Designing the woodenwidget Plurt


The Plurt lightweight yurt with the Hoopy.

One day I will try to write a book about design. I find the subject so fascinating and although I never formally studied it I have a good eye and apparently quite an imagination. Plus I love to problem solve, to find solutions to insolvable issues. I believe there is always a way to get something to work. Sometimes the solution is a bit complicated. Other times it is simplicity personified.

There are two ways to design something, either start with a blank sheet or use what is to hand or easily available. The first option will bring the fantastic vision of the designer to reality but it will take a very long time and be very costly. The latter option is more practical and sensible but may end up compromised aesthetically. As someone who designs creations that must be practical and easy to make I cannot take the first option, instead I need to find simple and inexpensive solutions which do not spoil the overall look of the finished product.

Sometimes however common materials come together and get repurposed so well that one can only admire the simplicity of it all. That was my goal with the Plurt. But also I wanted to add another dimension to the design brief which was minimal waste. This means using a material in its entirety or thinking carefully about what to do with any waste left over.

It just makes sense to me to use materials as they come and not have to work on them. It speeds the process and reduces waste to zero. I would make the Plurt using thin ply around a wooden frame with bonded XPS to act as insulation and to add strength to the panel. From working out how much a piece of ply could be bent I reasoned I’d need 6 wall panels which would create a 5 metre diameter circle 1.2m high. The ply would not even need to be cut down.

plurt (2)

The six curved wall panels and door frame.

Of course creating curved shapes is much harder than making straight ones but as a boatbuilder with many decades of experience making things that aren’t flat is as easy as kiss my hand (as Jack Aubrey would say) to me. A jig would be required but as there are 6 panels to make it would be time well spent making one. It would allow the making of 6 identical panels which would ensure a good fit and if they were all the same then they could be interchanged allowing for different configurations with windows and such.

The only downside to a jig is what do you do with it after the panels are made? In the case of the Plurt the wall panel jig is modified and becomes the jig for the roof panels. Once all the panels are made the jig itself becomes a table or a bed for your Plurt and the curved cut out pieces are turned into a free standing shelf unit. No waste.


The shelves made from the jig. Strong and stable.

Each wall panel has locating pegs which not only hold the panels in the right place ensuring a good fit and seal but they also help to distribute some of the load from the roof. Once the panels are put together clamps pull them tightly together. The clamps act in much the same way as the wire that encircles a traditional yurt but as the entire roof of the Plurt only weighs 150 kilos, the forces on the walls is considerably less.

Designing the roof panels was much more challenging. They are flat but trapezoid in shape and have slightly angled sides so that they sit against each other well. They all fit in to a standard 700c bicycle wheel at the top. Traditional yurts have a large and heavy wooden crown and all of the many roof poles are fitted in to it. It’s one of the reasons why putting up a traditional yurt is not a one man job. I wanted to find a better way.


See how the tenon sits beautifully in the wheel rim and cannot lift out.

A 700c wheel is about as big a bicycle wheel as you can get and it is a very common size. In designing I find there is a lot of luck. For example when I was doing my experiments to see what I could do with the roof I needed to get the tenon in the end of the roof panel to sit inside the bicycle wheel rim. I had imagined needing to cut a special shaped tenon but as it happens a simple rectangular shape sits beautifully inside the rim when the roof is at its final 26 degree angle. Nice when that happens.

Apart from getting the size and shape of the roof panels right there were many other issues, for example, how would I fill in the inevitable gap between the roof and wall panels created when you place a flat object on a curved one? How would I waterproof the 15 joins? That’s 40 metres of potential leaks! I did some quick experiments and discovered that a simple guttering made from garden hose worked just fine and even held itself in with friction alone. I was very happy to solve this one so easily.


A worm’s eye view of the roof.

There are many reasons to use garden hose as guttering. It fits in with the low cost and easy to find ethos for a start but it even works on an aesthetic level too as the gutters mimic the poles that you would find on a traditional yurt and helps to break up the large wooden surface. Bottom line? It looks good. It looks right. Other advantages to this system apart from the low cost are that these same gutters can be used to carry cables to the roof should you want a fan or lighting up there. But best of all it allows for a nice margin of error. If the roof panels are not 100% and there are some gaps between them, it really doesn’t matter because the hose just expands slightly to take up the slack.

Calculating the size and shape of the roof panels was not easy but harder than that was trying to minimise waste. For a long time I juggled with different sized panels, more or less panels and different roof angles until I finally found the compromise that would allow a good roof angle but just as importantly the right size that would allow the off cuts from one side of the panel to be used on the inside. A traditional yurt typically has a roof angle of 30 degrees so I was not unhappy about getting to 26 degrees.


The 15 roof panels leaning against the walls.

Initially I imagined the roof panels would also have some kind of locating device to ensure that they couldn’t move or slip plus they would help to spread the forces but due to the way the panels are fitted when the roof goes up, it just wouldn’t work, so that idea had to be abandoned. It seemed to me that once the whole thing was assembled and sitting right, the tops of the panels would all lock in to the bicycle wheel at the top and the supporting knees would lock in the lower end and once the clamps were added to the lower end it would have sufficient stiffness.

The next issue was how to assemble the roof on top of the wall panels. Traditional yurts often have a central scaffolding which is later removed so I decided to mount the wheel on a post in turn clamped to a simple step ladder. Obviously the wheel has to be mounted centrally to ensure the roof has equal overhangs. There is always a little margin for error of course but not much. Part of the compromise of the minimal waste design meant that there would not be much overhang so it had to be placed pretty accurately. the solution to this was to use a plum bob hanging from the post and pointing at the very centre of the yurt base.

Once the wheel was centred and at the correct height another advantage to this system came to light. Originally I assumed that the panels would be pushed in to the wheel from the outside and as they are triangular the gap would close as they went home but an unseen advantage of the tenon at the end of the roof panels is that they can be dropped in and rest on the inner rim of the bicycle wheel before being pushed home. This means the panel is supported while you adjust the last few millimetres.


Note the roof panel centrally placed over the door with added rain diverter.

Bear in mind that it’s one thing to make a structure like this but it is quite another to find so simple a way that it can be documented and explained for the builder. Seems that luck was on my side once again. The first roof panel is fitted centrally above the door frame. This was not my decision but arrived at because if there was a join at the door there would be a gutter dripping water on you as you went in when it was raining.

To keep the weight even on the bicycle wheel, the roof panels are assembled one at a time and the next panel is fitted opposite the last one and so on until all the panels are fitted. The roof panel that fits opposite the first one just happened to line up with the end of one of the wall panels so it is an easy thing to describe. The roof panels go in until there are two spaces left. One of the gaps will be too big for a panel and the other too small. A roof panel is fitted to the larger gap and then the rest of the panels are just shuffled across a few millimetres until the last gap is just big enough for the last panel to be dropped in.

At this point the panels are clamped together. There is one clamp at every join nicely hidden at the lower end. Now the roof is pretty solid and yet still not actually fixed to the walls and this is where the knees come in. They have slots in them so they can be adjusted. Once the roof is up, the knees are loosened and a shim is placed behind them so that the next  time the Plurt is disassembled all you have to do is remove the shim, the panel will drop down slightly opening the gap and the tenon will still be resting on the bicycle wheel rim. Now it can be pushed up and lifted out. As the roof panels are only 10 kilos each it’s not a hard thing to do.


The adjustable plexi dome fitted through the hub of the 700c bike wheel.

The next job is to fit the dome but how can you do this when the bicycle wheel is on a step ladder and has a bolt fitted through it? You can’t. first the scaffolding needs to come down. Once the roof is free standing the mounting bolt fitted to the wheel can be removed making space for the dome which is fitted using a threaded lead bar. But how do you get on the roof to drop it in? You can shimmy up the roof and place it in that way as the roof can easily take the weight but there is a better way.

Simply remove one panel. The roof stays up just fine even with the odd panel removed. Then get the ladder and pass the dome through the big gap and drop in into the hub of the wheel. Now the panel can be refitted and reclamped.


Fitting the guttering in to the grooves in the roof panels.

Now the Plurt is erected but not fully assembled. The internal guttering needs to be fitted. It’s an easy task but does take a little while as there are 15 x 2.5 metre length gutters to fit.


Sometimes the best solutions are the simplest. Foam pipe insulation fits perfectly.

The gap between the roof and wall panels was still unsolved at this point. Sometimes when you do not have a solution it’s best to stop thinking about it and often this way a solution will appear and that is exactly what happened. Turns out the perfect material for the job was some foam pipe insulation. It’s a neutral grey colour and can be forced in to the gap making a very respectable seal. It’s perfect and fits the easy build ethos. It’s cheap and readily available and comes in one metre lengths which, believe it or not, is exactly the distance between the roof panel knees so they didn’t even need cutting down to fit. Sometime the design gods are on your side. A perfect solution for what had appeared a really difficult problem to solve. it’s often this way with design. Often what you think will be the hardest problems to solve turn out to be the easiest and vice versa.


In the Plurt you can fit opening windows anywhere.

Now that the Plurt is fully assembled and weather tight it’s time to fit the windows. This is just much easier to do when the walls are in place and firmly held. Unlike a trad yurt which has an interior lattice the Plurt can have opening windows anywhere in any panel. This is important because a yurt needs good ventilation. As the wall panels are identical you can move the window lay out around if you fancy a change. Not something you can do on most yurts.

The Plurt has many advantages over a traditional yurt and perhaps the biggest is that it does not have a fabric skin. No doubt it’s a good watertight system but it doesn’t allow easily for openings and will eventually rot away in the sun. The Plurt is simply painted plywood so when the paint is getting a bit tired, all it needs is another coat and it’s good for the next few years. replacing a fabric covering for a 5 metre dia yurt is a costly undertaking to have to make every ten years whereas a tin of paint doesn’t cost much in comparison.

The advantages keep coming. If you wanted to make a traditional yurt yourself you’d probably come unstuck at the fabric stage. Unless you have a large clean space and an industrial sewing machine you simply won’t be able to make the covering and having done some heavy duty canvas sewing I can tell you that sliding around a 20 sq metre piece of cloth is very hard, even the professionals struggle. This simple fact alone means that a diy yurt is out of the question for even the most resourceful of people. And consider how hard it is to fit such a cover to a framework standing 3 metres high. Doing away with the fabric is the first step in being able to offer a diy yurt.


The very cosy and bright Plurt interior.

Another advantage to having no fabric is that it’s easier to have windows that open. Not only that they can be made of glass which is much nicer to look out of than the transparent plastic fitted to most yurts. You can even fit salvaged windows in the Plurt.

None of the panels are long nor heavy so it makes the Plurt much easier to assemble, even alone and it makes it easier to transport and store. Less materials means less cost. The Plurt typically costs about half that of a similarly sized traditional yurt.

Despite the Plurt’s light construction it does not feel flimsy at all. It does not move when the wind blows and the insulation dampens the noise of the rain although you always know it’s raining when you’re in a Plurt.

As I said at the beginning of this post, design is a fascinating subject and it’s amazing how by choosing certain important criteria before you start can have such an effect on a design. There really is very little waste when you build a Plurt and that is a good thing. We all need to be more conscious of how we use the planet’s valuable resources and reducing waste and building in wood is a good place to start.

How many other small dwellings can be built so cheaply and be taken apart in minutes? And on top of all that there’s the element of being in a circle, the calm that it brings compared to straight lines and right angles. All in all I am very pleased with how the Plurt has turned out. It’s a lovely space to spend time in. It could be used to live in, or as an office, a kid’s playroom, a yoga space or even rented out to earn a living from it. Having created the design and spent months perfecting the plans it’s now up to others to decide how they will use their Plurt.

Why Plurt you ask? It’s PLywood yURT but later I discovered that it also stands for P.L.U.R.T. Peace, love, unity, respect and trust. Nice. Plurt it is.


A very customised Fliptail 7


Woodenwidget has been selling plans to build folding boats for over ten years and in that time many builders have been kind enough to share their attempts. It is always a delight to see how builders finish their boats. Some copy the plans exactly, even down to the same type of wood and fabric. Most personalise their craft in some way but do not deviate from the plans structurally but every now and then an ambitious builder lets their imagination run wild.

Alex is one of these builders. He owns a rather lovely small classic yacht and he wanted to try and match a dinghy to it. His emails were intriguing and right up to the end I had no idea what he was up to because he said he didn’t want to send any pics until the boat was finished.

So when I finally saw the pictures I was so impressed. He was worried that as the designer of the Fliptail I might be somehow offended by his modifications but nothing could be further from the truth. The pleasure I get from seeing what people do with the design is very heartening.

What you see below is basically the email I got from Alex with his comments and pictures which explain it all much better than I could.


Hello Benjy,

As promised, some construction photos and final results of my Fliptail 7, "Foal", tender to my 19′ Ralph Stanley sloop, "Bucephalus".

First off, I want to say explicitly that none of the diversions I made from your plans were because I felt the design was in any way flawed. All the significant changes were because I’m fussy about wanting something aesthetically just so, or to fit her into her very specific role as a tender. Your plans were excellent, far better thought out than my modifications, and any difficulties I encountered in the construction were entirely my own doing, as would be any failures of the vessel as I built her.

The first big change I made, of course, was making her a peapod, double-ended. This was pretty straightforward: figure out the midpoints of all the longitudinal elements, backbone and hoops, and then mirror the bow section. For instance, two stems:


You can’t have too many clamps:


With the hoops, I went a little further off piste. I knew I wanted a concave sheerline, and the best way to accomplish that in this situation was to have hoops with a fair curve, and then angle them down slightly. So the hoops are a slightly different shape than the typical Fliptail’s; significantly, there is no absolutely straight section, they are curved throughout. They still start from the same principles and basic dimensions, though –and they still take a lot of clamps:


Turns out you can’t get 10mm plywood over here. Even high end wood suppliers stock in 1mm increments up to 9mm, then jump to 12mm. For the sake of lightness, as well as because 12mm wouldn’t fit when folded up, I used 9mm okume. It turns out it’s pretty flexible stuff, when you’re sitting on it, but I think it’ll be fine. Again, I slightly tweaked her plan view, even beyond making her double-ended, to fit the differently-curved hoops:


Given that she will be used in a salt water environment, all hardware is either brass or bronze. This includes 1/4"-20 bronze carriage bolts for the floorboard supports, since I didn’t have a full 10mm of ply to countersink for machine screws. (As an aside, reconciling your metric instructions to my SAE working habits and materials suppliers was a challenge in its own right):


A minor change of detail, on the keel cheeks, to remove a sharp point:


I couldn’t bear to let the offcuts off the ends of the hoops go to waste, and since I wanted something a little curvier at the ends of the hoops than the cedar wedges specced in the plans, to show the sheer better, I used the offcuts thus:


Foal’s construction sequence has been considerably different from that you outline for the typical Fliptail. I have done a lot of pre-fitting of parts, since I’m detailing them differently, which means I need to assemble, disassemble, adjust, check the fit, and only then start varnishing and painting, once they’re shaped as I want them. The complexities of changing her to have a concave sheer made things even more difficult, as everything had to be assembled, scrutinized, and adjusted many times to be sure the curves were coming out right. It has definitely slowed the process, but I think the results are worth it.

To match her parent vessel, Foal is painted blue-grey inside, with bright trim.

Lower hoops in place:


Stem detail. The oak bearing pads, laminated to the stem and sternpost, have their outside faces canted out about 6°, to angle the upper hoops downward and create the concave sheerline. Angling the hoops down reduces her freeboard by a couple inches, but the geometry also works to give her sides a bit more flare:


Since I had the ash offcuts on hand, and it makes nice detailing when finished bright, the floorboard supports and upper hoop supports are of ash instead of cedar. Since the assembly process was slow anyhow, I took the time to laminate the brace-blocks onto the floorboard supports, instead of screwing them on, and shaped some curves to get rid of some weight and lighten them visually. I also glued a pad across the end grain where the lower hoops bear, to protect the end grain from splitting. In this photo, the bolts all have yet to get their nylock nuts, and be trimmed to length:


Sternpost detail. I left the pads long to create a fairlead, either for a towline, for towing Bucephalus, or for rowing out a kedge, or taking a warp ashore for mooring or warping in. The hoops also sit an inch lower on the sternpost than on the stem, to enhance her sheer:


Upper hoops in place, with temporary vertical supports to fine tune the sheerline. Angling the upper hoops down to create the sheer had the effect of flaring the sides, which angled the bottoms of the upper hoop supports inward, and in turn meant the floorboard supports needed to be shortened:


You can see the 12° of "deck camber" a bit better here:


Final vertical supports in place, but not yet varnished. There’s only so much you can do to add sheer to a 7′ boat, but from a little distance, she does have the tiny bit I had hoped for:


I have an unreasonable hatred of barrel-bolts. To avoid using them, I instead shaped the heads of #6 screws on my lathe to remove the flare of the heads, and screwed them into the endgrain of the vertical supports. I then drilled into the floorboard supports and lined the sockets with open-ended Chicago screws left over from the floorboard hinges. The Chicago screws have a slight crown to the end, which provides both a strike plate for the screws/pins and a little more clearance to allow water to drain off. It takes stretching the fabric a bit to get the verticals into place, but they hold their position quite well:


I also re-invented the detail of how to locate the floorboard supports on the lower hoops. Instead of notching the lower hoops, I drove a 1/4" oak dowel through the keel, to serve as a stop for the floorboard supports when they are swung into place: (designer’s comment: Normally a small flat is cut on the upper side of the hoop and the floor support locates in the slot. This is so that when rowing you can put your feet on the supports and force against them. I worry that Alex’s solution is not ‘idiot proof’ but so long as he is aware of the issue, his solution is fine)


Covering the modified hoops went just fine. To match Bucephalus –and my other dinghies– I used dark green fabric for the bottom panels and white for her topsides. For sealant I used up a couple partial tubes I had on hand of both 3M 4200UV and 3M 5200, aka "demon snot". In keeping with her bronze and brass hardware, for salt water resistance I used 1/4" monel staples, for longevity:



Foal together with Bucephalus’s "home waters tender", Toggle:


For a rubrail I did as I had done on Toggle, and used a length of 3-strand spun dacron (1/2"), long-spliced into a loop. I first routed a 1/4" radius cove 3/16" deep into the upper hoop, positioning the cove at the top edge of the hoop so that the rope would stand proud of the wood both on top as well as to the outside. This is because I’ve found that when coming along side a larger boat in any sort of a chop, a dinghy will tend to sort of scoop its rail up and into the larger boat, not just bang against it sideways, so padding along the top edge is warranted as well. (Also, bronze nylock nuts are now in place.):


With the cove shaped, I stapled the top edge of the topsides fabric into the cove, so that the fabric turned over the bottom edge of the cove. Ideally this will ease some of the point-loading against the staples. The rope rubrail then covered the edge of the fabric and is secured in place by 3/4" #6 bronze round-head screws: you insert the screw in between two strands of the rope, and then drive it *through* the third strand and into the rail, so that when it is driven home, the head disappears beneath the first two strands. Be sure to use round-head screws for this, if you try it: flat-head screws tend to frazzle the rope as you’re driving them. There are slight bulges where the screws are, but those tend to get less obvious over time:


At stem and stern I left bights of rope long enough both to let the hinges work and to use as lifting handles:


The aft bight is longer, to work as a sling for rowing out an anchor:


To take the strain of lifting, the rubrail is also seized to the upper hoops close to the stem and sternpost:


The sternline is just girth hitched to the stern lifting becket, since it’s unlikely to take much strain, or even to get much use, but for a little bit of "bling" the painter is girth hitched to a bronze captive ring:


As for where to put the oarlocks, it turns out your written instructions were, for me, exactly right: they work well immediately abaft the aft upper hoop supports. With two people in the boat, each sitting at the extreme ends, that position even works (sorta; adequately) to row stern-first and keep the boat a little better balanced on her waterline. Contrary to what you indicate in the instructions (yet again!), I installed the oarlocks to the inside of the rail rather than to the outside: I lose a couple inches breadth of effective rowing position, but it removes any chance of dinging the boat I’m coming along side of:


And Foal has now had a preliminary bout of sea trials. I don’t have any photos of her under way, as I was the photographer and could not both row and operate a camera, but here’s proof that she does float:


She also makes a very tidy package, with a pair of 6′ oars. I’m still working out the best way of lashing her for stowage, so please excuse the painter and sternline macrame:


I am happy to confirm that she is surprisingly stable, and rows far, far better than I had expected (even with the rowlocks just clamped in place). I admit, I expected her to handle like an overturned umbrella in a duck pond, but that expectation was completely unjust, and her handling is stellar. I think she will be an even better tender than I had hoped for:



Many thanks for all your support throughout the build. Let me know, now or in the future, if I can provide any details of Foal’s build, or the suppliers of her materials, for Wooden Widgets’ library of information.

All the very best,





Hoopy wooden bicycle


The Hoopy lightweight wooden bicycle that you can build yourself.

Ever fancied building yourself a wooden bicycle? Well now you can. Woodenwidget offer the Hoopy, a funky, one size fits all, lightweight, easy to build wooden bicycle. Right from the start, the Hoopy has been designed to be as easy to build as possible requiring no specialist tools or materials.


Unique internal crank keeps clothes clean and looks great.

The plywood frame weighs from just three kilos and although it might look complicated to build, it really isn’t. The frame can be marked, cut out and glued ready for varnish is just a couple of days with little more than a drill and a jigsaw.


Gorgeous wooden rims from Made of Slovenian beech. Wooden headlight and leather clad forks too.

The hollow frame has many advantages. It can be used for storage or even batteries and a motor for an electric conversion. No mudguards are needed. You can even cut your own designs in to it but the most unusual advantage is the central drive chain. No more greasy clothes or ripped trousers! It looks better too! The low shape makes it suitable for everyone as it’s so easy to mount.


This Hoopy even has wooden dust caps for the valves!

About £100’s worth of new parts must be bought but the rest can be from salvaged bikes or bought second hand. With care, it should be possible to build a basic single speed Hoopy for about £250 but as you can see from the photos it’s possible to spend a lot more!


Brass dropouts, gold chain, wooden pedals.

The seat height is fixed yet works for the majority of adults, a taller rider simply sits a bit further back on the long seat. But if the builder preferred a more tailored fit it is possible to adjust the seat height during the build. That same seat is very comfy and long enough to easily carry a passenger. Or even a small third one on the wide crossbar.


You can even cut your own shapes if you don’t like the ‘hoops’

As bikes go it might surprise you. It’s very smooth, swift and comfortable especially with the 2 speed auto hub and wooden rims. It’s nicely balanced and going ‘no hands’ is no problem at all.


This one has no cables and a two speed auto hub.

Plans are available from Woodenwidget and if you buy Foldavan plans, you can get £10 off any of the other plans. Woodenwidget will also plant five trees on your behalf. At just £30 the plans are extremely good value with 30,000 words, over 200 pictures and 130 pages. Plus a huge amount of extra info about woodworking, tools and varnishing etc.

To find out more visit


Even the back light is made of wood.


Hoopies on the beach.