8-foot Culler Oars

I took a notion to build a pair of oars when I found a whole pile of nearly-clear 2x8 spruce lumber at the local building supply store (Kent). This is the type of material Pete Culler suggests using in his book "Boats, Oars and Rowing". 

The two boards I got looked great when I picked them up and so I loaded them into my wood stack (beneath some other stuff) and hoped that they would not develop a severe twist. This hope was dashed a few weeks ago when I pulled them out to find them not only cupped, but badly twisted. 

I decided that for a first set of oars, I'd just wing it with the available materials. A good trick for straightening twisted boards is to face laminate them such that the twists and cups oppose each other. My boards were symmetrically warped (with cups facing each other), which made them ideal candidates. I face-laminated the boards using WEST 105 + cedar wood flour filler, then clamped it all up and hoped for the best. 

Figure 1: Clamp it up!

After un-clamping my boards, I was very pleased to see that they had straightened almost perfectly and possessed no discernible twist or cup. I squared up the edges with a plane and snapped a chalkline to mark the centre of the lamination. The line was offset from the glue seam by about 1/8" at the greatest deviation. Not too bad!

Figure 2: Checking the edge for square. A ways to go yet on the left. 

Figure 3: The chalk line is a good way to check for straightness.

Next step was to produce a template from the dimensioned drawings. Culler's oars have a tapered shape along the loom from the grip to the throat, and then (obviously) they widen again at the blade. It's important to make the template symmetric since it is going to be traced on 4 sides of 2 oars and used as a reference. I grabbed some 1/4" MDF and cut out the template on the bandsaw. 

On the narrow side of the oar, the taper is straight from the grip all the way to the blade tip -- for reasons to be explained later -- and therefore no template is necessary. 

Figure 4: Template, showing approximate scale on the boat hull

I decided to cut both oars from the same stock and marked out the templates opposite each other on the face of the lamination. Next the oar blanks were sawn apart using a hand ripsaw I picked up a couple of years ago. 

Figure 5: Clamping the template in place

Figure 6: Separating the blanks with a ripsaw

I set up the bandsaw with a 3tpi ripping blade and set to work rough-shaping the oar blanks. This took a considerable amount of time on my 3-wheeler bandsaw, but it was a hell of a lot quicker than trying to do it with a chisel and saw cuts. I even managed not to break the blade! I think the secret when ripping is to just let the blade do the cutting and don't try to feed the material too hard. If you feed to hard then it just overheats the blade and risks breakage or scorching the wood. When this step is complete, there is still a lot more wood to remove, but at least things are starting to look oar-shaped. 

Figure 7: Three tooth per inch rip blade all set up. 

Figure 8: First cut complete

Figure 9: All cuts complete.

To get the taper on the narrow side of the oar, I lined the blanks up on the bench, calculated offsets from the centreline 
(by hand like a fuckin boss) and marked this out.

Then, being a tough guy, I figured I'd just knock out the undesirable material using my drawblade and a plane. I did not anticipate how much work this was going to be and quickly abandoned my tough-guy hand tool plan in favour of the bandsaw. The bandsaw of course made short work of the taper cutting.

Figure 10: Straight-tapers marked

Figure 11: Feelin pretty tough with my hand tools

Figure 12: Definitely questioning whether toughness is the right approach. Sore arms.

Figure 13: Maybe the bandsaw aint so bad. 

Nest task was to shape the looms. This is traditionally done using a device called an "eight-siding gauge". You can get very fancy with these gauges, but for a one-off project I went with the simplest possible version.


  1. Drill 2 holes in a piece of scrap wood -- just a tiny bit wider than the widest part of the loom. 
  2. Divide the distance between those 2 points into 3 parts, using proportions 5 - 7 - 5. Opinions vary as to the exact ratio that works best, but apparently this is what was used by sparmakers back in the day. 
  3. Two more holes go at the dividing lines. 
  4. Hammer nails through the end holes, and put screws into the 2 middle holes so that the points are juuuuuust peeking out. 

To use it, hold the eight-siding gauge on an angle so that the outer nails make contact with the outside of the oar's loom and the screw points dig into the wood. Slide the gauge up the loom, ensuring the outer nails stay in contact and score the loom with the screw points. If you manage to keep good contact (which requires that you change the angle of the gauge for a tapered loom), the entire loom will be divided in 3 parts along its length, at a constant proportion of 5 - 7 - 5. 

Figure 14: Eight-siding gauge, showing nail pins on the outside and screws inside

I did this for all 4 sides of each loom and then darkened the score marks with pencil. A block plane worked nicely to clean off the undesired material, leaving me with 2x octagonal oar looms. 

You can see the upper part of each oar loom is left blocky and square. This is not strictly necessary, but comes highly recommended by Pete Culler as a means of counterbalancing the outboard portion of the oar with some added weight inboard. I think it looks pretty damn cool too!

Figure 15: Score marks shown at right, one corner of the loom removed at left. 

Figure 16: Both looms after eight-siding.

The last major piece of shaping was the blades. Recall earlier I mentioned that there was a reason for the straight taper along the narrow side. The taper makes the loom a little bit wider perpendicular to the plane of the blade. It's barely noticeable up near the grips, but you can see a pronounced ellipse shape in the throat, and the centre of the blade is considerably thicker than what is called for at the blade edge. Considering that the bending moment will be acting perpendicular to the plane of the blade, more thickness is advantageous here.

The throat is already rough-shaped, and the blades receive a straight taper from the centreline to the edge. There was no way around using the plane for this part, but thankfully there was not much material to remove.

Figure 17: Blades all ready for shaping on the bench

Figure 18: Blade shaping complete. Arms tired. 

At this point, I did about all the bladed tools were good for and switched over to sand paper. The most arduous part of the whole process was shaping the looms and grips. The eight-sided shape needs to be reduced down to a mostly-round shape and the easiest way to get there is to start sanding. 

The basic technique is to lay a piece of belt sander paper (40 grit or so) down over top of the eight-sided piece. Stand over top of the piece and grab both ends of the belt, then whip that sucker back and forth for all you're worth. The piece will quickly *start* to look round, and then very slowly come into a true round shape. This process is guaranteed to murder your arms, so best to do it in parts. 

Figure 19: Started with the grips to build confidence.

It's a really good idea to check periodically with a set of hashes where you're at with the sanding process. There were several times I thought I was doing good, and discovered with the hash marks that I was missing spots. 

You want to switch the position of your hands from time to time too. First start with up and down strokes with lots of pressure, bending the sanding belt sharply over the oar loom. Then switch to a wider grip with the belt bent at a more obtuse angle, using side-to-side strokes (these hurt). 

Continue until you've satisfied yourself either a) that it is round enough, or b) you cannot sand any more. 

Figure 20: Hash marks along the loom indicate where I missed. 

Figure 21: Before and after

Figure 22: Both looms done (2 sessions' worth of sanding)

With the looms completed, it was time to work on the final blade shape. Pete Culler stresses the importance of thin-as-possible blades for reducing outboard weight. Thinner blades are also a bit more efficient and have a more aerodynamic profile when they are feathered in a headwind. 

Culler's plans essentially call for removal of material to either side of the spine at the centre, and a blade cross section that thins rapidly as you move outward toward the blade edge. There are lots of tools that can do this work; in production runs, it can be carved quickly using a backing (rounded) plane or spokeshave. In one-off projects like this, I was advised to make a shaped sanding block and glue belt sander paper to it -- the argument being "you will spend 10x more time making or sourcing a tool than just doing it with sandpaper" -- or somesuch. Credit to the Woodenboat builders' forum for the help!

I carved a scrap block of 2x4 to approximately the shape I wanted, and glued on sandpaper with 3M Super 77. This was wrapped tightly for a few hours to make sure it would hold fast. Once the paper was firmly attached to the block, I made a pushing block out of another piece of scrap and attached it to the top. 

Figure 23: Making the curved sanding block 

Figure 24: Sanding block with pusher attached to the top

The procedure for sanding is straightforward. Just mark a line at the centre and don't cross it. Hatch marks are very useful for identifying high points and generally keeping track of what is getting worn away the quickest. I found the block tended to sand a groove first and leave a ridge to either side. Once I had a good depth, I applied some side pressure toward the centre to sharen the spine. Last, I re-hatched the whole thing, spun the block around and worked with the flat side to remove the outer ridge.

I found quite a few errors in the first oar as I got more practiced on the second, and then found errors on the second as I got more practiced refining the first oar. I would definitely recommend switching back and forth several times to check and fix mistakes. Time here is probably <10 minutes for the first pass on each of 8 blade faces. Double that for the refinements afterward. The final result is not perfect, but looks pretty decent. I might revisit them before finish-sanding just to be sure I'm happy (and when my arms recover). 

Figure 25: Oar blade ready for sanding. Line at centre and hatches on the face.

Figure 26: Final result. Not perfectly symmetric, but it should work well enough.

The oars are now ready for finish sanding before paint/varnish. I snapped a few pics of the mostly-completed oars to show the whole thing and to give a sense of scale relative to the boat. 

Figure 27: Two completed oars

Figure 28: Next to the boat for scale. 

For sanding of the blades and looms, I turned to my trusty random-orbit sander. This was a good tool for tidying up my blade fillets and removing the cross-grain sanding marks left over from when I rounded the looms from octagon to ellipse. I have a lot of practice doing this type of sanding on round surfaces but still had to be very careful not to create flat ridges along the length of the looms. 

Grip shaping came next. Someone on the woodenboat forum suggested that I round the ends of the handles for when I want to hang my thumb over the ends. Couldn't agree more, and that's exactly what I did. 

I also filleted the tops of the square blocks using a disc sander and some 120 grit sandpaper wrapped around a dowel. They aren't perfect at the transition from grip to block, but I don't dare sand away any more material at this point. Might try to make it a bit better by hand, but at this point they are perfectly usable. 

To finish off a long night of sanding, I went over the whole issue with 220 grit sandpaper as a final surface prep and they are looking flippin' fantastic. 

Figure 29: Fillet on the top of the block 

Figure 30:  Rounded ends on the grips & completed fillets

And my new professionally-made oars arrived from Lunenburg. They are 8' workboat oars sold by the dory shop. 

The dory shop oars are not as dainty as the ones I made, but they are real beauties! Solid spruce and very sturdy. They sport a thick, untapered loom, and a thick, broad blade. They are definitely not designed with Pete culler in mind but will probably prove to be the more sensible thing to have on hand while rowing around rocky shorelines. 

Very interested to see the difference in performance between the vaunted Culler oar and a more contemporary design. I'll definitely be using the latter as-is this season, but there's lots of wood for making adjustments if I find them too heavy to handle. 

Figure 31: My oars at left, Dory Shop oars at right. 

For tip guards I used a couple of little strips of white oak left over from the rub strip on the boat. I put a great big gob of epoxy on the end of the oars and stood them up on the oak strips. A few passes with the disc sander to remove excess and my delicate softwood oars are now ready to smash off some rocks!

Figure 32: Tip guards completed.

I finished the oars using Epifanes brand clear varnish -- left over from the boat. Since it was going on bare wood this time I followed the instructions to the letter. First coat that went on was 50%/vol turpentine. After this I reduced it to 25%/vol, then 15%/vol, and then an extra two coats of varnish at full strength. 

Prior to the first coat and between each subsequent coat, I sanded with 220 grit, per label instructions. After a few weeks, they were ready to go. 

You can see in Fig 33 that I didn't varnish the handles. This was a strong caution by Pete Culler. He says to do these with a few coats of boiled linseed or tung oil. The bare wood texture is apparently much easier on the hands and won't cause blisters as readily as a varnished surface.

Figure 33: First coat of varnish!

If I remember, I'll post one last pic that shows the leathering process. Leathers protect the oars in the locks, and it looks pretty sharp too!

No comments:

Post a Comment