In my opinion, the most import thing is to prevent water intrusion, everything else is pretty easy to repair.  You seem to believe that damage to the fiberglass skin would be difficult to deal with, but I don’t believe that’s the case.  Given the options you outline, I’d go with the ‘bigger holes’ approach.

Peter Nyberg
Coventry, CT
s/v Silverheels (1988/2016)

> On Aug 2, 2020, at 7:39 PM, Chris on LBI <cknell at vt.edu> wrote:
> 
> This topic is a continuation of a different topic titled “"Repair of Spider
> Cracks at Mast Step Tabernacle”. The subject matter of that topic quickly
> strayed from gelcoat repair to how to replace the tabernacle screws after
> they pull loose. I thought it would be good to highlight this subject matter
> with a new topic heading. I thank Graham Stewart, Rodger Pihlaja, Mike
> Riter, and Rick Lange all of whom weighed in on the original thread.
> 
> Last week while lowering the mast, the three screws holding the tabernacle
> to the block above the cabin let loose when the mast was at an angle of
> about 45 degrees. A shroud had fouled and the resulting (unintended) forces
> pulled the screws out. In reading on this list/forum, I was relieved to
> learn that these screw connections were designed to fail to avoid more
> expensive damage to the plate and the cabin top. It also seems that the
> repair is very easy to accomplish. However, in asking advice regarding the
> repair and in doing some research, I'm undecided as to how the repair should
> be performed since I have conflicting information. I've started this new
> thread to focus specifically on reconnecting these three screws to the block
> below the tabernacle plate so that the sacrificial nature of the junction is
> preserved.
> 
> The three screws on my boat are 3/16” stainless steel, 1” long, with 10
> threads per inch. The mounting block below the tabernacle appears to be wood
> covered with fiberglass and gelcoat. The combination fiberglass and gelcoat
> appear to be about 1/4” thick. The stainless steel base of the tabernacle is
> 1/8” thick. The holes in the fiberglass are larger than the major diameter
> of the screws.  Based on this information I estimate that about 5/8” of the
> screws were threaded into the wood block. I read somewhere on this
> list/forum that the block is typically pine. Using an online tool 
> 
> (https://www.engineeringtoolbox.com/wood-screws-allowable-withdrawal-load-d_1815.html
> ) that calculates pullout force, I estimate that each of the three screws
> would pull out at about 70 pounds of upward force; 210 pounds total.
> 
> I had planned to make the repair by filling the holes with epoxy and then
> drilling holes in the epoxy to receive the screws. Since there was some soft
> wood around the perimeter of the holes, I scraped it out as best I could,
> and noted that the new epoxy “plug” would be larger in diameter than the
> holes in the fiberglass.  That is, some of the epoxy plug would be
> underneath of the fiberglass. This led me to wonder at what force the plug
> would release from the surrounding wood because if it did release, it would
> transfer the load to the fiberglass above the plug. Apparently the way to
> estimate this force is to take the vertical surface area of the wood/epoxy
> plug and multiply it by the shear factor of a wood/epoxy joint (obviously an
> estimate). A few sources online use 800 psi as this shear factor. So using a
> 5/8” deep hole with a diameter of 5/16”, I get a pullout force of about 500
> pounds per plug; 1500 pounds total. 
> 
> I assume that the pullout force of a screw from an epoxy plug is greater
> than 500 pounds but I can't find a way to estimate this. Although, if it
> were a steel pin (without threads), the shear factor would be estimated at
> 1600 psi. So with the threads the shear factor would be much great than this
> and thus the screw withdrawal force would be much greater than the plug
> withdrawal force. That is, the plug will pull out well before the screw
> pulls out of the plug. Unless of course it is a shorter screw. But without a
> way to calculate it, I would not know what length screw to use to achieve
> the desired pullout force.
> 
> While a few days ago I was all set to start mixing up a batch of epoxy, now
> I'm not so sure. Using these calculations based on the assumptions
> indicated, the planned approach would change the original design in two very
> significant ways. First, the total pullout force would increase from 210
> pounds to 1500 pounds. Second, with the epoxy repair, in the case of
> failure, the load would be transferred to the underside of the fiberglass
> which I assume the original design is intended to protect. That is, the
> original design sacrifices the wood/screw connection, the screws pull out,
> and the fiberglass is left unharmed.
> 
> So now I am left contemplating four different options: Wood Slivers, Bigger
> Holes, New Holes, Shorter Screws
> 
> Wood Slivers
> The idea would be to partially fill the holes with epoxy, and then pack the
> holes with wood slivers that would intentionally weaken the epoxy with the
> hope that the epoxy/screw junction fails before the wood/epoxy junction of
> the plug. That is, the screws would pull out without damage to the
> fiberglass. Obviously this pullout force can't really be calculated, but it
> seems like it should be less than the wood/epoxy junction of the plug, but
> there is some risk that the plug could pull out first and transfer the load
> to the fiberglass.  One very important unknown to this approach is the
> quality of the wood/epoxy junction. If the wood surrounding the epoxy is
> weak, the plug may pull out of the hole before the screw pulls out of the
> plug. This would transfer the load to the fiberglass. So there are risks to
> this approach. Filling the hole with epoxy and encapsulating the slivers in
> epoxy should prevent water infiltration.
> 
> Bigger Holes
> This idea is to drill larger holes in the fiberglass, so that if the epoxy
> plug were to pull out, it can without damaging the fiberglass. The holes
> would have to be sealed to prevent water infiltration, perhaps with gelcoat.
> The force to pull the plugs out is likely much greater than the original
> pullout force (1500 lbs vs. 210 lbs), but maybe this is not a bad thing; if
> a minor accident occurs (less than 1500 lbs), everything remains intact and
> I still go sailing that day. It would seem that this approach would protect
> the fiberglass as intended (although maybe a stainless steel rail gets bent
> or broken).
> 
> New Holes
> The original holes could be filled with epoxy and abandoned. Three new holes
> would be drilled in the tabernacle base plate and the block, being sure to
> make holes in the fiberglass larger than the major diameter of the screws.
> This would closely replicate the original design. Care would have to be
> taken to prevent water infiltration. It would be interesting to get
> confirmation from Stan that the factory drills holds in the fiberglass that
> are larger than the screw diameter.
> 
> Shorter Screws
> The holes would be filled with epoxy, drilled, and new, shorter screws
> threaded into the epoxy plug. The problem with this approach is that I don't
> have a way to calculate the pullout force of the screw from the epoxy. 
> 
> Since I am a novice at repairs of this sort, I again look to the list/forum
> for advice. I suspect that in a case like this, group-instinct trumps
> calculations.
> 
> 
> 
> -----
> Long Beach Island
> --
> Sent from: http://rhodes-22.1065344.n5.nabble.com/