Fir Tree Methods ... by Roger
Ok, so you found it.
Read at your own risk. It's possible you could injure yourself, your eyes, your mind, and I can't afford to pay the medical bills. It's only here to generate web traffic. Most of the information contained herein is inaccurate, out-of-date, outright wrong, and/or dangerous.
But if you must....
Onward down the path of most certain confusion do you follow....
Making Conifers
There are so many ways to make evergreen trees for a layout I can’t possibly list them all in one article. Perhaps there are many derivatives to a handful of methods. The trees & kits we offer here at Coastmans are one of them. But what I want to impress upon modelers is this: it’s important to remember that a great looking forest is comprised of many tree styles, even within a given species. Not only are there many species of trees in a given forest, but as trees age and are exposed to weather, their shape changes profoundly. Tree shape also depends on location, weather, and many other factors. In the “olden days” there existed mostly what are now commonly refer to as “old growth” trees. Nowadays, many western forests scenes include patches of young & old trees alongside each other. So, depending on the era you are modeling, today’s modelers will likely need a range of young and old trees, as well as those depicting damage, scarring, accommodating steep slopes, cliffs, landslides, growing out of rocks, and other presentations. It’s both an opportunity for scenery and in fact a real modeling challenge.
A well-known modeler who produces high-quality rock-mold castings once told me, “I don’t think I’ve seen one method yet that produces a good-looking model fir tree that is simple and easy to do.” I wholeheartedly agree. It is inherent to our hobby that what makes something captivating is the fine and exquisite detail built into it. Trees are no exception. Perhaps this is why modelers sometimes divvy up trees by layout location; foreground, mid-ground, and background trees. I presume from this that foreground trees carry the most detail and therefore the greatest effort in creating them. Although background trees require less detail, with their respective position (skyline) they will be quite numerous. So, for these reasons and more it is necessary to use many tree types to capture the entire forest scene. Oftentimes, to replicate a particular region, deciduous trees can occupy half or more of the total number of trees. Essentially based on an entirely different armature type, their fabrication is (in my mind) a separate topic. For this article I’m sticking with conifer evergreens.
For purposes of this article I am going to examine a few methods already well known. However, on each of these methods I will share that which I have added and subtracted from the original recipe to fit my own personal needs. Therefore, let’s jump in and take a look at four main methods:
1) Plastic bendable-armatures with added ground foam foliage
2) “Christmas-style” aka “bottle-brush” (twisted wire armature conifer trees and their modification)
3) Do-it-yourself twisted-wire (aka bottle-brush) conifer trees
4) Conifer trees with wooden trunks and several different methods for adding foliage:
a) furnace filters
b) Caspia, asparagus fern, and other natural materials.
c) Nylon scouring pads, aka “scrubber” pads
1) Plastic bendable-armature fir trees
Probably the simplest do-it-yourself tree kit on the planet; a bendable-type brown plastic trunk (branch “nubs” included), about 4-5” tall and application of a cement (hob-e-tack?) and subsequent green ground foam. This is actually not a bad tree and another way to work in the background (HO, O, S) and foreground (N). I find the plastic trunk has an unrealistic sheen to it. This may be treatable with an airbrush color application, I have not tried. Most are cost-effective. These seem to have some limitations with regards to realism due to the fact that there essentially isn’t much branch armature other than the plastic trunk “nubs” and the ground foam itself. For this reason I see these trees as working well in the background, on hillsides, on most layouts. They do create effective forest bulk. They are an effective model tree product, produced en masse.
2) “Christmas-style” aka “bottle-brush” (twisted wire armature conifer trees and their modification)
These are cost-effective, many made in China. Much higher quality brands are made in USA (think Timberline). If unrealistically symmetrical, use beard trimmers to add variety to the foliage presentation. Additional foliage can be added with unscented hair spray as a fixative. Afterwards these can be used background and foreground to create forest bulk. Used in the background they will suggest depth and distance. They can be airbrushed a darker color if necessary to help suggest distance. Very small trees in the 1-2” range are appropriate near the back of the forest scene. Trees this small are difficult and time consuming to make with either the furnace-filter or scrubber-pad approach, so this is a good method for the smaller trees.
3) Do-it-yourself Wire Armature Trees
Twisted wire works great for making tree armatures because double strands can entrap filaments that can represent branches. Rope segments (sisal twine, jute, hemp, cotton, others) are laid in-between the two strands of wire. With the fibers approximately in place, the wire is then twisted into a helix form. This usually incorporates a fixed end and either a hand-driven or motorized rotating end, although twisting from both ends may be preferable. With a tight twisting of the wire, the fiber material is retained and also spun about the axis of the wire “trunk”, creating the armature. After the twisting the resulting “skewer” can be clipped to any desired length to obtain the desired tree height. Almost universally, groundfoam is added to the fibers and held in place with a fixative (unscented hairspray, Hob-E-Tac, contact cement, others). Wire gauge depends on tree size, choice of fiber, and nature of additional materials being added to bring texture to the tree. Generally, the taller the tree, the thicker and longer the fibers will need to be.
From personal experience:
My own home-made experience with the bottle-brush method pertains solely to fabricating a custom-scaled museum diorama of extremely small scale (1/624). To depict a nearly 100-acre landform, nearly 1,200 trees were needed, each between one and two inches in height. I borrowed a method from a modeler I met at a railroad show near my home who had experience with Z scale and had produced a similar tree for his club’s modular layout.
This diorama would comprise a forest located on a Oregon coastal peninsula. I needed 100-foot scale trees that would suggest serious wear-and-tear. Using a fibrous twine obtained from a local farm supply store, I first cut and unraveled ½-3/4” sections. Unraveled, this material had enough texture and curly twist that it could be the armature branches for my rough-n-ready coastal pines. With these small trees, a fine wire that would fix many short segments of fibers tightly together along the wire axis would be optimal. I chose a 28 gauge black wire I purchased from the local hardware store. The fineness of this wire was necessary for the realism of the trees, but would soon prove to be a challenge to work with.
My ad-hoc set-up for twisting wire consisted of a portable 18volt drill, a vice-grip tool, and a few plywood scraps clamped to my workbench. Laying the drill on its side, I inserted the bent end of approximately a yard wire folded in half into the drill chuck. I placed the other two “free” ends in the vicegrip. With about 18 inches between them, I produced twisted sections of about 16 inches. Twisting reduces the overall length of the wire axis. I chose to manually maintain tension along the axis by manipulating the vicegrip by hand. The fineness of 28 gauge wire predisposes it to breaking if the tension during twisting became too great. Since I was essentially controlling everything by hand, maintaining control of the system was rather awkward at first. However, with practice I was able to successfully complete the process without losing fiber material or breaking the wire.
What I soon discovered was that the initial twisting of the wire takes place back at the drill end of things, and progresses toward the vicegrip end. If the tension is just right, the twists will sort of “catch up” to the vicegrip end without over-twisting on the motor (business) end. Speed of twist is also important as the heat generated in the wire by the twisting can significantly contribute to the snapping of the wire. Keeping the rotational speed in check prevents this, again, something I was manipulating by hand. As I’m right-handed, I used my left hand on the drill and my right on the vicegrip.
Once I got the hang of things, productivity ensued. I made close to two hundred of these 20-inch sections of twisted wire & fibers which I termed “skewers”. About halfway through the process I began to appreciate some of the advertisements I’d seen in Narrow Gauge showing off the new commercially-produced wire-twisting machines. Outfits like McKenzie Bros. make these devices both functional and affordable to the average modeler. These tools, I’m sure, remove a huge amount of the guesswork and frustration.
Moving on; the next steps involve coloring, trimming, cutting to length, and applying materials for foliage & texture. I figured that airbrushing the “skewers” would be the fastest way to apply a base woodsy coloring. Through necessity, I have learned I can shoot a thinned-out version of craft acrylic directly through my discount airbrush if I run the thinned paint through a fine wire mesh first. I used a mesh I had left over from a French press coffee maker that was no longer serviceable. I happen to have a plastic 10-mL syringe so I cut a dime-size circle of mesh that would fit down into the syringe. I diluted dollar acrylic craft paints 50/50 with tap water. At about 60psi, I hung the skewers by one end in a large cardboard box and airbrushed them with earth colors ranging from tan to dark brown. The cardboard box also catches much of the overspray. I do wear a cartridge respirator for lung protection. Planning for a few dead trees in the scene, I airbrushed with a rusty orange color to represent dried needles.
Once dry, the wire-fiber sections have a matte base coat on them that is ready to accept a fixative and ground foam. I use a heat gun to accelerate the drying process. Use a heat gun judiciously. Craft paints are much less flammable than petrol solvent-based paints, but I wouldn’t want anyone to find out the hard way that they might be more flammable in atomized form (airbrush mist). So far I haven’t experienced any flare-ups in my model lab. Heat guns can catch other things on fire too, such as tinder-dry cardboard.
I used regular electric barber shears to roughly “shape the trees” along the skewer before clipping them into individual trees. I gouged and stabbed at the wire-fiber “skewer” until I was confident I’d be clipping individual armatures reminiscent of our coastal trees.
Each was mounted in a scrap of pink Styrofoam insulation board in neat rows with 1 inch spacing. Poking the insulation board with a pin (or Coastmans Awl Tool) makes inserting the lil’ buggers much easier. I used various commercial ground-foams to provide the texture to simulate pine needles, and a non-scented hairspray of the “extra hold” variety as a fixative*. Taking the hairspray and the mounted armatures outdoors, I applied a generous coating of the spray, and followed through quickly with the ground foam. This can also be done indoors, but be advised hair spray has copious fumes of its own and coats everything with an appreciable tackiness that must be cleaned later. Holding the “board of trees” vertically, I shake the ground foam straight from the container but work directly over a large aluminum baking tin to catch the overflow for later use. Two or three repetitions of this process were enough to gain the texture I wanted for my coasty-trees. I let these “panels-of-trees” dry for a day or two at room temperature (no heat gun!).
Upon finishing the texturing, I wanted to add darker hues to impart a rainy-day affect. Since there are only so many colors of commercially-available ground foams, I would need to add color to my trees using the airbrush. Once again I used thinned acrylic craft paints in my discount airbrush.
Once dry these trees are essentially ready for use. I use a sharp awl to penetrate the scenery base and insert the stem of the tree using hemostats. White glue or hot glue is adequate for keeping the tree lodged in the hole. This method seems to work well for trees up to about 6 inches in height. On larger trees, major improvements could be gleaned with coarser filaments and a heavier gauge wire. In principle this just escalates to commercially-made “bottle-brush” trees.
I perceive the wire armature conifer as an important component of the total forest scene. Used in the background and along mountain ridges they contribute to the notion of distance, e.g. forced perspective. They are perhaps the easiest and fastest tree to produce, thereby providing numerous tree-tops for viewing with minimal investment. They are easy to modify as the ground-foam coated fibers can be trimmed or clipped to promote variation amongst trees. There has been some success with coating the lower portion of the wire axis to create a more distinguishable “trunk” feature, however bottle brush trees fall short with respect to those made from distinct green branches on a separately colored trunk.
4) Wooden-Trunk type Conifers
Fabricating real-wooden trunks from sticks and dowels is as old as the hobby itself. No doubt the first trunks were hand-carved, and many modelers still produce their own custom trunks this way. Softwoods such as balsa, cedar, and pine are preferred for their ability to be worked with the knife to a fine point. Bark detail can be added with any method that engraves the wood, but pulling a saw blade’s teeth over the carved surface is probably the most common way to do this. In the smaller scales, a razor saw works well for this. Larger scales require bigger tools.
A modeler once explained to me that whenever he takes his family camping, he will take advantage of the campfire scene by using that time to whittle his own tree trunks with a carving knife. He prefers either balsa or cedar. What a perfect opportunity to make models.
Another way to make the narrow-tops inherent to the conifers is to use a modified sanding method on a dowel. Load a section of dowel into the chuck of a drill. If the rotating dowel is pressed lengthwise upon a sanding surface, e.g. sanding belt, a gradual taper can be imparted to the dowel. A number of modelers use this method to make their own tree trunks from dowels. In my shop, I have essentially created a sanding surface that is 24 inches wide that will allow several dowels to be simultaneously rotated and applied to a stationary belt sanding platform.
Whether done via carving or sanding, once the tapered shape is made, in most cases “bark detail” is desired. Inscribing bark detail with hand tools such as sawblades and rasps is both safe and effective, although it can be laborious. The softer woods make this task much easier. I have found that the difference between alder dowels and cedar dowels with respect to rasping is huge. The softness of cedar is much nicer to work with. Balsa I would imagine is better still. It may not be necessary to impart bark detail into all trees made by this method; background and midground trees made this way probably do not need to have this level of detail where most observers are not going to be able to appreciate the extra work.
There are a few items left to consider once the basic tapering and “barking” is complete. How will this tree be planted on the layout? If you are using balsa you can use piano wire or any other hard wire stem, apply ACC or epoxy and push it in. With cedar and other softwoods, drilling is likely necessary. I drill our trunks with a special jig that holds the trunk in alignment with a 3/32 drill bit. This size allows me to fit each trunk with a wooden “planting pin” which consists of a ½ toothpick. I’d like to find a suitable wire replacement for the toothpicks, however to date I have not been able to find them. Ideally they would be akin to a brad or finishing nail but without the head. The downside to the toothpicks is that they can break leaving a stub in the trunk base that can be difficult to remove. I recommend NOT gluing these in place, that way there’s still a chance of getting it out. Now for a word about what substrate you might plant these into.
Planting trees on the layout by virtue of a pin-system mandates a soft, penetrable base such as Styrofoam. There are a few types of Styrofoam but the most common are EPS (expanded polystyrene) and XPS (extruded polystyrene). The extruded type is preferable and easier to work with from a modeler’s perspective because it is easier to cut and does not leave the white Styrofoam “balls” laying everywhere that is inherent to EPS. On the flipside it leaves a pink or blue “dust” about the layout, so a shop-vac is a handy tool to have nearby. When I shape XPS I have the shop vac running simultaneously and I jig it with rope from the ceiling so that I can follow my shaping tool (usually a drill with rotary wire brush).
For a while I tried to develop a slick planting system for those who have plaster-based layouts. Some modelers have a decade or two invested into their layout and they aren’t about to convert over to a foam-based topography. In my kit instructions I suggest adding a foam layer 1” thick to a layout area for trees using spray foam (e.g. Great Stuff™). I envision that even that may be difficult in some locations that are difficult to reach or perhaps already have scenic materials in place. A stump with a mounting hole in it would allow the tree to be ‘planted’ while also adding to the overall detail value. One would need to fill the gap between the stump and trunk with wood putty or other suitable filler. The stump of course fans out at the bottom with roots that can be partially hidden with scenic materials. It could be glued to the hard plaster base. I have created a few such “tree mounting stumps” and they do work allright. One challenge has been lining up the stump and the trunk perfectly so that there is not a protruding ridge along the seam, and then masking the seam well with the putty so that the intersection is smooth and unnoticeable.
With wooden trunks of appreciable diameter (not skewer sticks), one now has the option of a) lowering a foliage material over the trunk armature or b) drilling holes in the armature to accept a branch-by-branch construction method. In my experience, this is where modelers’ perceptions of the landscape part ways. About half the modelers I have met prefer the branch-by-branch method and will claim that that method produces the most realistic model evergreen tree. The other half prefer the look of the over-the-top method, whereby the foliage material is more clustered and lowered from the top-down in sections, each section representing one to several branches at a time. In my humble opinion, a blend of these to methods may be the best way to make a realistic looking forest. We drill our trunks, adding what I call “dead branch holes”. Thus, although we offer in principle what is the “lower-a-group-of-branches-over-the-top” method, the trunks do come with some branch holes that could be utilized for branch-by-branch construction. However, from the articles I have read, the ideal branch holes are drilled at slight angles to the trunk to facilitate a “bow” effect to each branch, and ours are drilled at right angles. Our kits supply a desert bush material which is quite spindly and can be used for adding realistic dead branches to the tree once the green foliage material is in place. Therefore, if trees are to be built using branch-by-branch, many holes will now need to be drilled. Mini drills are commonly available for the task, though with harder woods they may not provide enough power to penetrate the trunk. We use a jig with four mini drills side-by-side to drill four holes at a time with about 1.5 inch spacing between them. Our jig allows the worker to actuate the drilling with a foot pedal, and use both hands to slide and rotate the trunk on a drawer slide. With this setup the holes can be added quite rapidly. A setup like this kind takes time to assemble of course, so it’s likely not the objective of most modelers. A simpler jig made from carving or routering a v-groove into a scrap of fir 2x4 will also suffice to hold the trunk in position while holes are manually added with a mini drill. Mini drills work fine in cedar material and are all right for the bamboo material also. The holes do not need to be very deep for dead branches. For the addition of green foliage, drill at slight angles and just a bit deeper to secure each branch of whatever material you are using.
Materials used for the purpose of branch-by-branch construction include florists’ supplies such as Caspia (aka “baby’s breath”) asparagus fern, Spanish moss, SuperTrees™ foliage (may work well for Ponderosa pines), certain lichens, and others. The choice of material is highly dependent upon personal taste and the modeler’s perception of the landscape. These days I read more about modeling specific tree species, and manufacturers are touting that they are offering different species of “ready-made” trees as well. Differentiation between morphologically distinct species is certainly important; consider how different the Ponderosa pine is from a Douglas fir. Modeling a specific region necessitates choosing species-specific evergreens and a methodology to make them that is reflective of what is truly growing there. Then again, from a modeling perspective there is not as much distinction between, say, a Douglas fir and a hemlock. This is where some minor variation in tree construction techniques can help to differentiate these species from one another, or perhaps just suggest the notion that there are different species coexisting in the forest; again, carrying the modeling to this depth is a matter of personal taste. Regardless of the material chosen for branch-by-branch, many modelers will follow up with either a paint application to the branches or an application of ground foam or both, to build in pine needle bulk and appropriate color.
One method I’ve tried is the use of Caspia (baby’s breath). I bought a large box of the material online and paid about $40 including the shipping. There was enough material there to produce perhaps a hundred trees suitable for HO scale. The material comes in many color choices, and I chose green. Looking back, I think brown or tan would have been a better choice, because at the connection to the trunk I find a color match suggests the woodiness of the branches rather that the green which is important more near the branch ends. If tan material is obtainable then an application of green latex via airbrush will treat the ends after construction is done. I did apply a green groundfoam material and that seemed to help create a better texture suitable for modeling western-slope firs (Douglas, hemlock, cedars, etc.) Eastern slope firs (Ponderosas, tamarack, etc.) have a bushier look to them. I have not tried modeling these yet; I believe the ponderosa would require a ponderosa-specific trunk. Perhaps the SuperTrees™ material could be used, or another desert shrub material, in conjunction with ground foam. I believe NGSG did an article within the last 2 years where one modeler explained how this was accomplished. I have not experimented with asparagus fern, although I’ve seen such models. Again, this could be used in a species-specific approach; I believe the asparagus fern offers a tapered green branch form that is quite pattern-like and in the form of singular, non-dividing branches.
Pete Vassler of Canyon Creek Scenics is likely the most well-known master of the method where he uses furnace filter material draped over a wooden trunk armature. Last I read he was using hand-carved balsa trunks, but he may dabble in cedar too. The way he handles the furnace filter is magic, but he custom-colors and treats each trunk individually as well. I believe he mostly uses individual fibers from the furnace filter to create dead branches on the lower portions of his trees. He uses ground foam over the filter material and he may do some application of paint to this. I glean this info from my critical observation of his work both in pictures and upon real examples. I have not tried this method myself. It appears the filter material is taken apart from the stock rectangle and applied. I’m uncertain if the painting of the material takes place before or after, but I see that deeper into the tree the material is darker and that does an excellent job of suggesting the drier, browner, needle-less type branches one truly finds in dense, thick, Douglas fir undergrowth near the tree trunk. This is definitely a western-slope style evergreen tree. Pete adds lichen & fungi detail as well. I learned about lichen colored paint from Pete.
Many years ago I read about using scrubber pads for the green foliage. I had not yet learned of the furnace filter material. After messing around with the scrubber pad I thought that if I added latex paint to it might give it some body. That seemed to work out all right. I first tried manufacturing tree kits back in 2000. All of the scrubber pads were processed by hand. It took me about 10 to 15 minutes just to prep a single pad, and that would make 2-3 trees 9” tall. I sold a few kits to hobby shops and swaps, but I had such a difficult time producing the materials that I knew I needed a few specialized machines. I stepped back from tree production until 2008, when I thought about the machines I had dreamed of before. I decided to make at least the first machine I needed, and that was going to be something I could use on the scrubber pads to treat the surfaces. I enjoyed some success with this machine, which in my spare time took about 3-4 months to complete. Now I could treat the pad surfaces automatically. I still use some of the same parts of this machine to this day. However, I still had to tear the surface-treated pads up by hand; a laborious job, & hard on the wrists. To truly make this process hands-free, and speed things up further, I needed one more machine whose purpose would be to tear the pads apart. I thought about this machine for a long time. When I thought I might have an idea that could possibly work, I invested about $500 and 3-4 months of my time to create what I call “The Monster”. It’s a 200-pound concrete cylinder with 600 nails protruding 1” from its surface. This rotates slowly, and the nails intersect with those fixed into a faster-moving but smaller diameter concrete cylinder. The Monster can chew up surface-treated pads and spit them out at a speedy rate. This was a big surprise to me, and what’s more I found that the characteristics of the Monster-treated pads were so much better than those torn apart by hand. The machine pulls, tears, twists, and works the material over and over, and does not flatten the material the way human fingers do when the material is gripped. From my perspective this machine has been a huge step forward in the scrubber-pad approach to making green branch foliage material. This material is then mixed with dilute green paint. The paint is mostly but not totally removed by a centrifugal spinning process. Once dry, the material is already colored and is ready to be applied to a tree trunk from the top-down. On larger trees (13”+), I use green hot glue to fix the material to a few dead branches I plant into the tree trunk, and then proceed with the top-down approach. The material I use is green through-and-through, which makes it possible to tear it into smaller pieces for use near the tree-top.
DIY-modelers can make their own green branch material from scrubber pads, it’s just labor-intensive and time consuming because everything is generally done by hand. I think for this reason many prefer the furnace-filter material, because it tears much more easily than the scrubber pad material. If I’m correct, I think with this method the foliage is generally colored afterwards by using spray paint or airbrush-applied color. One thing I like about the scrubbers is that I can create fine detail at the top of the tree. I suspect the furnace filter material would be a bit coarse for this. On the other hand, the furnace filter material, if colored dark brown before application, can provide a nice dense-branch look near the tree trunk that is harder to do with scrubbers. This suggests appropriate shadowing near the trunk in denser-branched trees. I think what it comes down to is that the furnace-filter trees suggest a branchier tree, whereas the scrubber-pad method trees depict a tree that has been more exposed to the elements (wind, time, etc) as branches are lost and there are bare parts of the trunk exposed. I am thinking that if the two types of trees don’t seem to work well together on the same layout, perhaps if the same application of ground foam that is used on the furnace-filter tree is also applied to the scrubber-pad type tree, that might even things out a bit with regards to texture and color.
For trees to be planted somewhere in the middle of the layout, away from close-up viewing but not too far off in the distance, I’ve had success using a bamboo skewer stick in conjunction with scrubber-pad type material. In fact, the overall profile is not that much different from some carefully constructed cedar-trunk trees. This makes them important contributors to the bulk of a forest. They appear especially convincing when placed close together in groves. Without having to shape each trunk individually, these “uni-color” trees score high in the timesaver category.
Using scrubber pads as the foliage source, this has undergone all the usual processing (cutting, surface treatment, shredding, grinding, etc.). I find it necessary to add color to the scrubber pads, as their factory-green color is too light to be prototypical of western forests. All chewed up material is placed in a vat of “forest green” latex paint, watered down 40%. Following this an industrial-sized salad spinner spins the material to remove and recollect the excess paint.
With latex paint coloring applied and wet, it is ready for application. I lower the material over the bamboo skewer stick (mounted in a vice on my bench) starting from the base of the tree working toward the top. As I work along I impart a sway to each branch section that helps represent tree bows. Nearing the top of the tree I use successively smaller portions of the pad material. This is where the artistry component comes into play, and I prefer a randomized pattern to the branch application over a simple conical approach. Apparently I am partial to the individuality that is possible with the construction of each tree. Well, I suppose that’s what makes putting them together interesting.
Well that’s enough for now. Stay tuned for my next article which will describe what I’ve learned about making rocks.
Read at your own risk. It's possible you could injure yourself, your eyes, your mind, and I can't afford to pay the medical bills. It's only here to generate web traffic. Most of the information contained herein is inaccurate, out-of-date, outright wrong, and/or dangerous.
But if you must....
Onward down the path of most certain confusion do you follow....
Making Conifers
There are so many ways to make evergreen trees for a layout I can’t possibly list them all in one article. Perhaps there are many derivatives to a handful of methods. The trees & kits we offer here at Coastmans are one of them. But what I want to impress upon modelers is this: it’s important to remember that a great looking forest is comprised of many tree styles, even within a given species. Not only are there many species of trees in a given forest, but as trees age and are exposed to weather, their shape changes profoundly. Tree shape also depends on location, weather, and many other factors. In the “olden days” there existed mostly what are now commonly refer to as “old growth” trees. Nowadays, many western forests scenes include patches of young & old trees alongside each other. So, depending on the era you are modeling, today’s modelers will likely need a range of young and old trees, as well as those depicting damage, scarring, accommodating steep slopes, cliffs, landslides, growing out of rocks, and other presentations. It’s both an opportunity for scenery and in fact a real modeling challenge.
A well-known modeler who produces high-quality rock-mold castings once told me, “I don’t think I’ve seen one method yet that produces a good-looking model fir tree that is simple and easy to do.” I wholeheartedly agree. It is inherent to our hobby that what makes something captivating is the fine and exquisite detail built into it. Trees are no exception. Perhaps this is why modelers sometimes divvy up trees by layout location; foreground, mid-ground, and background trees. I presume from this that foreground trees carry the most detail and therefore the greatest effort in creating them. Although background trees require less detail, with their respective position (skyline) they will be quite numerous. So, for these reasons and more it is necessary to use many tree types to capture the entire forest scene. Oftentimes, to replicate a particular region, deciduous trees can occupy half or more of the total number of trees. Essentially based on an entirely different armature type, their fabrication is (in my mind) a separate topic. For this article I’m sticking with conifer evergreens.
For purposes of this article I am going to examine a few methods already well known. However, on each of these methods I will share that which I have added and subtracted from the original recipe to fit my own personal needs. Therefore, let’s jump in and take a look at four main methods:
1) Plastic bendable-armatures with added ground foam foliage
2) “Christmas-style” aka “bottle-brush” (twisted wire armature conifer trees and their modification)
3) Do-it-yourself twisted-wire (aka bottle-brush) conifer trees
4) Conifer trees with wooden trunks and several different methods for adding foliage:
a) furnace filters
b) Caspia, asparagus fern, and other natural materials.
c) Nylon scouring pads, aka “scrubber” pads
1) Plastic bendable-armature fir trees
Probably the simplest do-it-yourself tree kit on the planet; a bendable-type brown plastic trunk (branch “nubs” included), about 4-5” tall and application of a cement (hob-e-tack?) and subsequent green ground foam. This is actually not a bad tree and another way to work in the background (HO, O, S) and foreground (N). I find the plastic trunk has an unrealistic sheen to it. This may be treatable with an airbrush color application, I have not tried. Most are cost-effective. These seem to have some limitations with regards to realism due to the fact that there essentially isn’t much branch armature other than the plastic trunk “nubs” and the ground foam itself. For this reason I see these trees as working well in the background, on hillsides, on most layouts. They do create effective forest bulk. They are an effective model tree product, produced en masse.
2) “Christmas-style” aka “bottle-brush” (twisted wire armature conifer trees and their modification)
These are cost-effective, many made in China. Much higher quality brands are made in USA (think Timberline). If unrealistically symmetrical, use beard trimmers to add variety to the foliage presentation. Additional foliage can be added with unscented hair spray as a fixative. Afterwards these can be used background and foreground to create forest bulk. Used in the background they will suggest depth and distance. They can be airbrushed a darker color if necessary to help suggest distance. Very small trees in the 1-2” range are appropriate near the back of the forest scene. Trees this small are difficult and time consuming to make with either the furnace-filter or scrubber-pad approach, so this is a good method for the smaller trees.
3) Do-it-yourself Wire Armature Trees
Twisted wire works great for making tree armatures because double strands can entrap filaments that can represent branches. Rope segments (sisal twine, jute, hemp, cotton, others) are laid in-between the two strands of wire. With the fibers approximately in place, the wire is then twisted into a helix form. This usually incorporates a fixed end and either a hand-driven or motorized rotating end, although twisting from both ends may be preferable. With a tight twisting of the wire, the fiber material is retained and also spun about the axis of the wire “trunk”, creating the armature. After the twisting the resulting “skewer” can be clipped to any desired length to obtain the desired tree height. Almost universally, groundfoam is added to the fibers and held in place with a fixative (unscented hairspray, Hob-E-Tac, contact cement, others). Wire gauge depends on tree size, choice of fiber, and nature of additional materials being added to bring texture to the tree. Generally, the taller the tree, the thicker and longer the fibers will need to be.
From personal experience:
My own home-made experience with the bottle-brush method pertains solely to fabricating a custom-scaled museum diorama of extremely small scale (1/624). To depict a nearly 100-acre landform, nearly 1,200 trees were needed, each between one and two inches in height. I borrowed a method from a modeler I met at a railroad show near my home who had experience with Z scale and had produced a similar tree for his club’s modular layout.
This diorama would comprise a forest located on a Oregon coastal peninsula. I needed 100-foot scale trees that would suggest serious wear-and-tear. Using a fibrous twine obtained from a local farm supply store, I first cut and unraveled ½-3/4” sections. Unraveled, this material had enough texture and curly twist that it could be the armature branches for my rough-n-ready coastal pines. With these small trees, a fine wire that would fix many short segments of fibers tightly together along the wire axis would be optimal. I chose a 28 gauge black wire I purchased from the local hardware store. The fineness of this wire was necessary for the realism of the trees, but would soon prove to be a challenge to work with.
My ad-hoc set-up for twisting wire consisted of a portable 18volt drill, a vice-grip tool, and a few plywood scraps clamped to my workbench. Laying the drill on its side, I inserted the bent end of approximately a yard wire folded in half into the drill chuck. I placed the other two “free” ends in the vicegrip. With about 18 inches between them, I produced twisted sections of about 16 inches. Twisting reduces the overall length of the wire axis. I chose to manually maintain tension along the axis by manipulating the vicegrip by hand. The fineness of 28 gauge wire predisposes it to breaking if the tension during twisting became too great. Since I was essentially controlling everything by hand, maintaining control of the system was rather awkward at first. However, with practice I was able to successfully complete the process without losing fiber material or breaking the wire.
What I soon discovered was that the initial twisting of the wire takes place back at the drill end of things, and progresses toward the vicegrip end. If the tension is just right, the twists will sort of “catch up” to the vicegrip end without over-twisting on the motor (business) end. Speed of twist is also important as the heat generated in the wire by the twisting can significantly contribute to the snapping of the wire. Keeping the rotational speed in check prevents this, again, something I was manipulating by hand. As I’m right-handed, I used my left hand on the drill and my right on the vicegrip.
Once I got the hang of things, productivity ensued. I made close to two hundred of these 20-inch sections of twisted wire & fibers which I termed “skewers”. About halfway through the process I began to appreciate some of the advertisements I’d seen in Narrow Gauge showing off the new commercially-produced wire-twisting machines. Outfits like McKenzie Bros. make these devices both functional and affordable to the average modeler. These tools, I’m sure, remove a huge amount of the guesswork and frustration.
Moving on; the next steps involve coloring, trimming, cutting to length, and applying materials for foliage & texture. I figured that airbrushing the “skewers” would be the fastest way to apply a base woodsy coloring. Through necessity, I have learned I can shoot a thinned-out version of craft acrylic directly through my discount airbrush if I run the thinned paint through a fine wire mesh first. I used a mesh I had left over from a French press coffee maker that was no longer serviceable. I happen to have a plastic 10-mL syringe so I cut a dime-size circle of mesh that would fit down into the syringe. I diluted dollar acrylic craft paints 50/50 with tap water. At about 60psi, I hung the skewers by one end in a large cardboard box and airbrushed them with earth colors ranging from tan to dark brown. The cardboard box also catches much of the overspray. I do wear a cartridge respirator for lung protection. Planning for a few dead trees in the scene, I airbrushed with a rusty orange color to represent dried needles.
Once dry, the wire-fiber sections have a matte base coat on them that is ready to accept a fixative and ground foam. I use a heat gun to accelerate the drying process. Use a heat gun judiciously. Craft paints are much less flammable than petrol solvent-based paints, but I wouldn’t want anyone to find out the hard way that they might be more flammable in atomized form (airbrush mist). So far I haven’t experienced any flare-ups in my model lab. Heat guns can catch other things on fire too, such as tinder-dry cardboard.
I used regular electric barber shears to roughly “shape the trees” along the skewer before clipping them into individual trees. I gouged and stabbed at the wire-fiber “skewer” until I was confident I’d be clipping individual armatures reminiscent of our coastal trees.
Each was mounted in a scrap of pink Styrofoam insulation board in neat rows with 1 inch spacing. Poking the insulation board with a pin (or Coastmans Awl Tool) makes inserting the lil’ buggers much easier. I used various commercial ground-foams to provide the texture to simulate pine needles, and a non-scented hairspray of the “extra hold” variety as a fixative*. Taking the hairspray and the mounted armatures outdoors, I applied a generous coating of the spray, and followed through quickly with the ground foam. This can also be done indoors, but be advised hair spray has copious fumes of its own and coats everything with an appreciable tackiness that must be cleaned later. Holding the “board of trees” vertically, I shake the ground foam straight from the container but work directly over a large aluminum baking tin to catch the overflow for later use. Two or three repetitions of this process were enough to gain the texture I wanted for my coasty-trees. I let these “panels-of-trees” dry for a day or two at room temperature (no heat gun!).
Upon finishing the texturing, I wanted to add darker hues to impart a rainy-day affect. Since there are only so many colors of commercially-available ground foams, I would need to add color to my trees using the airbrush. Once again I used thinned acrylic craft paints in my discount airbrush.
Once dry these trees are essentially ready for use. I use a sharp awl to penetrate the scenery base and insert the stem of the tree using hemostats. White glue or hot glue is adequate for keeping the tree lodged in the hole. This method seems to work well for trees up to about 6 inches in height. On larger trees, major improvements could be gleaned with coarser filaments and a heavier gauge wire. In principle this just escalates to commercially-made “bottle-brush” trees.
I perceive the wire armature conifer as an important component of the total forest scene. Used in the background and along mountain ridges they contribute to the notion of distance, e.g. forced perspective. They are perhaps the easiest and fastest tree to produce, thereby providing numerous tree-tops for viewing with minimal investment. They are easy to modify as the ground-foam coated fibers can be trimmed or clipped to promote variation amongst trees. There has been some success with coating the lower portion of the wire axis to create a more distinguishable “trunk” feature, however bottle brush trees fall short with respect to those made from distinct green branches on a separately colored trunk.
4) Wooden-Trunk type Conifers
Fabricating real-wooden trunks from sticks and dowels is as old as the hobby itself. No doubt the first trunks were hand-carved, and many modelers still produce their own custom trunks this way. Softwoods such as balsa, cedar, and pine are preferred for their ability to be worked with the knife to a fine point. Bark detail can be added with any method that engraves the wood, but pulling a saw blade’s teeth over the carved surface is probably the most common way to do this. In the smaller scales, a razor saw works well for this. Larger scales require bigger tools.
A modeler once explained to me that whenever he takes his family camping, he will take advantage of the campfire scene by using that time to whittle his own tree trunks with a carving knife. He prefers either balsa or cedar. What a perfect opportunity to make models.
Another way to make the narrow-tops inherent to the conifers is to use a modified sanding method on a dowel. Load a section of dowel into the chuck of a drill. If the rotating dowel is pressed lengthwise upon a sanding surface, e.g. sanding belt, a gradual taper can be imparted to the dowel. A number of modelers use this method to make their own tree trunks from dowels. In my shop, I have essentially created a sanding surface that is 24 inches wide that will allow several dowels to be simultaneously rotated and applied to a stationary belt sanding platform.
Whether done via carving or sanding, once the tapered shape is made, in most cases “bark detail” is desired. Inscribing bark detail with hand tools such as sawblades and rasps is both safe and effective, although it can be laborious. The softer woods make this task much easier. I have found that the difference between alder dowels and cedar dowels with respect to rasping is huge. The softness of cedar is much nicer to work with. Balsa I would imagine is better still. It may not be necessary to impart bark detail into all trees made by this method; background and midground trees made this way probably do not need to have this level of detail where most observers are not going to be able to appreciate the extra work.
There are a few items left to consider once the basic tapering and “barking” is complete. How will this tree be planted on the layout? If you are using balsa you can use piano wire or any other hard wire stem, apply ACC or epoxy and push it in. With cedar and other softwoods, drilling is likely necessary. I drill our trunks with a special jig that holds the trunk in alignment with a 3/32 drill bit. This size allows me to fit each trunk with a wooden “planting pin” which consists of a ½ toothpick. I’d like to find a suitable wire replacement for the toothpicks, however to date I have not been able to find them. Ideally they would be akin to a brad or finishing nail but without the head. The downside to the toothpicks is that they can break leaving a stub in the trunk base that can be difficult to remove. I recommend NOT gluing these in place, that way there’s still a chance of getting it out. Now for a word about what substrate you might plant these into.
Planting trees on the layout by virtue of a pin-system mandates a soft, penetrable base such as Styrofoam. There are a few types of Styrofoam but the most common are EPS (expanded polystyrene) and XPS (extruded polystyrene). The extruded type is preferable and easier to work with from a modeler’s perspective because it is easier to cut and does not leave the white Styrofoam “balls” laying everywhere that is inherent to EPS. On the flipside it leaves a pink or blue “dust” about the layout, so a shop-vac is a handy tool to have nearby. When I shape XPS I have the shop vac running simultaneously and I jig it with rope from the ceiling so that I can follow my shaping tool (usually a drill with rotary wire brush).
For a while I tried to develop a slick planting system for those who have plaster-based layouts. Some modelers have a decade or two invested into their layout and they aren’t about to convert over to a foam-based topography. In my kit instructions I suggest adding a foam layer 1” thick to a layout area for trees using spray foam (e.g. Great Stuff™). I envision that even that may be difficult in some locations that are difficult to reach or perhaps already have scenic materials in place. A stump with a mounting hole in it would allow the tree to be ‘planted’ while also adding to the overall detail value. One would need to fill the gap between the stump and trunk with wood putty or other suitable filler. The stump of course fans out at the bottom with roots that can be partially hidden with scenic materials. It could be glued to the hard plaster base. I have created a few such “tree mounting stumps” and they do work allright. One challenge has been lining up the stump and the trunk perfectly so that there is not a protruding ridge along the seam, and then masking the seam well with the putty so that the intersection is smooth and unnoticeable.
With wooden trunks of appreciable diameter (not skewer sticks), one now has the option of a) lowering a foliage material over the trunk armature or b) drilling holes in the armature to accept a branch-by-branch construction method. In my experience, this is where modelers’ perceptions of the landscape part ways. About half the modelers I have met prefer the branch-by-branch method and will claim that that method produces the most realistic model evergreen tree. The other half prefer the look of the over-the-top method, whereby the foliage material is more clustered and lowered from the top-down in sections, each section representing one to several branches at a time. In my humble opinion, a blend of these to methods may be the best way to make a realistic looking forest. We drill our trunks, adding what I call “dead branch holes”. Thus, although we offer in principle what is the “lower-a-group-of-branches-over-the-top” method, the trunks do come with some branch holes that could be utilized for branch-by-branch construction. However, from the articles I have read, the ideal branch holes are drilled at slight angles to the trunk to facilitate a “bow” effect to each branch, and ours are drilled at right angles. Our kits supply a desert bush material which is quite spindly and can be used for adding realistic dead branches to the tree once the green foliage material is in place. Therefore, if trees are to be built using branch-by-branch, many holes will now need to be drilled. Mini drills are commonly available for the task, though with harder woods they may not provide enough power to penetrate the trunk. We use a jig with four mini drills side-by-side to drill four holes at a time with about 1.5 inch spacing between them. Our jig allows the worker to actuate the drilling with a foot pedal, and use both hands to slide and rotate the trunk on a drawer slide. With this setup the holes can be added quite rapidly. A setup like this kind takes time to assemble of course, so it’s likely not the objective of most modelers. A simpler jig made from carving or routering a v-groove into a scrap of fir 2x4 will also suffice to hold the trunk in position while holes are manually added with a mini drill. Mini drills work fine in cedar material and are all right for the bamboo material also. The holes do not need to be very deep for dead branches. For the addition of green foliage, drill at slight angles and just a bit deeper to secure each branch of whatever material you are using.
Materials used for the purpose of branch-by-branch construction include florists’ supplies such as Caspia (aka “baby’s breath”) asparagus fern, Spanish moss, SuperTrees™ foliage (may work well for Ponderosa pines), certain lichens, and others. The choice of material is highly dependent upon personal taste and the modeler’s perception of the landscape. These days I read more about modeling specific tree species, and manufacturers are touting that they are offering different species of “ready-made” trees as well. Differentiation between morphologically distinct species is certainly important; consider how different the Ponderosa pine is from a Douglas fir. Modeling a specific region necessitates choosing species-specific evergreens and a methodology to make them that is reflective of what is truly growing there. Then again, from a modeling perspective there is not as much distinction between, say, a Douglas fir and a hemlock. This is where some minor variation in tree construction techniques can help to differentiate these species from one another, or perhaps just suggest the notion that there are different species coexisting in the forest; again, carrying the modeling to this depth is a matter of personal taste. Regardless of the material chosen for branch-by-branch, many modelers will follow up with either a paint application to the branches or an application of ground foam or both, to build in pine needle bulk and appropriate color.
One method I’ve tried is the use of Caspia (baby’s breath). I bought a large box of the material online and paid about $40 including the shipping. There was enough material there to produce perhaps a hundred trees suitable for HO scale. The material comes in many color choices, and I chose green. Looking back, I think brown or tan would have been a better choice, because at the connection to the trunk I find a color match suggests the woodiness of the branches rather that the green which is important more near the branch ends. If tan material is obtainable then an application of green latex via airbrush will treat the ends after construction is done. I did apply a green groundfoam material and that seemed to help create a better texture suitable for modeling western-slope firs (Douglas, hemlock, cedars, etc.) Eastern slope firs (Ponderosas, tamarack, etc.) have a bushier look to them. I have not tried modeling these yet; I believe the ponderosa would require a ponderosa-specific trunk. Perhaps the SuperTrees™ material could be used, or another desert shrub material, in conjunction with ground foam. I believe NGSG did an article within the last 2 years where one modeler explained how this was accomplished. I have not experimented with asparagus fern, although I’ve seen such models. Again, this could be used in a species-specific approach; I believe the asparagus fern offers a tapered green branch form that is quite pattern-like and in the form of singular, non-dividing branches.
Pete Vassler of Canyon Creek Scenics is likely the most well-known master of the method where he uses furnace filter material draped over a wooden trunk armature. Last I read he was using hand-carved balsa trunks, but he may dabble in cedar too. The way he handles the furnace filter is magic, but he custom-colors and treats each trunk individually as well. I believe he mostly uses individual fibers from the furnace filter to create dead branches on the lower portions of his trees. He uses ground foam over the filter material and he may do some application of paint to this. I glean this info from my critical observation of his work both in pictures and upon real examples. I have not tried this method myself. It appears the filter material is taken apart from the stock rectangle and applied. I’m uncertain if the painting of the material takes place before or after, but I see that deeper into the tree the material is darker and that does an excellent job of suggesting the drier, browner, needle-less type branches one truly finds in dense, thick, Douglas fir undergrowth near the tree trunk. This is definitely a western-slope style evergreen tree. Pete adds lichen & fungi detail as well. I learned about lichen colored paint from Pete.
Many years ago I read about using scrubber pads for the green foliage. I had not yet learned of the furnace filter material. After messing around with the scrubber pad I thought that if I added latex paint to it might give it some body. That seemed to work out all right. I first tried manufacturing tree kits back in 2000. All of the scrubber pads were processed by hand. It took me about 10 to 15 minutes just to prep a single pad, and that would make 2-3 trees 9” tall. I sold a few kits to hobby shops and swaps, but I had such a difficult time producing the materials that I knew I needed a few specialized machines. I stepped back from tree production until 2008, when I thought about the machines I had dreamed of before. I decided to make at least the first machine I needed, and that was going to be something I could use on the scrubber pads to treat the surfaces. I enjoyed some success with this machine, which in my spare time took about 3-4 months to complete. Now I could treat the pad surfaces automatically. I still use some of the same parts of this machine to this day. However, I still had to tear the surface-treated pads up by hand; a laborious job, & hard on the wrists. To truly make this process hands-free, and speed things up further, I needed one more machine whose purpose would be to tear the pads apart. I thought about this machine for a long time. When I thought I might have an idea that could possibly work, I invested about $500 and 3-4 months of my time to create what I call “The Monster”. It’s a 200-pound concrete cylinder with 600 nails protruding 1” from its surface. This rotates slowly, and the nails intersect with those fixed into a faster-moving but smaller diameter concrete cylinder. The Monster can chew up surface-treated pads and spit them out at a speedy rate. This was a big surprise to me, and what’s more I found that the characteristics of the Monster-treated pads were so much better than those torn apart by hand. The machine pulls, tears, twists, and works the material over and over, and does not flatten the material the way human fingers do when the material is gripped. From my perspective this machine has been a huge step forward in the scrubber-pad approach to making green branch foliage material. This material is then mixed with dilute green paint. The paint is mostly but not totally removed by a centrifugal spinning process. Once dry, the material is already colored and is ready to be applied to a tree trunk from the top-down. On larger trees (13”+), I use green hot glue to fix the material to a few dead branches I plant into the tree trunk, and then proceed with the top-down approach. The material I use is green through-and-through, which makes it possible to tear it into smaller pieces for use near the tree-top.
DIY-modelers can make their own green branch material from scrubber pads, it’s just labor-intensive and time consuming because everything is generally done by hand. I think for this reason many prefer the furnace-filter material, because it tears much more easily than the scrubber pad material. If I’m correct, I think with this method the foliage is generally colored afterwards by using spray paint or airbrush-applied color. One thing I like about the scrubbers is that I can create fine detail at the top of the tree. I suspect the furnace filter material would be a bit coarse for this. On the other hand, the furnace filter material, if colored dark brown before application, can provide a nice dense-branch look near the tree trunk that is harder to do with scrubbers. This suggests appropriate shadowing near the trunk in denser-branched trees. I think what it comes down to is that the furnace-filter trees suggest a branchier tree, whereas the scrubber-pad method trees depict a tree that has been more exposed to the elements (wind, time, etc) as branches are lost and there are bare parts of the trunk exposed. I am thinking that if the two types of trees don’t seem to work well together on the same layout, perhaps if the same application of ground foam that is used on the furnace-filter tree is also applied to the scrubber-pad type tree, that might even things out a bit with regards to texture and color.
For trees to be planted somewhere in the middle of the layout, away from close-up viewing but not too far off in the distance, I’ve had success using a bamboo skewer stick in conjunction with scrubber-pad type material. In fact, the overall profile is not that much different from some carefully constructed cedar-trunk trees. This makes them important contributors to the bulk of a forest. They appear especially convincing when placed close together in groves. Without having to shape each trunk individually, these “uni-color” trees score high in the timesaver category.
Using scrubber pads as the foliage source, this has undergone all the usual processing (cutting, surface treatment, shredding, grinding, etc.). I find it necessary to add color to the scrubber pads, as their factory-green color is too light to be prototypical of western forests. All chewed up material is placed in a vat of “forest green” latex paint, watered down 40%. Following this an industrial-sized salad spinner spins the material to remove and recollect the excess paint.
With latex paint coloring applied and wet, it is ready for application. I lower the material over the bamboo skewer stick (mounted in a vice on my bench) starting from the base of the tree working toward the top. As I work along I impart a sway to each branch section that helps represent tree bows. Nearing the top of the tree I use successively smaller portions of the pad material. This is where the artistry component comes into play, and I prefer a randomized pattern to the branch application over a simple conical approach. Apparently I am partial to the individuality that is possible with the construction of each tree. Well, I suppose that’s what makes putting them together interesting.
Well that’s enough for now. Stay tuned for my next article which will describe what I’ve learned about making rocks.