. According to the manual for the MR603, there are three methods of getting the receiverinto programming mode:
Use the P1/P2 power-on jumper to select option 5 (5-flash) to enter programming mode
Switch the transmitter on and hold channel 2 and channel 4 controls at high or low extremes and then switch the receiver on; these are the F1 and F2 buttons on a Tx20v2 and the Selecta switch and S2 button on a Tx22X; the receiver LED will flash rapidly shortly after a signal is received from the transmitter, centre one or both channel 2 & 4 controls (e.g. release the buttons on Tx20v2 or the S2 button on Tx22X. Note: the receiver always goes through an initialisation sequence after power is applied, so you must hold the ch2/ch4 control positions until a signal is received and the LED flashes rapidly.
Switch the transmitter and receiver on (they must be bound), wait 5 seconds without touching any controls and then tap out morse SOS (... --- ...) on the transmitter bind button (or toggle the gear switch high to low for a stick transmitter)
dots (…) will be a quick press of the button and must be less than 1 second in duration
dashes (---) must be greater than 1 second and shorter than 5 seconds – 2 seconds is a reliable time
the time between each dot or dash must be less than 5 seconds
I used the third method, tapping out SOS with the bind button, as this seemed like the easiest. It's also completely different to how Deltang receivers were put into programming mode, and so I wanted to give it a try.
When the receiver is in programming mode, its LED should start flashing once every second or so (known as 1-flash). If it doesn't work the first time, try putting the receiver into programming mode again.
4. Reprogramming the receiver
We are now ready to reprogram the receiver by using the code we extracted from the programming table.
This is achieved by using the Direction switch on the transmitter.
Moving the Direction switch to the Reverse position, increments the flashing value by one (ie changes 1-flash to 2-flash, or 2-flash to 3-flash, etc.)
Moving the Direction switch to the Forward position tells the receiver to accept a value for that particular setting and move on to the next setting
So, our 1-flash, indicates the receiver is initially set to receive values for table 1. But the servo settings are in table 2, so we need to reset the 1-flash to 2-flash. To increase the flash from one to two, we flick the direction switch on the transmitter to the Reverse position. When the direction switch is released or returned to the mid-point, we should now see the flashing increased to two flashes, pause, two flashes .... etc (ie 2-flash)
NOTE: If it accidentally goes to 3-flash, carry on incrementing the flashing with the reverse switch until it reaches 7-flash. Incrementing it again should recycle the flashing back to 1-flash. You can't go back to 1-flash by flicking the direction switch to the Forward position!
Once you have changed it to 2-flash, click the direction switch on the transmitter to the Forward position. This tells the receiver to accept that value, store it, and get ready to receive the next value in our code list.
The next value in the list is 3, indicating we want to reprogram the output from Pad 3. So, we now need to keep incrementing the flashes using Reverse on the Direction switch. Again, if the flashes go beyond 3-flash, we can cycle round until we reach 3-flash again.
As before, when we see 3-flash, we click Forward on the Direction switch to show we want to accept this value.
We now keep repeating the above process so we have 1-flash, 1-flash and 1-flash for each of the next three settings.
The number of flashes for each setting will depend on how the receiver had been set up when you bought it. Some, but not all, will have a default value of 1-flash. However, if they do not, you will need to cycle round the flashes until you once more reach 1-flash.
When you have completed the full sequence of programming the values, the final click on Forward on the Direction switch will make the received LED show a continuous, non-flashing light. If you lose track or want to check the values, switch off the receiver and go back to the start of the programming process.
6. Wiring everything up
of course, the receiver could be programmed whilst it is already wired-up, but I tend to do the re-programming before burying the receiver inside the loco.
Some of my smaller diesel locos use only 2S battery packs, which give 7.4v. As the diesel soundcards require between 5v and 8.4v, a 2S battery pack is OK. The card can be wired into my usual loco circuitry, with a connection from P3 on the receiver to either of the signal wires (usually white) on the soundcard.
However, most of my locos are powered by 3S li-ion battery packs using either 14500 (AA sized) or 18650 cells. These provide, nominally, 11.1v and so the supply to the soundcard needs to be reduced. The easiest and most cost-effective way of doing that these days is via a step-down buck converter - available from eBay or AliExpress.
For example:
At less than £1GBP per unit, they are a lot cheaper than trying to make your own. Their advantage is they are adjustable, turning an input of 4.75 - 23v into an adjustable output of 1 - 17v. Using a digital multi-meter, I adjusted the trimpot to give an output of 5.5v and wired this into my normal loco circuit.
The output from P3 of the Micron MR603 receiver is connected to any of the signal leads (usually white but sometimes orange) on the soundcard.
The diesel engine sound should now vary in pitch and tone in relation to the speed of the loco.
A few years ago, I produced a video showing how Deltang receivers could be reprogrammed. The technique is very similar to that used by Micron receivers. If you need further clarification on the programming process, this video might help.
If you need any further information or help, post a comment below, and I will get back to you.
Introduction
