Between this chair, the guy making transparent wood, and the absolute state of the marble machine x project, I've become quite invested in the creative struggle as seen on youtube, in a way I hadn't experienced since the french guy decided to turn his cooking channel into a step by step guide on how not to make a pasta machine.
In addition to the comment on Matthias Krantz, just don't watch his most recent video to start out. So many problems and he didn't even include them all in the video, he might still have them in his discussion board though
@@3x3CustomTamar Really Great try! IF (BIG IF) you attempt to make it again, how about cable through holes drilled in the angled pieces? You could tighten it as necessary with a turnbuckle in the bottom piece. You don’t need to secure the slats to the fabric. Secure the slats permanently to the short angle pieces, not to the fabric or cable. Anyway, respect your trying to solve this physics build !
Now, I suck at maths, however, this seems like a situation that might benefit from some: The entire top half of the chair is a lever, with a fulcrum around the corner of the middle joints, with the other side of the lever being the face opposite the corner that's acting as a fulcrum and is pulling on the strap. For this chair to work, the force of the lever where the face meets the strap has to be lower than the elastic deformation of the strap itself. Assuming you weigh 60KG, that you sit half a meter from the fulcrum, and that the distance from the fulcrum to the strap is 4.5CM, the force at the strap is approximately 6,500 newtons. The Venom strap advertises a mere 1,800 pounds of *breaking* strength, which is just 8,000 newtons (there are two straps so you probably won't break the strap, but clearly we are exceeding its elastic deformation point). So no matter how much pre-tension you put on this system it's never going to work (as you surmised). Either a redesign is needed or material must be used that won't deform significantly from these forces, or even twice these forces (since if you sat down quickly you'd have momentum, and you can lean back in the chair). Basically, stapling the straps will not work. Granted, I may have made some mistakes in there or oversimplified the situation in some way. It's been a heckin' long time since mechanics A levels for me. Anyway, I love this idea, and I really hope you find a way to make it work. The fact that you kept trying to make it work after 6 attempts is inspiring!
i'm not nearly as educated on topic as I'm only on second year of architecture studies but based on inspecting the Miesrolo chair i think you're absolutely right. it can be clearly seen that the straps used are much stronger even tho the leaver is shorter so it has potentially lower stress. they appear to be much wider and made of non elastic jeans-like material and glued in multiple layers so it's so much stronger
"I suck at math" *proceeds to discuss several physics principles and concepts, all requiring math* Don't sell yourself short man, you clearly have a good handle on things. I agree with you, the geometry of the chair is working against her since her body has so much leverage against those straps. The material needs to be flexible but have an incredibly high tensile strength if you want to keep the strap width and thickness reasonable. I would say that getting rid of the straps all together and using pinned joints that have their movement limited by the geometry of the chair segments could work but then the back support wouldn't stay in tension. It's a very tricky problem...
I disagree that these straps won't work. Your math is correct, however the 1800 pound breaking strength is for 1" straps, but it looks like she is using 2" straps which are rated for double the breaking strength. And like you mentioned, she has 2 straps for a total of 32000 newtons of breaking strength. With the 6500 newtons of load, I don't think we would be at the point of plastic (permanent) deformation, even with double load. The issue I see that I outlined in a different comment is the amount of strap that is allowed to stretch. The entire length of strap can stretch in the current design, and can be allowed to show itself in only one joint. If she were to attach the strap to the blocks by staples, or even better - glue, the stretchable length of the strap will be greatly reduced (the glued sections of strap will not stretch any appreciable amount).
@@nickdibart the existing straps may still work, but surely you can agree that a new material with a higher plastic deformation point would benefit the design as a whole.
HEADS UP!: **Leather is super stretchy.** -- Climbing shoes made from leather usually stretch much more than their synthetic counter parts. Cow hide is just thick human skin which is supple and dynamic. Slackline and highline can have as little as 2.5% stretch. Maybe there are industrial materials that are even more static than this but inherently there will be some stretch with woven materials. Stainless cables will have far less stretch. Maybe make an expanding end point so as you are setting up the chair, the cables can be stretched with mechanical force (a lead screw pulling both cables taught). I get the appeal of using wood and cloth to make an impossible chair but maybe that is impossible.
@@kingkarlito Maybe. Because of the joint solely relies on friction, there is a fair amount of tensile force needed from the cables. What might work better is skipping on the rolling portion of the chair and just make it a folding chair with some fancy origami tricks (aluminium sheet + custom hinges/piano hinges will optimise strength and the project's mass)
@@dragade101 The rolling chair idea is still Really good for portability and storage, but yes she needs a material with basically zero tensile give and enough strength to hold. Small steal cables running the length of the strap fabric is probably the right direction.
It seems to me that this sort of thing is really where a mechanical engineering degree comes in handy to make the right material choices and to calculate the structural aspects of the design. I *think* it could in principle work but I but that's assuming ideal (ie theoretical) materials. Maybe there are materials that are close enough but my suspicion is that all the structural elements will end up having to be metal instead of wood/textile. All the rigidity comes from tension and the lack of stretch/compression of the materials. Kind of like how we pair concrete with steel rebar. Steel has low stretch for a given tension, and concrete doesn't compress.
@@simon7719 kind of. Sometimes its best to just experiment and iterate several times. You could try to calculate the needed forces but why not just over build and test your design! There likely is a tension that would these blocks safely and it. Maybe what would help is that each block has some registration keys, especially for the back pieces. That once the blocks are drawn together and held fast, they cannot hinge away from each other (that simply they could be parallelograms with a keyed surface to help them not slide off of each other). The back of the chair is probably the hardest area as you are not loading it the same as the front arc is being loaded. Finding the right segment shape is probably the key here.
The issue may be with the design itself, specifically the back support. You do NOT want the tension evenly distributed, as the forces are not equal. The back has barely any forces acting on it compared to the bottom curve, so when you increase the tension, all that happens is that the back (where the webbing runs at the inside of the curve) is curved inwards - and since the webbing is not blocked by the wooden blocks like it is at the large bottom curve, it just releases from the surface of the maple until it is obstructed by the walnut. Thus, any additional tension will just continue to force the back curve inwards, until it collapses and releases the tension of the webbing like it does at 23:08 - notice how the entire chair drops at bit. I understand the benefit of leaving the segments to slide around freely, but I think you have to at the very least attach the webbing at the base of the back support, before it curves up. That way you can tighten the bottom part of the webbing A LOT indepentently from the back, where you would want a lot less tension. That way, the base of the chair would even keep its shape if the back support collapsed.
You have a good point. The strapping needs to be bonded at each join to prevent the highly loaded areas from transferring that tension to the lesser loaded areas and altering the shape. However the webbing at the thickness and width used is just not stiff enough for the loads. I think a groove and a mostly matching rib on the back of the slats would pinch the webbing enough to isolate each joint. With that and a few extra layers of webbing and it may work. I would need the material data to be more sure.
@@court2379 Very good points here. I would also look into much wider strapping, or even using something like canvas or similar material that would completely run from one side to the other. This would spread the tension over a much greater area and hopefully reduce the tendency of the material to stretch. The use of a thin steel cable would also reduce the stretching issue.
The static designs I've seen on the Web use some method of making the sides effectively thicker in the front-back direction in order to increase stiffness. One method is to form a triangular truss structure with a third side between the original two. That can be done in a number of ways. Another method is chords, that is, some kind of cable connecting distant points on each of the two sides. And absolutely, there are at least three sections in the chair, with different forces acting on each, and each section and their forces should be analyzed and addressed accordingly in the design.
@@court2379 I don’t think you need to bond the strapping at every point, but for every change in the load. Especially the forward bend and the chair surface need to have 2 independent belts to distribute the weight loads on the structure.
Having a quick scroll down the comments, I didn't see this suggestion so thought I'd share. Why not do away with the strapping entirely and make a mechanical hinge connection between each block much like a metal watch band?
Came here to say this, I think with the current design the strapping will always be a failure point as it will stretch over the entire surface of the strap. A mechanical hinge at each point will eliminate the stretching over the 5ish foot length of the design. I think she’s on the right track though! The dowel alignment to avoid twisting is frankly brilliant. Don’t give up!!!
i think you just need make the blocks take more stress with a box joint and that would be sufficient i do think a mechanical connection would look better
The straps can still be used, but they need to be fixed to each piece of wood. The current design allows the elasticity of the whole strap to come into play - even if it only stretches by 2%, over the 1m length that's 4cm which can open up at any joint. If the strap is constrained then the same 2% elasticity becomes insignificant on each joint.
I think it is the positioning of the tension that is causing the issue. Use wire running through holes in the CENTRE of each block (this will stop it folding over. Make some kind of screw adjustment to allow the tension to be adjusted. Fantastic effort btw, admire your perseverance
Just going to add some thoughts from an engineering perspective: You talked about how you selected your strap materials based on them being advertised as strong and lightweight, but strength was never your limiting factor. Deformation (i.e. stretching) was, and that's a function of stiffness. Stiffness is literally defined as the ratio between an applied force/stress and the resulting deformation/strain, and while strength and stiffness are of course highly correlated, it's certainly not universal that a strong material will be stiff nor that a stiff material will be strong. More to the point, plastics, even if they're strong, generally do not have high stiffness when compared to metals, ceramics, or other construction materials (which is why we don't have plastic houses or bridges; they are similarly constrained by limitations on deflection as much as strength). With that in mind, the hinge idea others have suggested is probably a good one, as metal will be a stiffer material to begin with, and the hinge will also be a bulkier, thicker, and thus stiffer interface than any strap could be. One other idea would be to look into unidirectional carbon fiber , which can be bought in "tape" widths that come on spools and could be swapped out more or less directly for the plastic strapping you've tried. Carbon fiber is one of the stiffest materials in the world in tension (significantly stiffer than even steel), yet without the epoxy reinforcement that its usually coupled with, bare carbon cloth is nearly as pliable out-of-plane as any other webbing would be. Sandwiching a few layers of unidirectional carbon together and leaving them bare would almost certainly be your best bet to make the strap-based design work. As a bonus, it's even black, so it'll match the look you were initially going for! Just be careful and do some research before buying to ensure you're getting real carbon cloth rather than a carbon-fiber "look" in some cheaper material, and that it's not preimpregnated with epoxy.
I would agree but don't think you need to go to the expense of Carbon Fiber when the metal banding strap should provide enough tension without the stretching. Just ensure that you are tight when first applied, maybe even over tight and let it stretch a little and it should work just fine ....you my need a couple of straps to equal the width of your current banding stretchable material...at least it would not hurt to try it.
As another engineer, I had the same thought. The issue here is the stress/strain curves for the materials selected for the tension component. Nordic_Nerd is right on that basically no plastic will have little enough strain to work, and leather is going to actually be worse in that regard. I think the only feasible options for the loads you want (holding up a whole adult with essentially just tension over about 2 meters with less than a cm of strain) are going to be metal or carbon fiber. The CF route will look cool, but the material is expensive, and be difficult to work with, and you also need to figure out how you're going to protect it from outside loads since it's generally quite fragile in the non axial tension direction. I think the best bet is actually to just run ~1/4" steel cable down the inside of the wooden blocks as the load bearing line, and then you can look at maybe just adding aesthetic strapping over top (imo it would look pretty cool just as stained wood).
Could pre-stretching the poly or nylon strap for an extended period of time solve the problem? Could inlaying metal to reduce the overall amount of wood (a more compressible material) reduce the amount of excess tension that causes the chair to roll up while in use? Would using shorter lengths of whatever material and dividing the chair into sections i.e. base, leg, seat, and back limit the stretch in any one section and thus the overall tension as well? I think that running a steel cable internally (maybe through sleeves to prevent it tearing through the edges) might be the best bet. If so I'd look at complete cable systems for stairs and railings with tensioners that are provided.
My daughter's boyfriend build a chair like this and had the same issue you were having. We solved it by using a chain similar to a bike chain in place of the straps (looked closer to mini tank treads so that the screws could anchor through it. He also used hardwood as the pine he originally used deformed slightly under compression.
Was going to recommend the same, not sure what the stretch would be for a bike chain, but I'd guess similar or less than steel wire (which is about a tenth of low stretch webbing). Personally I'd put it in a chanel under the webbing, then you won't see it when the chairs up and it might stop crumbs and dust falling in it. Fixing it at various points would also limit the stretch in specific areas. Currently the spare material from the back and base can spread through the chair to make gaps, if the material is fixed either side the stretch between each joint is limited to a percentage of the ~15mm between the slats, rather than the stretch for the whole chair being a percentage of the ~3000mm of webbing in the chair. This might cause problems with rolling it up if the support material isn't inline with the corners it hinges on, but that should be a simpler problem (e.g. rounding the corners slightly).
The beauty of your channel, above and beyond the fact that you are an amazing self taught woodworker, is that you allow us all to see your vulnerability with the highs and lows of each project. Anyone who has spent any time in the shop, whether they want to admit it to themselves or not, has been down the same path. Thank you for allowing us to go on this journey with you!!
@@3x3CustomTamar what John said above is by FAR one of the reasons I love your channel (pretty sure I've said this before). It goes a LONG way to making the rest of us see our failures as a lesson learned. Please do not stop with that format of videos, it's very refreshing and makes the rest of us feel like normal humans who make mistakes in the shop too.
I was thinking about metal hinges instead of strapping too, though finding ones that are strong enough to support an adult while being small enough for the design might be an issue.
If a 200lb person sits on that 18 inch diameter, the max moment in the chair is about 150 ft-lbs. For two straps , that's about 75 ft-lbs each. If the blocks are 2 inches thick, the strap tension needs to be about 900 lbs to support that weight. The design on the strap is not about strength, but stiffness. Nylon's modulus of elasticity means that even if it does not break, it will stretch too much. If the user "bounces", the loads can easily double or more. Steel strapping (like used for attaching heavy cargo to pallets), or hinges could provide the required strength, with minimal elongation under the loads here. It would also make the chair much safer.
This is so much work and I think you are almost there! 1: The nylon strapping is going to stretch. A non stretchy material like Kevlar or Stainless steel wire would be better - frankly SS wire would probably be the easiest. 2: The various blocks are snapping back when you sit on the chair because the lines of force are acting around a fulcrum made by the block corners. This is always going to be a risk. To get around this I would drill holes in the middle of the blocks (Through where the beautiful dowles are) and run stainless steel wires down the middle. 3: With this solution of course you wont be able to just roll the chair up because when the SS wire is tight, rolling the chair will require the wire to stretch, which it wont. 4: To get around this you will need the ability to release the SS wire when you want to roll the chair. This can be done with Quick Release type levers which can also allow ajustment of the wire tension as the wood settles. 5: Even with this I am unclear due to lack of experience with wood, if the wood itself will have the necesarry compressive strength to resist the very high forces that its subjected to. 6: I would try this myself but I dont have a wood shop :-) 7: I love your videos, please keep up the good work!
Amen. There are so many videos on TH-cam of people being successful. It’s really inspiring to see someone fail and keep at it. You inspire me. Thanks for sharing.
Leather will,stretch as much, or more than webbing. I love how you’re trying to solve all the problems and wish I had a useful suggestion to get it to work
Look into kevlar reinforced belting, like motorcycle drive belts. They are VERY strong and have like ZERO stretch to them since the have to work under enormous tension without slipping. Might be a bit pricey tho. Not sure. Maybe find a Harley dealer/shop and see if they have any old belts they would give you.
My first thought also was: Well.. if tension-flex is the problem.. maybe going for something that doesn't flex? I am 100% not smarter than anyone who already saw and commented the previous videos and especially not smarter than YOU xD - but I had seatbelt in my mind. But probably I am just naive and it flexes as strong as any of the already tested materials. Also I can't explain from an engineering perspective, but I would fix/screw the belt to the blocks at least where the sitting-area is. That would take tension from the upper/back part, where it isn't really needed/helpfull. Love your enthusiasm with this project
I would personally go with carbon fibre but without any resins (you just need tensile strength) and localize mounting in a way that focuses on stress points. In order to prevent that thing from collapsing you need material that just don't stretch. I'm also not really sure if compressive forces are taken into account, I know it's hard wood but it doesn't mean it wont compress a little. Some of the parts of this chair work like vise. I would seriously think about adding steel plates on wood to prevent that compression.
I posted the idea of steel rope on the first or second video, even received a like from her and thought maybe she would do it. But for some reason, she keeps going back to weaker materials.
With steel wire in segments (lower part, bend 1, seat, bend 2 and back) then the flex will not distrubate the full length. Then add the belt on top for estistics
The problem is always going to be the tension needed to support the weight, anything withthatmuch tension will always caust it to be unsafe and unstable. Like sitting on the string of a longbow. The single connection from ene to ene will never work.
@@wirelesmike73 really the issue is one of leverage. The back acts as a lever and part of the block lower down becomes a fulcrum, creating something with tremendous mechanical advantage. So the forces in that convex part at the base are huge. I kind of suspect that if you solved the stretching problem, it would just crush the wood blocks.
@@wackyvorlon If the whole thing is strung together by one strap or cable or roap, the more weight or pressure put on it the more the ends are going to curl in on themselves. The flex caused by stretch in whatever is used does nothing but make that tension build up to the point of springing into the direction it wants to go. That is dangerous, because while that sprung tension may make it seem like it's strong and stable, once the stress limit is reached it creates an immediate failure that is compounded by the sudden release of all of that potential energy releasing all at once. Doing away with what is the equivalent of pointing a giant slingshot back at yourself will lead to the more stable solution. The only good reason to build up that kind of potential energy is if you want to move or prepell something automatically. That's not necessary for a roll up chair unless you want it to roll it's self back up on its own as soon as you stand up.
What if: Make the blocks themselves into a chain, and ditch the strap entirely. Cut each block into thirds along the direction the strap currently runs, then shift the centre pieces along by half a block and anchor so that each centre piece now connects the two neighbouring sections. Should fix both the tension and torsion issues if the fit is tight enough. Could be very pretty too with a different wood or just stain for the centre. For the floppy ends: a short nib of doweling in between each section that is just enough to grip those pieces prone to flopping but still easily popped out when rolling up. (Plus wont be as dangerous without the end point tension anyway) Loving your work! I think this is the first time I’ve ever been engaged enough to actually comment on a yt vid hah
I know it's easy to submit ideas when you haven't done the work yourself, but I have a suggestion. I love how you've eliminated a lot of the forces with the end to end stretch around the dowels, but the issue isn't the tension (in my opinion) but the stretch in the strap, and a lack of local clamping forces. If this makes sense; try a heavy duty thick lashing strap material. It's thicker and heavier which is a downside, but it has a much, much higher tensile strength and less stretch by length. I would also consider adding in a bar/dowel to make a matching concave/convex shape between the walnut slats and the maple blocks, so that the strap is held harder the more weight is added. You should test your materials for stretch before committing them to the design. If you weigh down one end and fix the other up high, you can physically measure the stretch before strapping it to your lovely chair, saving a lot of time and effort.
I have a few things that may help you if you decide to make another model of this chair: Firstly, you could probably use metal hinges (door hinges) where the segments connect instead of a strap as that wouldn't stretch. Secondly, a seatbelt may be a viable option for a strap, and they may be free if you get them from a scrap yard. Thirdly: Check out the video by Josh Wright in which he makes a similar chair. It holds weight, but doesn't look as good as yours. Here is a link to the video: th-cam.com/video/MuodNlDKce4/w-d-xo.html Also, this is an absolutely amazing video, and an excellent idea! Keep up the fantastic work - your projects are a great inspiration for us all! 😁👍If you do find a way to make it work, please sell some online if possible, since many of us (myself included) would love to purchase this.
Seatbelts do have high tensile strength but they are meant to stretch a bit under high forces so that crashes are safer. In a bad crash it's recommend to replace your seatbelt because they don't rebound from the stretch, but it usually doesn't get replaced because a lot of the time the car is totaled at that point
You could sew seatbelts together with a heat shrink layer in the middle. Stretching the seatbelts will cause heat, and the heat shrink layer will shrink to compensate. You could also use a braided steel cable (like ocean fishing line)
You’re right, that lower bend needs to be able to take a much higher load than the rest and with the straps it can transfer that stress anywhere along the chair. I bet hinges would work, isolating the stress, but you’d need some really heavy hinges for the stressed areas. Fantastic work, you’re so creative and always give me lots of ideas for my projects.
Before I read your comment, hinges were what came to mind. No stretch, and each joint is independent instead of coupled to the rest. There is some elegance to the straps that is lost though. Definitely changes the feel of the chair. This is a tough project! Things worth doing are often not easy. If they were easy, everyone would do them.
@@jaustinpage It would probably only need metal hinges at a few of the most bendy parts near the knees. And for the straps, I think something like braided steel cable is probably going to work better than any cloth/leather/nylon strap. You could run a leather strap over the top of load bearing steel cables for asthetics. There's a whole field of engineering and design called tensegrity, which is all about exploiting cables and tension for structural integrity, so it's definitely possible.
Loving this. I have two thoughts that hopefully help. 1. Having the strap/cable/whatever not be attached at several points means that any slack is 'sent' to the highest tension spots (the front) and will collapse any inverted sections (the back rest) 2. The front bend of the chair will always compress when sat on, so it needs to look 'wrong' and slope forward when not under load in order to settle into the correct shape when sat on. If the chair looks right before you sit on it, it's going to sink down and lean back when you do.
On 1: I had the same thought. However, realistically, for any chance of this to work and actually be stable enough, I think about the only option is to buy a metric F-ton of hardware and use strong metal hinges, possibly steel ones, between all the blocks. I think there's probably not any strapping material that would actually work. On 2: It might have been easier to create a roll-up couch table. In fact, maybe some of these failed chairs might still be salvageable towards that.
I agree with the second advise, but instead of " wrong" slope, in the first curve use 2 different types of angles on the wood, one inwards and one outwards, to counter each other and transfer the compressive and tensile strength accordingly without creating the gaps in between the wood blocks. That should ensure that the shape stays same more easyli when sat on and in standby position. As right now when sat on all the force and pressure is just hold by the straps and almost nothing is transmitted through wood because of the gaps in-between the blocks. So I think the straps would work, both kinds but with different blocks design for the first curve
Crazy Idea: have a channel inside the wood you can run the strapping in, then have a tensioner at the top or bottom. (Something like a come-along / ratchet strap). That way the strapping will always be pulling the wood sections directly together, and if it's not taught enough, just do a few more clicks on your tensionser. If you want to roll it up, release the tension and roll.
Physically thinking this would also take away a lot of force on the strap. Because the angle is not as far away, the force is not that strong, that pulls at the strap
The strap shouldn't be a structural element when the chair is upright. The strapping is flexible by design, and it isn't fair to the material to expect it to be flexible and elastic in one setting and rigid under tension in another. The big issue is the forces involved on the curve because the seat is a giant lever. The forces are not distributed equally along each part of the curve, instead there is a fulcrum at the middle part of the curve which has to withstand most of the forces placed on the lever. These forces are dynamic because the forces should increase as a person leans back in the chair. Increasing the internal friction between blocks might be a good idea even if it implies that they should interlock with a complex shape, use powerful magnets, or require a metal insert between blocks. There is an elegant solution here, but it will take a good amount of experimentation.
I hadn't seen this suggestion so ill throw it in. Along with the improved strapping methods many have suggested, designing the chair so that it, when weighted, rests in the horizontal seating position. Essentially design it so when you initially unroll it, the section you sit on is at a ~30* upward angle as opposed to horizontal. Then when its weighted it has room to flex to the desired seating position.
@@jhay440 ratchet strapping in general might be the solution, cause those straps, especially the 2t rated ones only stretch when you put as heap of stress and force on em
Correct. Also stronger straps - less stretchy. Kevlar or steel perhaps. Let them slip except at the ends where you have adjustment. The wood will compress so you will need to regularly tighten it. Take out all but two of the slats at the bottom
I love that you tried again. I think you were so so close! I should try making one before commenting more, but I think had you stapled/fixed the straps to each individual section like you did the first time you'd have nailed it. With that free-moving strap and the amazing amount of tension you managed to get on it, it seemed to create a bow-and-arrow like effect. I love to watch this youtube CHEST'ER make furniture. He doesn't sew anything and just staples the crap out of his pieces. If you ever feel the urge to try again, maybe cover 100% of the front with fabric stapled on, then cover that up with another layer of fabric to make it look nice (perhaps even leather as you mentioned). Regardless, these videos are the best and we all feel part of the struggle. It's really exciting, thank you for sharing and keep up the great work!
Just watched your second attempt. You mentioned leather a couple of times and I'm pretty sure it would not be a good choice. Think about a new pair of shoes and how they stretch with wear. Love the idea of cable mentioned below. Maybe adding something so that the pressure points can have more strength. Possibly second shorter piece of cable. Please don't give up, I'm sure there's a way.
I've been thinking on this too. my thoughts are to 'isolate' the tension between each pair's connection. currently you're allowing the tension to distribute and focus in the worst place, like under where your knee would be. there is created enough slack there so that ANY separation creates a lever action instead of a continuous form through the touching faces of the two parts. there is also the matter of keeping the lower circle from collapsing while keeping the app circle from exploding and what this means for how you stabilize them. I wonder if it might work better if you didn't glue the dowels in and also rounded the meeting edges of the sections. this would allow the segments to hinge around the dowels and always pass the forces through that line. also maybe use a steel thread cable - like a bicycle brake cable - and include a cam action on the end to tension the chair after unrolling
I think the simplest solution would be just give it a leg to stand on. Something similar to a foldable walking stick. Sure, it would need a lot of segments and it would not be stable on it's own.. but if the chair was stable on its own (just not able to bear the load), the leg like that could potentially hold, or at least help to hold, the load...
I really admire your persistence. Chairs are one of the most difficult items of furniture to make due to the forces applied. I think your instincts are correct that the current design has some inherent safety issues that may not be correctable simply by using alternative materials. Seems a design tweak is the best course of action. Good luck. Love your videos!
Love the honest attempts you make and how you show us what you learned along the way. And while the chair didn't come together, I wish you much success. Please keep up that positive spirit!
I'm also emotionally invested in this trilogy now. I don't think leather would cut it, as leather is also elastic. I would try metal hinges (flush with the wood) instead of a band of some sort. Metal hinges doesn't need to be tensioned and are strong. Also I would add one more bending segmented part so that there are support both left, right and in the middle of the chair (like three spines).
There's this saying that goes "If you can't make it perfect, make it adjustable" and I feel like that applies here. So maybe some kind of tensioning system would be in order?
Perhaps, the top of each block could be grooved, with the strap plates having a corresponding key. Similar to your dowel setup. The strap would effectively be clamped onto each block individually, the tension would not transfer through the length of the strap causing the blocks to collapse. It would be like a fabric hinge at each joint, being much stronger than staples or screws.
I love that you tried this again and you DEFINITELY did a great job working at the issues. I thought the dowels were brilliant. I feel like leather would stretch too much as well though. Maybe not the first time, but certainly after a bit of use. I have some straps for my kayak rack that have steel cable woven inside the strap. It is meant as a theft deterrent so that you cant cut them, but it would also provide tensile strength without requiring the same extreme tension to overcome the flex of the strap. Who knows, but an excellent project regardless.
Stapling the members is definitely on the right track. Right now the members are separating because the strapping is not fixed to the individual members so they're allowed to slide freely. To compound this, you get variable tension along the length of the strapping due to the geometry of the chair. Increasing the tension will just make it tighter everywhere, but it'll still be variable and create gaps. Stapling would constrain the members and create tension where you need it to keep the joints tight, but I am concerned that the tension forces may exceed the holding power of the staples so a more robust holding method may be needed. The other issue that I see is that even if you resolve the gaps, I think this solution either won't hold your weight, or will be very unstable without a brace. This is because you're trying to support a vertical load with a horizontal tension. This would require infinite tension so what happens is that as you sit in it, the angles change so the tension forces are more vertical. On top of this, the wood is designed for a fixed angle, but in reality the angles will vary with weight and movement. Due to leverage, I suspect the tension forces can be quite high, and possibly damage the straps and/or the fasteners for the straps. If you were to provide a brace that you can wedge in after it's unfolded, I think this would greatly reduce the stress that the tension members would see. The Ikea Poang chair is a very similar shape (that obviously doesn't fold up) made with steam bent plywood. You'll notice that the seat is at about a 30 degree angle. They obviously did this without a brace, but I think steam bent wood is a significantly better material for this chair shape (with the obvious flaw it doesn't fold up). Even then the chairs have a high failure rate due to the tension forces.
simple idea that might help: make separate strap dedicated only to the main curve, without the back and the bottom. That way, there would be no places for the strap to collapse the chair under tension. The main strap would be used to hold the chair together and would not experience that much tension
I'm really glad you stuck to this. We can't always see where we are going. Sometimes you just have to keep slamming your face in a wall until you find a way around the turn. My two cents are on having two different tension systems. One for the back, and one for the legs. In my humble opinion, it appears that the gap between the slats on the back and the structure may be your problem. The two parts of this, pretend and purely academic, tension system could potentially be joined with a different layer of structure on the top or bottom of the profile, or the inside of the alignment. The mechanical advantage of the legs on the tension side must be incredible. I'm just throwing it out there that you have a LOT of force. I think the leverage forces are 20:1... 30:1? If you weighed 100 lbs, I'd bet another 2 cents, that is a ton of force in total. Again, I'm totally speculating, and not reading other comments on purpose. You are kicking ass... There is nothing wrong with being ambitious as long as you can afford to be
If you feel like trying again, I faced a similar issue in the past and ended up going with a kevlar thread. After a very quick look, you can get kevlar webbing, from their specs it has a tensile breaking strength of 5500lb and more importantly an elongation break of 2.4%, I'm sure it can also be found in many different sizes and strengths
I thought of this too. Kevlar is almost impossible to break and even more resistant to stretching. I think Kevlar has the strength to sustain her weight without needing to be under too much tension
Great second attempt! Love the dowel idea to reduce rotation. If I put on my armchair engineer hat, thinking about the forces, this design is an uphill battle. Focusing on the point you’re seeing the most stretch, at its simplest you have a lever. The inner radius of the block being the pivot, a force of ~130lbs ~18” away (when sat in as designed), and the strap opposing it 2” from the pivot (the other side of the block). That ends up being a 9x force multiplier, turning 130lbs into 1,170lbs (or 5200kN) of force on the strap. (18” / 2” = 9x the force. 9 x 130lb = 1,170lbs) As for the stretch, I haven’t modeled it, but I suspect even a 1% stretch would significantly alter the shape of the design. Especially when it only has the two attachment points where the 1% of the whole length is able to be concentrated on one pivot like you saw, as opposed to the idea you have of attachment at each segment reducing the 1% to a single 4” segment of webbing. Based on what you experienced by adding tension, that too isn’t a solution because it adds a inherent spring that adds danger of collapse. So to both hold it’s shape unloaded, but also not collapse when sat on, the strapping needs to have almost 0 stretch. Difficult when applying 5kN of tensile force. So in my mind, in order to make this work (or at least give it its best shot) you need to reduce the leverage on the strap, and increase strap tensile strength. The first can be improved by using thicker blocks. A small change of going from a 2” thick block to a 3” thick block decreases the leverage by moving the pivot further away from the strap reducing forces from 9x to 6x and 780lbs. Going to a 4” thick block drops it to 585lbs (2600kN) pulling on the strap - still a lot. For aesthetics, you could even have varying thickness over the length since forces decrease towards the top and at the tail of the base. As for near zero stretch at these tensile loads, two ideas come to mind. Layering the webbing literally multiplies its capability. I haven’t found the specs of the viper webbing you used, but 1” tubular nylon webbing has a breaking force of 4,000 lbs, with stretch happening at MUCH lower forces. Layering 2, 3, or 4 layers with individual segment attachment, might reduce stretch enough in combination with thicker blocks - while still maintains rollability. The other no stretch idea is to use metal, metal strapping has almost twice the tensile strength of the strongest webbing. A relatively thin metal strapping might give you the strength required with almost no stretch, while still being able to somewhat roll up. You could possibly use it in conjunction with the webbing. This is a hard problem because of the amount of leverage exerted. It’s like having a 20” plank on a pivot point 2” from the end, a person standing on the long side, and a piece of webbing just 2” from the pivot trying to hold the person up. That’s big forces. But there’s a chance it could work with these (and other modifications I haven’t thought of).
@@3x3CustomTamar Good effort, and thanks for sharing even if it doesn’t work. I know people are asking you to keep going with this, if you don’t, that’s cool too. There’s other things you can spend time on that we’d also love to watch.
This is a really good explanation of the physics at work here and great suggestions for attempt 3. I like the idea of the varied width but that would be pretty complex maths to work out without a program that can do the maths for you 🤔
Either metal band like Josh Wright used or aramid fiber cord or strap (Kevlar & other brands) for ultra low stretch. If strap stretches 0.1" at the outside of a 2" block, & chair back is 15" from front curve that's 0.1"/2" = 5% deflection; .05 x 15" = 3/4" sag per block joint with that stretch. Check Josh's video, he's close to solving this. Is it a great chair at that point? Opinions may vary on that. Valiant effort
I love the tenacity on this project. In case you tackle this again: If steel cable was run through the middle of the blocks it may equalize the tension/compression and prevent the bottom end from wanting to curl back up. Add a locking cam for the final tensioning could make adjustments easier to.
No, a cable running through the middle wouldn't contribute to the load in any meaningful way. It would only be able to add an axial load between joints. In bending, all the load is experienced at the top surface and at the bottom surface of each joint. I actually disagree with the conclusion of the video that there was too much tension. It still looks to me that there isn't enough and that the webbing simply has too much stretch (which is what the weave is designed to do, no matter what the marketing says). I also suspect that the maple isn't supporting the compressive loads as rigidly as needed and it's deflecting too much as well. I think the solution is to oversize each angle by a few degrees to counteract the deflection that will be experienced during compression loads. More tension would probably also work, but at some point, the compressive displacement will become a significant factor too.
@@joranbooth5529 in the video, the end of the chair at the bottom curled back under because the tension from the strap being located on the bottom of the chair blocks pulled it back. If the tension was centered in the blocks it probably would not want to curl back since the tension would be equal between the faces. A cable would just be easier than a strap to run this way. Alternately, if the strap was pinned to not freely stretch between more than two blockers, it probably would not curl back under, but getting the tension right would be tougher.
Wood blocks are the actual structure, the strapping just work to keep joints together but remain flexible so it can be curled up to be put away, which is why it worked better when she actually glued it... So cable would work, as it would function the same and would be a stronger material than the strap... Personally, I would shape the blocks similar to folding table hinge so they lock together in the chair position and use a railing cable so it can be easily tension/adjusted...
If you ran the belt/cable through the center of the blocks, you would effectively half the leverage you get, which would only cause more stretch. It's better to isolate the belt connection for each joint by fixing it to the segments, so this "curling" effect doesn't occur.
@@CatNolara Nonsense, load on a curving structure doesn't work that way. The fabric was never handling the whole load by itself, just holding the blocks in place so they can handle most of the load, otherwise gluing it wouldn't have stabilized it as much as it did... Besides, steel cables are a lot more rigid and a massive amount stronger than fabric.
Maybe try building a proof of concept of just the main curve. Part of the stress is that you have disassemble and reassemble an entire chair everytime you want to adjust (even the relatively stable base. Maybe try breaking it into 3 separate problems and solving each one, then decide how to best connect those 3 parts.
Based on my older belts, I doubt leather would work either -- it does stretch eventually. I see that several folks have suggested steel cables, and that sounds like the best bet; they are used architecturally, and in tensegrity pieces (although I don't know how much weight most of those would support).
@@stanimir4197 yes, I realize that, but dyneema rope shouldn't stretch much more than steel cable. So there should be no advantage to using steel, except for aesthetic looks. I don't know about the belts, never used them. Anyway, I think this design is just impossible.
@@stanimir4197 can you point me to a source for that, AFAIK dyneema rope breaks at 3% deformation. I would be surprised if stretch at normal loads would be a magnitude bigger than steel
Tamar, instead of just the two points securing the webbing, you need to have every slat be a secure mount point for the webbing. (but not with staples). I would create complimentary grooves/grips between the back piece and the top piece to grip the webbing at each slat. The wood would act like a vice at each point preventing the stretch you are seeing without having to over-tension the webbing. But use at your own risk. No guarantees since the forces created are deceptively large. In fact, I don't think any wood is strong enough. It will either crack or crush at the fulcrum points. Good luck, though. I'm impressed with your tenacity.
Mitch, I totally agree! I am a chair builder by trade and I find these videos about trying to "save" a bad design by some kid on the interwebs rather cringe-worthy. 2 minutes with a tape measure, a calculator and a stress-strain table for nylon webbing will show that this concept is doomed from the start: the stuff is just way too flexible (and influenced in its properties by temperature and humidity) to allow this concept to work. What amazes me is that she doesn't learn. A lot of people have given very competent input but she's continuing without making use of the insights people are sharing here...
I love that you share your failures along with your successes. it's so motivating to see you go through the highs and lows and keep learning and trying new ideas!!! super excited to see the next version :)
Piano wire, routed through holes in each block, drilled near the "outer" surface of the bend in which that block resides (in other words the surface that is put into tension as a result of that bend). Also, place lugs at the top of the chair, around which the wire is wrapped, thatcan be tightened/loosened for assembly/roll-up respectively (similar to the lugs that are used to tune guitar strings). Just make sure fingers can't get in the way.
One thing i mentioned when you tried to sit in the chair at around 23:09 the reason it looses tension and collapses is not that the belt is too long, but because it is attatched to the backrest. Basically, as tension builds up in the lower curve of the chair, the string pulls the backrest towards the front. To solve this, you might consider using two(four) seperate belts instead of one: one to only keep the tension in the lower part of the chair, and a second one to support the backrest. This would definitively need further prototyping, but I hope this helps :) Anyways, great video and great effort, really enjoyed the journey, thanks!
Tamar: if you put 50Kg at 50cm from the frons joints and they are about 5cm thick you will have a tension of 500Kg in the strap! The rest of the chair could be done as you did but the front joints at least have to have the above ball park calculations in mind.
Your calculation is essentially correct but I think the tension on the straps will be much higher because - most people are heavier than 50 kg, some are MUCH heavier - 50 cm is on the low side, especially when leaning back Plus, the thinner the pieces of wood, the worse it gets. And let's not forget that the wood has to withstand a similar amount of pressure.
Pushing the materials this hard requires doing the math. I'm sure you can find an engineer to help you specify the right strapping. I also agree that it's important to connect the strap to each piece so that the ends don't decide to curl and release the tension.
I wonder about the distribution of forces vs the structural constraints for each section. I'm not a mech engineer, but I think some modeling would be a good idea. The collapsing sections seems to say that the strapping/tension needs to be segmented more robustly.
Exactly. And when you just pressfit the straps to the wood, the moment one strap begins to slip ever so slightly, the force on the neighbouring pieces increase. In the end you will always have a big force on the ends of the chair. I dont think leather will be better - on the contrary. Leather once was skin, and skin is very elastic - has to be. Those straps are made to be unelastic and they still were too elastic.
Agreed. It is just a simple calculation of leverage. Just pick a joint and consider all the parts between the load and the joint as fused. Then it is a large lever vs the 2in from the pivoting corner to the webbing. Thus putting a huge load on the webbing. Tamar, Every material is essentially a spring. You need a spring that doesn't stretch much to maintain the shape of the chair. Polymer webbing just doesn't have the stiffness you need in the sizes you are using. Leather would be far worse. Even if it is strong enough, it is the stiffness that needs to be high. Think of it like lifting a load with a rubber band and a string. Both will carry the load, but the rubber band will stretch much farther than the string does. There are two ways to increase that stiffness. Select a material that is stiffer (a higher modulus of elasticity) or make the spring bigger (increase the cross section). You might get enough stiffness by having the webbing span the entire chair width. I could run the numbers for you if you get me the exact material data. Otherwise I am thinking you will need to use spring steel strips or high strength steel wire (a small chance something like Amsteel rope would work). It will be a little tricky though as wood is relatively weak and we have to not exceed it's strength in crushing as well. In general making the supports the full width of the chair will make a large difference. The dowels would be better moved right to inside corner to pivot around. That would distribute the stress over a larger area instead of the small inside corners contacting. The thicker the chair the better as that reduces the load on the webbing dramatically. You already made this one thicker which was a good idea. With the current design you might have been closer if you used four or so layers of webbing. I am a mechanical engineer and can help you design one that works if you want to contact me.
@@3x3CustomTamar So it seems to me that because the webbing is only stopped from sliding by the dowels at the ends, everything in between can slide. Because of the leverage created by each weight bearing angle and the compounding of all of those angles, the webbing slides and pulls the ends back in on each other. Securing the webbing to each individual wood block (two flush mounted screws under the slats maybe? Or possibly just putting the screws that hold the slats through the webbing...) would solve for that sliding. Then, the most likely points of failure would be the holes made in the webbing. The webbing might be strong enough in its own to deal with that. Alternatively, they could be be reinforced with more box x stitches, which might also mean doubling the straps (admittedly more tedious, but in for a penny, in for a pound right?) I think that if the screws went inside the box on each side of the x, then that could (maybe..?) be the best way to reinforce it. Admittedly, there's a whole bunch of 'maybe' baked into this suggestion, but just some brainstorming from thoroughly invested fan.
@@alistairlacaille yep, I think you describe the problem. Plus, the whole thing should be assembled with some tension on the straps so they stretch less when in use.
@@3x3CustomTamar okay conpletely randomly came across this - 5:30 into the below video, Vousair Chair (or something, on mobile can’t check) - they leave it under tension and use a strap to keep it in chair form… th-cam.com/video/NtXK5ZgQ54w/w-d-xo.html
Great work! You’re so close! How about reducing the problem further and focusing on a single C section instead of the full S curve? A smaller prototype uses less material and time so you can iterate faster. You can also drop requirements and add them back later. For example, can you make a rigid chair out of keystone blocks or other segmented shapes? What improvements to its shape make it more stable? Once that works, then can you add linkages to make it bend in one or two segments? Then can you make it roll up and still work?
i love that you didn't quit even though you wanted you kept going until the final result. i was sitting at the edge of my chair rooting for you every time you made a change. either way keep up the great work, i love watching all your videos. even though i have not started wood working yet, you inspire me. Keep it up
I just want to let you know that while this project is probably your arch nemesis and it is taking a serious mental toll on you, as a viewer this is the kind of content we like to see. It is worth it, continue! I too have a project where I keep failing so I can relate 100% to this!
I think that "Bicycle chain" imbedded in a groove of each slat would solve two problems at the same time. It does not stretch, and it has side bending resistance.
@@PlanWithoutChance motor bike chain might work. It's all about the sheer force load that the lugs and plates can take. I don't know their rate loading at all. But it would certainly be flexible enough! Make the chair hella heavy to shift around though...
The strapping needs to be joined securely from piece-to-piece, not simply at the ends. Consider bicycle chain along a groove instead of material. It can be positively locked to each block. 🛠🔄
27:43 the solution could be using *door hinges* instead of elastic material. That way the individual pieces wouldn't influence each other and the backrest wouldn't curl up because of the seating area. Also, by adding some magnets, the pieces could be held in place when the chair is in it's chair shape.
You could even use a heavy-duty sash-chain or roller-chain (bicycle chain), and fit that into a recess cut into your wooden blocks. That'd probably provide the _least_ amount of stretch, while remaining flexible. Roller-chain is sort of an "Industrial Lego" and has it's own standardized sizing system. So, it can be easily had in sizes above mere bicycle chain.
I love how real your videos are. People tend to hide their struggle in videos, the way you go about finding solutions is really commendable. Love your work, you’re just amazing!!!
Hi, the toughness of your meshy material comes with the flex. You should use something not stretchy, like lifting slings or slings with steel wiring in it. Look for forklift stuff. Additionally, when you tension the straps make sure they're screwed down to every single piece, so they can't move along the whole length.
I’m gonna suggest looking into dyneema. It’s 4% stretch when under full load. I think one of the big issues you’re facing is that the straps stretch too much. Then maybe add a second strap along the back that you can clip into place to offset the front tension. A strap that prevents any blocks from folding over like the bottom block did.
Dyneema is a brand name for UHMwPE, so I wouldn't expect the results to be much different. Even if it adds more than just a bit of weight to the chair, I think either cable or hinges are the way to go on this one.
I feel like having 2 additional, shorter straps that focus on the seat, and then the lower bend section where most of the tension is generated would help to keep the tension localized where it is needed. That way instead of having 1 strap that is distributing the tension throughout the entire chair you would have a short strap around the lower bend that takes the majority of the force and localizes it to that section with an additional strap that stretches from the butt of the seat around to the middle of the base (right around center of gravity) that takes a little bit more of the force. That would leave the full-length strap with a third of the of the overall tension at most (and hopefully a lot less if she can get the other two to be more taut).
with a design like this more localized tension would definitely help, also I'm not sure if it applies here but with some materials and rope you need to pre stretch it to achieve its intended mechanical properties but sometimes that makes it more elastic.
metal hinges would be my first thought. would solve the side to side motion issue and might give more stiffness, which seems to be important for this chair since even a little flex can move the furthest part a noticeable distance.
you want the tension to be focused on each joint individually, that way there is no chance of it pulling the bottom up or collapsing the back, the straps sliping through each wooden brace compounded the tolerances and also pulled from other areas. not all the forces are equal, so more tension is required for the bench bend, in comparison to the straight sections
I love this channel so much! Thank you for making all of us that work with wood feel better about mistakes. Metal cable routed into the pieces instead of the straps could be helpful and possibly hidden? Either way though, thank you for the emotional rollercoaster that was these two pieces of content 😊
I was thinking the same thing. Consider using something like bicycle break cables that are sent through drilled holes. You could do two per side which would also help with the side to side motion. The nice thing about metal cables is you can easily adjust them.
Use door hinges to hold the pieces together you’d need one for each joint, they swing in the motion you want whilst being sturdy enough not to break under tension (you’d need to cut a channel in each piece to compensate for the width of the pin on the hinges) but I don’t see why it wouldn’t work
I thought about this when first seing the design in the first video: One of the issues is just that you're dealing with huge forces: If you think about the lever of sitting in the center of the seat part with let's say 60kg (all just a back of the envelope calculation), you have maybe a lever of e.g. 1:10 at the front, which means the straps must hold about 600kg (and so the very back edges of the wooden blocks that are under compression). Furthermore this lever means that a tiny gap in these joints will massively deform the rest of the chair. And as there is virtually no material with 0% stretch (or deformation under pressure) so I doubt that this will work no matter what materials are used short of wire cables and steel blocks. In my opinion it is just a flawed design that looks cool in the rendering but no practical considerations went into it. But I really like how you show your process and all the thoughts you put into your experiments - I learn something new every time! And I'm always looking forward to your next projects!
My quick calculation is similar- 300kg per strap. That means 300kg compression on the wood blocks which is very high for wood compressed perpendicular to grain.
More armchair engineering: I think the key is reducing that 1:10 lever. Your weight offset from the front of the chair creates a moment (force x distance) which has to be counteracted by an equal moment from the strap to the inner edge of the blocks. The longer that distance (height of the blocks), the less force it has to exert to get the same moment. Also, you should attach the strap to each piece rather than just the ends as you mentioned. Stretch (strain) is a percentage of distance, so the longer the distance between strap supports the more the stretch that can accumulate at your stress points and deform your geometry.
Yeah that's what I thought huge forces and minimum tolerance for stretch... I have no clue how to calculate that (finite element analysis?) but the forces involved may call for high tech materials... Carbon fiber or something like that with extreme tensile strength without stretch...
As a thought experiment, imagine replacing the webbing with recessed metal hinges at the junction of each block to its neighbor. This arrangement of rigidity and a tight turning radius are what you’re trying to replicate. I know it would violate your keen aesthetic to use hinges; but any material flexible enough to act as a hinge will likely also tend to stretch. There might be a solution using a tensioner to adjust out any gaps that develop. Or maybe a hybrid approach that places metal hinges only on the corners that are stress points. Not sure what the answer is but thought that might help you visualize the problem.
I love your perseverance and I learn a lot from you, thank you. I would like to make a suggestion about your strapping, I use something similar to tie down loads and a fresh strap has a lot of give, so just to limit your variables I suggest you pre stretch your strapping.
Hi Tamar, just wanted to make a suggestion. I think adding the dowel between the pieces is creating a leverage point. Instead of 2 flats coming together and being compressed and forced into each other, the dowel is creating a round edge that the pieces can pivot around and in turn increasing the stress on the webbing/straping. Hope this makes sense. Really enjoying your channel. Thanks.
You need a solid connection btw the strapping and the wood at each section (glue, a mess of fasteners). That way, the straps act as individual hinges rather than allowing all the stretch to accumulate in one or two joints. Strain is measured in inches per inch. Shorter initial length results in shorter overall for a given tension.
Yes, thats what I was going to do after figuring out the correct strap length. but it just kept stretching... then I got hurt. so decided it was wise not to continue with this iteration.
Thank you so much for showing the bits that don't work out. I feel your frustration but it helps me so much when I screw up not to beat myself up too much. Your videos are fantastic.
Hey. This is really cool. Back in college we took a jab at a very similar project, pretty much the same. The way you used dowels was very smart. And architect friend of ours mentioned maybe we could CNC each piece, and I think that would also transfer the tension to the wood instead of the strap, since we couldn't get the straps materials to work either. Id love to see a calculation on just how much force that material would need to resist and still be flexible. Good luck.
Came here to say this. The problem is that the tension required to support weight through that structure is enormous. You currently have the blocks in compression and the strap in tension. Solution is either move the tension into the blocks or use a material that is better suited to the task for tension. Leather is not going to do it I’m afraid. Consider tank straps? Very thick, not so bendy, but definitely overkill and should do the job. Alternatively, use some steel cables, and hide them under the straps you currently have? That way it will be aesthetically the same, but the strap will serve as a lid and the cables will carry the tension. Good luck! And thank you for bringing us along for the ride!
I agree, throughout this I've had the feeling that the tension should be borne through the wood not the strapping. Think maybe go back to that clever computer design programme and redesign the shape, thickness and depth of the sections. Also maybe a scale model would be helpful, remember that's how the artisans used to bring forth their creations. The perseverance involved in this project has been admirable, (if distressing feeling the pain) to watch. Sincerely hope the conundrum is solved.
Seat belts can withstand a lot of force and still be flexible. That's exactly what they're made for and it's exactly what I would have tried. And seat belts don't stretch. So, you hopefully will cut down on stretching by using it, also. You can buy it in bulk/rolls to use for whatever you need.
Try using dyneema, I use it for sailing. It's stronger than steel cable, a hell of a lot lighter and is fantastic because it's pre-stretched, and wont stretch more. Can also be spliced and tightened to create tension if needed
I was thinking along the same lines ie too much stretch. My thought was flat woven metal strap like is used for ground straps but I don't know if that would be a good look. I like the chain idea better. Another possibility would be stainless steel cable with a threaded piece on the end. The threads would allow tensioning.
my thought too! With all this tension I feel like wood and plastic alone will not be sufficient, you need some metal, maybe even try a experimenting with some sort of hinge between each piece.
it may also help to drive a screw through the chain at particular points to isolate tension from collapsing other parts of the chair and limiting the amount of stretch available at any particular joint. Thicker blocks of wood in the front section would reduce the how much chain stretch would affect the overall shape.
Since it’s bending so far when assembled, could you design it “wrong” (leaning more forward) so that when you sit, it stretches to the preferred design? Not sure if it would work, but just a thought I had while watching.
I've watched two successful implementations of this chair on YT if you're interested in checking those out. Z-Shaped pieces seem to be the key. Pallet straps might be a better band
Adjusting the shape of the segments definitely seems like a much more elegant solution than chasing down the issue of the strapping. Tamar does have a CNC, and while using it may not make for quite as fun of a woodworking video, it would allow the production of unique shapes that do the work of holding all the angles in place.
Yes, I'll simply toss my 2 cents here as well. I'm not a carpenter but IMHO the problem between your chair and the ones that work are that the other chairs pieces lock into place (with gravity) when they are unfolded. The strap connecting them simply holds the pieces together for when it is rolled up. Yours doesn't lock in place at all. The strap is what's holding it up and gravity (plus weight) is causing it to fail. If you redesign the wooden pieces so that they lock when it is unfolded and gravity (plus weight) reinforces the structure, then you'll have conquered this project as well.
I thought the stabilizing dowels were a good idea. Did that part of the experiment work? My suggestion for the next attempt would be to test the stretchiness of various strap materials before building anything. My instinct says that leather will stretch even more than nylon straps.
I saw another builder do something with an overlap joint in the wood. I am thinking maybe a ship lap type overlap, but with a back angle on the edges so that it sits in a pocket once unrolled and puts some of the tension force onto wood. Hard to describe but i think there has to be a way so that the strap isn't taking on all the forces, the wood has to interlock some how. Kind of like a two layer french cleat. Beautiful choice in wood, i like the asthetic of how this design looks.
You need to pre-stretch the strapping. A forklift helps with an anchor point on the ground. I have this same issue when installing Dynema cable in cranes. I use the hydraulic force of the winch cylinder to stretch it out during installation but every once in a while I have to return to the crane and redo the splice and sewing.
This is also the case for Dyneema sailboat rigging. They even sell pre-stretched (annealed) Dyneema for this reason. 'Course you can always get the cheaper stuff, tie one end to a tree and crank away on the other end with a winch.
Could try a "locking" design one piece locks into the next when unfolded and put the strap all the way around the ends with tension keeping them back a bit and leather will not work it will tear And dry rot
A very interesting and challenging project. I appreciate that you have shared your efforts to address each problem as you worked through the design and material issues. Thank you!
What if instead of a chair, you had the same rolling idea, but without the back to support, well, your back. What if instead when you assemble it, it resembles an empty cylinder or, a log thats empty. Then you don't have to worry about the back being strong enough to support someone's weight.
Looking at the design, shouldnt the strap "switch" sides somewhere around the seat area and be on the outside of the bend? Because one continuous strap will be tensed by the lower part and put "curling" force on the top part making it collapse.
the side with the strap is the side everything curls into, and as you can see by the chair shape, all points on the chair expect the same curling direction for this design to hold.
I completely agree, but it defeats her goal of making it roll up. The switching sides would definitely make it more structurally sound but I guess it's a toss-up between two goals for her project
Thank you for sharing such great project!! Although it didn’t ended up ideally, it’s still a beautiful woodworking furniture! I got a suggestion, I think connecting each blocks with something like door hinges might help with the strength and reduce some flexibility.
It's super frustrating when you try, try and try again feeling like you're tantalisingly close to making something. Big respect for sticking at it when it got so frustrating !
How about those Metal Straps, that are used to secure heavy equipment on pallets? Those won't have any stretch, roll up without an issue and be strong enough to not snap!
@@WhoAmi2357 I was thinking hinges, as they are rigid, and would still allow the rolling, but I feel that kinda defeats the purpose of having a single continuous strap running the whole length
@@cezarcatalin1406 I commented on this separately, but I was thinking steel banding strap used in shipping on pallets. It could be used just like plastic strapping, but would have very little stretch and a lot of strength.
You could always just make a strap between each piece instead of throughout the chair. Then there would only be the stretch between each individual piece to contend with.
What about larger cable segments? I.e. a cable going from bottom end to the beginning of the load bearing curve, a cable for the curve itself, and then a third cable for the back pieces
@@kfurgie999 That would also be exactly the same thing, all else being equal. Maybe you want to adjust the strength of the strap segments depending on their load. That would make sense and can be achieved with using a different material or just more material (wider strap or multiple straps).
@ the difference is that in this design the tension transfers between them because the straps can actually slide beneath the planks holding them onto the blocks. this is also what the staples she mentioned in the video are about: fixing the straps so they do behave like many tiny straps
OK, love this! Tried hard and did not succeed... very few woodworkers show this, but it is part of every woodworker's life... Absolute A+ for effort. I learned alot watching this...
Motorcycle chains! The wood already has some flex, so you don't want strapping that stretches, too. Motorcycle chain would also resist torsion, providing additional stability.
And more weight... I think at some point, the idea of this "roll up" thing, is that it can be portable. Too much weight would affect that portability. Don't you think?
![SAFEPLANET - ดาวเคราะห์ [ THE PLANET ]](http://i.ytimg.com/vi/4X1XL3I_k3c/mqdefault.jpg)
Between this chair, the guy making transparent wood, and the absolute state of the marble machine x project, I've become quite invested in the creative struggle as seen on youtube, in a way I hadn't experienced since the french guy decided to turn his cooking channel into a step by step guide on how not to make a pasta machine.
You need to check out Mattias Krantz, in that case.
We learned a lesson at the end of that one though, if you want a pasta machine done right, mail it to This Old Tony.
Can I get a link to that pasta machine one?
This comment is gold. I'd like to add the channel "Stuff Made Here" if you're interested in these sort of creators!
In addition to the comment on Matthias Krantz, just don't watch his most recent video to start out. So many problems and he didn't even include them all in the video, he might still have them in his discussion board though
I'm emotionally invested in this trilogy now
Hahaha me too
@@3x3CustomTamar Really Great try! IF (BIG IF) you attempt to make it again, how about cable through holes drilled in the angled pieces? You could tighten it as necessary with a turnbuckle in the bottom piece. You don’t need to secure the slats to the fabric. Secure the slats permanently to the short angle pieces, not to the fabric or cable. Anyway, respect your trying to solve this physics build !
i shouldn't have scrolled down while watching, as this comment (it being a trilogy) spoiled it for me.
@@3x3CustomTamar A guy named Josh Wright found a way to make it work, maybe you could try to take inspo from his idea?
Same!
Now, I suck at maths, however, this seems like a situation that might benefit from some:
The entire top half of the chair is a lever, with a fulcrum around the corner of the middle joints, with the other side of the lever being the face opposite the corner that's acting as a fulcrum and is pulling on the strap. For this chair to work, the force of the lever where the face meets the strap has to be lower than the elastic deformation of the strap itself. Assuming you weigh 60KG, that you sit half a meter from the fulcrum, and that the distance from the fulcrum to the strap is 4.5CM, the force at the strap is approximately 6,500 newtons. The Venom strap advertises a mere 1,800 pounds of *breaking* strength, which is just 8,000 newtons (there are two straps so you probably won't break the strap, but clearly we are exceeding its elastic deformation point). So no matter how much pre-tension you put on this system it's never going to work (as you surmised).
Either a redesign is needed or material must be used that won't deform significantly from these forces, or even twice these forces (since if you sat down quickly you'd have momentum, and you can lean back in the chair). Basically, stapling the straps will not work.
Granted, I may have made some mistakes in there or oversimplified the situation in some way. It's been a heckin' long time since mechanics A levels for me.
Anyway, I love this idea, and I really hope you find a way to make it work. The fact that you kept trying to make it work after 6 attempts is inspiring!
i'm not nearly as educated on topic as I'm only on second year of architecture studies but based on inspecting the Miesrolo chair i think you're absolutely right. it can be clearly seen that the straps used are much stronger even tho the leaver is shorter so it has potentially lower stress. they appear to be much wider and made of non elastic jeans-like material and glued in multiple layers so it's so much stronger
"I suck at math" *proceeds to discuss several physics principles and concepts, all requiring math*
Don't sell yourself short man, you clearly have a good handle on things. I agree with you, the geometry of the chair is working against her since her body has so much leverage against those straps. The material needs to be flexible but have an incredibly high tensile strength if you want to keep the strap width and thickness reasonable. I would say that getting rid of the straps all together and using pinned joints that have their movement limited by the geometry of the chair segments could work but then the back support wouldn't stay in tension. It's a very tricky problem...
I disagree that these straps won't work. Your math is correct, however the 1800 pound breaking strength is for 1" straps, but it looks like she is using 2" straps which are rated for double the breaking strength. And like you mentioned, she has 2 straps for a total of 32000 newtons of breaking strength. With the 6500 newtons of load, I don't think we would be at the point of plastic (permanent) deformation, even with double load.
The issue I see that I outlined in a different comment is the amount of strap that is allowed to stretch. The entire length of strap can stretch in the current design, and can be allowed to show itself in only one joint. If she were to attach the strap to the blocks by staples, or even better - glue, the stretchable length of the strap will be greatly reduced (the glued sections of strap will not stretch any appreciable amount).
suck at maths huh?
@@nickdibart the existing straps may still work, but surely you can agree that a new material with a higher plastic deformation point would benefit the design as a whole.
HEADS UP!: **Leather is super stretchy.** -- Climbing shoes made from leather usually stretch much more than their synthetic counter parts. Cow hide is just thick human skin which is supple and dynamic.
Slackline and highline can have as little as 2.5% stretch. Maybe there are industrial materials that are even more static than this but inherently there will be some stretch with woven materials.
Stainless cables will have far less stretch. Maybe make an expanding end point so as you are setting up the chair, the cables can be stretched with mechanical force (a lead screw pulling both cables taught). I get the appeal of using wood and cloth to make an impossible chair but maybe that is impossible.
this, steel cable or chain running through holes in the pieces is what will get her there. anything that will stretch will end up buckling like this.
@@kingkarlito Maybe. Because of the joint solely relies on friction, there is a fair amount of tensile force needed from the cables.
What might work better is skipping on the rolling portion of the chair and just make it a folding chair with some fancy origami tricks (aluminium sheet + custom hinges/piano hinges will optimise strength and the project's mass)
@@dragade101 The rolling chair idea is still Really good for portability and storage, but yes she needs a material with basically zero tensile give and enough strength to hold. Small steal cables running the length of the strap fabric is probably the right direction.
It seems to me that this sort of thing is really where a mechanical engineering degree comes in handy to make the right material choices and to calculate the structural aspects of the design. I *think* it could in principle work but I but that's assuming ideal (ie theoretical) materials. Maybe there are materials that are close enough but my suspicion is that all the structural elements will end up having to be metal instead of wood/textile. All the rigidity comes from tension and the lack of stretch/compression of the materials. Kind of like how we pair concrete with steel rebar. Steel has low stretch for a given tension, and concrete doesn't compress.
@@simon7719 kind of. Sometimes its best to just experiment and iterate several times. You could try to calculate the needed forces but why not just over build and test your design! There likely is a tension that would these blocks safely and it. Maybe what would help is that each block has some registration keys, especially for the back pieces. That once the blocks are drawn together and held fast, they cannot hinge away from each other (that simply they could be parallelograms with a keyed surface to help them not slide off of each other).
The back of the chair is probably the hardest area as you are not loading it the same as the front arc is being loaded. Finding the right segment shape is probably the key here.
The issue may be with the design itself, specifically the back support. You do NOT want the tension evenly distributed, as the forces are not equal. The back has barely any forces acting on it compared to the bottom curve, so when you increase the tension, all that happens is that the back (where the webbing runs at the inside of the curve) is curved inwards - and since the webbing is not blocked by the wooden blocks like it is at the large bottom curve, it just releases from the surface of the maple until it is obstructed by the walnut. Thus, any additional tension will just continue to force the back curve inwards, until it collapses and releases the tension of the webbing like it does at 23:08 - notice how the entire chair drops at bit.
I understand the benefit of leaving the segments to slide around freely, but I think you have to at the very least attach the webbing at the base of the back support, before it curves up. That way you can tighten the bottom part of the webbing A LOT indepentently from the back, where you would want a lot less tension. That way, the base of the chair would even keep its shape if the back support collapsed.
You have a good point. The strapping needs to be bonded at each join to prevent the highly loaded areas from transferring that tension to the lesser loaded areas and altering the shape. However the webbing at the thickness and width used is just not stiff enough for the loads. I think a groove and a mostly matching rib on the back of the slats would pinch the webbing enough to isolate each joint. With that and a few extra layers of webbing and it may work. I would need the material data to be more sure.
@@court2379 Very good points here. I would also look into much wider strapping, or even using something like canvas or similar material that would completely run from one side to the other. This would spread the tension over a much greater area and hopefully reduce the tendency of the material to stretch. The use of a thin steel cable would also reduce the stretching issue.
The static designs I've seen on the Web use some method of making the sides effectively thicker in the front-back direction in order to increase stiffness. One method is to form a triangular truss structure with a third side between the original two. That can be done in a number of ways. Another method is chords, that is, some kind of cable connecting distant points on each of the two sides.
And absolutely, there are at least three sections in the chair, with different forces acting on each, and each section and their forces should be analyzed and addressed accordingly in the design.
@@court2379 I don’t think you need to bond the strapping at every point, but for every change in the load. Especially the forward bend and the chair surface need to have 2 independent belts to distribute the weight loads on the structure.
the point is longger the webbing, the longger webbing stretching.
Having a quick scroll down the comments, I didn't see this suggestion so thought I'd share. Why not do away with the strapping entirely and make a mechanical hinge connection between each block much like a metal watch band?
Came here to say this, I think with the current design the strapping will always be a failure point as it will stretch over the entire surface of the strap. A mechanical hinge at each point will eliminate the stretching over the 5ish foot length of the design.
I think she’s on the right track though! The dowel alignment to avoid twisting is frankly brilliant. Don’t give up!!!
Yep, think that will do it!
i think you just need make the blocks take more stress with a box joint and that would be sufficient
i do think a mechanical connection would look better
The straps can still be used, but they need to be fixed to each piece of wood. The current design allows the elasticity of the whole strap to come into play - even if it only stretches by 2%, over the 1m length that's 4cm which can open up at any joint. If the strap is constrained then the same 2% elasticity becomes insignificant on each joint.
I think it is the positioning of the tension that is causing the issue. Use wire running through holes in the CENTRE of each block (this will stop it folding over. Make some kind of screw adjustment to allow the tension to be adjusted. Fantastic effort btw, admire your perseverance
Just going to add some thoughts from an engineering perspective:
You talked about how you selected your strap materials based on them being advertised as strong and lightweight, but strength was never your limiting factor. Deformation (i.e. stretching) was, and that's a function of stiffness. Stiffness is literally defined as the ratio between an applied force/stress and the resulting deformation/strain, and while strength and stiffness are of course highly correlated, it's certainly not universal that a strong material will be stiff nor that a stiff material will be strong. More to the point, plastics, even if they're strong, generally do not have high stiffness when compared to metals, ceramics, or other construction materials (which is why we don't have plastic houses or bridges; they are similarly constrained by limitations on deflection as much as strength).
With that in mind, the hinge idea others have suggested is probably a good one, as metal will be a stiffer material to begin with, and the hinge will also be a bulkier, thicker, and thus stiffer interface than any strap could be. One other idea would be to look into unidirectional carbon fiber , which can be bought in "tape" widths that come on spools and could be swapped out more or less directly for the plastic strapping you've tried. Carbon fiber is one of the stiffest materials in the world in tension (significantly stiffer than even steel), yet without the epoxy reinforcement that its usually coupled with, bare carbon cloth is nearly as pliable out-of-plane as any other webbing would be. Sandwiching a few layers of unidirectional carbon together and leaving them bare would almost certainly be your best bet to make the strap-based design work. As a bonus, it's even black, so it'll match the look you were initially going for! Just be careful and do some research before buying to ensure you're getting real carbon cloth rather than a carbon-fiber "look" in some cheaper material, and that it's not preimpregnated with epoxy.
I would agree but don't think you need to go to the expense of Carbon Fiber when the metal banding strap should provide enough tension without the stretching. Just ensure that you are tight when first applied, maybe even over tight and let it stretch a little and it should work just fine ....you my need a couple of straps to equal the width of your current banding stretchable material...at least it would not hurt to try it.
As another engineer, I had the same thought. The issue here is the stress/strain curves for the materials selected for the tension component. Nordic_Nerd is right on that basically no plastic will have little enough strain to work, and leather is going to actually be worse in that regard. I think the only feasible options for the loads you want (holding up a whole adult with essentially just tension over about 2 meters with less than a cm of strain) are going to be metal or carbon fiber. The CF route will look cool, but the material is expensive, and be difficult to work with, and you also need to figure out how you're going to protect it from outside loads since it's generally quite fragile in the non axial tension direction. I think the best bet is actually to just run ~1/4" steel cable down the inside of the wooden blocks as the load bearing line, and then you can look at maybe just adding aesthetic strapping over top (imo it would look pretty cool just as stained wood).
She could use dyneema no? I use it when climbing for strong anchors that don't need to stretch cause it basically doesn't, its very strong.
Could pre-stretching the poly or nylon strap for an extended period of time solve the problem?
Could inlaying metal to reduce the overall amount of wood (a more compressible material) reduce the amount of excess tension that causes the chair to roll up while in use?
Would using shorter lengths of whatever material and dividing the chair into sections i.e. base, leg, seat, and back limit the stretch in any one section and thus the overall tension as well?
I think that running a steel cable internally (maybe through sleeves to prevent it tearing through the edges) might be the best bet. If so I'd look at complete cable systems for stairs and railings with tensioners that are provided.
@@Blink1826000 i was just about to comment about big steel cable (mechanical engineer student)
My daughter's boyfriend build a chair like this and had the same issue you were having. We solved it by using a chain similar to a bike chain in place of the straps (looked closer to mini tank treads so that the screws could anchor through it. He also used hardwood as the pine he originally used deformed slightly under compression.
Was going to recommend the same, not sure what the stretch would be for a bike chain, but I'd guess similar or less than steel wire (which is about a tenth of low stretch webbing). Personally I'd put it in a chanel under the webbing, then you won't see it when the chairs up and it might stop crumbs and dust falling in it.
Fixing it at various points would also limit the stretch in specific areas. Currently the spare material from the back and base can spread through the chair to make gaps, if the material is fixed either side the stretch between each joint is limited to a percentage of the ~15mm between the slats, rather than the stretch for the whole chair being a percentage of the ~3000mm of webbing in the chair. This might cause problems with rolling it up if the support material isn't inline with the corners it hinges on, but that should be a simpler problem (e.g. rounding the corners slightly).
Yeah, I was thinking about some sort of steel mesh. not sure it would roll up very well.
Think that is a much better idea than leather.. leather stretches a lot. Leather would be terrible.
The beauty of your channel, above and beyond the fact that you are an amazing self taught woodworker, is that you allow us all to see your vulnerability with the highs and lows of each project. Anyone who has spent any time in the shop, whether they want to admit it to themselves or not, has been down the same path. Thank you for allowing us to go on this journey with you!!
It’s all part of the process. The moment of the big win is all worth it 👍
@@3x3CustomTamar what John said above is by FAR one of the reasons I love your channel (pretty sure I've said this before). It goes a LONG way to making the rest of us see our failures as a lesson learned. Please do not stop with that format of videos, it's very refreshing and makes the rest of us feel like normal humans who make mistakes in the shop too.
This build is like nature trying to build the human spine.
"I have way more time invested in this than I thought I would,"
* Looks at scoliosis* Its still not safe to sit in lol
And just like the spine, the bones are shaped uniquely to support the bending and form a locking mechanism so it won't be over bent like the chair...
@@abysmalvoid9136 mortal kombat begs to differ..
Unlike her nature said “eh, good enough. They can have bipedalism or a good spine not both”
Maybe metal hinges or steel cable, the advantage of the cable is that you can make the tension adjustable.
The hinges (in place of the strap) sound great.
I came here to recommend replacing the webbing with cable as well.
I was thinking about metal hinges instead of strapping too, though finding ones that are strong enough to support an adult while being small enough for the design might be an issue.
Hinges feel like the right tool for this.
If a 200lb person sits on that 18 inch diameter, the max moment in the chair is about 150 ft-lbs. For two straps , that's about 75 ft-lbs each. If the blocks are 2 inches thick, the strap tension needs to be about 900 lbs to support that weight. The design on the strap is not about strength, but stiffness. Nylon's modulus of elasticity means that even if it does not break, it will stretch too much. If the user "bounces", the loads can easily double or more.
Steel strapping (like used for attaching heavy cargo to pallets), or hinges could provide the required strength, with minimal elongation under the loads here. It would also make the chair much safer.
This is so much work and I think you are almost there!
1: The nylon strapping is going to stretch. A non stretchy material like Kevlar or Stainless steel wire would be better - frankly SS wire would probably be the easiest.
2: The various blocks are snapping back when you sit on the chair because the lines of force are acting around a fulcrum made by the block corners. This is always going to be a risk. To get around this I would drill holes in the middle of the blocks (Through where the beautiful dowles are) and run stainless steel wires down the middle.
3: With this solution of course you wont be able to just roll the chair up because when the SS wire is tight, rolling the chair will require the wire to stretch, which it wont.
4: To get around this you will need the ability to release the SS wire when you want to roll the chair. This can be done with Quick Release type levers which can also allow ajustment of the wire tension as the wood settles.
5: Even with this I am unclear due to lack of experience with wood, if the wood itself will have the necesarry compressive strength to resist the very high forces that its subjected to.
6: I would try this myself but I dont have a wood shop :-)
7: I love your videos, please keep up the good work!
Great respect for your tenacity.
"The master has failed more times than the beginner has even tried"
-Stephen McCranie.
Amen. There are so many videos on TH-cam of people being successful. It’s really inspiring to see someone fail and keep at it. You inspire me. Thanks for sharing.
Leather will,stretch as much, or more than webbing. I love how you’re trying to solve all the problems and wish I had a useful suggestion to get it to work
Yep, leather would be WAAAYYYYY worst, even if taken from one of the less stretchy body part.
Look into kevlar reinforced belting, like motorcycle drive belts. They are VERY strong and have like ZERO stretch to them since the have to work under enormous tension without slipping. Might be a bit pricey tho. Not sure. Maybe find a Harley dealer/shop and see if they have any old belts they would give you.
My first thought also was: Well.. if tension-flex is the problem.. maybe going for something that doesn't flex? I am 100% not smarter than anyone who already saw and commented the previous videos and especially not smarter than YOU xD - but I had seatbelt in my mind. But probably I am just naive and it flexes as strong as any of the already tested materials.
Also I can't explain from an engineering perspective, but I would fix/screw the belt to the blocks at least where the sitting-area is. That would take tension from the upper/back part, where it isn't really needed/helpfull.
Love your enthusiasm with this project
@@goomStar I think the strap she used has less flex than the seat belts.
I would personally go with carbon fibre but without any resins (you just need tensile strength) and localize mounting in a way that focuses on stress points. In order to prevent that thing from collapsing you need material that just don't stretch. I'm also not really sure if compressive forces are taken into account, I know it's hard wood but it doesn't mean it wont compress a little. Some of the parts of this chair work like vise. I would seriously think about adding steel plates on wood to prevent that compression.
It’s going to be really hard, the forces involved are considerable.
Steel strapping is probably closer to what you’d need.
I posted the idea of steel rope on the first or second video, even received a like from her and thought maybe she would do it.
But for some reason, she keeps going back to weaker materials.
With steel wire in segments (lower part, bend 1, seat, bend 2 and back) then the flex will not distrubate the full length.
Then add the belt on top for estistics
The problem is always going to be the tension needed to support the weight, anything withthatmuch tension will always caust it to be unsafe and unstable. Like sitting on the string of a longbow. The single connection from ene to ene will never work.
@@wirelesmike73 really the issue is one of leverage. The back acts as a lever and part of the block lower down becomes a fulcrum, creating something with tremendous mechanical advantage. So the forces in that convex part at the base are huge. I kind of suspect that if you solved the stretching problem, it would just crush the wood blocks.
@@wackyvorlon If the whole thing is strung together by one strap or cable or roap, the more weight or pressure put on it the more the ends are going to curl in on themselves. The flex caused by stretch in whatever is used does nothing but make that tension build up to the point of springing into the direction it wants to go. That is dangerous, because while that sprung tension may make it seem like it's strong and stable, once the stress limit is reached it creates an immediate failure that is compounded by the sudden release of all of that potential energy releasing all at once.
Doing away with what is the equivalent of pointing a giant slingshot back at yourself will lead to the more stable solution. The only good reason to build up that kind of potential energy is if you want to move or prepell something automatically. That's not necessary for a roll up chair unless you want it to roll it's self back up on its own as soon as you stand up.
What if: Make the blocks themselves into a chain, and ditch the strap entirely.
Cut each block into thirds along the direction the strap currently runs, then shift the centre pieces along by half a block and anchor so that each centre piece now connects the two neighbouring sections.
Should fix both the tension and torsion issues if the fit is tight enough.
Could be very pretty too with a different wood or just stain for the centre.
For the floppy ends: a short nib of doweling in between each section that is just enough to grip those pieces prone to flopping but still easily popped out when rolling up. (Plus wont be as dangerous without the end point tension anyway)
Loving your work! I think this is the first time I’ve ever been engaged enough to actually comment on a yt vid hah
I know it's easy to submit ideas when you haven't done the work yourself, but I have a suggestion.
I love how you've eliminated a lot of the forces with the end to end stretch around the dowels, but the issue isn't the tension (in my opinion) but the stretch in the strap, and a lack of local clamping forces.
If this makes sense; try a heavy duty thick lashing strap material. It's thicker and heavier which is a downside, but it has a much, much higher tensile strength and less stretch by length. I would also consider adding in a bar/dowel to make a matching concave/convex shape between the walnut slats and the maple blocks, so that the strap is held harder the more weight is added.
You should test your materials for stretch before committing them to the design. If you weigh down one end and fix the other up high, you can physically measure the stretch before strapping it to your lovely chair, saving a lot of time and effort.
I have a few things that may help you if you decide to make another model of this chair:
Firstly, you could probably use metal hinges (door hinges) where the segments connect instead of a strap as that wouldn't stretch.
Secondly, a seatbelt may be a viable option for a strap, and they may be free if you get them from a scrap yard.
Thirdly: Check out the video by Josh Wright in which he makes a similar chair. It holds weight, but doesn't look as good as yours. Here is a link to the video: th-cam.com/video/MuodNlDKce4/w-d-xo.html
Also, this is an absolutely amazing video, and an excellent idea! Keep up the fantastic work - your projects are a great inspiration for us all! 😁👍If you do find a way to make it work, please sell some online if possible, since many of us (myself included) would love to purchase this.
Seatbelts do have high tensile strength but they are meant to stretch a bit under high forces so that crashes are safer. In a bad crash it's recommend to replace your seatbelt because they don't rebound from the stretch, but it usually doesn't get replaced because a lot of the time the car is totaled at that point
i was going to say door hinges as well.
My thought as well, using hinges metal or wooden included in the pieces instead of the dowel inserts.
You could sew seatbelts together with a heat shrink layer in the middle. Stretching the seatbelts will cause heat, and the heat shrink layer will shrink to compensate. You could also use a braided steel cable (like ocean fishing line)
@@32Rats If each hinge flexes 1/32, and they will flex at least that much, there will be far too much slack.
You’re right, that lower bend needs to be able to take a much higher load than the rest and with the straps it can transfer that stress anywhere along the chair. I bet hinges would work, isolating the stress, but you’d need some really heavy hinges for the stressed areas. Fantastic work, you’re so creative and always give me lots of ideas for my projects.
Before I read your comment, hinges were what came to mind. No stretch, and each joint is independent instead of coupled to the rest.
There is some elegance to the straps that is lost though. Definitely changes the feel of the chair.
This is a tough project! Things worth doing are often not easy. If they were easy, everyone would do them.
@@jaustinpage It would probably only need metal hinges at a few of the most bendy parts near the knees. And for the straps, I think something like braided steel cable is probably going to work better than any cloth/leather/nylon strap. You could run a leather strap over the top of load bearing steel cables for asthetics. There's a whole field of engineering and design called tensegrity, which is all about exploiting cables and tension for structural integrity, so it's definitely possible.
@@guaposneeze you could even pretty easily create ruts in each piece so that the cable sits flush under the strapping
I think shorter straps would work as well. Like if the strap was just going from 1 block to the next it would basically be a hindge
Spoiler: this chair will be the key to understand the Mould Effect.
Loving this. I have two thoughts that hopefully help.
1. Having the strap/cable/whatever not be attached at several points means that any slack is 'sent' to the highest tension spots (the front) and will collapse any inverted sections (the back rest)
2. The front bend of the chair will always compress when sat on, so it needs to look 'wrong' and slope forward when not under load in order to settle into the correct shape when sat on. If the chair looks right before you sit on it, it's going to sink down and lean back when you do.
On 1: I had the same thought. However, realistically, for any chance of this to work and actually be stable enough, I think about the only option is to buy a metric F-ton of hardware and use strong metal hinges, possibly steel ones, between all the blocks. I think there's probably not any strapping material that would actually work.
On 2: It might have been easier to create a roll-up couch table. In fact, maybe some of these failed chairs might still be salvageable towards that.
I agree with the second advise, but instead of " wrong" slope, in the first curve use 2 different types of angles on the wood, one inwards and one outwards, to counter each other and transfer the compressive and tensile strength accordingly without creating the gaps in between the wood blocks. That should ensure that the shape stays same more easyli when sat on and in standby position. As right now when sat on all the force and pressure is just hold by the straps and almost nothing is transmitted through wood because of the gaps in-between the blocks. So I think the straps would work, both kinds but with different blocks design for the first curve
Crazy Idea: have a channel inside the wood you can run the strapping in, then have a tensioner at the top or bottom. (Something like a come-along / ratchet strap). That way the strapping will always be pulling the wood sections directly together, and if it's not taught enough, just do a few more clicks on your tensionser. If you want to roll it up, release the tension and roll.
Physically thinking this would also take away a lot of force on the strap. Because the angle is not as far away, the force is not that strong, that pulls at the strap
The strap shouldn't be a structural element when the chair is upright. The strapping is flexible by design, and it isn't fair to the material to expect it to be flexible and elastic in one setting and rigid under tension in another. The big issue is the forces involved on the curve because the seat is a giant lever. The forces are not distributed equally along each part of the curve, instead there is a fulcrum at the middle part of the curve which has to withstand most of the forces placed on the lever. These forces are dynamic because the forces should increase as a person leans back in the chair. Increasing the internal friction between blocks might be a good idea even if it implies that they should interlock with a complex shape, use powerful magnets, or require a metal insert between blocks. There is an elegant solution here, but it will take a good amount of experimentation.
Isn't it exactly what other woodworking guy - Josh wright did that didn't work?
Overengineering underperforms.
Really doesn't need to be that complicated.
Yes this helps to solve the issue of torque on the strap causing it to atretcu
I hadn't seen this suggestion so ill throw it in. Along with the improved strapping methods many have suggested, designing the chair so that it, when weighted, rests in the horizontal seating position. Essentially design it so when you initially unroll it, the section you sit on is at a ~30* upward angle as opposed to horizontal. Then when its weighted it has room to flex to the desired seating position.
@@jhay440 ratchet strapping in general might be the solution, cause those straps, especially the 2t rated ones only stretch when you put as heap of stress and force on em
Correct.
Also stronger straps - less stretchy. Kevlar or steel perhaps. Let them slip except at the ends where you have adjustment. The wood will compress so you will need to regularly tighten it. Take out all but two of the slats at the bottom
@@Chris-ZL Steel was one one idea I had as well. Leather won't work as it stretches by quite a bit.
i would glue all the strap to all the wood
@@Wtfinc she tried that in attempt one, problem isnt the securement of the straps but how much they stretch
I love that you tried again. I think you were so so close! I should try making one before commenting more, but I think had you stapled/fixed the straps to each individual section like you did the first time you'd have nailed it. With that free-moving strap and the amazing amount of tension you managed to get on it, it seemed to create a bow-and-arrow like effect. I love to watch this youtube CHEST'ER make furniture. He doesn't sew anything and just staples the crap out of his pieces. If you ever feel the urge to try again, maybe cover 100% of the front with fabric stapled on, then cover that up with another layer of fabric to make it look nice (perhaps even leather as you mentioned). Regardless, these videos are the best and we all feel part of the struggle. It's really exciting, thank you for sharing and keep up the great work!
Just watched your second attempt. You mentioned leather a couple of times and I'm pretty sure it would not be a good choice. Think about a new pair of shoes and how they stretch with wear. Love the idea of cable mentioned below. Maybe adding something so that the pressure points can have more strength. Possibly second shorter piece of cable. Please don't give up, I'm sure there's a way.
I've been thinking on this too. my thoughts are to 'isolate' the tension between each pair's connection. currently you're allowing the tension to distribute and focus in the worst place, like under where your knee would be. there is created enough slack there so that ANY separation creates a lever action instead of a continuous form through the touching faces of the two parts. there is also the matter of keeping the lower circle from collapsing while keeping the app circle from exploding and what this means for how you stabilize them. I wonder if it might work better if you didn't glue the dowels in and also rounded the meeting edges of the sections. this would allow the segments to hinge around the dowels and always pass the forces through that line. also maybe use a steel thread cable - like a bicycle brake cable - and include a cam action on the end to tension the chair after unrolling
do you think hinges between each piece would work to mitigate the compounding effect?
I think the simplest solution would be just give it a leg to stand on. Something similar to a foldable walking stick. Sure, it would need a lot of segments and it would not be stable on it's own.. but if the chair was stable on its own (just not able to bear the load), the leg like that could potentially hold, or at least help to hold, the load...
I really admire your persistence. Chairs are one of the most difficult items of furniture to make due to the forces applied. I think your instincts are correct that the current design has some inherent safety issues that may not be correctable simply by using alternative materials. Seems a design tweak is the best course of action. Good luck. Love your videos!
Love the honest attempts you make and how you show us what you learned along the way. And while the chair didn't come together, I wish you much success. Please keep up that positive spirit!
I'm also emotionally invested in this trilogy now. I don't think leather would cut it, as leather is also elastic. I would try metal hinges (flush with the wood) instead of a band of some sort. Metal hinges doesn't need to be tensioned and are strong. Also I would add one more bending segmented part so that there are support both left, right and in the middle of the chair (like three spines).
Hinges also have the benefit that some slack in a less critical hinge does not translate to the the ones getting the most stress
Im thinking Kevlar strips instead of leather or metal
Love yor perseverance Tamar; you're continually learning & that's good. Thank-you.
Always learning!
There's this saying that goes "If you can't make it perfect, make it adjustable" and I feel like that applies here. So maybe some kind of tensioning system would be in order?
That was my thought too. And that way you could put straps on both sides, which would make it more stable, like muscles in a human body.
Perhaps, the top of each block could be grooved, with the strap plates having a corresponding key. Similar to your dowel setup. The strap would effectively be clamped onto each block individually, the tension would not transfer through the length of the strap causing the blocks to collapse. It would be like a fabric hinge at each joint, being much stronger than staples or screws.
Honestly? This might work. Couldn't be a dowel, but it might really work for the bottom curve at least.
@@polerin idc how she attaches the strap to each segment, as long as she does it at all. Imo its THE reason this doesnt work.
I love that you tried this again and you DEFINITELY did a great job working at the issues. I thought the dowels were brilliant. I feel like leather would stretch too much as well though. Maybe not the first time, but certainly after a bit of use. I have some straps for my kayak rack that have steel cable woven inside the strap. It is meant as a theft deterrent so that you cant cut them, but it would also provide tensile strength without requiring the same extreme tension to overcome the flex of the strap. Who knows, but an excellent project regardless.
Definitely something to look into
Stapling the members is definitely on the right track. Right now the members are separating because the strapping is not fixed to the individual members so they're allowed to slide freely. To compound this, you get variable tension along the length of the strapping due to the geometry of the chair. Increasing the tension will just make it tighter everywhere, but it'll still be variable and create gaps. Stapling would constrain the members and create tension where you need it to keep the joints tight, but I am concerned that the tension forces may exceed the holding power of the staples so a more robust holding method may be needed.
The other issue that I see is that even if you resolve the gaps, I think this solution either won't hold your weight, or will be very unstable without a brace. This is because you're trying to support a vertical load with a horizontal tension. This would require infinite tension so what happens is that as you sit in it, the angles change so the tension forces are more vertical. On top of this, the wood is designed for a fixed angle, but in reality the angles will vary with weight and movement. Due to leverage, I suspect the tension forces can be quite high, and possibly damage the straps and/or the fasteners for the straps. If you were to provide a brace that you can wedge in after it's unfolded, I think this would greatly reduce the stress that the tension members would see.
The Ikea Poang chair is a very similar shape (that obviously doesn't fold up) made with steam bent plywood. You'll notice that the seat is at about a 30 degree angle. They obviously did this without a brace, but I think steam bent wood is a significantly better material for this chair shape (with the obvious flaw it doesn't fold up). Even then the chairs have a high failure rate due to the tension forces.
simple idea that might help:
make separate strap dedicated only to the main curve, without the back and the bottom. That way, there would be no places for the strap to collapse the chair under tension. The main strap would be used to hold the chair together and would not experience that much tension
yeah seperate strapping for the back and straightaway
I'm really glad you stuck to this. We can't always see where we are going. Sometimes you just have to keep slamming your face in a wall until you find a way around the turn.
My two cents are on having two different tension systems. One for the back, and one for the legs. In my humble opinion, it appears that the gap between the slats on the back and the structure may be your problem.
The two parts of this, pretend and purely academic, tension system could potentially be joined with a different layer of structure on the top or bottom of the profile, or the inside of the alignment.
The mechanical advantage of the legs on the tension side must be incredible. I'm just throwing it out there that you have a LOT of force. I think the leverage forces are 20:1... 30:1?
If you weighed 100 lbs, I'd bet another 2 cents, that is a ton of force in total.
Again, I'm totally speculating, and not reading other comments on purpose. You are kicking ass... There is nothing wrong with being ambitious as long as you can afford to be
If you feel like trying again, I faced a similar issue in the past and ended up going with a kevlar thread. After a very quick look, you can get kevlar webbing, from their specs it has a tensile breaking strength of 5500lb and more importantly an elongation break of 2.4%, I'm sure it can also be found in many different sizes and strengths
I thought of this too. Kevlar is almost impossible to break and even more resistant to stretching. I think Kevlar has the strength to sustain her weight without needing to be under too much tension
I was also thinking of Kevlar or CF
Great second attempt! Love the dowel idea to reduce rotation.
If I put on my armchair engineer hat, thinking about the forces, this design is an uphill battle. Focusing on the point you’re seeing the most stretch, at its simplest you have a lever. The inner radius of the block being the pivot, a force of ~130lbs ~18” away (when sat in as designed), and the strap opposing it 2” from the pivot (the other side of the block). That ends up being a 9x force multiplier, turning 130lbs into 1,170lbs (or 5200kN) of force on the strap. (18” / 2” = 9x the force. 9 x 130lb = 1,170lbs)
As for the stretch, I haven’t modeled it, but I suspect even a 1% stretch would significantly alter the shape of the design. Especially when it only has the two attachment points where the 1% of the whole length is able to be concentrated on one pivot like you saw, as opposed to the idea you have of attachment at each segment reducing the 1% to a single 4” segment of webbing. Based on what you experienced by adding tension, that too isn’t a solution because it adds a inherent spring that adds danger of collapse. So to both hold it’s shape unloaded, but also not collapse when sat on, the strapping needs to have almost 0 stretch. Difficult when applying 5kN of tensile force.
So in my mind, in order to make this work (or at least give it its best shot) you need to reduce the leverage on the strap, and increase strap tensile strength.
The first can be improved by using thicker blocks. A small change of going from a 2” thick block to a 3” thick block decreases the leverage by moving the pivot further away from the strap reducing forces from 9x to 6x and 780lbs. Going to a 4” thick block drops it to 585lbs (2600kN) pulling on the strap - still a lot. For aesthetics, you could even have varying thickness over the length since forces decrease towards the top and at the tail of the base.
As for near zero stretch at these tensile loads, two ideas come to mind. Layering the webbing literally multiplies its capability. I haven’t found the specs of the viper webbing you used, but 1” tubular nylon webbing has a breaking force of 4,000 lbs, with stretch happening at MUCH lower forces. Layering 2, 3, or 4 layers with individual segment attachment, might reduce stretch enough in combination with thicker blocks - while still maintains rollability.
The other no stretch idea is to use metal, metal strapping has almost twice the tensile strength of the strongest webbing. A relatively thin metal strapping might give you the strength required with almost no stretch, while still being able to somewhat roll up. You could possibly use it in conjunction with the webbing.
This is a hard problem because of the amount of leverage exerted. It’s like having a 20” plank on a pivot point 2” from the end, a person standing on the long side, and a piece of webbing just 2” from the pivot trying to hold the person up. That’s big forces. But there’s a chance it could work with these (and other modifications I haven’t thought of).
Ugh. Yeah. It’s a hard nut to crack. Ha
@@3x3CustomTamar Good effort, and thanks for sharing even if it doesn’t work. I know people are asking you to keep going with this, if you don’t, that’s cool too. There’s other things you can spend time on that we’d also love to watch.
This is a really good explanation of the physics at work here and great suggestions for attempt 3. I like the idea of the varied width but that would be pretty complex maths to work out without a program that can do the maths for you 🤔
I would think ss wire rope would work and look good.
Either metal band like Josh Wright used or aramid fiber cord or strap (Kevlar & other brands) for ultra low stretch. If strap stretches 0.1" at the outside of a 2" block, & chair back is 15" from front curve that's 0.1"/2" = 5% deflection; .05 x 15" = 3/4" sag per block joint with that stretch. Check Josh's video, he's close to solving this. Is it a great chair at that point? Opinions may vary on that. Valiant effort
I love the tenacity on this project. In case you tackle this again: If steel cable was run through the middle of the blocks it may equalize the tension/compression and prevent the bottom end from wanting to curl back up. Add a locking cam for the final tensioning could make adjustments easier to.
No, a cable running through the middle wouldn't contribute to the load in any meaningful way. It would only be able to add an axial load between joints.
In bending, all the load is experienced at the top surface and at the bottom surface of each joint. I actually disagree with the conclusion of the video that there was too much tension. It still looks to me that there isn't enough and that the webbing simply has too much stretch (which is what the weave is designed to do, no matter what the marketing says). I also suspect that the maple isn't supporting the compressive loads as rigidly as needed and it's deflecting too much as well.
I think the solution is to oversize each angle by a few degrees to counteract the deflection that will be experienced during compression loads. More tension would probably also work, but at some point, the compressive displacement will become a significant factor too.
@@joranbooth5529 in the video, the end of the chair at the bottom curled back under because the tension from the strap being located on the bottom of the chair blocks pulled it back. If the tension was centered in the blocks it probably would not want to curl back since the tension would be equal between the faces. A cable would just be easier than a strap to run this way. Alternately, if the strap was pinned to not freely stretch between more than two blockers, it probably would not curl back under, but getting the tension right would be tougher.
Wood blocks are the actual structure, the strapping just work to keep joints together but remain flexible so it can be curled up to be put away, which is why it worked better when she actually glued it...
So cable would work, as it would function the same and would be a stronger material than the strap...
Personally, I would shape the blocks similar to folding table hinge so they lock together in the chair position and use a railing cable so it can be easily tension/adjusted...
If you ran the belt/cable through the center of the blocks, you would effectively half the leverage you get, which would only cause more stretch. It's better to isolate the belt connection for each joint by fixing it to the segments, so this "curling" effect doesn't occur.
@@CatNolara Nonsense, load on a curving structure doesn't work that way. The fabric was never handling the whole load by itself, just holding the blocks in place so they can handle most of the load, otherwise gluing it wouldn't have stabilized it as much as it did... Besides, steel cables are a lot more rigid and a massive amount stronger than fabric.
Maybe try building a proof of concept of just the main curve. Part of the stress is that you have disassemble and reassemble an entire chair everytime you want to adjust (even the relatively stable base. Maybe try breaking it into 3 separate problems and solving each one, then decide how to best connect those 3 parts.
Based on my older belts, I doubt leather would work either -- it does stretch eventually. I see that several folks have suggested steel cables, and that sounds like the best bet; they are used architecturally, and in tensegrity pieces (although I don't know how much weight most of those would support).
Dyneema would be lighter and more flexible than steel and more than strong enough
@@xerxis100 look at 16:26 - the strap is made of 'dyneema'
@@stanimir4197 yes, I realize that, but dyneema rope shouldn't stretch much more than steel cable. So there should be no advantage to using steel, except for aesthetic looks. I don't know about the belts, never used them. Anyway, I think this design is just impossible.
@@xerxis100 steel elongation is order(s) of magnitude lower than UHMWPE.
@@stanimir4197 can you point me to a source for that, AFAIK dyneema rope breaks at 3% deformation. I would be surprised if stretch at normal loads would be a magnitude bigger than steel
Tamar, instead of just the two points securing the webbing, you need to have every slat be a secure mount point for the webbing. (but not with staples). I would create complimentary grooves/grips between the back piece and the top piece to grip the webbing at each slat. The wood would act like a vice at each point preventing the stretch you are seeing without having to over-tension the webbing. But use at your own risk. No guarantees since the forces created are deceptively large. In fact, I don't think any wood is strong enough. It will either crack or crush at the fulcrum points. Good luck, though. I'm impressed with your tenacity.
yes a lot of shorter pieces of webbing would probably limit the flexing, steel wire maybe.
UHMWPE is almost as slippery as PTFE (Teflon)
Mitch, I totally agree! I am a chair builder by trade and I find these videos about trying to "save" a bad design by some kid on the interwebs rather cringe-worthy. 2 minutes with a tape measure, a calculator and a stress-strain table for nylon webbing will show that this concept is doomed from the start: the stuff is just way too flexible (and influenced in its properties by temperature and humidity) to allow this concept to work. What amazes me is that she doesn't learn. A lot of people have given very competent input but she's continuing without making use of the insights people are sharing here...
I think if you just replace the webbing with door hinges, it should work and be easier to mount to boot.
I absolutely love the honesty in these videos. It is so encouraging. Thank you for being so authentic.
Glad you enjoy them!
Consider: adding locking mechanisms, you can make a pretty simple one that just needs a tiny bit of rope and a loop, and hook
I love that you share your failures along with your successes. it's so motivating to see you go through the highs and lows and keep learning and trying new ideas!!! super excited to see the next version :)
Glad you like it!
Piano wire, routed through holes in each block, drilled near the "outer" surface of the bend in which that block resides (in other words the surface that is put into tension as a result of that bend). Also, place lugs at the top of the chair, around which the wire is wrapped, thatcan be tightened/loosened for assembly/roll-up respectively (similar to the lugs that are used to tune guitar strings). Just make sure fingers can't get in the way.
One thing i mentioned when you tried to sit in the chair at around 23:09
the reason it looses tension and collapses is not that the belt is too long, but because it is attatched to the backrest.
Basically, as tension builds up in the lower curve of the chair, the string pulls the backrest towards the front.
To solve this, you might consider using two(four) seperate belts instead of one:
one to only keep the tension in the lower part of the chair, and a second one to support the backrest.
This would definitively need further prototyping, but I hope this helps :)
Anyways, great video and great effort, really enjoyed the journey, thanks!
Tamar: if you put 50Kg at 50cm from the frons joints and they are about 5cm thick you will have a tension of 500Kg in the strap!
The rest of the chair could be done as you did but the front joints at least have to have the above ball park calculations in mind.
Your calculation is essentially correct but I think the tension on the straps will be much higher because
- most people are heavier than 50 kg, some are MUCH heavier
- 50 cm is on the low side, especially when leaning back
Plus, the thinner the pieces of wood, the worse it gets. And let's not forget that the wood has to withstand a similar amount of pressure.
Pushing the materials this hard requires doing the math. I'm sure you can find an engineer to help you specify the right strapping. I also agree that it's important to connect the strap to each piece so that the ends don't decide to curl and release the tension.
I wonder about the distribution of forces vs the structural constraints for each section. I'm not a mech engineer, but I think some modeling would be a good idea. The collapsing sections seems to say that the strapping/tension needs to be segmented more robustly.
Exactly. And when you just pressfit the straps to the wood, the moment one strap begins to slip ever so slightly, the force on the neighbouring pieces increase. In the end you will always have a big force on the ends of the chair.
I dont think leather will be better - on the contrary. Leather once was skin, and skin is very elastic - has to be. Those straps are made to be unelastic and they still were too elastic.
Agreed. It is just a simple calculation of leverage. Just pick a joint and consider all the parts between the load and the joint as fused. Then it is a large lever vs the 2in from the pivoting corner to the webbing. Thus putting a huge load on the webbing.
Tamar,
Every material is essentially a spring. You need a spring that doesn't stretch much to maintain the shape of the chair. Polymer webbing just doesn't have the stiffness you need in the sizes you are using. Leather would be far worse. Even if it is strong enough, it is the stiffness that needs to be high. Think of it like lifting a load with a rubber band and a string. Both will carry the load, but the rubber band will stretch much farther than the string does.
There are two ways to increase that stiffness. Select a material that is stiffer (a higher modulus of elasticity) or make the spring bigger (increase the cross section). You might get enough stiffness by having the webbing span the entire chair width. I could run the numbers for you if you get me the exact material data.
Otherwise I am thinking you will need to use spring steel strips or high strength steel wire (a small chance something like Amsteel rope would work). It will be a little tricky though as wood is relatively weak and we have to not exceed it's strength in crushing as well. In general making the supports the full width of the chair will make a large difference.
The dowels would be better moved right to inside corner to pivot around. That would distribute the stress over a larger area instead of the small inside corners contacting. The thicker the chair the better as that reduces the load on the webbing dramatically. You already made this one thicker which was a good idea. With the current design you might have been closer if you used four or so layers of webbing.
I am a mechanical engineer and can help you design one that works if you want to contact me.
I cannot wait to see this chair design finally get completed…
I am keeping my eye on this project.
Haha. One day. I will get it.
@@3x3CustomTamar So it seems to me that because the webbing is only stopped from sliding by the dowels at the ends, everything in between can slide. Because of the leverage created by each weight bearing angle and the compounding of all of those angles, the webbing slides and pulls the ends back in on each other. Securing the webbing to each individual wood block (two flush mounted screws under the slats maybe? Or possibly just putting the screws that hold the slats through the webbing...) would solve for that sliding. Then, the most likely points of failure would be the holes made in the webbing. The webbing might be strong enough in its own to deal with that. Alternatively, they could be be reinforced with more box x stitches, which might also mean doubling the straps (admittedly more tedious, but in for a penny, in for a pound right?) I think that if the screws went inside the box on each side of the x, then that could (maybe..?) be the best way to reinforce it. Admittedly, there's a whole bunch of 'maybe' baked into this suggestion, but just some brainstorming from thoroughly invested fan.
@@alistairlacaille yep, I think you describe the problem. Plus, the whole thing should be assembled with some tension on the straps so they stretch less when in use.
@@3x3CustomTamar okay conpletely randomly came across this - 5:30 into the below video, Vousair Chair (or something, on mobile can’t check) - they leave it under tension and use a strap to keep it in chair form… th-cam.com/video/NtXK5ZgQ54w/w-d-xo.html
Great work! You’re so close! How about reducing the problem further and focusing on a single C section instead of the full S curve? A smaller prototype uses less material and time so you can iterate faster. You can also drop requirements and add them back later. For example, can you make a rigid chair out of keystone blocks or other segmented shapes? What improvements to its shape make it more stable? Once that works, then can you add linkages to make it bend in one or two segments? Then can you make it roll up and still work?
change the C to an O and it will work really well because the leverage is greatly reduced.
i love that you didn't quit even though you wanted you kept going until the final result. i was sitting at the edge of my chair rooting for you every time you made a change. either way keep up the great work, i love watching all your videos. even though i have not started wood working yet, you inspire me. Keep it up
Awesome to see a return to this project! Greatest follow up to a past video this year!
I just have to solve this. Ha
I just want to let you know that while this project is probably your arch nemesis and it is taking a serious mental toll on you, as a viewer this is the kind of content we like to see. It is worth it, continue! I too have a project where I keep failing so I can relate 100% to this!
And I will keep trying until I figure this out! 💪
I think that "Bicycle chain" imbedded in a groove of each slat would solve two problems at the same time. It does not stretch, and it has side bending resistance.
I agree with cycle chain. And put a tensioner at one end of each chain to allow for the small amount of stretch over use.
I was also thinking chain. Not sure bicycle chain would work under that much tension though - the plugs between links aren't that strong really.
@@NikolaHoward Nice idea! Chain can be beefed up to motorcycle chain.
@@PlanWithoutChance motor bike chain might work.
It's all about the sheer force load that the lugs and plates can take. I don't know their rate loading at all.
But it would certainly be flexible enough!
Make the chair hella heavy to shift around though...
a 530 motorcycle chain should be strong enough, they don't snap when doing wheelies on 200kg motorcycle + rider
The strapping needs to be joined securely from piece-to-piece, not simply at the ends. Consider bicycle chain along a groove instead of material. It can be positively locked to each block. 🛠🔄
Hah, I commented pretty much the exact same thing and then saw your comment.. I really think it's the way to go
I just made a much longer comment saying essentially the same thing...
27:43 the solution could be using *door hinges* instead of elastic material. That way the individual pieces wouldn't influence each other and the backrest wouldn't curl up because of the seating area.
Also, by adding some magnets, the pieces could be held in place when the chair is in it's chair shape.
OMG, this is such a simple fix. You paint it Gold and put it in a museum as a work of art, with a sign saying don't touch... 👍👍👍
Old seatbelt material imho
That's what the second material was
😂😂
You could even use a heavy-duty sash-chain or roller-chain (bicycle chain), and fit that into a recess cut into your wooden blocks. That'd probably provide the _least_ amount of stretch, while remaining flexible. Roller-chain is sort of an "Industrial Lego" and has it's own standardized sizing system. So, it can be easily had in sizes above mere bicycle chain.
I love how real your videos are. People tend to hide their struggle in videos, the way you go about finding solutions is really commendable. Love your work, you’re just amazing!!!
Awesome to hear. We all go through these things. It’s all part of the process
Hi, the toughness of your meshy material comes with the flex. You should use something not stretchy, like lifting slings or slings with steel wiring in it. Look for forklift stuff. Additionally, when you tension the straps make sure they're screwed down to every single piece, so they can't move along the whole length.
I’m gonna suggest looking into dyneema. It’s 4% stretch when under full load. I think one of the big issues you’re facing is that the straps stretch too much.
Then maybe add a second strap along the back that you can clip into place to offset the front tension. A strap that prevents any blocks from folding over like the bottom block did.
Dyneema is a brand name for UHMwPE, so I wouldn't expect the results to be much different. Even if it adds more than just a bit of weight to the chair, I think either cable or hinges are the way to go on this one.
steel cables stretch more than dyneema
I feel like having 2 additional, shorter straps that focus on the seat, and then the lower bend section where most of the tension is generated would help to keep the tension localized where it is needed.
That way instead of having 1 strap that is distributing the tension throughout the entire chair you would have a short strap around the lower bend that takes the majority of the force and localizes it to that section with an additional strap that stretches from the butt of the seat around to the middle of the base (right around center of gravity) that takes a little bit more of the force. That would leave the full-length strap with a third of the of the overall tension at most (and hopefully a lot less if she can get the other two to be more taut).
That’s a terrible idea.
with a design like this more localized tension would definitely help, also I'm not sure if it applies here but with some materials and rope you need to pre stretch it to achieve its intended mechanical properties but sometimes that makes it more elastic.
THANK YOU for showing how an expert can go through projects that don't turn out as hoped. It helps us beginners feel like we are not alone in this.
metal hinges would be my first thought. would solve the side to side motion issue and might give more stiffness, which seems to be important for this chair since even a little flex can move the furthest part a noticeable distance.
you want the tension to be focused on each joint individually, that way there is no chance of it pulling the bottom up or collapsing the back, the straps sliping through each wooden brace compounded the tolerances and also pulled from other areas. not all the forces are equal, so more tension is required for the bench bend, in comparison to the straight sections
Screws will concentrate the force in that area and can rip thru the webbing if the compression between the slats loosens.
I love this channel so much!
Thank you for making all of us that work with wood feel better about mistakes.
Metal cable routed into the pieces instead of the straps could be helpful and possibly hidden?
Either way though, thank you for the emotional rollercoaster that was these two pieces of content 😊
I was thinking screw the fold up issue and stick some bent rebar in it instead. Still a good looking chair !
I was about to say the same thing. Metal cable.
I was thinking the same thing. Consider using something like bicycle break cables that are sent through drilled holes. You could do two per side which would also help with the side to side motion. The nice thing about metal cables is you can easily adjust them.
Use door hinges to hold the pieces together you’d need one for each joint, they swing in the motion you want whilst being sturdy enough not to break under tension (you’d need to cut a channel in each piece to compensate for the width of the pin on the hinges) but I don’t see why it wouldn’t work
that is exactly what I was about to suggest - put the hinges on the side you want to fold into - the same side as the cross boards - that will do it
This was my thought as well, but I'd put the pin facing out so you don't have to compensate for it.
Exactly what I would suggest - the bands are just too flexible to ever create a strong enough connection.
I thought about this when first seing the design in the first video: One of the issues is just that you're dealing with huge forces: If you think about the lever of sitting in the center of the seat part with let's say 60kg (all just a back of the envelope calculation), you have maybe a lever of e.g. 1:10 at the front, which means the straps must hold about 600kg (and so the very back edges of the wooden blocks that are under compression). Furthermore this lever means that a tiny gap in these joints will massively deform the rest of the chair. And as there is virtually no material with 0% stretch (or deformation under pressure) so I doubt that this will work no matter what materials are used short of wire cables and steel blocks. In my opinion it is just a flawed design that looks cool in the rendering but no practical considerations went into it.
But I really like how you show your process and all the thoughts you put into your experiments - I learn something new every time! And I'm always looking forward to your next projects!
Yeah. I’m starting to think it just won’t work. But I also don’t want to give up. Ha
My quick calculation is similar- 300kg per strap. That means 300kg compression on the wood blocks which is very high for wood compressed perpendicular to grain.
The one example I have seen of this chair where someone could physically sit in the chair they used 4 steel cables under the fabric.
More armchair engineering: I think the key is reducing that 1:10 lever. Your weight offset from the front of the chair creates a moment (force x distance) which has to be counteracted by an equal moment from the strap to the inner edge of the blocks. The longer that distance (height of the blocks), the less force it has to exert to get the same moment. Also, you should attach the strap to each piece rather than just the ends as you mentioned. Stretch (strain) is a percentage of distance, so the longer the distance between strap supports the more the stretch that can accumulate at your stress points and deform your geometry.
Yeah that's what I thought huge forces and minimum tolerance for stretch... I have no clue how to calculate that (finite element analysis?) but the forces involved may call for high tech materials... Carbon fiber or something like that with extreme tensile strength without stretch...
As a thought experiment, imagine replacing the webbing with recessed metal hinges at the junction of each block to its neighbor. This arrangement of rigidity and a tight turning radius are what you’re trying to replicate.
I know it would violate your keen aesthetic to use hinges; but any material flexible enough to act as a hinge will likely also tend to stretch.
There might be a solution using a tensioner to adjust out any gaps that develop. Or maybe a hybrid approach that places metal hinges only on the corners that are stress points.
Not sure what the answer is but thought that might help you visualize the problem.
Inset the hinges, put a strap over there for esthetics and there you tricked thousands of people! :D
@@boomerangfreak that's what I would do, but I'm not much of a "maker". I'm that lowliest of TH-cam comment-makers; an engineer!
@@danielleohallisey4218 we all need engineers! Who else are we supposed to yell at for their impossible to make designs! ;)
@@boomerangfreak always happy to serve a useful purpose! :D
I love your perseverance and I learn a lot from you, thank you. I would like to make a suggestion about your strapping, I use something similar to tie down loads and a fresh strap has a lot of give, so just to limit your variables I suggest you pre stretch your strapping.
Hi Tamar, just wanted to make a suggestion. I think adding the dowel between the pieces is creating a leverage point. Instead of 2 flats coming together and being compressed and forced into each other, the dowel is creating a round edge that the pieces can pivot around and in turn increasing the stress on the webbing/straping. Hope this makes sense. Really enjoying your channel. Thanks.
You need a solid connection btw the strapping and the wood at each section (glue, a mess of fasteners). That way, the straps act as individual hinges rather than allowing all the stretch to accumulate in one or two joints. Strain is measured in inches per inch. Shorter initial length results in shorter overall for a given tension.
Yes! This!!!
Securing each block to the strapping, creating individual hinges
Yes! Absolutely. I believe this is a good track to follow
Yes, thats what I was going to do after figuring out the correct strap length. but it just kept stretching... then I got hurt. so decided it was wise not to continue with this iteration.
29:52 as an engineer, this emotion just speaks to me deeply even though no words were told at all.
Haha
Been there so many times
i hate that it didn’t work but i REALLY love these videos
Haha at least I didn’t fail at that part… 😂
Thank you so much for showing the bits that don't work out. I feel your frustration but it helps me so much when I screw up not to beat myself up too much. Your videos are fantastic.
Hey. This is really cool. Back in college we took a jab at a very similar project, pretty much the same. The way you used dowels was very smart. And architect friend of ours mentioned maybe we could CNC each piece, and I think that would also transfer the tension to the wood instead of the strap, since we couldn't get the straps materials to work either. Id love to see a calculation on just how much force that material would need to resist and still be flexible. Good luck.
Problem is that I only gave myself a week to figure this out. This is going to require months of testing…ha
Came here to say this. The problem is that the tension required to support weight through that structure is enormous. You currently have the blocks in compression and the strap in tension. Solution is either move the tension into the blocks or use a material that is better suited to the task for tension. Leather is not going to do it I’m afraid.
Consider tank straps? Very thick, not so bendy, but definitely overkill and should do the job.
Alternatively, use some steel cables, and hide them under the straps you currently have? That way it will be aesthetically the same, but the strap will serve as a lid and the cables will carry the tension.
Good luck! And thank you for bringing us along for the ride!
I agree, throughout this I've had the feeling that the tension should be borne through the wood not the strapping. Think maybe go back to that clever computer design programme and redesign the shape, thickness and depth of the sections. Also maybe a scale model would be helpful, remember that's how the artisans used to bring forth their creations. The perseverance involved in this project has been admirable, (if distressing feeling the pain) to watch. Sincerely hope the conundrum is solved.
Seat belts can withstand a lot of force and still be flexible. That's exactly what they're made for and it's exactly what I would have tried. And seat belts don't stretch. So, you hopefully will cut down on stretching by using it, also. You can buy it in bulk/rolls to use for whatever you need.
@@JC-11111 she already used seatbelts on Episode 1, btw :/
The effort you've put in is inspirational! Chair looks so cool. Maybe steel hinges. Or brass maybe.
defo NOT brass!
@@g60force why?
You need a vertical stabilizer under neath
And some kind of locking mechanism on the sides to keep it from trying to role up on you
Congratulations on getting this as far as you have, while maintaining a surprisingly sunny disposition. You are a real trooper.
Try using dyneema, I use it for sailing. It's stronger than steel cable, a hell of a lot lighter and is fantastic because it's pre-stretched, and wont stretch more. Can also be spliced and tightened to create tension if needed
Highly recommend this as its tensile strength is so strong, only problem with this is it can sheer at a super low force.
@@maximearmand8903 could you somehow insert it into a tube to protect from outside damage? but would it still fit into the chair?
Dyneema was my first thought too.
Nice project Tamar... Love the shape.... Make channels in the blocks and use motorcycle chain instead of the strap
I was thinking along the same lines ie too much stretch. My thought was flat woven metal strap like is used for ground straps but I don't know if that would be a good look. I like the chain idea better.
Another possibility would be stainless steel cable with a threaded piece on the end. The threads would allow tensioning.
my thought too! With all this tension I feel like wood and plastic alone will not be sufficient, you need some metal, maybe even try a experimenting with some sort of hinge between each piece.
it may also help to drive a screw through the chain at particular points to isolate tension from collapsing other parts of the chair and limiting the amount of stretch available at any particular joint. Thicker blocks of wood in the front section would reduce the how much chain stretch would affect the overall shape.
leave the "motorcycle" chain in the bike. If chain would be used the best bet would be a single strand # 40 roller chain with attachment tabs.
@@TimeLapseRich how did I forget those existed?!
Since it’s bending so far when assembled, could you design it “wrong” (leaning more forward) so that when you sit, it stretches to the preferred design? Not sure if it would work, but just a thought I had while watching.
this is what she continually does by adding tension to the strap but it just can't support the load of a whole bouncing human
Try it with steel strapping. Everything with weaves will stretch, and leather stretches too.
She is back with a vengeance!
Edit: the wood won. But that was some great effort!
Haha!
Suman! nice seeing u here
@@austinbowles oh hey! Small world. Haha.
I've watched two successful implementations of this chair on YT if you're interested in checking those out. Z-Shaped pieces seem to be the key. Pallet straps might be a better band
Josh Wright
Adjusting the shape of the segments definitely seems like a much more elegant solution than chasing down the issue of the strapping. Tamar does have a CNC, and while using it may not make for quite as fun of a woodworking video, it would allow the production of unique shapes that do the work of holding all the angles in place.
Yes, I'll simply toss my 2 cents here as well. I'm not a carpenter but IMHO the problem between your chair and the ones that work are that the other chairs pieces lock into place (with gravity) when they are unfolded. The strap connecting them simply holds the pieces together for when it is rolled up. Yours doesn't lock in place at all. The strap is what's holding it up and gravity (plus weight) is causing it to fail. If you redesign the wooden pieces so that they lock when it is unfolded and gravity (plus weight) reinforces the structure, then you'll have conquered this project as well.
I thought the stabilizing dowels were a good idea. Did that part of the experiment work?
My suggestion for the next attempt would be to test the stretchiness of various strap materials before building anything. My instinct says that leather will stretch even more than nylon straps.
Yea. That was the only part that worked. Ha
I agree. Leather will stretch.
I saw another builder do something with an overlap joint in the wood. I am thinking maybe a ship lap type overlap, but with a back angle on the edges so that it sits in a pocket once unrolled and puts some of the tension force onto wood. Hard to describe but i think there has to be a way so that the strap isn't taking on all the forces, the wood has to interlock some how. Kind of like a two layer french cleat.
Beautiful choice in wood, i like the asthetic of how this design looks.
You need to pre-stretch the strapping. A forklift helps with an anchor point on the ground. I have this same issue when installing Dynema cable in cranes. I use the hydraulic force of the winch cylinder to stretch it out during installation but every once in a while I have to return to the crane and redo the splice and sewing.
This is also the case for Dyneema sailboat rigging. They even sell pre-stretched (annealed) Dyneema for this reason. 'Course you can always get the cheaper stuff, tie one end to a tree and crank away on the other end with a winch.
Obrigado por mostrar as dificuldades para desenvolver um produto, parabéns!
@@NOLAMarathon2010 Congratulations, you can copy and paste
I actually do... im from Portugal!
wut?
@@vascorfael7223 So... too early to ask for our gold back? 😂😂😂😂😂😂
@@NOLAMarathon2010
i don't speak portuguese either, but what i know of spanish allowed me to get the gist of it.
Could try a "locking" design one piece locks into the next when unfolded and put the strap all the way around the ends with tension keeping them back a bit and leather will not work it will tear And dry rot
A very interesting and challenging project. I appreciate that you have shared your efforts to address each problem as you worked through the design and material issues. Thank you!
Glad you liked it!
What if instead of a chair, you had the same rolling idea, but without the back to support, well, your back. What if instead when you assemble it, it resembles an empty cylinder or, a log thats empty. Then you don't have to worry about the back being strong enough to support someone's weight.
Looking at the design, shouldnt the strap "switch" sides somewhere around the seat area and be on the outside of the bend? Because one continuous strap will be tensed by the lower part and put "curling" force on the top part making it collapse.
the side with the strap is the side everything curls into, and as you can see by the chair shape, all points on the chair expect the same curling direction for this design to hold.
I completely agree, but it defeats her goal of making it roll up. The switching sides would definitely make it more structurally sound but I guess it's a toss-up between two goals for her project
Thank you for sharing such great project!! Although it didn’t ended up ideally, it’s still a beautiful woodworking furniture!
I got a suggestion, I think connecting each blocks with something like door hinges might help with the strength and reduce some flexibility.
It's super frustrating when you try, try and try again feeling like you're tantalisingly close to making something. Big respect for sticking at it when it got so frustrating !
How about those Metal Straps, that are used to secure heavy equipment on pallets?
Those won't have any stretch, roll up without an issue and be strong enough to not snap!
Basically hinges if we fix the metal strap to each pieces.
@@WhoAmi2357 I was thinking hinges, as they are rigid, and would still allow the rolling, but I feel that kinda defeats the purpose of having a single continuous strap running the whole length
I think this idea gives the best possibility of success without a significant design overhaul.
You could try some straps made for off-road vehicle recovery. There are versions that have zero stretch.
Still too stretchy.
We need steel cables at least.
@@cezarcatalin1406 i agree. I'd like to see that myself. I think it would look even cooler too.
@@cezarcatalin1406 Thats' what I was thinking too, but steel cables will cut into the wood so it'll still sag a lot.
@@GrayHatter plastic coated cable, and tight tolerance on the holes
@@cezarcatalin1406 I commented on this separately, but I was thinking steel banding strap used in shipping on pallets. It could be used just like plastic strapping, but would have very little stretch and a lot of strength.
You could always just make a strap between each piece instead of throughout the chair. Then there would only be the stretch between each individual piece to contend with.
I was having a very similar thought
What about larger cable segments? I.e. a cable going from bottom end to the beginning of the load bearing curve, a cable for the curve itself, and then a third cable for the back pieces
And or strap segments
@@kfurgie999 That would also be exactly the same thing, all else being equal.
Maybe you want to adjust the strength of the strap segments depending on their load. That would make sense and can be achieved with using a different material or just more material (wider strap or multiple straps).
@ the difference is that in this design the tension transfers between them because the straps can actually slide beneath the planks holding them onto the blocks. this is also what the staples she mentioned in the video are about: fixing the straps so they do behave like many tiny straps
OK, love this! Tried hard and did not succeed... very few woodworkers show this, but it is part of every woodworker's life... Absolute A+ for effort. I learned alot watching this...
All a part of the Process!
Motorcycle chains! The wood already has some flex, so you don't want strapping that stretches, too. Motorcycle chain would also resist torsion, providing additional stability.
And more weight... I think at some point, the idea of this "roll up" thing, is that it can be portable. Too much weight would affect that portability. Don't you think?