thompsondockside.com

Hello all!

It's been a long time since I posted anything related to my boat. I won't go into everything that's happened personally or on the project right now, but suffice to say that I'm back on task and making good progress.

In short, I've completely given up on steam bending my new frames. I've gone to laminating them now. Of the 60 frames in my boat, I'm replacing all but the front four. That makes 56 new frames, or 28 matched pairs, and as of this writing, I have all new frames made up for (starting from the rear) frame stations 1 through 13, plus 15, 16, and 20. If you're counting, that means I'm well past the halfway mark! Here are a few of them in a pile.



I've had a lot of revelations, inspirations, and a few duds along the way, and I plan to do a fairly extensive write-up once I'm on the water. One of my brighter realizations was that these frames are matched pairs, as I said above. Idea #1 was that I would only have to set up my bending form once for each pair, then use it twice to make the two frames. Another light bulb then flashed, and I realized that if I could rip my laminating stock double width - plus a little for trimming - I could cut my glue-ups in half! That's what I've been doing for the most part, as shown here (photo taken after removing most of the clamps):



I clean up the squeeze-out of epoxy on the top edge with a block plane and on the outer face of the bend, and I can then rip the double frame into two single frames using either the bandsaw or the tablesaw. Finally, I clean up the new cut edges and ease the inside corners with a 1/8" roundover bit in the router. Here's that same one after some cleanup:



And here are two finished frames, a six-ply and a five-ply:




The rearmost 10 or so frames were done using six thinner plies, due to the more extreme bend that they had to take on. From there forward, I'm using five plies, and I may switch to four as I move on toward the bow.

My methods for ripping the laminating stock are another whole topic I'll cover later.

I just wanted to stop by and show some of what I've been up to!

Bill, looking great!

Those are some really nice looking frames.

Nice work Bill, and I really like the fact that you can make two frames at once. A few questions for you:

How are you capturing the shape of the frame? Is your boat upside down with planks off? Is it upright and you are taking the shape off the interiror?

Are you steaming the thin layers of wood, or bending them dry around the form?

Looks good!

What type of adhesive are you using?

Andreas

PS: Here are some white oak boat frames glued with phenol resorcinol adhesive:

Thanks for the nice comments... those are some incredible frames! What's that from? I also didn't notice at first that they're loaded on the back of a truck!

I only want to do this once (actually, I didn't even want to do it once! ), and I've read enough anecdotal evidence that some epoxies don't agree with white oak, possibly in the presence of salt water. So -- I bit the bullet and went with Smith's Oak & Teak (O&T) epoxy. It's expensive, but in the scheme of things, it's probably a drop in the bucket, right? I had initially bought a 2-pint kit, just for odds & ends work, and I've since bought two more 2-quart kits that I'm expecting to take me to completion.

The Smith's epoxy is supposed to be non-blushing, which is nice. It's also a 1:1 ratio by volume, which makes measuring very easy. The O&T version also has a fairly long working time of 60-90 minutes, but then sets relatively quickly (4 hrs) with complete cure within 24 hrs. It takes me about an hour working alone from the time I start mixing to the time I finish clamping. Now that my basement has warmed up a bit, I'm finding that my mix is just starting to thicken slightly by the time I'm coating the last surface. Not really gelling, but maybe just not quite as thin as when I start.

This O&T epoxy is also pretty thick in general. The resin is about like a thin syrup, but the hardener is more like molasses. Maybe not that bad, but close. One downside is that they don't sell "mustard pumps" for the cans it comes in. My solution was to go to the dollar store and buy a set of ketchup & mustard squeeze bottles. I later bought another pair made for glue from a local wholesaler of cabinetry hardware & supplies. These are made of clear/translucent plastic and a little bit bigger. My mustard & ketchup bottles didn't come with caps, but these did.

As I did each of the first several frames, I would draw a mark on the bottle so that I could gauge how much epoxy I was using each time. I could then calculate how much I'd need to complete the project. I'm measuring & mixing the epoxy in the little cups that come with liquid laundry detergent, and I'm spreading it with 1" disposable chip brushes. I trim the bristles down by about 1/2" for better stiffness and less waste. The cured epoxy pops out of the laundry cups, ready for the next batch.

Sure is nice seeing some new wood going into the boat!

Those big frames are glued laminated timber made at Peshtigo, WI by Unit Structures, Inc. The picture is from 1952. They were for a 165 ft. wooden AMS class minesweeper being built at Burger Boat in Manitowoc.

My father was in charge of the marine laminating work at Unit Structures in the 1950s. The laminating plant has been owned by him since 1972 and has been called Sentinel Structures, Inc. since that time.

The Thompson family of Thompson Bros. Boat Mfg. Co. were 50% owners of Unit Structures, Inc. from its origins in 1934 until August 1962.

The Thompson Boat lumber storage building can be seen in the background of this photo. The windows are covered up with white boards.

Andreas

thegammas wrote:Nice work Bill, and I really like the fact that you can make two frames at once. A few questions for you:

How are you capturing the shape of the frame? Is your boat upside down with planks off? Is it upright and you are taking the shape off the interiror?

Are you steaming the thin layers of wood, or bending them dry around the form?

Peter,

I was going back over this old thread and saw that I had not answered your questions! I think I started answering a while back but my browser crashed and I forgot about it after that. Hope it's not too late to help you, but maybe others can benefit regardless. Here we go:

1. Capturing the shape of the frame. This has been one of the biggest challenges for me, since my boat had nearly completely lost the shape that it should have had. I knew I couldn't just pull a frame out and duplicate what was there before. One of my first steps (after flipping the boat and removing the garboard planks... plus the first narrow strake on the starboard side) was to rebuild the keelson straight & true. The old one was hogged, bowed, sagged, and rotten beyond belief. Putting the new keelson in place gave me a "target" for the inboard end of the new frames to aim for. I felt that the sides and gunwales of the boat had held their shape well enough to use as a reference. It was the bend of the chine through to the keel that I had to recreate from thin air.

My first technique that met with moderate success was a shape duplicating jig built over Memorial weekend 2009. I joined a couple of legs of 3/4" x 4" birch ply into an "L" shape and drilled two rows of 1/4" holes. Through these holes I ran 1/4"-20 carriage bolts to hold a series of 1/2" thick plywood fingers, slotted for adjustability and secured with washers & wingnuts. A photo will help here.



I did two rows of holes simply for greater range of adjustment... you can see some fingers mounted to the inside of the "L" and some to the outside. Notice that some of the fingers are retracted completely where they're not useful at the time. From here, I would tighten down all the wingnuts, SHOOT A PHOTO, then carry the jig to my bending table and transfer the positions of the fingertips onto the form.

One issue with using this jig for capturing a frame's shape is that the bending form requires the inside curve of the frame, and this captures the outside. In most places, I'm also measuring to the outside of the planking, so in all cases, I had to subtract the thickness of the frame or the combined thickness of the frame+planking. Shooting a photo earlier would let me easily remember how much I needed to subtract for each finger. Nonetheless, this technique let me get close enough to the proper shape to make a go of it. In some cases, I made a new frame, stuck it in the boat, and then had to go back and adjust my bending form: in 1/4" here, out 1/2" there, etc. to achieve the final intended shape.

If you go this route, it's also important to remember that the planking touches the frame at the laps in most cases, but at the turn of the bilge the frame hits in the middle of a strake and the laps are floating a bit off the frame. You also might need a different plywood L for the bow and for the stern, due to the varying hull cross section.

2. Position of the boat. If you've read this far, you know the boat is upside down with enough planking removed to be able to get at the work area. I also had to remove as much planking as was needed to achieve a new fair curve. Some of the old planking was holding the old hogged shape and would have prevented me from achieving the right curve.

3. Inside or outside the boat? This jig could also be used on the inside of the boat by reversing the fingers to point to the outside of the L. From the inside, you could actually capture the exact shape of the bending form by putting the jig fingers directly against the inside curve of the existing frame.

4. Cold or steamed? One of the main reasons I went with laminated frames was to avoid steaming altogether. By using a 6-ply frame at the rear where the bend is the sharpest, and a 5-ply from frame 11 forward (counting from "1" at the rear), I knew by experimenting that I could easily bend the wood cold and glue it up. I was tired of firing up the steam box, waiting for it to boil, cooking the wood, bending (aka breaking) my stock, waiting for it not only to cool but also to settle (several days, from reading & from personal experience).

Other tips. Once I decided to laminate, my biggest challenge became obtaining my bending stock. My frames are about 0.57" thick by 1-1/8" wide. A 6-ply frame would need plies 0.95" thick, and a 5-ply frame required plies 0.114" thick. It happens that 0.95" is almost exactly 3/32", but 0.114" doesn't divide well in sixteenths, thirty-seconds, or even sixty-fourths. I spoke with some of my contracting suppliers: cabinet shops, custom millwork & moulding shops, and none were really set up to handle ripping stock thinner than 1/8". I was also personally loath to rip that much (approx. 300 pieces) myself. Until...

...At least four revelations hit. One was that I could use a thin-kerf circular saw blade and reduce my waste by over half. I had tried ripping oak with my regular tablesaw blade and its 1/8" +/- kerf. It just about killed me to think that over half of my lumber was being turned to sawdust, not to mention what I would lose through offcuts and other losses. A thin-kerf "framing" circular saw blade makes a kerf of almost exactly 1/16", and the center hole is the same 5/8" as a tablesaw blade. It's 7-1/4" diameter vs. 10", so you do have a reduced depth of cut. However, you're plowing through a cut half the width and with greater torque (smaller diameter blade), so the cuts go incredibly fast and are glue-ready smooth. My favorite blade is a 24-tooth Freud Diablo available at the big-box stores for $8-$10. I was well over halfway through my ripping before I felt I needed to replace it. It's probably still fine for general use in the circular saw, but not for ripping white oak.

The next revelation came in early 2009. It was about how to build a jig for advancing my tablesaw's rip fence for each cut. There was no way my eyes could handle shifting the rip fence 3/32" for each 6-ply cut, much less trying to guess at a 0.115" for the 5-ply ones. I had come across techniques for making plywood gears a while earlier, and I decided to put the technique to use for this jig. Rather than go into it here, I'll just post a link to a YouTube video I uploaded a while back. This thing works like a charm! See http://www.youtube.com/watch?v=Eotsqt5IcIY to get an idea of what I built. The video is a bit rough - I think I used my cell phone to record it, but it does show what I was able to achieve.

A third revelation was the fact that I could use oak stock of 2-1/4"+ thick and cut/glue two matching frames (port/stbd) at the same time. To rip this material, I did have to make a cut halfway through the board, flip, and cut again from the other side. There were some challenges in aligning the two cuts, but I've worked through those. More details on that at another time.

Fourth and finally, after gluing up a few frames in oak that were not quite right and having to re-do them, another idea hit. In engineering, we would often try to do a rapid prototype as a test or proof of concept before investing the resources to build the real thing. In this case, that meant using cheap wood and glue to test a new bending form setup before committing valuable oak & epoxy. I used my new ripping jig on the tablesaw to cut laminating strips from a construction-grade pine/spruce 2x6. I carried a stack of these and a bottle of Titebond II over to the boat. I spread glue on about five strips and, with a frame removed, I was able to shove the stack directly into the boat where the frame should go between the hull and the sheer clamp. Then I used Kreg screws (very sharp, self-drilling wood screws) through the original screw holes in the planking to pull the laminations tight against the inside of the hull and also tight to each other. On the keel end, I clamped the stack to the keelson and clamped on a temporary straightedge where the garboard belonged.

After just an hour I could pull out this new sample frame and use it directly to set up my bending form table. Again, I would typically glue up a cheap prototype frame of pine and make any adjustments needed before committing to my oak and epoxy.

While this may sound tedious or excessive, I didn't have to do it for every frame station. My technique was to do this once for every 4 to 6 frames, and then I could actually interpolate the bending form's shape between those. In the long run, it's quicker and cheaper to do a few pine prototypes with Titebond than to wait 12-24 hours each time with epoxy.

I hope someone out there can benefit a little from some of these tips. That's really what this forum is all about for me. Feel free to follow up with questions, and I'll try to answer!