Designs for a Strong Autopilot Drive Mount / Shelf?

[beetle]

The autopilot ram will place significant loads on whatever it is attached to. The ram manufacturer should have published maximum axial load. You can determine load placed on the supporting shelf by measuring the height of the lever arm (distance from mount base to where the ram attaches the the mount) and apply max axial load on that lever arm. An alternative is to look at the ram mount thru-bolt diameters, most likely stainless steel acting in single shear, and work out the breaking points of those bolts - that will tell you how strong the attachment shelf/mechanism those bolts attach to needs to be.

From a design perspective, a shelf will be strongest whem the shelf plane in the same plane as the ram (keep all loads in shear across the plane of the shelf). A flat plate shelf will be weakest when loaded is applied at 90 degrees to the shelf, and strongest when load is applied in the same plane as the shelf.

Plywood is a good core material, but not terribly strong. Also think in terms of what that shelf is going to attach to - all the load absorbed by the shelf is going to be transferred to the hull in some fashion. Spread out load as much as possible. As an example, if the ram can apply 1500 pounds of force, and you can spread that force over 100 square inches then you're asking each square inch to handle 15 pounds. That's a fair bit of force to deal with - the more you can spread the load the less likely the shelf and attachment structure is to fail.

Do NOT orient the ram such that it attaches squarely to the side of the hull - this places all that load at 90 degrees to the relatively flat hull panel, and a flat panel is least strong when loaded at 90 degrees (same consideration about keeping loads in plane for the ram's mounting shelf applies to attaching the shelf to the hull).

I built my autopilot ram shelves using 3/4" marine plywood core with 85 ounces of fiberglass top and bottom (as if building a miniature cored boat deck) - simply 5 layers of 0-90 glass and epoxy top and bottom of the shelf. The ram orients fore & aft, the shelf is glassed to the hull stringers and not the hull itself. Make absolutely certain that the plywood core does NOT touch any part of the boat - this will lead to hard points and cause flexural failure of the hull as it flexes and the plywood does not. Place a 1/2" layer of structural foam (divinycel is san example) between the plywood and the hull, the fiberglass shelf skins cross the foam and adhere to the hull - all the load is taken by the fiberglass skins.

Those are some thoughts - first thing is to sort out what the loads are you're asking the shelf and hull attachment to handle.

- rob

[SailRedemption]

Originally Posted by funjohnson

I wouldn't wait for the fillet to cure before applying the glass. If it's cured, then any imperfection in the fillet, unless filled, will create a void in the glass. If you glass before the fillet kicks, you can lightly push the fiberglass along the fillet and get a perfect bond on that critical joint. The bigger the fillet the better. Stick with your cloth, there is no need for mat with your epoxy.

Treat the plywood like a core... make it like the bulkhead next to it.

I hate to be that guy, but If you haven't already, you're going to want to address those thru-hulls while you are in there.

Matt

I agree, I only mentioned let it cure first because it looks like it's on a slant and it's easier to glass something that may not want to slide or sag when you try to put glass on it. But yes, definitely better to glass before it fully cures, great point either way. He could wait until the epoxy starts to cure when it's stiff but not hard and you would still get the chemical bond between the glass matt and fillet.

[beetle]

Originally Posted by SailRedemption

He could wait until the epoxy starts to cure when it's stiff but not hard and you would still get the chemical bond between the glass matt and fillet.

Epoxy resins take about a week to lock up, until then they are considered 'green' and will establish a chemical bond with the next layer of epoxy.

If you only wanted a mechanical bond (as you point out, a mechanical bond is significantly weaker than a chemical bond) you'll want to wait a couple of weeks for the first layer of epoxy to completely cure before applying the next layer.

Part of the problem is that any epoxy going onto the hull from the shelf is only going to have a mechanical bond (unless you built the shelf into the boat while building the boat), so I wouldn't worry much about whether or not the epoxy and glass on the shelf has (or doesn't have) a chemical bond to the fillet - either way it's only a mechanical bond to the hull material.

- rob

[Tessellate]

Great advice here! If anyone else has more pictures, keep them coming.

A couple new learnings:

- SailRedemption makes a good point - I did sand the hull attachment points first but didn't do so nearly enough. I got to "speckled" fiberglass, which is paint and fiberglass rather than sanding down for a longer time.

- I used Home Depot birch plywood, which I stupidly realize now is junky plywood. I did my best to select the piece with the fewest voids and knots, and cut out the best section, but there's a big difference between that and quality marine ply.

- Both Halien and beetle suggest using balsa or a structural foam between the plywood and the hull surface, to alleviate hard spots. How do you do this? How do you cut it and how do you shape it? And are we talking only about a 1/8" buffer?
It's hard for me to imagine the hull flexing a lot across a 4"-12" section, but I guess it's possible.

Has anyone made a shelf out of G10 fiberglass board? Expensive, but it might allow me to use 1/2" thickness instead of 3/4". A problem I'm dealing with is my installation space is so tight that I didn't want to build up too much fiberglass on top of the shelf because it could cause the drive arm to hit the cockpit floor. I have only ~1/4" clearance between the drive arm and the cockpit floor. It may be the tightest below decks autopilot install ever - the space around the rudder stock on this boat is pretty terrible. So I'm trying to make the best of very limited space.

[beetle]

Originally Posted by Tessellate

How do you do this? How do you cut it and how do you shape it? And are we talking only about a 1/8" buffer?
It's hard for me to imagine the hull flexing a lot across a 4"-12" section, but I guess it's possible.

1/8" is too little, 1/2" is about max you'd use around the main bulkhead to contact the hull. I went with 1/2" as it is easy to work with and convenient to purchase. 3/8" would work as well.

To do the work, cut lengths of your foam (I use divinycel) in flat ribbons the thickness that you want (3/8", 1/2"), the height should be the thickness of your core plus the radius of your fillet - you're going to 'cut' the fillet radius directly into the foam core material after the shelf and core are bonded to the hull. I use 1/2" radius for the fiberglass bend is nice and use that where possible, 1/4" radius will work, I wouldn't go much smaller radius than that (the entire point of the radius is to avoid stress risers in the fiberglass layup that will create localized loading - a 90 degree bend is exactly what you want to avoid).

Take your plywood core, fit it to your attachment points (hull, stringer, where it goes), take your core to a bandsaw or saber saw and remove the thickness of your foam. Take your plywood and foam back to the boat, re-install to check fit. The foam will bend relatively easily, especially the thinner it gets. When you're happy with the result, spooge (semi-technical term for epoxy and microballoon mix) the foam to the hull, the plywood to the foam (usually in one pass, it's easiest), let harden.

Next day come back with your Dremel tool or other favorite radius sanding device (drum sander on a flexible drive shaft attached to a low RPM 1/2" drill is ideal) and carve the radius into the foam. If the foam height was correct and you spooged the plywood to the foam on foam-centerline, no further epoxy fillet work is necessary - that takes some care and practice.

Then do your glass laminate layup over the plywood/foam/hull.

Originally Posted by Tessellate

Has anyone made a shelf out of G10 fiberglass board?

Yes. G10 is total overkill as a core material as regards weight, cost, and structural qualities. G10 as core needs to be grabbed both sides, simply glassing to one side without encapsulating it is asking for failure as G10 is enormously stiff compared to a hand-built fiberglass layup - the G10 will crack away from the glasswork. G10 is not the best core material due to stiffness differential between the G10 and the encapsulating glass - G10 is sometimes used as core material for shroud chainplates when wrapped in carbon, otherwise G10 is a good backing plate substitute for metal but not so good as coring material. You could thru-bolt a G10 shelf to a fiberglass 'L' bracket if there's absolutely no space for any glass-work to encapsulate the G10 - in this case all the G10 is doing is substituting for a metal plate; I would use metal for an autopilot if bolting the shelf in place - G10 is softer than metal as regards how stainless steel bolts can wallow-out drilled holes.

What you might consider is going with thin core mateiral (1/8" plywood) and thick fiberglass work on each side of the core in order to minimize the total thickness of the resulting shelf. The core does nothing except make the overall glass laminate lighter for a given stiffness (the core separates the skins to make the laminate sandwich stiffer for a given weight); no one says you need any core at all - though if you're doing a laminate you would like something to support the weight of the fiberglass & epoxy while it is curing.

Originally Posted by Tessellate

I have only ~1/4" clearance between the drive arm and the cockpit floor. It may be the tightest below decks autopilot install ever - the space around the rudder stock on this boat is pretty terrible. So I'm trying to make the best of very limited space.

Is there any chance you can rotate the ram mount to face the underside of the cockpit floor and bolt it to the floor? Could you rotate the ram around so it faces forward - or off at some interesting angle that gives you more vertical room to work while still remaining at 90 degrees to the rudder stock?

- rob

[Halien]

I used 6mm of balsa strips cut to fit and glued to the edges of the plywood. I sealed them with epoxy, tacked the mount to the hull, added fillets, and then the tabbing. In my installation the balsa acts like a standoff between the mount and the hull, so the ram's force is transmitted to the hull through the substantial tabbing rather than the edges of the mount itself. The foam is a better choice if you have it, but balsa does the trick too.

My son in law is a math teacher. I don't remember the numbers, or even how he got to them, but transferring the load through the tabbing rather than the edges of the mount spreads the load over a pretty big area. I've used my contraption in all kinds of weather for a couple of years with no problem. The only thing I worry about is a situation where the rudder is being bashed on rocks while the ram is trying to steer. That would suck, but that situation is an outlier.

Hope this helps.

[RaymondR]

Don't muck about with more fibreglass, get a piece of 1/4" alluminium cut the same size as the plywood shelf and glue and screw it down onto the plywood. Use big SS washers on the underside of the plywood. I use an epoxy putty from the plumbing department of the hardware to glue alloy down. It is sold as a putty to seal the joints of cast iron sewage pipes.

[Tessellate]

Originally Posted by beetle

... () ...
- rob

Thanks for taking the time to write such detailed advice, very helpful!

There's no way I can rotate the ram that much unfortunately. The motor casing is quite large and I'm in a pretty confined space.

I could do a stainless steel angle bracket like in my 2nd photo in post #4, but I suspect the bulkhead wouldn't hold up to twisting / compression force in that direction.

[olaf hart]

I just mounted mine upside down under the cockpit floor, four SS bolts and some mastic and good to,go..