Galilee's Rabbet

[CDR_Ret]

Hi all. After a hiatus of nearly six months, I have been able to get back to working out some of the details of Matthew Turner’s brigantine Galilee.

 

While attempting to clean up the rabbet along the keel, stem, and sternpost in DELFTship, I ran into some issues that call into question the original G.C. Berger plans I obtained from the Smithsonian. This isn't the first time this has happened. Check out my research and design log on this ship. So the following is a series of technical questions about rabbet lines and garboard strakes.

 

Let’s start with a description of the rabbet at the dead flat. The following diagram was created by the scientist-engineers at the Department of Terrestrial Magnetism, Carnegie Institution, Washington, D.C. (DTM) sometime around 1904–5, when they were planning the conversion of Galilee from a merchant ship in the Tahiti trade to a magnetic research vessel. They needed to know where all the ferrous hardware was located in the hull to figure out the magnetic constants for the ship. For that reason, they took fairly good measurements of the hull, though their drafting skills could have used some work … Note that the inner rabbet line lies at the point where the frame intersects the face of the keel. This is a key factor in the following discussion.

 

The back rabbet and rabbet as shown in the first diagram do not conform to the shape shown in many references. In most diagrams, the back rabbet line lies on the face of the frame, which intersects the rabbet at a right angle. Instead, in this diagram, the top surface of the garboard is beveled at the inner rabbet line, giving the inner edge of the garboard a point. The top surface of the garboard strake ends at the inner rabbet line, the “point” is at the back rabbet line, and the bottom corner is at the outer rabbet line. I’m not sure how the non-naval DTM engineers knew to draw the rabbet profile like this unless they obtained that detail from Matthew Turner himself.

 

According to the DTM drawing, the garboard is about 6” thick. However, the vertical dimension of the rabbet as seen in the side profile view from the outer to inner rabbet lines should vary as the angle of the frame increases or decreases along the keel. The flatter the frame, the closer the distance between the inner and outer rabbet lines should approach the thickness of the garboard. The steeper the frame (e.g., toward the bow or stern), the wider the outer to inner rabbet line distance should be. This diagram illustrates the geometry:

© Department of Terrestrial Magnetism, Carnegie Institution, Washington DC

Curiously, the Berger plans don’t indicate a significant divergence of the outer (solid) and inner (dashed) rabbet lines except at the extreme ends of the keel:

© Smithsonian Institution

Now, let’s consider the thickness of the garboard in the first diagram. As mentioned earlier, the garboard strake is 6 inches thick. The lower hull planking is 3 inches thick. This makes the garboard twice the thickness of the adjacent planking. According to Rules for the Construction and Classification of Wood Ships (ABS, 1921), garboards must have the full scantlings for at least three fifths of the length amidship. Granted these rules were promulgated thirty years after Galilee was built, they seem to agree more-or-less with other contemporary references I have found. So, if the garboard strake stays the same thickness along most of its length, the inner rabbet line is going to curve upward as the frame angle increases. Only if the garboard thins gradually as it approaches the stem and stern areas, will the inner rabbet line stay more or less parallel to the outer rabbet line.

 

Another factor is that several views of the Berger drawings show dashed lines inset at about 3 inches from the sides of the keel as shown below. These seem to indicate the depth of the back rabbet line. If the garboard strake’s thickness remains the same for most of the length of the ship as noted above, then the back rabbet line’s depth from the sides of the keel would also vary as the frame angle changes. (Another possibility is that the dashed lines show only the back rabbet line inset at the stem and stern, though they are visible in the plan view of the keel as well.)

© Smithsonian Institution

 

So, either Berger erred in drawing the inner rabbet line as a straight line for as long as he did along the keel, or the garboard changed in thickness as its angle to the keel changed to maintain that straight line. I'm in a quandary as to what the right answer is.

 

Does anyone have thoughts or critiques of this analysis?

 

A good resource for this question is this video of cutting a rabbet along the keel of a modern 150-foot wooden ship using traditional techniques. Also, check out the other links in the left margin of the video page!

 

Thanks for the assist.

 

Terry

Edited March 11, 2020 by CDR_Ret

[Bob Cleek]

Living a couple of blocks away from her last resting place, I watched the tides wash over Galilee's keel for many years. I should have taken a closer look, but it's too late now. (Her transom is (was?) preserved at Ft. Mason,  Golden Gate Recreation Area, San Francisco.) 

 

I'll offer a few observations, 

 

1.   How could those DTM guys know the keel to garboard faying surface was pointed, unless they pulled the garboard, which they wouldn't have done because they had no need to do it if they were only looking to identify the fastenings. Turner's own yard built the ships he designed and he personally supervised the building. I would hazard to guess he did not draw any design details remotely close to what the DTM guys drew. His crew knew how he engineered his vessels and didn't need to be told how to lay out a garboard seam. Ergo: It's a pretty safe bet that aside from the placement of the keel bolts visible to the DTM surveyors, the rest of the drawing depicting what they couldn't see is likely some lubber's fantasy.

 

2.  No self-respecting wooden ship builder would drive keel bolts any closer to the garboard seam than they absolutely had to. That arrangement looks to me like it's sure to leak in short order. It's clear from the DTM drawing that whoever drew it didn't know shipbuilding because all of the drifts were drawn as being driven straight down. Drifts are always driven at opposing angles so as to lock the timbers together. 

 

3.  It's very hard to believe any wooden ship builder of that time, and particularly one as practical as Turner, would ever find a justification for building a "pointed" garboard and rabbet. It would be difficult and expensive work for no benefit, for all the reasons you noted. For what it's worth, this old wooden boat guy has seen more garboard seams and rabbets than he'd like to remember, "in the flesh" and on paper, and I've never, ever, see one a "pointed" one.

 

4.  With a nod to Howard I. Chapelle's tremendous contributions to the preservation of the American maritime heritage, the HAMMS lines are notorious for errors. Don't forget HAMMS was a WPA project designed to make work for unemployed architects, surveyors, photographers, and draftsmen and not all of them had specifically maritime trade skills. We can't be too hard on them, but we have to remember they are "secondary sources" and not primary historical records.

 

I'd say you'd be safe in using "common practice at the time" to "fill in the blanks" where there are gaps or "inexplicables" in the available record. Chapelle was quite forthright about doing the same in his work. Sometimes one has to extrapolate.

[CDR_Ret]

Thanks, Bob.

 

I'm in complete agreement with your assessment of the DTM draftsmen. During the field assessment, the engineers may have been able to obtain some verbal description of the hull's construction to add to what could have been visually verified from inside the hull. But trying to put all that together into dimensional drawings may have included some guesses.

 

As I noted in several posts in my Galilee thread, Berger's plans, or at least the subject, seem to predate the HAMMS project, which occurred in 1936–37. That idea may be in error. Berger's plans show her as a brigantine, but by the early-30s she was a an aging three-masted fishing schooner and had many alterations to support the fishing industry. According to her history documented in R.A. Stradford's Brigantine, Schooner, Houseboat: Journeys of the Galilee, the vessel was purchased by an ex-pat British captain named John Quinn, who with his wife used it as a houseboat, beached in the Sausalito mudflats from 1934 until the late 1950s.  

 

The Berger plans include a note thanking J. Quinn as the last owner of the Galilee (as well as Ray Bowes) for assistance in the production of the plans. However, the plans show the ship as a brigantine, so he must have used the ship-as-houseboat mainly to confirm her overall dimensions and arrangement. His plans, which are now part of the HAMMS archive (and are copies of copies of copies ...) look very similar to C.G. Davis's Rudder magazine drawings from 1899, including the erroneous transom shape.

With the ship moored and/or in the mud, I'm not sure how Berger could have validated much of the keel configuration.

 

So, I think I will do the best I can to map out the rabbet assuming a more-or-less rectangular garboard strake that tapers from 6" to 3" thick in the vicinity of the bow and stern.

 

Terry

Edited March 11, 2020 by CDR_Ret

[druxey]

It seems very atypical the way it was drafted. I'd concur with Mr. Cleek. And wrestling a 6" thick plank bow and stern would be a mighty task: it would have to be tapered in thickness.

Edited March 11, 2020 by druxey

[wrkempson]

Terry,

 

I think your observations in the original post are entirely correct.  I might add that there are some conventions in ship plan drawings that are not to be found in the actual construction of the ship.  Rabbet lines are one such convention.  The dotted lines on the body plan are indeed indicators of the depth of the rabbet, but the actual lines would be quite fiddly (read: almost impossible) to loft and would be of no value at any rate.  On the profile plan the "flat" stretch of the inner rabbet line is not in fact flat; but since it varies from the width of the garboard plank by fractions of an inch there is no good purpose in lofting the actual shape of the line in a basic drawing.  I have lofted this line myself and can tell you the amount of curvature is virtually nil (but only virtually, it is real nonetheless) and is possible only in CAD or some such overly precise drawing tool.  At any rate, the finally shape of the rabbet must be cut in situ so drawing it with precision is not productive.  

 

So, your inner rabbet line ultimately should be lofted on the basis of the how the frame line is finished off at the keel, another process in the drawings that is done more by convention than by geometric accuracy in the older drawings.  As you point out, the inner rabbet line derives from the intersection of the frame line (equals the inside face of the planking) and the plane of the side of the keel.  I have seen modern drawings where these lines are done accurately.  I don't know about the era in question for your ship, but suspect there was still a bit of simplification going on.  Now, add to all this the taper of the keel fore and aft and you have quite a donnybrook.  

 

I have no conclusions for you in all this, just some observations to agree with your final game plan.  Berger's drawing may be neither in error nor mistaken.  More likely it simply follows the conventions of the day which the men in the yard would know to adjust during construction.  None of what I have to say relates to the shape of the garboard.  Finally, the notices made of the DTM plan ring true to my inexpert understanding.  It looks like it may have been drawn using accepted conventions and a modicum of guesswork (regarding actual bolt placement, garboard shape, etc.).

 

Usual disclaimers inserted here,

 

Wayne

[CDR_Ret]

Thanks, Wayne.

 

Excellent summary of the problem. I haven't done much of this kind of research, so everything you said makes sense.

 

Appreciate the input.

 

Terry

[CDR_Ret]

Well, I've spent some of my time during the past two weeks self-isolating from coronavirus exposure to sort through the rabbet question.

 

After taking a look at the DELFTship model as I left it last Fall, I realized that there was very poor fidelity between the model and the original Berger lines plan. So I basically decided to work over the model lines again to attempt as much as possible to approach the original lines. After several iterations, I realized that the lines in the halfbreadth, elevation, and body plans simply were not compatible, so I did  the best I could to at least approach the hull shape that produced fared station, waterline, and buttock curves. Considering that I basically had to build the aft end and transom from scratch using reference photos and the real ship had a 5-inch hog, the results were pretty gratifying.

 

As a result of this work, I am posting several other new items pertaining to the Galilee plans elsewhere in the forums.

 

All that to say that I needed a fairly stable set of station lines so I could determine the angle of the garboard strake to identify the rabbet lines. The results are shown below. As we discussed, the back rabbet line curves upward along the middle run of the keel and becomes nearly flat at the ends before reaching the stem and sternpost. The thickness of the strake is full thickness at 6 inches while it thins down to 3 inches, which is the hull planking thickness, at the ends of the hull.

 

Terry