: You will be able to do harder bends with coated
: wood.

be interested why you say that. i have only done one way tests with bare wood vs what i call thin seal coats and found insignificant differences but with only about 2 samples each for 2 diff ply types at 1/8" thickness the sample size is small.

in subsequent construction, i have been thru 5 or 6 catastrophic brittle panel failures as well as 3 or 4 partials, but i am guessing but fairly sure would be fairly similar in unsealed ply. in other words the ultimate is very similar with thin sealcoats.

however, if better(thicker) sealed/coated ply only was used i just do not know what the results would be, but if i was to use a thicker coat, i would just include some .75oz or 2oz cloth and i am fairly sure that altho stiffer, would allow some max bends for the similar thickness. i didn't take it that far on one yak. on the other which had severe twist of abt 45-60deg over 8" or so, i just didn't think was possible without glassing the outside of this area and removing the 2 inner veneers. that turned the ply into almost cloth even without heat.(the end idea is to just add a few more layers of glass on the interior of this area to bring it back to the adjacent stiffness/thickness ).

so basically, if i was to use a seal coat of any real thickness, the advantages of including a cloth layer also just seem too big to ignore. but i have not tested that to know.

but hell i'll guess abt thick seal only:
- for an inside curve where the thick sealed surface is on the inside, as the fibres are filled and resistant to compression, there will be quicker catastrophic failure of the outside unsealed fibers than if thin sealcoat. if a cloth layer is used, there will still be catastrophic outside failure, but the inner sealed/glassed layer will not be broken stay integral.
- for an outside curve where the thick sealed surface is on the outside, as the fibres are filled and stronger in tension, there will be a slightly tighter bend than the unsealed condition. possibley a bigger more forgiving bend if glassing used and especially if the panel perimeter is had sharp or discontinuous profile curves.

and if strapping is used, with careful placement and thinnish panels it wont matter a damn what thickness of sealant is used. and if glass is used as well, you'd get the best, most predictable result.

so for me, wherever i knew that about the max of the bendability was to be reached on 3-4in wide panels, i double strapped the outside. like when i was getting repeated failures, i brought in the strapping. was real quick, probably should have done it as a matter of course.

however if maximum twist is involved, it is a much harder call. most twist failure i have been getting propogates from the edges. and thru a twist the edge undergoes both convex and concave bends that strapping just won't follow. one could make u-shaped metal straps to cover this, but if complex outline curvatures are also part of the panel perimeter, even those straps won’t be able to give a consistent edge. so under these extreme case, most likely just glassing the edges would achieve the restraint objective desire. a guess. if waterbased glue is used as the matrix, could get the bend and then remove – i’ve done a little of that. or more simply just glass the whole area in question and see if that would work. i took the easy way out with the inner veneer removal - no guts!

anyway, some ideas
-mick