Making EVA terrain boards


Using interlocking exercise mats to make terrain boards

I've been asked to write something about exactly how I made my terrain boards, so here it is...

A large 1:300th scale moderns game under way, using 10 interlocking EVA foam-based boards.


1. Background: why terrain boards - and EVA boards in particular?

2. Getting started: choice of EVA boards; materials required

3. Design: board flexibility

4. Construction: the hard part!

5. Variations


Most moderns figure games (by which I include WW2) require a significant amount of terrain to get the right degree of cover for troops to shelter behind, but getting all this arranged on a table top can take a great deal of time; and the final result often looks a bit tatty as well.  Using pre-set (or partially pre-set) terrain boards can greatly ameliorate these problems. 

The main disadvantages of using terrain boards is that they have to be constructed beforehand, and you need quite a lot of them if you are to have a decent variety of terrain - playing over the same battlefield over and over again will quickly become very boring. 

There are also technical issues to consider, such as warping and weight.  These two however, can in practice be greatly mitigated by using EVA foam mats as the base for the boards.  These can be obtained very easily on-line, as they are sold for use as light exercise mats.  Typically, they are square, 10 mm thick and 600 mm a side (that's 3/8" thick and 2' a side for Americans), and have interlocking edges so they fit and lock together; one face is usually dimpled, and one is flat.  Ideally, for wargaming purposes, both sides would be flat, but mats like this are less common; the dimpled side is useless for gaming purposes, so needs to be the bottom face.

Such EVA mats are also fairly cheap - something like 5 dollars US each at the moment.  You'll probably spend far more money on flock, glue, varnish, and other stuff making your terrain look pretty than on the actual cost of the foam underlays, to say nothing of your much more valuable time!

EVA foam mats are also light: 6 of them together will weigh only 2 kg.  So weight isn't an issue at all.  You do need to consider the issue of storage space for them - especially once you start adding hills and things so they become more than 10 mm thick.  But of course, stand-alone terrain and tablecloths also take up space, even if not quite as much... 

Furthermore, because they are flexible, warping isn't a real issue.  They do warp, in the sense that if you stack them together, they may not lie completely flat, but that doesn't really matter when you come to fit them together.  The fact that they interlock means any curvature in one board can be reduced when it is tiled with another board, which won't have the same curvature.  So you get a battlefield that actually lies quite flat.

Getting started

So what materials are you going to need?  First up are the EVA boards themselves.  I simply ordered mine through, since I live in Japan, but you can find them pretty well-much anywhere, and since they are so light, postage isn't much of an issue. 

In order to have a decent variety of terrain, and not to limit re-playability, you'll need at least twice as many boards as your "usual" gaming table size, and three times as many will definitely be better.  I've currently got 16 planned, as a I usually play on a 6 board-sized table (6' by 4'; 180 by 120 cm).  Of these 16, 8 are finished, and 2 more in a state sufficient for gaming (if not yet completely "finished"); I will probably plan some more to bring the number up to 20 or more. 

Note that these sixteen boards stacked together come 30 cm high, so take up twice as much vertical space as the original flat mats would by themselves.  That's 4 cubic feet of space, so not inconsiderable. 

When I bought my mats, there wasn't quite the variety of colours available that there seem to be now, so I had to undercoat mine before flocking, etc.  Since the average 300 ml can of spray paint will cover only half a dozen boards, the price of several cans of spray paint is something to consider.  What's more, paint can wear off.  The edges of the boards, and the corners in particular, can expect to see lots of wear in time, so they might need touching up in the future.  So if you can find mats that are already green, or perhaps even better, mid-brown, go for them, even if they are more expensive.  They'll perform better in the long run.  As you can see above, mine started off a not very useful cream colour.

I intended to flock my boards - trying to avoid the "tabletop cloth" look.  That meant purchasing a whole bunch of ground flock to model fields, etc.  Because I game my moderns in 1:300th scale, the flock also needs to be very fine; most flock is made for people making model railways, and is not fine enough for my purposes. 
I got mine from Tomix, who specialize in model railway stuff here in Japan at 1:150 scale, and theirs ("Tomix Color Powder") is finer than most ranges I know (they tend to be geared to those working mostly at 1:87 scale or similar).  They also have a good variety of colours (10 or so different shades of greens and browns).  Of course, I mix my flocking to provide a more realistic mix of colours, but the more hues in the mix, the better, generally speaking.  Flock isn't all that expensive - the Tomix stuff is the equivalent of US 2 dollars for a (small) bag, and to flock 10 boards, I have gone through less than 10 bags.

The flock of course needs to be stuck down - and stuck down very well if it isn't to come off.  This requires a lot of PVA glue.  I use reasonably thick stuff to stick it down, and then spray very dilute stuff later on top to seal it in.  I've probably gone through nearly a litre, I think, in making these boards.  I bought a cheap plastic sprayer to do the job - sold in gardening stores for spraying pot-plants with water or what-not.  It cost me maybe a couple of dollars?  Finally, the boards need to be varnished to seal everything in.  This requires a matt spray varnish of similar coast to spray paint, or perhaps more, given you will want a few coats.

In terms of construction tools, you are going to need a (largish) sharp knife to carve the foam.  A large "Stanley" box-cutter type blade will do the job.  However, carving through foam very quickly dulls a knife edge.  Things went much more smoothly when I went out and bought a knife sharpener (useful for the kitchen too!) to just renew the edge of the blade rather than replace it with a fresh blade. 

Tank country: a gently rolling landscape
Carving is necessary to shape not only hills but also lowered water courses - one of the main reasons I wanted to try out EVA was the prospect of having rivers sunken into the playing surface rather than sitting proud of them!  It was this photo above posted on Rhys Batchelor's blog "Badly Lead Men" that was the inspiration for me to think in terms of EVA...

Ultimately, the cost in materials probably comes to not much more than 10 dollars or so a board.  What they do cost is time.  A lot of time...


Terrain boards need to be interesting to be useful.  And by interesting, I don't mean visually interesting - despite that being the major reason for their existence; I mean tactically interesting.  Ideally each board will have its own personality - even if it's as boring as "this one's the flat one"!

They also have to fit together.  This requires quite a bit of planning.  While it would be technically possible for hills to span a board join, in practice, this would be so tricky that it wouldn't be worth it.  So any hills need to be small enough they fit into only one board.  At the ground scale I usually play at, 60 cm represents 2400 metres, so we're not going to be fitting any mountains in, but smaller "hills" should be OK.  The main consideration in fitting the boards together is how linear features - roads and rivers - will be represented.  More on this below. 

Another consideration is the vertical dimension.  Hills can be simply made by building up more EVA layers, and gluing them with an appropriately flexible glue (glue that is used to glue shoe soles does the ticket well).  Woods are more tricky.  I want mine to something better than just carpet off-cuts (not least because off-cut carpeting is essentially non-existent in Japan!), so I'll be looking at Woodland Scenics "clumped foliage".  But this stuff will be a bit too fragile (and compressible) to be permanently attached to the boards, so I plan on making woods separate terrain pieces that can be added on top of the laid-out boards.  Hedgerows and the like will be permanently modelled, however.

Another aspect that needs to be considered is the ground/model-scale discrepancy.  The reason I play 1:300th is that this problem is much less worse than when playing the larger scales.  I simply can't get over the sight of tanks lined up literally track-to-track when people play WW2 games using 20 mm (or larger) figures.  Often the shooting range of a tank is only a dozen times as long as the model! 

Above, you can see Spearhead played in 20 mm scale, taken from Robin Sutton's excellent blog, "Wargamingnz" (here's the link to the game report)Some people are perfectly OK with this degree of figure-to-ground scale distortion.  Unfortunately, I'm not one of them  :-(
1:300th scale reduces these problems - but it absolutely does not eliminate them.  If we want to model a village, it will need some buildings.  Do we use the ground scale for these buildings (say 1:3000) or the model/figure scale (1:300)?  That's still a whopping 10 times difference!  (I ignore the third option - something in between, since that just looks silly from both ends).  Now, if we use the ground scale, it looks more than a bit weird, at least to my eye, when our model soldiers tower over the buildings they are supposed to be sheltering behind. 

On the other hand, if we use the figure scale, our model village will be able to accommodate just 1 or 2 buildings, and not the one or two hundred it should have.   Our model roads will also be 10 times as wide as they should be compared to the surrounding scenery.  I think this is the better choice, myself, although I do limit the widths of rivers to be more like the ground scale, so what looks like a small stream is actually representing a decent river, and what looks like a small river is actually a major waterway.  Otherwise a 100 metre-wide river in real life, that should be only 2.5 cm across using the ground scale, will have to be modelled as 10 times as wide if using the figure scale.

Another thing that needs to be considered is the time period.  Now while natural features don't change much with wargaming time period, the same isn't true of man-made features.  Models of built-up areas really need to look quite different if they're supposed to be depicting towns from 1815, 1915, or 1985.  For this reason, my boards don't come with integral villages and towns.  Instead, I have hollowed out slots for them, in which I can place different models depending on what era I am gaming.  This also fits in with the rules I tend to use (Spearhead / Modern Spearhead) where built-up areas are modelled as square "blocks" of terrain.   

So my built-up areas are all on 80-mm square bases that fit into 81-(or so) mm square hollows.  Some look more like farmsteads, of course, or even isolated barns: these represent small "villages", like the two shown in the picture to the left; others look more like portions of towns.

Roads are also similar.  If you're into Napoleonics, you'll need dirt roads; if a Cold War gamer, you'll want tar seal.  WW2 is a bit trickier, since this would depend on place, but with tar seal being more usual for the sorts of roads that are likely to be significant enough to affect a game.  What you really don't want is a mix, as this will greatly complicate things, so you should probably choose based on what your next-favourite era is: WW1 or before - go for dirt; Moderns, or no other period: go for tar seal. 

And this brings me back to how our linear features: roads and rivers - are going to meet up at the board edges.  The simplest scheme would be to have them cross the board edge at the centre of the board edge, so that each board will match any other board (assuming each one has four road entrance/exits).  Unfortunately, there are two problems with this scheme.  One is that it leads to a somewhat stereotyped road net, as the entry and exist points are the same for each board, although this can be disguised somewhat with cunningly bent road sections between the joins. 

Perhaps more importantly, it doesn't allow roads and rivers to coexist on the same basis.  Any rivers will have to go somewhere else, because the join points are already occupied with roads.  This leaves us with two options: the river layout has to be such that it divides the river-containing boards into two mutually-exclusive sets (A and B below to the left) that must be arranged quincunx-fashion, or we have to live with the fact our river boards can only be arranged in one general direction, and not two (see 1 and 2 below, versus 3).

Now, if I was gong to do this all over again, I might actually go with the second option.  If you throw in a couple of boards with only 3 road joins, not 4, and make the other ones far from straight, you can get a not too unreasonable looking road net.  And it means you actually get more options in arranging your (non-river) boards.  But as it is, I went for the first option.  And because I did, I realised that since my river boards were now organised into two sets, I could invert this, and organise my non-river boards in the same way, so the roads were off-centre, while the rivers took the centre slots.  I figured this would give a better-looking round net.

So I planned my boards like this:

I've used dark blue for wide rivers (over 30 mm - one base width - across), and light blue for narrow ones (less than 30 mm across), but I've made sure both are the same width (30 mm) where they join, enabling a narrow river to broaden out into a wide one, if desired (since the joins are actually somewhat staggered due to the interlocking nature of the mats, a notionally 30 mm-wide can actually be made slightly wider or narrower than that in reality).

The road joins, where present, are 1/3 (or 2/3) the way along each edge.  In hindsight, all of the boards with water features should have had 4 road joins (roads are more likely to follow rivers) - the fact that 3 of the 8 only have 3 joins greatly complicates things.  Fortunately, the bottom right board isn't finished yet, and I have been able to give it an extra road join, which will ease things considerably.

It is important to ensure the terrain features on each board are not generally arranged symmetrically and/or centrally.  This means extra variety can be obtained by simply rotating the boards.


Once the designs were finalised, construction could begin.  The courses of the roads and rivers were pencilled in on the boards, as well as the hills, including contours.  I didn't plan my hills actually being stepped, as such (or at least, not markedly stepped), but I needed to know how high and thus steep they would be - I figured one Spearhead contour per 5 mm of height. 

The rivers were the first things to be tackled.  The way I eventually did this, after some trial and error, was to vertically slice right through the board, to cut out the riverbed. 

The removed riverbed was then sliced horizontally (not the easiest of tasks - a very sharp and flat knife is required), so it was only 5 mm thick, instead of 10, and set aside.  The cut-out edges of the boards were then carved away to a depth of 5mm, to make sloping riverbanks, and the now-thinned river bed was then inserted back into the place it came from, but 5 mm lower.  To get a good fit, a clean vertical slicing action is important - the more deviation, the more shoe glue you are going to have to use to hold the reassembled board together, and the weaker it will be.  Having a properly meandering river will considerably strengthen the finished board, as the riverbed won't act like a folding seam in this case.

Once this was done, the hills were tackled.  Low hills (2 contours) and small folds (single contour) can be carved from a single sheet of mat, and glued onto the base mat - again, use shoe glue, as the resulting product needs to keep its flexibility. I overestimated how much "filling" potential the flock would have in covering carving lines and the like from this process, so in hindsight, I should have spent more time removing ridges and scallops that were produced as a result of carving the foam away.  And in particular, I should have paid a bit more attention to the edges, where the hill fades into the base mat. It really does need to be at as shallow an angle as possible. 

Higher hills need a second layer of EVA mat glued on top.  There is a definite limit to how high you can go and a) have a realistic slope, and also b) not have your troops slide off the hills when playing a game.  A slope of about 1 in 4 (25 %) is about as steep as can be managed; on the other hand, physically carving a slope that is gentler than 1 in 8 is tricky, so you don't have an awful lit of room to play with!

The slots for the built-up areas were made in a similar manner to the rivers - cutting right through the mat vertically, removing the cut-out section, slicing it horizontally to reduce its thickness, and then gluing it back into place, but about 2.5 mm lower than before. 

Once this had been done, I undercoated the mats in mid-green; as mentioned before, this step shouldn't be necessary if you have started out with green or light brown mats.  River surfaces were painted brown (no blue!) at this stage.  Next, I pointed in the roads directly on to the smooth tops surface of the mat; in the few places they crossed hills, I first smoothed out the hill surface with a mixture of PVA (white woodworker's glue) and very fine sand (actually chromatography silica, which is easier for me to get hold of, being a chemist). 

Flocking was next.  I made up several different batches to simulate different kinds of fields; etc.  The biggest batch ("grass") was a relatively light green, made by combining a light green flock, with a bit of light brown, some darker green, and a few small light stone ballast chips to help give a less even texture in places: this was the "default" flock, and would be used at the board edges, as well as anywhere else that didn't require something special.  The mix used for river banks was a bit darker, and had a lot more stone chips in it.  There were three other shades of green to represent various crops and what-not, as well as two kinds of brown to represent more or less freshly ploughed fields, and a light brown to represent wheat fields and the like.  There was also a more brownish green shade that I used to pick out top contours on hills to represent possible hull-down positions.

Like rivers, fields present a ground/figure scale conundrum, and as with the rivers, I opted to split the baby and compromise in between: they are rather too big for the ground scale, but much too small for the figure scale.

The flocking was applied by first painting on PVA in the places I wished the flock to go, dumping the flock onto the appropriate places, pressing it down, tipping the board up, and tapping it to shake off the excess flock.  A few small pieces of "clump foliage" were strategically placed in certain places (such as where field corners meet) to provide visual relief, and a few hedges were created, also using clump foliage; mostly about 5 mm high so as to not look too large in relation to the rest of the landscape, and also to prevent them from being too easily dislodged when the boards were handled, especially when stacked together. 

Too many hedges can be a bad thing, especially given I am aiming for a southern German/Bohemian look, where hedges are not very common.  They are mostly there to cut down on lines of sight, especially near the board edges, since there are, of necessity, no hills there, meaning there would otherwise be some rather inconvenient guaranteed long-range lines of sight wherever boards meet.

In this picture you can see some hedges to the left near the board join that stop people having a clean line of sight along that portion of the join.  You can also see a bit of a step in the hill where I didn't carve the thing very smoothly: flock can only smooth out things so much...

Once an entire board had been flocked, and given a few hedges here and there, I took it into the bathroom and sprayed it with dilute PVA (about 1:10 PVA:water) to coat the surface of the flock, and left it to dry for an hour with the extractor running; I repeated this every hour for the next four hours, giving 5 coats in total to thoroughly coat the top of the board.  Doing this in the bath means I can just use the shower hose to rinse of the spray escaping the mat edges down the sink.

After drying overnight, I then took the board outside and gave it a couple of coats of matt varnish.  All this coating does tend to give a slight warp to the board, but it's of not much consequence, as can be seen in the photos - once they are locked together, the warping pretty well much disappears.  And that completes the board, except for those with rivers, which need some extra foliage along the river banks in particular, as well as bridges adding (I've planned mine to be removable so they can be blown in-game by engineers), and further painting of the river surface (a darker colour in the centre, plus a more glossy varnish to give added "depth" - I've still to do this on even my initial boards).


I noticed the other week at my younger daughter's day-care they have some EVA mats that come covered in a felt-like material on one side.  If only I had known such a thing existed before, I'd have given serious thought to using these and dispensed with (most of) the flocking, and just sprayed the colours on.  It probably wouldn’t have looked as good, but would have been so much quicker.  Hmm, maybe I'll buy some for a set of boards specifically for ancients, since the terrain there isn't as small...

One thing I don't have at the moment is a set of boards featuring an autobahn.  These are not only reasonably dense in Germany, but are prime routes for exploitation in war-time, so it's quite a likely a random battle could feature one.  So in the future I intend to make some up that each feature an autobahn segment; this will exit in the centre position, taking up "the river slot". 

If I was going to start over again, one thing I would definitely change is the way I made my built-up areas.  I made most of these up long before I had the idea to make the terrain modules, so they were not planned as well as they could have been for their current role: the roads don't all come in at the same positions along the boards.  This means most can only be used on two or three of the boards at most; some are single use: very inefficient.  Ideally every built-up area should be able to be used in any and every slot.  This won't be possible due to some having three roads leading in and out, some two, but you should be able to get away with at most three different configurations, and not a dozen like I have!  So make sure your roads enter your built-up areas exactly in the centre of the edge of the built-up area, or of not that, at least the same distance off-centre (and orientation) every time.


  1. Brilliant stuff Luke, do you find the hedges and foliage tend to crush when the boards are stacked?

  2. No - partly because the hedges are only 5 mm high, so there isn't much to crush in the first place, but mostly because there are areas of higher elevation protecting them (i.e. don't put a hedge running across the very top point of the highest hill on the board...)

  3. And did you come up with a good way of making the woods?

    1. Well, I don't know about "good", but I have a way. A felt underlay to mark the area of the woods, with the edges flocked to disguise the fact there is felt there, and also to give an "undergrowth" feel. Then I add a top-story made of a lot more flocking ("clump foliage") that is glued down onto a layer of painted cotton wool, with lots of PVA, and the whole lot then matte-varnished to hold it together. A 15 mm or so gap between it and the felt allows stands to hide in here, and it can be removed to move them around. I initially used cut-down match-stick equivalents to hold the top-story up, but they bend away from vertical too easily (just being stuck into the cotton wool), so in the future I will use a twisted wire frame, like from champagne bottles.

  4. Not sure if this worked the first time...
    Luke, is the EVA heat malleable?, I am considering using a Butane flame to take off the obvious cut/carving lines..

    1. Hmm, that's an excellent thought. Why didn't I think of that? Well, not a butane flame though! EVA is not exactly fire-resistant, so that would likely end up with a... problem :-( However EVA is not only a thermoplastic, but the main component of the stuff that is used as the "melt" in glue-guns, so it should definitely respond to heating, if the heat is sufficiently mild. Ah, but I foresee a problem - I expect the stuff will become very sticky when you heat it - that8s the whole purpose in using it in a glue-gun after all! That could lead to quite a mess, especially on whatever tool you are using to apply the heat. I suggest using a hair drier. The only answer is to try it and see. I might not have time today to give it a try though, because I am off to Australia for a few days for a meeting, and I haven't even finished my presentation yet, let alone started packing!

    2. PS - Steve - if you are reading this, please e-mail me (lukeuedasarson at the inevitable so I can get hold of your e-mail address!

  5. Ok, tried a Butane flame...if you stay sufficiently far way you can 'very slightly' reduce the edges..But it's not worth it, a few seconds long and everything goes crispy and the chance of fire is waaaay too great. (Don't try this at home Kids) PS. to everyone still reading this, Luke is right, EVA blunts craft knives in a few cuts, next time I am in a knife fight I plan to use an EVA suit...this stuff is amazingly resilient, I went through a whole refill pack of blades just cutting hills as described by Luke..

    1. Yeah, buying a whetstone / kitchen knife sharpener halfway through the job was a wise investment. And it even sharpens kitchen knives :-)

      So, cutting is still the way forward...