Spindrift 12 build log

[Walt S.]

Guys,

 

After a hellish 6-7 weeks, I finally started making progress on the Spindrift again.  I put in the forward seat frame with quite a bit of extra-thick epoxy since the frame is shipped in one piece with the mast collar and the forward bulkhead beam and the whole piece was milled to be smaller so that you don't have to plane it to fit it.  There was a 3/4" gap between the frame and the bow that I filled with thickened epoxy.  Anyways, I set it in and went on another trip. When I returned, I found that the port gunwale had pulled away from the boat with some of the plywood fascia distorting the sheer line and causing the frame to settle in the distorted shape.  I'm debating pulling the whole frame out and resetting the gunwale.  

 

 

I'm not quite sure how this happened since the gunwales have been on for a long time with no problems.  My guess is that the fillet contracted as it cured.  I set it on in the morning so that the temperature was in the 60s but maybe it contracted overnight when the temperature dropped.  

[Wile E. Coyote]

If you clamp the gunwale back on does the shape return to how it was before?

 

Are your gunwales laminated, or solid one piece?

 

I'm amazed that the cured epoxy let go,  are you confident in your mix ratios etc?

 

I suggest contacting Graham for advice, this looks like a conundrum, but I bet it's not the first time this has happened.

 

Good luck,

 

Matt

[Chick Ludwig]

Looks like a pretty wide piece to bend around with a lot of stress on it. Looks like it "peeled" back from the bow. i wonder if the glue joint was "starved' of resin. Did you coat the wood with straight epoxy before adding the glue? If you don't do this, the wood could soak the poxy out of the glue leaving a "dry" joint.

[Walt S.]

Looks like a pretty wide piece to bend around with a lot of stress on it. Looks like it "peeled" back from the bow. i wonder if the glue joint was "starved' of resin. Did you coat the wood with straight epoxy before adding the glue? If you don't do this, the wood could soak the poxy out of the glue leaving a "dry" joint.

 

Chick, 

Yes, I pre-coated the gunwales and the plywood twice with straight epoxy before gluing.  The joint definitely has resin.  

 

It definitely peeled back from the bow.  Alan told me I was supposed to leave in the screws that anchor the gunwales to the breasthook.  I backed them out after 2 weeks because they were round-head and I hadn't countersunk nor put epoxy on the screws.  Welp.  I'm going to bend the gunwale back around and set it in with new glue and a counter-sunk screw.  

 

I'm not sure I could even get the forward seat frame loose even if I wanted to.  Given that the forward bulkhead temporary fillets head and so did the breasthook, there probably wasn't enough distortion to the bow to matter given the geometry.  I know nothing about naval architecture but it's probably better I just live with this.  

[Walt S.]

If you clamp the gunwale back on does the shape return to how it was before?

 

The sheer line seems to, yes.

 

Are your gunwales laminated, or solid one piece?

Solid

 

I'm amazed that the cured epoxy let go,  are you confident in your mix ratios etc?

Yes.  The epoxy was hard as a rock and required a router with an edge trimmer to remove the squeeze-out.

 

I suggest contacting Graham for advice, this looks like a conundrum, but I bet it's not the first time this has happened.

 

Thanks. I should've left the screws in.

[Alex]

I pre-bent my gunwales for about a week before fitting and had no problems at all. It is easy to do and I also did it for the other curved pieces inside the boat.

[PAR]

One of the images above shows a failure in the plywood glue line, not your epoxy glue line.

 

 

This would suggest the plywood had a defect in the offending veneers and likely just some bad luck. What plywood is this and where did you get it?

[Designer]

This ply came from us. We only use the finest European made okume ply made by the Allin company in France. http://www.mcilvain.com/new/wp-content/uploads/2012/09/BrunzeelFlyer_ipad.pdf.

 

I agree that the cross grain failed. There is a lot of stress at the end of bitter end of a solid gunwale on a Spindrift. When the first fibers failed, the load went to the next fiber which failed and it accelerated until the load started to diminish.

 

We devote a page to the knees in the plans. It says YOU MUST USE SCREWS OR THE JOINT WILL FAIL. 

 

The good news is that the glue is sound. This means that the gunwale can just be glued back and it will still be a good job. It will be more difficult now that he does not have the leverage that he had when he had the whole gunwale as he will be putting it on backwards. I would start by trying to bend it back in place with clamps between the gunwale and the side until I got to the breast hook. Fortunately the breasthook is just under 90 degrees so he should be able to get a large clamp near the forward end of the gunwale to the back edge of the other leg of the knee and put in the screws.

 

[Hirilonde]

One of the images above shows a failure in the plywood glue line, not your epoxy glue line.

 

This would suggest the plywood had a defect in the offending veneers and likely just some bad luck. What plywood is this and where did you get it?

 

Or it was simply stressed beyond reasonable expectation.  I had an issue with the forward section of my nesting Spindrift at the gunwale aft end of the section and it was Joubert ply.  I am skeptical to call it defective material.  It could have been rough handling.   Some times we never find the root cause. 

Graham just replied while typing, so maybe this is redundant.

[Walt S.]

This ply came from us. We only use the finest European made okume ply made by the Allin company in France. http://www.mcilvain.com/new/wp-content/uploads/2012/09/BrunzeelFlyer_ipad.pdf.

 

I agree that the cross grain failed. There is a lot of stress at the end of bitter end of a solid gunwale on a Spindrift. When the first fibers failed, the load went to the next fiber which failed and it accelerated until the load started to diminish.

 

We devote a page to the knees in the plans. It says YOU MUST USE SCREWS OR THE JOINT WILL FAIL. 

 

The good news is that the glue is sound. This means that the gunwale can just be glued back and it will still be a good job. It will be more difficult now that he does not have the leverage that he had when he had the whole gunwale as he will be putting it on backwards. I would start by trying to bend it back in place with clamps between the gunwale and the side until I got to the breast hook. Fortunately the breasthook is just under 90 degrees so he should be able to get a large clamp near the forward end of the gunwale to the back edge of the other leg of the knee and put in the screws.

 

Thanks, Graham.

 

The forward seat frame piece now isn't centered port/starboard.  Do you think it matters?  It looks like it stayed in place fore/aft.  

[PAR]

Agreed, possible just over stressed, though the rub doesn't look terribly bent on starboard. A Spanish windlass might help pull it back in, as will a bar clamp that's had it's top clamping foot drilled out for a small screw, which will dig into the wood and keep it from sliding.

[Walt S.]

Guys,

 

I tried bending it back by hand and it's no problem.  I'll sand the side and inside of the gunwale, epoxy a screw, put it through the gunwale slightly proud on the inside, and re-glue as before.  I'll bend it back by hand and use the impact driver to drive the screw into the breasthook. 

 

The gunwale had been in place for so long it is easy to bend. 

[Walt S.]

Guys,

 

I've been slowly filleting and taping the interior.  This is an older picture.  I've finished the forward two port/stbd compartments and am planning on doing the forward compartment tomorrow.  

 

My first attempt at filleting resulted in thinner fillets on the outer seam because I used a 1" diameter filleting stick.  I was thinking of sanding the outer seam and adding another layer of tape but Alan used a 1" diameter filleting stick on the CS15, so maybe I'll leave it alone. 

 

After I finish filleting, I suppose I could proceed one of two ways: flip the boat over, fair, glass, and paint then flip it back over and do the interior (seat tops, centerboard trunk, etc) -OR - I could finish the interior and then flip it over.  Does it matter which way you do it? 

 

 

[Chick Ludwig]

On a small boat like this, I'd probably flip it to glass it, then flip it back to finish, then flip again for final fairing, priming, and painting. You can avoid accidentally messing up your paint job this way. Do whatever feels good to you. I used a 1 inch radius tool on most of Turtler and Summer Breeze. 

 

Slip of the finger above---should have said 1 inch diameter.

[Walt S.]

On a small boat like this, I'd probably flip it to glass it, then flip it back to finish, then flip again for final fairing, priming, and painting. You can avoid accidentally messing up your paint job this way. Do whatever feels good to you. I used a 1 inch radius tool on most of Turtler and Summer Breeze. 

 

Thanks.  I think I'll do what you said.

 

My tool on the first set of fillets was 1" diameter, not radius.  Alan used the same size on his CS15.  Maybe I need another layer of tape. 

[Designer]

This method of construction is very forgiving.  If the fillets are too big you just added more weight and used more epoxy without gaining much. 

 

I like to make sure that the thickness across the filleted joint is slightly thicker than the ply thickness. The glass tape gives the tensile strength and the fillet fills any gaps and thickens the cross section of the joint gaining stiffness. I feel that if the joint is thinner than the ply you could get some flexing at the joint which could fatigue over time. The glass tape on each side adds to the joint stiffness as well.

 

As it is not possible to make a perfect fillet you could err to slightly thicker so that your minimum cross section is slightly thicker than the ply after you radius the outside. It is desirable to radius the outside of the joint because it is impossible to glass around a sharp corner without getting air bubbles under the glass. The rounded over corner makes the chine joint less susceptible to damage and just looks better. A 1/4" outside radius is about as tight as you can glass without a fight but 3/8" is better still and is more tolerant to less than perfect workmanship. 

 

As the angle of the joint gets more obtuse like as you work forward toward the bow you can lay the filleting tool over in the direction you are working and you will increase the cross section a  bit. You can change to a larger tool but you will get to where the the angle of the side bottom angle approaches 180 degrees, just blend out the fillet altogether.

 

Here is a drawing of the joint showing that with 1/4" ply, 3/8" outside round over and a 1/2" radius filleting tool you have a cross section of .32" which is just over 5/16". Of course it will vary as the angle changes but it is a good average.

[Chick Ludwig]

Sorry Walt, I meant to type "diameter".

I also do like Graham said above..."you can lay the filleting tool over in the direction you are working."

[Walt S.]

Graham,

 

Thanks. I'm inclined to just leave it alone.  The radius of the fillet is slightly larger than 1/4", so with tape on both sides it should adequate thickness.  The other ones are bigger.  If I  build another boat with the boys, it will be better.  

[Designer]

I think that you are okay.

[PAR]

For a consistent and decorative fillet, use a ball of the appropriate diameter, possibly on a stick. I have a few HDPE balls on a length of 1/4" dowel that I use to clean up fillets, like around the perimeter of a cockpit sole, for example. The primary fillet is made first, but a top coat of fairing compound is mixed and a light, thin, uniform layer applied over it to smooth out lumps and bumps. The ball doesn't change diameter as you drag the tool, like a dowel or tube will. With some practice and possibly tape bordering the fillet edges, you can be pretty neat and precise with this technique.

 

 

All the inside corners shown used a ball on a stick, except the centerboard case pivot plate (used a gloved finger) and the back of the centerboard case. You can see they're all the same size, uniform width, etc., etc., etc.