PVC pipe as a SOF kayak frame material?

Discussion in 'Boat Design' started by mitchgrunes, Jul 14, 2021.

  1. clmanges
    Joined: Jul 2008
    Posts: 506
    Likes: 99, Points: 28, Legacy Rep: 32
    Location: Ohio

    clmanges Senior Member

    You misquoted me--twice.
  2. mitchgrunes
    Joined: Jul 2020
    Posts: 110
    Likes: 17, Points: 18
    Location: Maryland

    mitchgrunes Senior Member

    I've been reading the first of the indicated books about mechanical engineering, J.E. Gordan's "The New Science Of Strong Materials". (Though I only found the second edition in the library - apparently first produced in 1976. Presumably somewhat out of date. Don't know if there were significant later revisions.)

    I don't know if it can help me design a kayak, but I think the history of science and technology are much more interesting than that of political battles and military conquests. And it clears up some of my misconceptions.

  3. mitchgrunes
    Joined: Jul 2020
    Posts: 110
    Likes: 17, Points: 18
    Location: Maryland

    mitchgrunes Senior Member

    One of the things I am getting out of that book, as well as looking things up, if I understand things correctly, is that PVC is not quite as strong and tough (per weight) as wood along-grain - but is substantially stronger and tougher cross-grain.

    So - given the variability of stresses from the sea environment, PVC may actually be a fairly decent structural material overall for a SOF sea kayak frame.

    I'm not knowledgeable enough to be sure, but so far, it looks that way.

    So I'm back to the original idea.

    But that is only weight for weight. One has to consider the available structures. PVC pipe - even the somewhat heavier schedule 80 (vs 40) pipe - has relatively thin walls. I should have thought of that when I noticed how light the pipe was.

    Some possible solutions:

    1. Wrap fiber+resin composite around the pipe, to make it thicker, as someone suggested. But, PVC has a much higher rate of thermal expansion than fiberglass. Which would create stress on the bond. There are adhesives designed to bond fiberglass to plastics that allow for some degree of thermal expansion, but it seems like a dubious solution. It might be good enough for the fittings, which are short. If only I could think of a sturdy way to thicken PVC pipe.

    2. Use spray foam (archinfos.com/library/lecture/read/142407-can-pvc-withstand-boiling-water) to strengthen the pipe. An interesting idea. Understanding how much strength and toughness I could get out of it means more reading.

    3. Add foam tubes (similar to pool noodles) around the pipe. I haven't found any references about that yet. Maybe it's impractical.

    4. Fill the pipe with something solid, like the bamboo stakes someone mentioned. But that won't strengthen the joints, and it can't fit too closely, else differential thermal expansion would become a problem. Plus, when they are forced into contact by bending, there would be stress concentration at the points of contact. Unless I pad the stakes...

    5. Use more stringers and cross pieces than I intended, perhaps to roughly match the weight (ugh) of a wood-frame boat. But, if I want a break-apart folder, that increases assembly time.
    Maybe I could make a weak straw and tape model, and play with it in a sink, to figure out how to space the pieces, to try to even out the stresses. But there are so many possible variations in wave height, direction and shape, I'm not sure that is practical.

    Also, the book mentions near-atomic scale "stress concentration" that I can't model. Other sources say that plastic pieces unlike wood pieces, shouldn't fit together too snugly, because of stress concentration. But PVC fittings fit together very tightly.

    I obviously could safely test rolling around inside the frame, with and without a skin, on the grass and the driveway, to see if it breaks. I could likewise practice rolls and rescues in easy water, and pull heavy objects across the deck or sides.

    But when all is said and done, the only practical way to test how a boat holds together under sea conditions is to take it into those conditions.
  4. Blueknarr
    Joined: Aug 2017
    Posts: 1,262
    Likes: 301, Points: 83
    Location: Colorado

    Blueknarr Senior Member


    Pvc pipe is too flexible.

    Take a sixteen foot length, jamb one end into a corner and see how far you can bend it.

    Way earlier in this thread you stated that you were drawn to pick pipe over wood because you lacked woodworking skills and tools. The tools and woodworking skills required for a software is minimal.

    Table saw or cabinet shop to turn big lumber into the toothpicks you will use.
    Sand paper or hand plane to round corners
    Miter box and saw
    Hand drill and bits
    Clamps lots of clampers

    Rethink using wood
  5. clmanges
    Joined: Jul 2008
    Posts: 506
    Likes: 99, Points: 28, Legacy Rep: 32
    Location: Ohio

    clmanges Senior Member

    Good to see you digging into some information, but it looks like you're still on the steep part of the learning curve ...
    You're missing some important differences.
    Think of wood as nature's fiberglass: cellulose fibers bonded together with lignin. This is what gives it its strength. We can do a lot of things with FG (and other such materials) because we're not constrained by organic growth patterns in the way we arrange the fibers. Remember, epoxy is only there to hold fibers together; it's pretty weak all by itself. And PVC doesn't have any fibers, and likely no appreciable grain structure. Think of PVC as being more like concrete: good enough in compression, but lousy under other kinds of stress.
    I think I did, but it wasn't to make it thicker, it was to put on the fiber matrix. If you put enough FG over the PVC pipe, you'd get a FG tube that would be strong enough that you wouldn't need the plastic pipe any more. FG tubes are done this way all the time, but the piece they're laid over (a mandrel) is then removed and reused.
    Pretty much the same problem: no grain structure.
    Just to add here: structural metal tubes don't have a fiber-matrix structure either, but they're in a different class; they have other properties instead. Keep reading; you'll see.
    This may all be a non-issue, depending on the way you make the joints and what material you use, but for a SOF boat, the bamboo would be fine without the pipe. There are people building bicycle frames out of bamboo and they work fine. They're just ugly because the frame joints have to be wrapped up in a lot of ... FG and epoxy.
    Again, this stuff is made for plumbing pipe, not for structural applications. I plumbed a small house with it once and found that, on horizontal runs, the pipe doesn't even support its own weight adequately; it has to be supported with hangers every few feet.
    Excellent idea.
    Consider building the full-scale test copy and loading it up to working weight--maybe a little extra--then getting someone with another boat to tow it around. Just a thought.
  6. portacruise
    Joined: Jun 2009
    Posts: 1,275
    Likes: 102, Points: 63, Legacy Rep: 218
    Location: USA

    portacruise Senior Member

    "But PVC fittings fit together very tightly."

    These PVC fittings are not designed for repeated assembly and disassembly in a structural application (female part in particular may have a section with a thin wall compared to PVC geodesic type). Females are a tapered joint unit and will not hold together as a Ridgid unit (like a metallic press-fit) when jammed together purely as a friction fit. Any kind of stress like inline pull, vibration or twisting can loosen the fitting, unless it is glued together as per intended use for leak proof plumbing. The fittings will likely remain mated but not locked in position without glue, when the external tension from a tight skin forces them to remain together. But side loads may result in fracture at the joint, depending on design and external forces associated with jostling of the flexible skin while under Sail.

    Geodesic Dome Connectors (GeoHubs) https://www.sonostarhub.com/collections/geodesic-dome-connectors-geohubs

    "If only I could think of a sturdy way to thicken PVC pipe."

    Any of those ideas for thickening that work in your writing would increase either the weight and/ or the diameter of the PVC tubing considerably, in order to have the proper strength. That results in a distortion from your stated desire for lightweight design and compacted width dimensions.
    Last edited: Jul 25, 2021
  7. alan craig
    Joined: Jul 2012
    Posts: 331
    Likes: 97, Points: 38, Legacy Rep: 14
    Location: s.e. england

    alan craig Senior Member

    Mitchgrunes - stop talking and start making! You'll learn more.
  8. Skyak
    Joined: Jul 2012
    Posts: 1,323
    Likes: 76, Points: 48, Legacy Rep: 152
    Location: United States

    Skyak Senior Member

    Of these solutions, only #1 increases bending stiffness -the big deficiency of PVC pipe. The fiberglass will take all the load because it is so much stiffer and the PVC will just keep it from buckling. If the PVC isn't rough epoxy won't stick. You will still have the weak PVC problem in the joints. Fiberglass over wood is vastly superior to FG over PVC, simply add fiberglass cross grain where needed.

    I recall seeing PVC pipe used to make a child's kayak on thingverse. It was not a folder and not expected to last long. While you are looking up material properties, check out "plastic creep" -PVC is arguably not even a solid on a sunny day.

    I can't resolve your safety criterion with your use of incapable materials. Logically, the answer is to learn how to work with capable materials. The entire point of engineering is to make things that don't fail -it is the sensible alternative to trial and error.
    My philosophy is there is nothing so expensive as that which doesn't work. But it seems we have reached the Confucian proverb -those who say it can't be done should not delay those who say they will try.
    mitchgrunes likes this.
  9. BlueBell
    Joined: May 2017
    Posts: 1,840
    Likes: 529, Points: 113
    Location: Victoria BC Canada

    BlueBell WOT

    How long will this thread go on?
    The OP has no business designing a kayak given the dialogue thus far.
    "Talk's cheap until you get the bill."
    Build it and then you will be able to recommend to others not to.
    Take lots of pictures and video though.
    Good luck.
  10. mitchgrunes
    Joined: Jul 2020
    Posts: 110
    Likes: 17, Points: 18
    Location: Maryland

    mitchgrunes Senior Member

    Because I am new to boat making, do not have an engineering background or practical experience, and am experimenting with an infrequently used material, I probably need a more conservative design, with a larger margin of safety.

    1. Instead of a 16" wide boat, the width of my hips, I will try 21", which is the distance between the outside of my knees I need to attain reasonably high forwards bend body flexibility.

    2. I will stay with cheap, lightweight 1/2" schedule 40 PVC pipe and fittings for the early prototype (which will only be tested in flatwater), but will migrate to 1" schedule 80 PVC pipe and fittings for the final boat.

    Published tables gives the larger pipe a tensile strength of 727 vs 264 pounds. (The change does not similarly improve burst pressure, or collapse pressure - but I suspect tensile strength is more relevant, because kayaks sustain stresses in many directions, and loads are presumably spread. I hope I have guessed that right.)

    I am about 150 pounds, but sea kayaks must sustain other loads. The foredeck sustains several hundred pounds of pressure during T-rescues. I'm not sure of the loads from waves, but a larger safety margin makes sense there too. In addition, as people here pointed out, strengths vary by temperature. Long term exposure to cold, below about 5 deg C, embrittles PVC, according to many sources.

    3. Furniture grade PVC (also called structural PVC), is substantially more expensive, but not substantially stronger at first, though long-term, its inclusion of an integral UV protectant keeps it stronger. I hope to do the same thing using a spray-on UV protectant, as I often have for fiberglass boats.

    uPVC pipe, and other more rigid PVC pipes, weakens less with high temperature, but they are more brittle and less flexible - which as discussed below is a bad thing on lightly built kayaks. Many sources say you need the more flexible variety to avoid breakage under flexion.

    4. I assume the strength of glue bonds in fittings, and the durability of friction fit fittings, for a take-apart folder, is greater for larger pipe.

    For early prototypes, I will stay with friction fit fittings, reinforced by duct tape, but for a final boat, I would migrate to glued and screw-together fittings, so wear won't make them come loose.

    5. For rough final version weight, assume three 17' stringers, plus 17' for cross pieces and cockpit, then, based on pipe weights, add 7 square yards of 18 oz/square yard waterproof PVC fabric skin, to get about 28.8 pounds, lighter than my current 30 pound SOF, though I've ignored fittings, deck rigging, etc. Acceptable.

    6. Dyson discusses the importance of high flexibility in lightweight kayaks, to prevent breakage, and to decrease the resonant pitching problem. Fixing resonant pitching requires the ends to flex vertically by the resonant waveheight. For the fast boats he discusses he estimates a 1' resonant waveheight for very strong fast paddlers in oncoming seas against a headwind. That's for a class of kayaks which like mine will have little rocker curvature, and low volume ends. My Caribou (more rocker and higher volume ends; does not flex much) has the greatest problems in oncoming seas with headwind in 2-2.5' waveheights. If my design falls in between, that may be more than the larger pipe will flex under the load of my weight. We will see. (Tradeoff: flexion takes energy so reduces flatwater paddled speed, though Dyson speculates that it may be faster in waves because it may store and release energy.)

    In any event, I have now collected enough info for an initial prototype.
  11. upchurchmr
    Joined: Feb 2011
    Posts: 3,266
    Likes: 246, Points: 63, Legacy Rep: 579
    Location: Ft. Worth, Tx, USA

    upchurchmr Senior Member


    I've built 8 SOF kayaks, with 80z material.
    No need for 18oz material, the lighter skin will take everything except a sharp knife.
    There are videos with hammers, bats, etc beating on the finished kayak with no damage that lets water in.
    Save some money and weight.

    Water based house paint works fine also.

    Looking forward to pictures.
  12. mitchgrunes
    Joined: Jul 2020
    Posts: 110
    Likes: 17, Points: 18
    Location: Maryland

    mitchgrunes Senior Member

    I give up. o_O

    1. I'm not sure that PVC won't develop fatigue failures under the conditions I want to use it.

    If I assume the equivalent load riding over waves on the 1" pipe is on the order of my weight, doubled to take temperature derating into account, that is about 300 pounds.

    That's more than the psi rated pressure changes in figure 4 of


    So, it is reasonable to suppose a failure after about 10,000 cycles (wave passages).

    If I assume a mean apparent wave frequency of 5 seconds, that would occur in 50,000 seconds. So a few paddling days could be enough to create failure.

    I know that is a very inaccurate calculation, based on imperfect understanding, and mis-use of not-quite-applicable data. But it's the best I can do with my background.

    2. To meet Dyson's criteria to suppress resonant pitching, assuming a resonant pitch waveheight of 1' - 2', the stringers must repeatedly bend vertically about 1' - 2' over the 8' half-length of the boat. That's a 12.5%-25% deflection. I assume that in rough weather with larger waves, or crashing onto the beach in shore break, they might bend somewhat more. But some PVC pipe makers publish a maximum of 15%.

    Also, it isn't clear that the published maximum deflection is intended for repeated (fatigue) conditions. I need a large margin of error.

    I thought about using a cross fitting with 1.25"-1.5" openings for the beam stringer axis, and 1" openings for the cross-piece. Then the 1" stringer beam pipe could slide through the large holes, to enhance flexibility. That's analogous to the designs that Dyson discusses for traditional baidarka, in which the flexibility of the wood was added to the stretch of the connecting ligaments, and ivory ball bearings were used to allow the members freedom to slide past each other. (No doubt one could find a way to connect the pieces of pipe together with flexible material analogous to ligaments and ball bearings - but that defeats the idea of an easy to build design.) Besides the pipe would still have to bend a lot. I'm also doubtful that PVC is designed to take frequent abrasion.

    3. There are references in the literature to people being severely injured when PVC sewer pipe cracks under load. I don't want to be injured if and when the frame of my kayak breaks. Perhaps wood fails more gradually.

    4. I vaguely recall hearing from a bicycle store that lightweight aluminum bicycle frames don't last very long, because aluminum doesn't take a lot of flexing. (Of course, that might have been ultralight racing gear - which is often designed to last only a few races.) That might mean aluminum isn't a good material for a flexible kayak either. But I'm not sure - tent poles do have to bend a little.

    5. Perhaps some sort of flexible wood does make more sense. (BTW, various people argue that one must not fiberglass or use epoxy on the wood of a kayak designed to flex, because that would suppress the desired flexibility. They advocate other methods of waterproofing the wood.)

    Sorry to have troubled you folks.
  13. gonzo
    Joined: Aug 2002
    Posts: 15,623
    Likes: 1,108, Points: 123, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    The tensile strength is irrelevant for your project. Minimum stiffness is the main constraint. The second is weight.
    clmanges likes this.
  14. mitchgrunes
    Joined: Jul 2020
    Posts: 110
    Likes: 17, Points: 18
    Location: Maryland

    mitchgrunes Senior Member

    As I said I have given up on buildig it myself.

    I don't think minimum stiffness is a factor for the boat I think I want. As I said, I wanted to use what Dyson said about flexing the ends at the resonant pitch waveheight - probably 1-2 feet at each ends - to suppress the resonant pitching that has been such a problem in my Caribou. And it would have flex that much hundreds of thousands or millions of times.

    Given such flexion, I thought tensile strength might be relevant. Also, from what I have read, tension is the weakest direction strength for most uniform composition amorphous plastics, like I think PVC pipe is. Since I couldn't find clearly explained PVC strength ratings along other directions (some companies give a maximum external pressure rating, but don't say whether it is top load pressure, or uniform pressure), I had to use tensile strength as the minimum generic strength estimate. Though perhaps fatigue is more significant than strength.

    I deliberately chose 1" schedule 80 PVC pipe to be overkill in initial strength. But, AFAICT, while 1" schedule 80 PVC is strong enough, PVC pipe was never intended to be flexed that much, that many times. (It might even be too stiff to do it once.) In contrast, some trees flex a lot in the wind.

    I'm still tempted to build a very crude PVC prototype to test whether the size and shape I think I want work for me - and then to lend to a custom builder as a model for the final boat. That's nuts, and I wouldn't dare test the prototype in open water, but I haven't found a boat sufficiently close to the size and shape I think I want, to risk paying someone to build.

    I wish I had the knowledge, experience, tools and confidence to build the final boat myself. I doubt I will find an affordable custom builder who has worked with hyper-flexible designs like I think I want.

  15. gonzo
    Joined: Aug 2002
    Posts: 15,623
    Likes: 1,108, Points: 123, Legacy Rep: 2031
    Location: Milwaukee, WI

    gonzo Senior Member

    It is improbable that the kayak pitches at the same frequency as the waves. Even if it occasionally happened, it would only be when the waves are within a very narrow range. Also, boats are a damped system and quite complicated. The boat would also have to be at a precise angle to the direction of the waves.
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.