Compression Post Design

I made a post a few months ago about the design considerations of converting the deck-stepped rig to keel-stepped on my Swiftsure 33 restoration project. After some careful thought and several conversations I've decided I would prefer to keep her in the deck-stepped configuration. The reason for thinking about this, however, is that due to improper sealing of hardware at some point, the compression post has rotted out and needs replacing. In the original construction, the 2 main bulkheads are fiberglass laminate cored with 3/4 in balsa. The aft bulkhead makes a 90 degree radiused turn to forward along the centerline, and another piece of 1 in plywood is glassed next to it directly below the mast on the port side. This seemed relatively scant to me upon first inspection, but I believe the load was well distributed between the bulkhead and doorframe for the head just forward of the mast step, as there is no sign of deformation after 60+ years. The only rotten section is directly below the mast, so I expect to cut out the ~10 in wide vertical section and sister in a new support. I could go to the local lumberyard and get a quote for a hardwood post, but I have plenty of marine ply offcuts already laying around that I could epoxy together to make a new vertical beam. Assuming this would be well tabbed back to the existing structure, what thickness should I be aiming for to have adequate buckling strength? There isn't much room for anything other than a butt or short rabbet joint so I expect it will need to be a bit thicker than the original to account for the stress at the joints.

 

My first thought is; do not use plywood for a compression member. Ply has fairly good grain structure on the surfaces but there is no telling what you have between the plies. Any laminated ply post would need to be outlandishly large to have reasonable confidence in the ability to carry the load over time. Ply will, sneakily, get interior wetness which affects strength dramatically. Other woods absorb moisture but not nearly to the same extent as ply.

To speculate on the dimensions of the solid wooden post, you need to make a pretty good guess about the actual compression load in worst case scenario such a really wicked sea states, and then add a generous safety factor. A variety of woods have a variety of compressive strengths. You are interested in the compression loads that apply to loads imposed on parallel grain structures. That is to say from end to end of the log.
Shear strength of high quality straight grained coastal fir is 800 psi when dry and 380 psi when wet. Longleaf yellow pine...1510/1040. Note that this is shear strength, not the same as compressive strengths which have much higher numbers.

To make the guessing game a little bit less mysterious, consider using a heavy walled, aluminum tube with end plates to spread the loads. With aluminum you can use slenderness ratio figures to select a suitable size. (Only if you have a good estimate of ultimate compression loading) Ally has consistently predictable compressive strength. In addition, it does not absorb moisture. Drill small weep holes at the bottom of the column.

 

Hi Charlie,
-Does the sister-ed plywood piece go all the way along the aft-to-fore wall of the head? Thickness of glassing around plywood?
-What are the dimensions of the mast section (length, width, thickness)?
-Is the headroom about 5'-6" (1.675m)?

Hi Forum,
-Is there a method to calculate or estimate compressive load at mast base based on rig particulars and sail area?

Seems like this one is a 'three trains left Chicago' sort of math problem. It is not a pure column in compression as it is braced by the bulkheads.

Column Capacity Calculator http://www.timbertoolbox.com/Calcs/columncalc.htm is an interesting calculator but you have to go find tables for the species of wood in question as Messabout mentioned.

 

Well my thought was that there are several examples of these boats still sailing with the original balsa/ply compression area intact. The original section is not free-standing, and the load is distributed between the aft bulkhead, head doorframe, head sink countertop/cabinet, passageway frame, and coachroof sides, hence my assumption that digging out the rotten core and replacing with a bit thicker section of ply would be adequate. I may attempt to use an aluminum tube as you stated and simply box it in with ply to get the original look, as the curved bulkheads are very distinctive and I would like to keep the look of them.

I haven't done the calculations for my particular boat but a quick google search yielded a shroud load of 6,000 lbs for a similar design. What would be a good safety margin to account for dynamic loads to translate into compressive force? Once I have that number I can just do a simple FEA simulation on different tube sizes to find a proper fit.

 

Tops,

Yes, it forms what is essentially a longitudinal bulkhead between the 2 mains along the centerline. In the sailboatdata side profile schematic, the blank wall surrounding the door to the head is one solid piece glassed to all surrounding structure, with an extra support below the mast inside the head of 1 in thickness. The lateral sections of the bulkheads are all intact and dry, and it is only this longitudinal part that needs addressing. I can measure the glass thickness next time I'm at the boat, but if I remember is less than 1/8th of an inch. Maybe thicker of the port side around the additional support as I have not cut in from that side yet. I will also have to get the mast dimensions, but it is a Lefiell spar from mid 90's, not sure of that has a standard dimension. Also, the distance from the coachroof to the bilge floor is closer to 5' 9".

 

Charlie, do these pictures come close to what is going on?

 

This is very close yes! A couple changes, there is no core in the coachroof, it's solid laminate about 3/8ths thick below the mast. Also, there are a few inches of gap between the lateral bulkhead and the joined ply support, it is attached only to the longitudinal bulkhead. Accounting for these changes, the two longitudinal core sections are exactly what needs replacing. I am guessing you are looking to run a load simulation? I can get exact dimensions for the mast section and laminate thickness by tomorrow evening. To be clear, my initial intention of using plywood was simply because I have some long and narrow offcuts sitting in my garage that would be a perfect fit in this application. If you are doing a simulation, I would love to know how it would compare to some commonly available boards. After looking through my photos I've ruled out any chance of a free-standing metal tube, as the placement of the keel bolt nuts wouldn't allow for a suitable plate in the bilge. I have about an inch to spare on either side of what you have in this schematic.

Also, can I ask what software you're using for this

 

The software is Autodesk Fusion360 Personal/Non-Commercial.

I am hoping the pictures add clarity to the discussion and am happy to change a couple dimensions to make them more realistic.

I am not qualified nor experienced in load calcs, sorry. I am curious what a 6 foot piece of the mast could bear in compression.

Since the original lasted so long, wondering if sound repairs and simple reinforcement are adequate versus a complete re-invention. Scarfed and beveled joints, good materials, proper working temps for glassing, fiberglass caps on the cores and resin-filled mounting holes...

 

Laminated ply posts are a good option. Many companies, like Gluelam and Roseburg, manufacture them. One of the advantages of engineered wood products is that the mechanical properties are published by the manufacturers, and take the guesswork away.

 

The configuration shown in post # 6 is more correct than the one in post # 8, which seems prone to bottom panel buckling.
The materials normally used for the core of sandwich laminates perform poorly in tension and compression. They are not suitable as supports for the mast. If the board under the mast seen in the picture is made of marine plywood, it may be correct but it would be advisable to also place vertical reinforcements to prevent the panel from buckling (not a double plate but a vertical reinforcement, perpendicular to the panel).
There is no problem in using FRP, its mechanical properties are perfectly determined as long as the supplier gives the actual fiber content of the material.

 

For clarification purposes, this is pretty much a perfect representation of what is going on. I have some limited design/FEA experience in solidworks, and I can get a free Fusion360 student license. If you would be willing to share this part file with me, I could input some more accurate dimensions and try to run some sims on this.

As for the design, that is what I was thinking as well. If I replace the ply and balsa cores with just ply and proper glass work, it should be stiffer than the original, and with proper sealing of the edges and hardware, moisture should not be an issue. Though, perhaps for the sake of peace of mind, it would be better to replace the balsa with ply, and use a solid board for the additional support on the port side. If I can get a simulation working I should be able to try it with some different materials and see what the results look like. I could also see what kind of difference it would make to close the gap between the support and lateral bulkhead. Additionally, if I relocate the sink to the center of the countertop, I could glass in some shelves between the later bulkhead and mast support, which should greatly reduce the chance of buckling.

 

TANSL

Thanks for your input, see my post above for my thoughts on stiffening the design pictured in post #8.

 

I am not sure what you intend to do to strengthen the panel so if I am wrong, please excuse me. If I have understood correctly and what you say is to place several shelves as in the picture, these shelves will not give you any additional resistance against the buckling of the panel.

 

Oh that's my bad. I meant to imply that I would add the vertical support perpendicular to the panel as you described, and THEN add the shelves to stiffen everything and give a more seamless look.

 

Charlie, I can give you step (*.stp) data right away, it will be solids and not fully parametric. You might be time ahead to start your own sketches since I did not have all actual dimensions in hand while drawing. I tried getting to Solidworks part data but my work software is too old to make the conversion from Fusion.