Piko Taurus: adding power buffer and red taillights

The Piko Taurus locomotive in G-scale I bought earlier this year is without any doubt a very good product. A conversion to DCC and adding sound functionality made the model even better and allowed me to enjoy it fully on my garden layout. Still, I wanted to give it a little bit of a personal touch and improve it further. That's how I came up with the following two extra modifications...

***

Modification #1 introduces a power buffer in a similar way I did it before for my Piko BR80 (more info here). The goal is to make the locomotive ride smoothly despite any dirt that could (and will!) accumulate on the track. Taurus is a large vehicle, so I have more freedom in choosing a capacitor that will compensate for any power disruptions. The largest - reasonably priced - I could find was 22000uF / 25V element. Its capacity is basically twice of what I added to my BR80 in the past.

The electrical part of the installation is very easy - the capacitor's pins need to be connected to the "plus" and "gnd" sockets of the decoder. And it's all described very nicely in the manual. Physical installation seems to be trivial, too, with all the empty space inside the Taurus...

...well, not exactly. The photo above shows my initial placement of the power buffer. This is what seemed like the natural choice to me. Unfortunately in this configuration, the chassis wouldn't close. I was forced to move the element and install it somewhere else. Only then I was able to close the model. The picture below shows my final location.

In my post about the BR80 upgrade I tried to come up with some "scientific" ways of deciding whether a power buffer of this size was useful at all. The results were promising back then. Now I can say I'm 100% positive the capacitors that I use do provide enough power to smoothen the ride...

My confidence comes from a mistake I've made during the upgrade. The manual is pretty clear about the need of disabling "analogue operation" if a power buffer is added. That's something I did not do. The side effect? Locomotive going forward at full speed once the DCC power is abruptly cut.

The decoder probably assumes it's running on a DC layout instead of DCC when the digital signal is suddenly gone, and consumes the capacitor's entire charge at once for a short but a very fast ride. This does look scary and it's something I disabled right away once I saw it (bit 2 on CV29). But it also proves the power buffer offers enough energy to actually push the model forward. In case of the Taurus engine, the locomotive was able to move by over 10cm. Nice!

***

Modification #2 is about adding red taillights. For some reason Piko's Taurus does not offer those. I really miss them, so I've decided to install my own set. I'm using regular 5mm LEDs with 1kOhm resistors. All soldered in the simplest possible way shown below...

The 5mm LEDs fit perfectly into the slots in the model's chassis. It almost feels as if Piko intended to include those but for some reason decided not to at the very last moment. Everything is ready - all I needed to do was to stick my elements into the holes. A tiny drop of hot glue gave me confidence my lights would not fall out.

Of course some simple extra wiring is needed. And suddenly the leftover cables from the DCC conversion became useful!

What I'm doing is basically bridging the front red LEDs with the back white ones, and vice versa. I do not need any extra control over my taillights, and it's OK for me that they will light up together with the opposite headlights (function F0 in DCC). I keep it simple!

My solution works right away. And now in addition to the strong headlights...

...I can also enjoy a little bit more subtle taillights.

I'm very happy with the two above changes. And I'm actually considering a third one already - adding a cabin light. I'm just not sure if I really need it or not... What do you think?

3D printed G-scale helipad with Arduino

My recently completed Bell UH-1 Huey helicopter was meant for the garden layout display right from the start. As I was thinking about a good area to place it, I came up with an idea of building a real helipad just for that purpose. But can you buy a G-scale heliport model? You probably can't. But you can always build or... 3D print one!

The shape of a helipad is not really that complex. It's just a round stand after all. The design of a 3D model didn't take long to complete and very soon I was printing the main part.

I wanted to make my helipad somehow special. And I also wanted to finally try the Arduino boards I had lying around for quite some time. Flashing positional lights is what I came up with. I quickly wired the Arduino with six blue LEDs and made them blink.

The connections are very easy. I simply used one output port of the board per each single LED. This way I do not need any power relays and the entire setup consists of only:

• Arduino Nano
• Six blue LEDs
• Six 1kOhm resistors
• 8V power supply

The program to manage the LEDs is not complex either. Actually it's so short that I can easily share it here. This is the complete Arduino code:

int LED_1_1 = 2;
int LED_1_2 = 3;
int LED_1_3 = 4;
int LED_2_1 = 5;
int LED_2_2 = 6;
int LED_2_3 = 7;

void setup() {
  pinMode(LED_1_1, OUTPUT);
  pinMode(LED_1_2, OUTPUT);
  pinMode(LED_1_3, OUTPUT);
  pinMode(LED_2_1, OUTPUT);
  pinMode(LED_2_2, OUTPUT);
  pinMode(LED_2_3, OUTPUT);
 
  digitalWrite(LED_1_1, LOW);
  digitalWrite(LED_1_2, LOW);
  digitalWrite(LED_1_3, LOW);
  digitalWrite(LED_2_1, LOW);
  digitalWrite(LED_2_2, LOW);
  digitalWrite(LED_2_3, LOW); 
}

void loop() {
  digitalWrite(LED_1_1, HIGH);
  digitalWrite(LED_1_2, HIGH);
  digitalWrite(LED_1_3, HIGH);
  delay(150);
  digitalWrite(LED_1_1, LOW);
  digitalWrite(LED_1_2, LOW);
  digitalWrite(LED_1_3, LOW);
  delay(150);
  digitalWrite(LED_1_1, HIGH);
  digitalWrite(LED_1_2, HIGH);
  digitalWrite(LED_1_3, HIGH);
  delay(150);
  digitalWrite(LED_1_1, LOW);
  digitalWrite(LED_1_2, LOW);
  digitalWrite(LED_1_3, LOW);
  delay(300);
  digitalWrite(LED_2_1, HIGH);
  digitalWrite(LED_2_2, HIGH);
  digitalWrite(LED_2_3, HIGH); 
  delay(150);
  digitalWrite(LED_2_1, LOW);
  digitalWrite(LED_2_2, LOW);
  digitalWrite(LED_2_3, LOW);
  delay(150);
  digitalWrite(LED_2_1, HIGH);
  digitalWrite(LED_2_2, HIGH);
  digitalWrite(LED_2_3, HIGH); 
  delay(150);
  digitalWrite(LED_2_1, LOW);
  digitalWrite(LED_2_2, LOW);
  digitalWrite(LED_2_3, LOW);
  delay(300);
}

I have installed the LEDs on the top of the helipad base. With small holes in the surface created with a hot needle I pushed the pins through.

The Arduino board and the wiring are hidden inside. I only hope this setup can survive outdoor conditions...

I have also 3D printed a bottom part which will seal the model and provide some level of weather resistance.

There are two extra elements I have printed in 3D:

• a white, flat H letter
• six transparent caps to protect the LEDs

Those were just glued to the base with a regular plastic cement.

Once powered up, it is blinking indeed!

And then I thought - what if it breaks? And I quickly realized I needed a way to easily remove and install again the helipad on my layout. That's when I recalled I had an empty filament spool lying around.

The solution: my helipad will be attached non-permanently to a permanent base.

I have 3D printed a plastic anchor and glued it to the bottom of the helipad. The anchor fits the hole in the spool perfectly. I can now just stick the helipad into the base and it holds well enough. No glue needed, and it can be easily lifted whenever I need.

OK, but if the helipad is elevated - how do the people enter the helicopter? They obviously need some stairs. And so I 3D printed those, too. Actually they came up so nice, I have decided to share them online: https://www.thingiverse.com/thing:2909311

The helipad is installed. It does not look bad at all. The lights blink at night. The only thing still missing is the helicopter...

I need to find a way to attach my Huey securely to the pad without making any permanent modifications to the model and I'm all set!

Weather resistant 3D printing? How about ASA...

I've been experimenting with 3D printing for a little over 6 months now. Many of the results of my work have found their place on my outdoor layout already. And some of them dramatically affected the way my miniature world looks.

The street lamps are the best example here. When equipped with an LED light source, they've become the essence of the night in my little town.

The street lamps were created in two steps:

• First I printed a set of classic lamps for the train station based on an external project.
• Then I designed my own modern lamps and filled the streets with them.

In all these attempts I've been applying the simplest printing technique and the filament that was easiest to use. That material was PLA and I was always wondering how weather resistant it was. Well, I do have some knowledge about that now...

As soon as the first hot summer days hit, my PLA-based lamps started failing. In particular they've proven not to be resistant to the heat at all. They bent towards the sun very quickly and created the following view...

What's easy to notice here is that the black lamp bent most, followed by the dark gray and then the light one. Black material probably absorbs the heat much more and is therefore affected to a larger extent.

The lamps are not the only example. I've printed a black piece of a pavement and placed a pair of figures on its surface. The two below photos were taken a few weeks apart. It's clear the pavement element warped in a wavy fashion and even caused the postman figure not to stand straight anymore.

This needed a solution since PLA was clearly not the right material to use for a garden layout. My investigation led me to another kind of plastic which was specifically designed for outdoor use. That material is called ASA.

Luckily ASA is already available as a 3D printer filament. It's not easy to get and it requires a much more complicated printing procedure. But the specification states its "glass transition" temperature is around 40° Celsius higher than PLA's and that should be enough to resist the sun heat. The Wikipedia page confirms those facts listing automotive as one of the industries that benefit from ASA use.

So once I got my first batch of ASA filament, I decided to re-print my black lamps. Those were affected most and should make some good test subjects. I also used this opportunity to play a little with the 3D design, but that's a completely different story.

My new ASA-based lamp posts were installed in the evening. In the photo below they can be seen in the front with the old, already bent light-gray PLA lamps in the background.

Over five weeks have passed. There were numerous hot days during that period. Let's see how the ASA plastic did in the summer outdoor conditions...

Well, they seem to be pretty straight. I can't really see any bending or warping here. And actually I can even see that the PLA based lamps in the back of the photo have continued deforming during that period. That would be a proof the test conditions were adequate.

So it turns out that a more weather resistant filament is not only needed for outdoor use but also makes an immediate and visible difference. PLA can still be used but only for objects featuring a more solid structure. The outhouse or the boat I've printed with PLA months ago are doing just fine. But lamps that are composed of a heavy head installed on a thin post are a subject to failure.

I'm switching to ASA printing now. I will be printing the rest of my lamps with this material only. And as for the lamps I already have... well... I guess they need to be replaced :(

Modern-looking 3D printed G-Scale street lamps

Finally! After weeks and weeks of planning, prototyping and 3D printing, my very own street lamps for my garden layout are finally here. And I have to say I'm truly proud to be able to present my creation to all of you.

Let me first share a little about how this project came to life. The reason why I even attempted designing my own road-side lamp model was very simple - I needed one for my 3D printer and couldn't find any. That's how I decided to create my own from scratch.

I've started by working in Blender. And since this phase took around 10 evenings, I've actually managed to learn the tool a little. It's fun working there, maybe except for the unpredictable Undo function. Solution: save often.

The design phase was not just about sitting in front of my computer. It required prototyping and a lot of it. Preparing a blueprint for something that looks good is one thing, but having it print well is something else.

Here are some of my early tests...

But I knew I was making progress all the time. And every next iteration of my lamp was better than the previous one. Two weeks later I finally had a prototype I was happy with.

My lamp is empty on the inside which allows wiring and resistor installation. The head features two micro-holes which fit a 5mm LED light source very well. It's really easy to use!

This is what my first fully functional piece looked like.

After that it was all about mass production. A single lamp takes about 3 hours to print (1.5 hour per side). I was printing only one per day which means I've spent around two weeks making those...

Wiring them up and gluing took another week. It could have been done faster if the time I spend on my hobby had not been limited...

Installation of the lamps on my road elements consumed most of my Saturday. Actually, I've finished the work when it was getting very dark already. In the end it turned out pretty well, because it allowed me to see the outcome of my actions very clearly. Here's what I saw once I turned the power on:

As we can see, placing the lamps straight is something I've yet to master. But it does look very, very good overall, especially for something I have made myself from scratch!

I have again shared my design on-line. If you want to try making the lamp yourself, here's the link: http://www.thingiverse.com/thing:2220722

I also want to bring up the topic of the cost...

1kg of printing filament can be bought for as little as $15. The lamp weights less than 15g which means we could easily make 65 pieces out of 1kg. The conclusion is that a single lamp's body costs around $0.23 (+electricity).

The model requires an LED, a resistor and some wire. 100 LEDs can be bought on Ebay for less than $3, 100 resistors for around $1. I've bought the cables locally and paid less than $0.15 per meter.

So putting the cost of the electricity, the glue and the solder aside, we can estimate one piece costs less than $0.50 plus your time. I honestly don't think there's a way to equip your layout with working lamp posts cheaper than this. How could I not recommend it then? Just go for it!

My first 3D printed G-Scale street lamps

I've had my 3D printer for a few weeks now and the time has come to start producing useful elements for my layout. This is a rather straightforward process and the key to success is a good blueprint for the item. Anyone can design anything they want but starting with your own creation is rather ambitious. Luckily there's a site called Thingiverse which holds a huge database of models prepared by amateur contributors.

Quick search using the "g scale" keyword revealed several design that could be of interest. A set of street lamps created by user SlashDev caught my eye and I decided to make it my first project.

I've downloaded the necessary files and made some modifications to the original design:

• I've made the lamps a little taller
• I've added an anchor allowing installation in a drilled hole
• I've designed a glass shade for the lamps

A free 3D editor Blender is the tool that allowed me to make all those changes.

After that, the printing was easy. It all worked right away. Each lamp is printed as two pieces which need to be glued together later on. Each half takes about 2 hours to complete using the high quality settings.

The light poles are empty inside which ensures the space for all necessary wiring.

A 5mm LED and a resistor fit perfectly inside the body.

A hot glue gun can be used to help hold everything in place and make the next step easier.

All that is left is gluing the two pieces together. The lamp is (almost) ready.

And it just works. At this point I've realized I wanted a glass shade to make the light more diffused.

Designing the globes was very easy. Each is just a half of an empty sphere. I have to say I really like what I've made here.

A two-arm variant is a similar story. It's equally easy to print and wire up.

There are a few challenges that should be mentioned:

My cheap printer is not perfect. And the two sides of each lamp do not fit each other 100%. This was however to be expected from o DIY machine, and they fit well enough to be perfectly usable.

The material I've used for printing is called PLA. It's very easy to use but probably not that weather resistant. Actually it's advertised as biodegradable. The time will show how well this kind of plastic can survive my outdoor conditions.

Gluing was not a piece of cake. The best glue for PLA is the cyanoacrylate. I was however afraid to use it considering how quickly it acts. I've used a regular plastic cement instead. It worked but required some patience. We'll see whether it holds in the long term...

Being a contributor by nature, I've decided to share my modified design for the lamps. The files necessary for printing can be downloaded here: http://www.thingiverse.com/thing:2064924

I've printed all the lighting I needed for my railway platforms. The next step is producing around 40 of roadside lamps. I'm going to design those by myself from scratch. And you can definitely expect another blog post covering exactly that :)

Cheap G-Scale street lamps from Ebay

The newly built city roads on my garden layout desperately needed some sort of street lighting. G-Scale lamps made by German or American companies are however really expensive. Taking into account the length of my roads, I knew I should be looking for a cheaper alternative. I have again turned to Ebay for ideas...

Ebay seems to be full of "G-Scale stuff". Unfortunately when it comes to street lamps, most of the offers feature items that are not really G-Scale. Lamp posts smaller than 10cm (4in) are barely taller than people figures. I was sure this would not look right...

That's when I found this auction:

It promised much, much taller street lamps. The price was a little higher but still acceptable. And so I purchased a set of 4 items for less than $15 including shipping. The merchandise arrived a few weeks later...

The packaging is typical for cheap items from China. The product is however well protected and everything arrived intact.

The set consists of four lamp posts and four resistors that we need to solder ourselves. Interestingly the seller sent me a wrong set of resistors but realized his mistake, contacted me and sent another package with the correct ones. Good service!

I really wanted to open the model to see what's inside. It is however well glued and I gave up as I was afraid of damaging it.

The model is made of plastic that feels rather cheap. On the other hand, it also feels like the low-quality material was used to produce a rather detailed and properly constructed item. Nothing falls apart and there's definitely no "crappy" feeling. Still - the cheap plastic brings doubts about the possible outdoor performance.

The lamp is indeed taller than other Ebay products. It measures around 18cm (7 in).

The resistors provided with the lamps are 150 Ohm. This means that at 16V a single lamp would consume around 100mA. This seemed like a lot to me...

And indeed - with the default resistor the lamps are very bright. I've decided to experiment a little and see what brightness I would consider valid. Here is a comparison of the light produced for different resistance values.

In the end I've decided to go with 470 Ohm resistors. After installing the lamps on the layout, I can confirm the decision was good. The light is still very bright but not blinding in a night setting.

Here's the setup just before dusk...

...and here it is right after the sunset.

One thing I do not like about the lamps is the fact that their light is not directed to the ground but to the sides. This means that even though they're very bright, the streets and the sidewalks are not really that well lit.

The overall outcome is still very satisfying. The lamps look nice and the part of the layout were they got installed became definitely much more interesting now. I'm going to wait till spring and see if they survive the winter. If they do, I will buy many more in order to light up all my roads.