Wednesday 30 July 2014

Homemade rotary tattoo gun from salvaged parts

Andrea has spent the last few weeks perfecting his homebrew tattoo gun, made from parts salvaged from most of the kind of equipment found around the hackspace. In particular, he found a simple servo-motor (from a large all-in-one printer/copier) and stripped off the encoder disk (sorry, CNC Paul, just didn't think to ask if it'd be any use to you!) and built a laser-cut frame around it.

The first design was made almost entirely "by-eye" and just happened to fit the components that made up the rest of the gun. After being dropped on the floor a couple of times, the flaws and weak points in the design soon became apparent, so a bigger, better, fancier design was put together.

The new design is not only made from super-awesome edge-light acrylic, and engraved with a cool retro-looking tattoo pattern itself, but extra bulk has been added around potential weak points.

The two pieces of acrylic are held together using M3 nuts and bolts. It's important that the nuts don't crush the acrylic, but maintain a nice, firm grip on it. So each bolt has no less than three nuts - one behind the head of the bolt, then another as a "spacer" and the third holds the acrylic tight up to the spacer bolt.

Using this arrangement, the nuts can be done up quite tightly, to grip the acrylic, without fear of the extra stress on the plastic causing it to bend and shatter under strain.
The handle (or grip) holding the needle is disposable, so a collar with grub-screw is used to attach this to the gun. And a simple rubber band applies enough pressure onto the needle to stop it flopping around as the rotary attachment drives it up and down, but without restricting its movement vertically.

The actual rotary head is an aluminium pulley, modified by adding an M3 bolt, slightly off-centre. The head of the tattoo-ing needle is connected to this, and as the motor spins, it causes the off-centre cam to drive the needle up and down, inside the grip.

Here's a video of the gun in action. Compared to the noisy clatter-clatter of a coil-based gun, this one is incredibly quiet.

The final result is quite impressive.

In operation, Andrea said that there was very little vibration compared to a coil-based gun. It's obviously much, much quieter, and there's also very little "splatter" while drawing lines. The lines are very precise and - apparently - "skin burn" is much less than when using a coil-gun.

The gun is also pretty handy at shading too. Using just a single colour ink, it takes a pretty decent gun to be able to create like a greyscale shading effect. Something which the rotary-junk-gun handled well too:

The crescent moon in this photo was completed in just a few minutes but proved, as a simple test, that the gun performs well as both a "liner" and a "shader".

Unsurprisingly, Andrea is quite pleased with the great performance from such a simple device - made relatively cheaply from a few parts lying around, that would otherwise have been confined to the tip.
Look out for more progress reports (and more members of the Nerd Club sporting new tattoos) in the coming weeks and months!

Sunday 27 July 2014

Die cutting with a 20 ton hydraulic garage press

Nick from CustomStuff asked a favour - she's made us plenty of playing/gaming cards for prototype projects like non-RFID poker cards in the past - so it's only fair we helped out.

Having got a rush order in, and having problems with her own cutting setup, we were asked to produce about a hundred playing card boxes, with little more than a pre-made die and some pre-printed patterned (and laminated) A3 card. What we needed was a way of squashing the die into the card with enough force to cut where it needed cutting, and to crease where it needed creasing.

Our answer was to put into use a piece of equipment we've had hanging around for a long time (in Paul's living room since last Xmas, when he used it to emboss some handmade leather gifts). It's a 20 ton hydraulic hand-operated garage press.

It took a few goes to get the process just right, but eventually we managed to get a good, clean cut on just about every one!

The process was to place a 1" thick steel plate (thanks to CNC Paul) on the press, then a sheet of mdf as a cutting mat (a regular cutting mat is just too soft, as we found from experience!) then the laminated card on top of the mdf, the die (cutting side down) on the card, then a second 1" thick steel plate on top.
Those steel plates are pretty darned heavy - each one weighs over 20kg - no mean feat to lift on and off the cutting surface, every single time!

With the steel sandwich in place, the press is lowered onto the top steel plate. A good few pumps and there was a slight - but noticeable - click, as the blade cut into the card. For the first few attempts, we stopped at this point, but found that some boxes didn't cut all the way through, all the way around. So in future cuts, we placed the die such that the cutting point of the bottle was off-centre, and pumped the handle almost until it wouldn't go any futher (without too much effort).

The result was a perfectly cut (and creased) box outline, every time.

After about 30 or so boxes, cutting performance fell slightly, with some corners of the box not cutting cleanly on every cut. This was fixed by replacing our mdf cutting mat. After cutting so many boxes, the mat started to fall apart in places.

A new piece of mdf, and cutting was as clean as the first time. After a few teething problems (and lifting a 20kg weight onto and off the press about a hundred times) we ended up with a stack of playing card boxes, each cut exactly, and each with crease lines in all the right places. It's nowhere near as quick as using a clicker press, and a lot more cumbersome, but we got through them all in just a few hours (maybe a bit more, including stopping for a brew and a chat in between every 10 sheets or so).

It was quite satisfying, not only to be able to return a favour, but to get such a successful result from something so crude and unlikely to work!

Tuesday 22 July 2014

When did eBay turn into Maplin?

For years we've turned to eBay for cheap, low volume components. Quite often searching for something on Farnell or RS returns three or four of the same thing from a supplier on eBay (unless buying in suitably large volume).

But over the last six-to-eight months, we've been receiving fewer and fewer parcels at Nerd Towers from eBay, and more and more coming from "proper" online retailers. Here's just one example:

We're looking to prototype some simple kits for the up-and-coming Brighton Mini Maker Faire. These kits will be used to create simple motorised drag racers. As ever, habit suggests we hit eBay and see how cheaply we can get hold of the little hobby motors.

any number of variations on 3v motor, miniature motor, 5v motor, 9v motor and so on returned similar results; the cheapest being £1.43 per piece

the cheapest "hobby motors" are £1.59 each

Maybe there are cheaper miniature motors on eBay, but finding them is not easy. Or maybe they're not there. But according to the level of search we used, a miniature hobby motor costs about £1.50

So we were quite amazed to find exactly what we were looking for on the Rapid Electronics website for just 53 pence!

That makes the cost of eBay-purchased products three times the cost of buying them directly from a supplier! But it's not just on electrical items. It seems that sellers are now passing on their high sellers fees and making the exact same products on eBay more expensive than in their own online shops! for example, sell 0.8mm birch ply online. It comes in 9x11 inch sheets, available both in their eBay store, and directly from their website.

In the eBay store, each sheet costs £3.25. But in their own online store, each sheet costs just £2.85

It's long been a running joke that when you buy anything from Maplin, you have to pay the "Maplin Tax". This is the extra you get charged for being able to walk into the store and pick something up off the shelf (if they actually have what you want in stock) instead of waiting for next day delivery from your favourite online retailer.

But there's very little point in an "eBay Tax". There's no benefit buying something at a higher price from one online retailer than from another - better to buy direct from the seller, and let them keep all of the money spent at their online store, surely?

These days, we're finding we use eBay more to browse products, then look to source them elsewhere. Which is quite ironic, given that high-street retailers have been complaining that this is how customers shop in their bricks-and-motar stores - find something they like in-store, then go home and buy the same thing off eBay!

eBay used to be a fantastic world-wide marketplace, where you could reliably go and pick up pretty much anything, easily and cheaply. Yet these days, it's increasingly difficult to find exactly what you want from the eBay search results, meaning it's not easy - and the prices are noticeably more expensive than buying from other sources, so they're certainly not cheaper. 

Maplin is also neither of these things (even in-store it can be difficult to find what you what because of either the surly staff, or the fact that it's not in stock) but at least they have the advantage of being convenient. When you need a 22uF capacitor today, you can pop in and pick one up (albeit paying about 50 times its value). eBay is slowly turning into the online version of Maplin - but without the added benefit of being convenient!

Monday 21 July 2014

Laser cut planks for Wild West 28mm terrain buildings

Here at Nerd Towers we love 28mm board gaming - and making up and painting the terrain as much as the miniature playing pieces and playing the games themselves.

A quick (but not-so-cheap) way of getting a really impressive looking gaming table up and running is through the use of laser-cut buildings, from companies such as or Sally4th etc. These buildings are, as often as not, lovely detailed, atmospheric pieces which look great when painted up.

The problem is, being familiar with laser cutting, and seeing how these buildings are made, they're a little bit "lacking" for many in the "Nerd Club Gaming Club". It's hard to say what it is, but scratch-built terrain, complete with little imperfections and slight errors always seem to have a little more "character" and (dare we say it) "realism".

I, personally, haven't played and enjoyed a video game since about 1989, when Jon Ritman and Bernie Drummond released Head over Heels on an affordable label. Not long after that came out, when I discovered that ZX Spectrums were not actually coded using a ZX Spectrum, I started writing my own software and simple games - and most computer games become exercises in moving sprites around a screen. I still don't play videogames, even 25 years later on, even after all the advances in technology. At the end of the day, I'd rather be making simple games than playing even the most immersive multi-player dungeon crawler.

Head over Heels - the last decent video game ever made?

And laser cut buildings on a board game terrain leave me cold for the same reason - rather than marvel at the periodic detail, things like finger joints and how the cutting kerf has been disguised catch my attention more than the model; I'm looking more at how it was put together, rather than enjoy the finished product.

And this is mostly, I feel, because most laser-cut buildings have very little construction needed - they're essentially flat-pack, prefabricated miniature houses. They have lots of large, flat surfaces, with some textures laser-engraved onto the surface to make them a little more interesting.

This is partly why we recently spent so much time hand-assembling a scratch-built roof for an Old West building from plasticard - to give the final model a little more texture and character. The problem with that approach was that it was really (really) time consuming. So we came up with a compromise of laser-cutting rows of slates and assembling them to create a textured roof, with overlapping tiles, but which took a fraction of the time to put together.

After successfully making our rows of roof tiles, we look the plastic-wooden-planking idea to the laser cutter to see how that would translate too. Instead of carving the card with a knife and then scratching woodgrain into the surface, we drew some slightly wobbly planks (to simulate them being warped and worn in the sun) and drew some free-hand lines over the surface (in Inkscrape) as grain.

(A variety of different worn/weathered planks; to create a straight edge, simply turn the plank upside down - so the wobbly bit is at the top - and overlap with another plank above. The two straight pieces at the bottom are for edging the sides of a "lapped plank" wall)

Our LS3020 laser cutter can etch and cut at the same time (or, rather, one immediately after the other) but the driver isn't as easy to use as the BuildBrighton white laser - you basically need to create a bitmap for etching (the woodgrain) and a vector image for the cut-lines. The NarlyDraw software that our LS3020 uses can only handle engraving from bitmaps, and cutting from vectors.

So this is where the fun starts:
It's easy to import a bitmap followed by a vector image, and line them up by eye. In fact, because our wood grain isn't critically placed, we could have done this. But we wanted to try out (and document) how to go about lining up the cut and etch images. So here goes....

Firstly, create the cut and etch drawing in Inkscape on two separate layers (one for the etching, one for the cutting). Place the etch layer above the cut layer, and make each layer visible and invisible, one at a time, to check that all the lines are on the correct layers!

Now draw a little cross (we used a 2mm x 2mm) and place it in the top left corner of one of the layers. Copy the entire cross and paste-in-place on the other layer. We now have a marker that we can use to align the two images in NarlyDraw. (If you're not confident of important the images to the same scale, you can always put another cross in the opposite, bottom-right corner, and use this as a guide to get both images to scale).

Hide everything except the etch layer, select everything and export the selection as a bitmap (Inkscrape exports to .png so we had to load this into Paint Shop Pro to turn it into a 1-bit bitmap: the one bit thing isn't necessary, but it's handy to do, to keep the file size down!) Now delete everything except the cut lines (or copy the cutlines and the cross to a new document) and save as .dxf.

With a .bmp (for etching) and a .dxf for cutting, simply import the two files into NewlyDraw, zoom in and make sure both layers line up on the same cross.

Lining up the cutting guide on both the etching and cutting layers. When both are perfectly lined up, the final result can be previewed easily in the NewlyDraw application

It was with some trepidation that we hit the "go" button, but shouldn't have worried!
NewlyDraw did a great job of etching the bitmap image first...

...before cutting out each plank from the surrounding wood:

Each set of planks takes only a couple of minutes to complete. They overlap nicely and sit quite flat, to create a realistic, weathered wooden wall for a shack or a shed

After a few of our sample offcut sheets had been fed through the laser cutter we had enough planks for the walls to build a small building

Another couple of goes through the laser cutter and we had plenty of roof tiles too

All that remains is to assemble a building using the bits and paint it up, so see whether the wood holds the grain pattern after a layer of acrylic has been applied and drybrushed  (etching it would be pointless otherwise!) or whether we need to use something more like an ink or a stain, and just keep the grain lines darker.

After all this messing about with laser cutters and trying out different approaches to etching and cutting on the same piece, it was actually quite late when we finished last night; so the build and paint begins in earnest this evening!

Sunday 20 July 2014

Laser cut slates for 28mm terrain roofs

After getting the laser cutter up and running at the new Nerd Club home in the Boiler Room Studios, we threw some simple designs at it and had it carving out roof shingles from some 1mm birchwood laser-ply at an impressive 32mm/sec, at just 15mA.

After the nightmare that has been the BuildBrighton laser cutter in recent weeks, it was really nice to just load up a file, hit go and have some super-intricate parts drop out of the cutting bed!
We've cut a mixture of "regular" and "irregular" rows of slates and even without the cutting and swapping around we expected to have to do, got some nice-looking roof panels pretty easily

We'll have to design and cut some ridge tiles, but the overall effect is far better than a laser-etched plank of 2mm mdf - and something we're looking forward to putting on our Wild West buildings.

Sunday 13 July 2014

UV exposure unit ready for fitting

About six weeks or so ago we pulled apart some UV acrylic nail art lamps with a view to creating an exposure unit for fixing images onto silkscreens (and for fixing solder mask to etched pcbs).

The project stalled for a while due to a number of things, but having recently got the laser cutter working, and successfully getting the CNC router to move about (albeit not exactly drawing shapes yet) we've been inspired to spend a bit more time kitting the unit out with stuff that "just works" so it's on hand, ready to be used, as the need arises.

With this in mind, we spent some time last night getting the UV exposure unit finished - well, the electrics for it, anyway

The controls are all wired up and at the minute we're running the whole thing straight off the mains. All that remains now is to line the bottom of the tray with reflective foil (to hopefully reduce any chance of "striping" as the unit won't actually sit very low down on the underside of the desktop). Once it's installed in our large desk we'll put a relay switch inline with the power supply, so we can control the exposure time using a homebrew digital timer (of course, made from a PIC and a max7219 and some 7 segment LEDs!)

Just to make sure nothing had got damaged during the rewiring, we powered it up to see that it all still worked. Amazingly, it did.

It was only afterwards we realised that had it gone horribly wrong, none of us actually knew were the fuse box is, in the Boiler Room Studios - we could have been left in the pitch dark, feeling pretty stupid. Thankfully it just worked; but we'll make sure we know where the fuses are before we do anything else with mains wiring - just in case......

Friday 11 July 2014

HPC LS3020 laser cutter up and running again!

A while back we secured a unit in a shared studio in Hove, for making messy things and generally doing the kind of things that are likely to either upset the family home or the neighbours (laser cutting, CNC routing, screen printing etc.)

Our LS3020 laser cutter from HPC has been in storage for a long time -maybe about 12 months or so - and it was to great delight that we got it connected up and working again last night.

Given the amount of hassle the laser cutter at BuildBrighton has caused in recent months, we were terrified that the mirrors might need realigning, the cutting lens could be damaged... any number of things that could have gone wrong with it, as it's been dragged around a few different places and never actually tried out, for a long time.

Amazingly, we switched on the old PC that was driving it last time, plugged everything in and hit "go". It cut our 3mm acrylic sheet at 15mA and at 16mm/sec, straight through, first time!

Because Nick, Charlie and Paul were also at the unit during the first test, they each got a yellow nameplate. The cuts are spot on - no beam spreading or half-cut lines; the cuts are nice and "tight" and well focussed. Some of the pieces needed pushing from the sheet - not because they were badly cut, but because the cutting was so precise, the parts had to be lifted exactly vertically to get them to come away.

Sometimes, on the BuildBrighton laser, everything falls away from the sheet as it's lifted off the cutting bed. While this is encouraging (at least the laser manages to get all the way through the material) sometimes it's only after two or three passes. And that means that sometimes the cuts are wider than they would normally be. Not only were all our cuts cleanly through the entire material, the cuts were so tight that the pieces were a little bit "squeaky" as they were removed from the sheet. At 16mm/sec we erred on the side of caution - there's a chance that we might even get away with cutting at a faster rate.

But for now, we're just thrilled that we have a working laser cutter once more. Admittedly, the extractor wasn't connected up properly during the test cut, and the studio did smell a bit like pear drops for a while, but as soon as that's all plumbed in, we'll be super-productive, laser cutting night and day, no doubt (for a while, at least!)

Thursday 10 July 2014

The internet continues to amaze

For years the internet has been amazing, and continues to amaze.
When I was first introduced to Compuserve over a hacked retail till modem running at 14.4kbs I struggled to comprehend what I was seeing. When, back in 1994, I was told "just type in whatever you're looking for", in my naivete, I didn't think to enter "tits" or "porn" - I just put "carrot soup". From memory there were about 36,000 results. My brain hurt. In 1994 the internet was amazing.

Years later, as I wrote entire DMS systems in HTML and Javascript (long before it was cool to call Activex objects AJAX) I was blown away by how anyone could not only contribute to the internet (not just the world wide web), but build massive, complex systems on it. In 1999 the internet was amazing.

Seeing those fantastic, simple animations and tiny file sizes when Flash hit the 'net was pretty amazing too - then it became a fully-fledged web publishing platform, for games, apps and even music and video online. It was all pretty amazing stuff -as was watching it all disappear again, as Web 2.0 rebooted the 'web - and almost every application became a browser-hosted application.

In short, the internet is pretty impressive. It was when it first came to public consciousness, and it still is today. More than anything, as well as being a massive repository of data and information, the internet has made interaction not just possible, but essential. Not stupid social media, and tw@tter updates, people showing photos of their morning breakfast toast-and-jam-(nom-nom), but being able to speak with people, literally from all over the world. Talking about anything. For any reason - sometimes for business, sometimes to buy, sometimes to sell, and sometimes just over a shared interest.

What amazes me more than anything about the internet are the people who use it. People who give up their time, expertise and experience - all for free. For no reward other than seeing a job well done. Lots of people use the internet to give things away: whether it's open source software, designs for the latest 3d printer, or - sometimes - something a little more personal.

I had one of my "isn't-the-internet-brilliant" moments last night, when I received an email from Martin (who runs He'd been reading this blog, seen we'd had problems laser cutting some design ideas, and rather than wait for the next painful installment of disappointment and broken laser tubes, he downloaded the designs and cut out some of our projects, so we could see the results straight away!

Without prompting, and for no reward (we had no idea he was doing it) Martin tried out one of our ideas and sent the results straight back. He even sent some comments and suggestions for possible improvement (yes, the tiles do look a little to far apart in the final cut!) and used different materials to demonstrate the different effects. And all because something had piqued his interest enough to want to contribute.

Our favourite has to be the (left-most) 0.8mm birch ply version: the tiles look about the right thickness (though the left-most 2mm mdf looks pretty good too, for a rough-built log cabin, say).

What was really amazing was that someone, somewhere, saw what we were doing and wanted to get involved. And just did.

Some days I really hate the internet.
I work with it every day. I build websites. I write complex software and use a mix of web- and non-web languages. I love the underlying technology, but some days I hate how it's abused. I hate adverts on just about every YouTube video. I hate that people allow them to be put on their videos (the person posting the videos can decide to turn them off if they want). I hate viruses and pirated software. And stupid in-jokes about cats, and massive video attachments in pointless emails, asking me to "share this".

And porn. Maybe it's considered weird - but I don't want porn on "my" internet. I want it to be a place of learning, shared interests, marvellous inventions and a community of people making everything better, one stepper-motor-driven device at a time!

But some days I really love the internet.
Thanks to Martin's email, today is one of those days.

Wednesday 9 July 2014

Laser cut roofing shingles for 28mm terrain

After the dismal failure of silicone mould making, but the tedious task of scratch building roof sections for our board game terrain, we decided to try laser cutting lengths of roofing tiles (rather than etching the entire, flat panel) to make our 3d models looks a bit more, well, 3d.

Using our scratch-built roof as a guide, we made each tile 3.7mm wide, 6mm high and spaced them 4.233mm apart. This means that a 4" strip of tiles can easily and cleanly be cut into 1, 2, 3 or 4 inch strips, to match any width of building.

Making the image didn't take very long at all using Inkscrape.
Our first roof section was just three inches wide, and we're unlikely to make any buildings more than 4" wide (they would take up too much room on our relatively limited 6x8" board sections) so 4" strips just seemed sensible. If ever longer runs are needed, it's no trouble to join two strips end-to-end to make longer runs of tiles/slates for a roof.

For a regular (modern, well-built, slate) roof the tiles work well when placed in a row and by simply overlapping (and offsetting) one row above the previous one. The red lines in the image below show where they can be cut to create a 4" wide section.

For a slightly more old-fashioned roof, just a single strip of repeating tiles gives quite a nice effect. We nudged a few up and a few down, and rotated some by just a few degrees. The tiles are all still roughly the same size (on our scratch-built roof, the tiles were all different sizes and fitted like  jigsaw to get them to overlap properly) but the end result is still a pretty rough-looking surface.

To avoid repetitive patterns, the higgledy-piggledy row of tiles can be cut off at different places, so the pattern starts on a different tile at different places throughout the roof section. The tiles can, of course, be flipped over, to make them lean in different directions. And then, there's always the possibility of mixing shorter sections of straight and wobbly tiles on the same roof to create an old slate roof - generally regular, but with the old slate slipped or out of place:

The designs appear to work pretty well onscreen.
All we need now is access to a working laser cutter to give them a try!
Here's a pdf of the roof strips - load into Inkscrape and copy/paste whichever parts you need:

The top two rows are ready for laser cutting. The bottom two rows allow you to ungroup each individual tile (and the connecting strip) to rotate/resize/respace as necessary. When done simply select the entire row of individual tiles as well as the connecting strip and select "path -> union" from the menu, or hit Ctrl & + to join all the shapes into one single shape.

Making silicone moulds from scratchbuilt terrain

It was an interesting exercise, albeit a pretty  fruitless one.
Without a degassing chamber, it was always going to be risky - though any bubbles that might have formed in the mould would simply become small bumps on the final, cast shape, which could easily be cut/sanded away from the dental plaster. As it turned out, the RTV silicone we used came out in a single, big lump, rather than a long, thin, fluid stream as the YouTube tutorial videos suggest.

Maybe it's just a lack of technique here - we treated the silicone like dental plaster and just wanted it out of the mixing cup before it set, but the set-up time for silicone is hours, not minutes like plaster!
A bit of banging on the table and quite a few bubbles came to the surface. This either meant we'd made an excellent mould and managed to get all the air bubbles to the surface, or the entire silicone mixture is likely filled with air bubbles throughout and the mould is already ruined!

After making up some silicone and leaving it overnight, taking the roof section out in the morning was quite exciting. The silicone poured into the mould as a thick, gloopy goo - nothing like the thin, flowing liquid-like properties of the dental plaster we use when casting pieces. We weren't even sure whether the silicone would pick up the details because it was so thick and gloopy.

The silicone worked really well.
In fact, it worked rather too well, creeping underneath our individual layers of tiles and making a thin film of silicone between each course on the roof. This made removing the roof from the mould quite tricky, without ripping the really thin layers of silicone.

The end result was a not-so-satisfactory mould. There's no way it can be used as it is. And our beautiful wobbly tiled roof section now has loads of little bits of pink poking through between the tiles. So it looks like we're going to have to make a new roof section as well!

We don't really fancy making another 3" wide roof section again from so many individual tiles. But simply engraving a sheet of mdf/plastic doesn't really give a very satisfactory result - the roof still looks like a flat sheet once the building has been constructed.

So maybe there's a middle way.... maybe we can draw and laser cut strips of shingles, all connected along the top edge, to glue on in overlapping fashion to create our roof effect - in this case it'd still mean gluing about ten strips (rather than lifting a complete roof if we stuck to just laser etching) but that's still less work than we put into making a single row our course of tiles on this roof section!

Tuesday 8 July 2014

Scratch building gaming terrain for Wild West board game

The laser cutting to make linka-like moulds at BuildBrighton was a bit of a failure last night. The laser cutter is, once again, out of action. We really need to spend some time down at our new unit getting our laser cutter up and running, so we can just crank this sort of stuff out!

In the meantime, after seeking some advice, we tried scratch building some Wild West style buildings. There are a number of companies online selling laser-cut buildings, which are great for modern games, or near-future type boards - but for old Western type games, they all look a little bit "regular".

This is a typical Wild West type building. It's weather bleached and a little bit higgledy-piggledy. The board planks are warped and all over the place. This is quite difficult to recreate with a sheet of uniform, laser cut, 2mm or 3mm mdf.

Someone at our local model store suggested coffee stirrers as planks.
It seemed like a good idea, so we set about making a wooden shack from actual wood

We started out making a basic framework for our building using some 3mm foamcore board. Then stuck a load of coffee stirrers to the front side and cut them to length. We deliberately cut them off the board, then placed the shorter lengths onto the foamcore. This helped to create the slightly wobbly, less-than-perfect edges along the sides of the building

It would have been easier to glue all the planks in place, then just run a knife blade along the edges, but because our wood is still quite regular (and a bit chunkier than the thin planks used in the building in the photo) we tried to emphasise the rough look of the building by making the edges irregular and a bit out-of-alignment.

Making the building was actually quite a nice process - far more "art-and-crafty" than the usual "engineering and construction" approach used (when building rooms from the Hirst Art moulds). While the glue dried, we taped the sides together and checked the overall size against one of our 28mm Wild West miniatures.

The front of the building was extended beyond the roof level, as was common with commercial buildings in Old West towns. The building is 3 squares wide and 2 squares deep. It's a little on the small side for an actual, to-scale building (but a bit too large to be an outhouse) but it looks fine against the miniature as a generic wooden building. Maybe a simple sign on the top part of the front would make this look like a believable "general stores" or something?

The roof for the building could be made from a number of materials.
A lot of buildings had simple wooden shingles, many had corrugated tin roofs, and occasionally a brick/stone building might have a slate/tiled roof. We couldn't  decide which to go with, and eventually settled on shingles.

Our local model shop has sheets of plasticard for a quid a pop which was perfect for trying out a few ideas. Before we actually started on our roof, we looked again at the reference photo and decided that our coffee stirrer building still looked a little regular. What we needed was an overlapping plank built building.

We cut some of the plasticard into strips. Luckily we didn't have a steel ruler to hand, and pressed just a little too hard with the knife blade, which created the exaggerated, wobbly edges. Horrible for a cut plastic edge, but perfect as some rough-cut wooden planks.

Along the length of each plank we ran the tip of a craft knife, to create a wood grain effect.
Although it's not easy to see this on the white plastic surface, once it's painted and drybrushed, the deep cut, along with the slightly raised edges on each side of the cut, will give a really strong, visible wood grain.

Perhaps it was the solvent fumes from our plastic cement, but once we had a decent stretch of "planked wall section" from plasticard strips, we had a crazy idea about using the same technique for our shingle roof. Except instead of using strips, we'd cut each strip into individual shingles, and glue them together to make a roof. What a bonkers idea!

We took some of the left-over strips from the wall section, and cut a few more, then sliced them into roughly rectangular shapes. Luckily everything in our Wild West reference picture was made from rough-hewn wood - slightly dodgy shingles with uneven sizes and wobbly edges are just what we need!

Making the roof was a slow, tedious job. Every layer or two we had to wait for the plastic cement to dry enough to make the previous row secure (it works by melting the plastics together, so the shingles are really fragile on the roof, until the glue has dried fully). But after a little while, the roof was complete.

In place it looks ok. The whole building looks fine. It needs a layer of paint and maybe it'll start to come to life. We're really pleased with the (painstakingly built) roof and the rest of the building looks not to bad - it certainly has more character than some of the rather flat, laser-cut western buildings available online.

Our plasticard isn't really thick enough to make the awning frame from, so perhaps some laser cut mdf just for that would suffice - then we can stick a rough-cut planked roof, or maybe some corrugated metal on top to finish the model. A large sign for the top part of the frontage, perhaps with a thin wooden frame would be more inkeeping with our slightly more regular house-build than a hand-painted sign (as per the reference photo) and that's our first western building done!

We'd like to have another go at making some more western style buildings, using the plasticard strips, as well as trying the odd brick building (maybe a bank or a hotel). But the thought of making more of those roof sections is a bit daunting.

So before we go sticking everything together, we might just dig out some silicone and have a go at making a mould of this roof section. If it's thick enough (our current roof section is at least as thick as the Linka wall pieces we made the other day, but much larger) it'd be far easier to just pour some (dental) plaster into a mould in a few minutes, than spend two hours or more gluing each individual roof tile in place!

Monday 7 July 2014

Linka gaming terrain moulds

Paul has been busy casting loads of Hirst Art pieces for our Space Raiders boardgame terrain - enough to make a few decent sized rooms at least! He also came across some alternative moulds from the late 70s/early 80s called Linka.

Just like Hirst Arts moulds, these are best used with dental plaster/stone, to make solid, robust wall sections. Unlike Hirst Arts, the walls are more like thin, flat panels, than blocks to make walls from.

The moulds cast really easily and the pieces, although thin, are relatively easy to get out without too many breaking and snapping during de-moulding (some of the long, thin sections are susceptible to breaking as the mould is flexed to get the other sections out)

One really nice element to the pieces is the way they can be connected either in a long, flat section, or at right angles, to make nice, sturdy wall pieces.

These moulds were originally created for Hornby 0-gauge and 00-gauge model railways. As such, they're not quite to the right scale for our 28mm miniatures.

The 0-gauge stuff looks fine against a 28mm western-styled miniature. The bricks may appear a little larger than if they were made to the same scale as the character, but they're not completely out-of-scale. Given that most 28mm miniatures have "cartoon" proportions (large heads, feet and hands) having larger-than-they-should-be bricks in a wall section isn't the end of the world.

The smaller 00-gauge stuff also looks ok, set against a 28mm character. In fact, the bricks look more in scale to the character than the 0-gauge sections.

But it's immediately obvious that the doors and windows are far too small for this size playing piece. What we need is something somewhere inbetween....

Using the Linka moulds as "inspiration" (we definitely did not just rip off their idea but make the wall sections a little larger) we drew some wall sections in Inkscape, but made them a little larger.

We kept the interlocking edges so multiple pieces could easily be connected together. However, we made each wall section almost as high as a playing piece. By creating some "full-height" and some "half-height" pieces, they can be stacked one on top of the other to create either single-storey or multi-storey buildings.

We made each wall section in 1", 1.5" and 2" sizes.

(In the image above, we've incorrectly labelled the 1/4" section as half-inch. Sizes are approximate: a one inch section is 7 bricks across, but a quarter-inch section is two bricks across. Obviously, four of these sections would be longer than a single, one inch section - the length, in inches, is really only a guide, to help determine how many playing squares each wall section would cover)

After placing the layouts alongside some 1" squares (our board game playing surface uses 1" squares) it became apparent that we'd need to do some jiggling about:

To place a door, for example, it would either have to go in the centre of a single 1" piece or to one side of a longer 2" piece. In either case, we'll end up with a door frame very close to one edge of a wall section, which is potentially a point of weakness. We can't place a door in the middle of a 2" piece, because then, when it is placed on the board, it would fall between two playing squares, and not line up with a single one.

An alternative to this is to create a 1.5" wall section, into which we can put our doorway, allowing about a quarter of an inch of material around the sides of it. In this way, we can put the door in the middle of the playing piece, but we'd need either quarter-inch or three-quarter-inch wall sections to go on either side, to create either a 2" or 3" long wall section.

As we couldn't decide which to choose, we drew both.

As we already have some liquid silicone and catalyst here, the idea to laser cut and engrave the brick pattern onto some 3mm acrylic and - assuming everything works as we hope - make a rubber mould of the final piece; this will allow us to make multiple wall sections quickly and easily from our dental plaster, without tying up the laser cutter for hours at a time!

Progress will be reported (successful or otherwise) in a later blog post......