We All Make Mistakes  
By Eric Staggs - Spokane, Washington - USA

Well, yes, it’s a beautiful design. I need it to do this, but it’s designed to do that. Oh, well I can just modify it to suit myself, right?

We have all fallen into that trap (haven’t “we”? Oh great, am I alone again?), second guessing seasoned designers and their collective talent. I fell deeply into this thinking I could easily, and without difficulty, make a fixed seat rowboat into a slider. Why not, right? It’s a simple conversion, just a little wood here, some fasteners there, sure that oughta hold. CRACK!

click to enlarge

Thames Rowing Skiff, Selway-Fisher. Want to know more about her? Article coming soon!

click images to enlarge

My wife wanted a sliding seat rowboat to change up her workout routine. How can I say no to a beautiful woman wanting to keep herself in shape? I designed what I thought to be an effective sliding seat setup by stealing some items from a Glen-L set of plans, some measurements out of shells I had rowed, and a little shadetree squint-and-build. While they actually are great sliding seats, I ran into a problem with the scantlings. I had epoxied the vertical supports to the 4mm ply thinking that it would only be subject to compressive stresses while being rowed. Who can argue with that logic! Hm, oh yeah, Mr. Murphy himself.

click to enlarge

After sanding the area, before damage

I had neglected to remember “compression only” when I built a small roller to lift the boat onto the dock (which was damaged by ice and is now a full 18” above the water line). Hey, just pull this up, tip it down into the cradle, I’m a genius! CRACK!

What happened later was the vertical bracing, being attached to the horizontal framing as well as the main frames in the design, were very strong. Stronger than the lamination glues in the BS1088, as a matter of fact. Over several launchings and retrievals, I managed to pull apart 2 small pieces of the ply right at the uprights. Its not a major issue as it isn’t structural, causes no rowing abnormalities, and they only separate when the boat is being launched or retrieved from the dock.

click to enlarge

After repair and first coat of epoxy/'glass

Now this is a curiosity of mine. Why not just leave it, it's not harming anything, just a little paint and glue… Have many of you fallen into that trap? Neglect it until it becomes a problem? Bah, if it gets bad, I will worry about it then. I had fallen too deep on that in another boat, so I pulled the trigger on a repair. After seeing how easy it is, I highly recommend just making the necessary repairs over whatever time is needed.

click to enlarge

Foot stretcher rails as sanded

In the photos you will notice I sanded the paint off to bare wood, added a small piece of fiberglass to spread the load a bit more, and solidify the damaged area. I also included a fair amount of epoxy to close the wound and prevent decay. Hey, look I have the sander in my hand, why not do that foot stretcher lift she asked for? Off I went again, sanding more paint, fabricating a set of blocks and new runners, and mere moments later, we have lift! Well worth the time.

click to enlarge

And being glued up

Without a doubt, just get it accomplished. I haven’t actually finished the project (still needs paint, as it took 2 days for the epoxy to dry in the balmy 39 degree F garage with a heater to assist), but just wanted to inspire someone, anyone, to take that nagging little error, correct it, and extend the life of your project well beyond a minor area of rot turning into a major project or a moderately sized bonfire.

Oh yes, as far as the boat? 17’ Thames Rowing Skiff, Selway-Fisher. Want to know more about her? Article coming soon!

Eric Staggs
Spokane, Wa




Stainless Screws