HMS Bellona 1760 by SJSoane - Scale 1:64 - English 74 gun, as designed

[glbarlow]

Amazing work, such beautiful moldings and curves

[SJSoane]

On 1/7/2014 at 6:19 PM, SJSoane said:

Hi everyone,

 

Thank you for your thoughtful contributions to this interesting question of the beam roundup. I am reconciled to averaging the roundup between midships and at the transom. But just for the record, I have attached an annotated drawing of the situation. The deck would droop if the radius for the beams is held constant.

 

I have also attached my photo of the 1760 Bellona model, showing that the quarterdeck decking clearly sits on top of the transom above the lights, and so the transom does not have a different roundup from the deck.

 

It may be druxey, Greg and Ed have the right idea; the roundup must have increased over the gunroom in the final set of beams in the quarterdeck. But more subtle than I think I want to address in my model....

 

Best wishes,

 

Mark

 

 

 

 

 

 

HI everyone,

 

Drawing lessons from the Bellona, part II.

 

I am quoting an entry I made in this log in 2014, when I and a number of people contributing to my questions were pondering why the roundup of the quarterdeck beams was actually higher in the captain's cabin than the roundup called for in the quarterdeck specifications.  

If the beams remained a constant roundup along the length of the quarterdeck, the crowns of the beams in the captain's cabin would have been almost 2 inches lower than they are actually drawn in the admiralty plans.

 

Now almost 8 years later I can answer this definitively, after carefully following David Steel's instructions for draughting the drawings. And Ed Tosti, Greg Herbert and druxey figured this out 8 years ago, but I didn't understand it at the time. Now I do.

 

Drawing the decks in sheer begins at the centerline of the ship and is an arc of a circle (only three points are specified to construct it; the heights at midships, and at the fore and aft perpendiculars. The height of the decks at the sides is NOT an arc of a circle, as I had previously assumed. One has to draw a cross section at midships, showing the maximum width of the deck and then raise a 5" line for the maximum roundup at this point. An arc of a circle is constructed on these three points. The width of every other beam is then drawn on top of this, giving a reduced height of roundup at every point. At the bow, when the width of the beam becomes zero, the roundup is also zero. 

 

This means that as the beams get rapidly narrower towards the bow, the roundup reduces and therefore the height of the beam at the side reduces just as rapidly. This causes the line of the tops of beams at the side--as seen in the inboard profile drawing-- to be a compound curve which rises more sharply at the bow than the stern and also more rapidly than the arc of the circle at the midpoint of the beams.  

 

 

This in itself is interesting, because the ports are a constant height above the beam tops AT THE SIDES, and therefore the ports' heads and sills are NOT on an arc of a circle, even though all of the rest of the sheer drawing (top timber, sheer line, etc.) are constructed as arcs of a circle. So there is a subtle mismatch between the ports and the rest of the sheer.

 

But finally, as answering the question at the top of this posting, the instructions in Steel clearly direct the draughtsman to change the roundup on the quarterdeck to match the roundup at the stern windows, which is steeper. Druxey had alerted me to this, but again I did not understand this until now. Here are quotes from Elements and Practice of Naval Architecture

 

p. 240

The heights of the decks must next be considered; for sometimes, in order to give depth for the lights, the decks are necessarily sprung abaft, and their round-up must be made conformably thereto.

 

p. 245

Thus may the round-up be taken at as many timbers as may be found necessary, and set below the underside of the deck at its respective timber in the sheer plan; then, a curve line passing through those spots will represent the deck at the side : but observe, that the decks are to have a sufficient round abaft, to correspond with the round-up of the stern above the lights.

 

 

And here we see it below. The green line represents the roundup of the quarterdeck according to its specified roundup for a beam at this point. The grey line above (and the orange line below, which is the bottom of the beam) is the roundup actually measured on the admiralty plans. In other words, the beams do rise higher than specified aft of midships, to align with the roundup of the stern windows.

 

druxey explained this as keeping the stern design feeling light and springing, not drooping. Another example in the eighteenth century of art and form over-ruling function and constructional efficiency or consistency.

 

 

 

 

 

No more drawing lessons now. Back to model construction!

 

Best wishes,

 

Mark

 

 

 

[SJSoane]

thank you, JR, for those additional references. I will try to track some down, and see if they give any greater insight into how the designer shaped the hull forward and aft of where the frame geometry still relied on the floor sweep. I am guessing it was a process of trial and error, and gradual experimental deviation from previous ship forms.

 

And thank you, Chuck and Glenn, for your kind comments about the project.  I still sit and admire the form of the Bellona's hull, and wonder at the skill of the designers who melded all of those complex curves together into something both functional and exceptionally beautiful.

 

Mark

[druxey]

The secondary pay-off is the increased headroom aft under the quarter deck! A couple of inches would make a difference between having to duck under the beams or not!

[SJSoane]

Hi everyone,

Well, it has been a long, long time since I last posted. A number of life-outside-the-shop issues took over, but I am back to the Bellona at last!

 

My plan of work now is to complete the planking and all outboard work, so I can paint the wales and friezes with the hull on its side. Then I can install the lower deck guns and proceed at last with the upper deck. With the lower deck guns installed, I would have no hope of turning the hull on its side for painting.

 

As I got closer to the top with planking, I realized that I would have to install the sheaves (main sheet and tack, fore sheet, sprit topsail sheet) in the sides before the next strake could go on. I thought this would be a fun, short break from planking. It turned out to be over a month of work. They are way more complicated than I ever realized. And I made several mistakes in their fabrication that meant starting over.

 

First the complication. The sheaves on the Bellona are neatly slotted between the sheer and waist rails outboard; and they slope at right angles to the side to arrive inboard just under the clamp. The inboard edge also swings up the fay to the underside of the clamp:

 

And the block in sheer is a symphony of angled lines. The top and bottom of the blocks correspond the sheer line of the hull at their respective locations; the sides are vertical; the sheaves are angled within the blocks to provide a fair lead for the lines:

How to construct these? Since I did not have any router bits that could cut a slot thin enough, I created the slots by laminating blocks, with a groove cut in the side of one with a mill bit. For the double sheaves I laminated a thin sheet in the middle of the block. While the blocks were still square and the slots were parallel to the top and bottom, I drilled the holes for the sheave pins.

 

 

At first, I tried mounting the blocks on my Sherline rotary table on the mill; thinking I could just dial in the appropriate angles in one go. But I simply could not visualize which way angles went relative to each other, and I also stupidly misread the scale on the rotary table, cutting .3 degree rather than 3 degree angles. I ruined several blocks until I discovered this. The sheave hole in the block at the bottom of the previous photo was supposed to be sloped and you can see it is parallel ....

 

To make life easier, I turned instead to using my angle blocks to set up the mill vise at the appropriate angle (see the angles against a square in the photo below). This helped me physically visualize which way things should be aligned. And then I used a mill cutter to cut first the ends and then the sides. the block at the bottom of the photo below is an example of the failed blocks relative to the correct ones with angles:

 

 

And then trimming to get the blocks to their correct sizes. I turned the sheaves in ebony, and used a tiny round file to ease the ends of the slots. Note how the tops of the blocks in the photo below have angled tops to the inner surface; this is to fay to the bottom of the clamp, as seen in the first photo posted above. Also notice how large these puppies are next to my captain. They are tiny in my model, massive on the ship. I can see the laminations on the blocks more than I would have liked; but these are all under the frieze painting and so hopefully this will disappear.

 

 

They are dressed proud where the next strake of planking will come up to them:

 

But the main sheet sheaves had to be cut into finished planking (it did not occur to me to deal with this when I was planking lower down). These were very laboriously cut into finished wood, with much trimming and fitting, trimming and fitting, to do this cleanly. It would have been better to think ahead before the planking was installed here:

 

 

All for now. On to the next strake of planking.

 

Mark

 

 

 

 

 

[druxey]

Welcome back, Mark! Those fixed blocks are tricky. looks like you've got them beat though.

 

Had you considered painting your friezes on thin paper and then gluing them to the ship's sides? Contemporary models show that this was often done, rather than painting directly on the wood. I found this much easier to accomplish, particularly when the surface was directed downward!

[mtaylor]

Very well done on the blocks, Mark.  Great to see you've returned.

 

Druxey makes a good point about the friezes.  This seems to be catching on a bit as there's a couple of models that have recently used this method.  

[SJSoane]

Hi everyone,

 

A long, long time away from the Bellona! Other life things have interfered, including getting Covid even though I had both vaccinations and the booster.

But hopefully things are getting going again.

 

I am up to the top plank in the waist, the Sheer Strake:

 

 

And I have come across an interesting question regarding the sheer and waist rails as I tackle this part of the ship. Siggi first alerted me to this, and then John directed me to the contract on the National Maritime Museum website for the Bombay Castle, which was based on the Berwick, which was a sister ship to the Bellona, and the contract Brian Lavery believes is closest to that for the Bellona.

 

Here is the puzzle. I have long assumed that the uppermost plank in the waist is the thicker Sheer Strake, and nailed over it is the uppermost moulded rail named the Sheer Rail; and that the Waist Rail is the moulded rail below, cut through by the ports as seen in my model above. This is how they are names in Steel's plate:

 

And Goodwin's book, The Construction and Fitting of an English Man of War, page 57, also labels them as Sheer above, Waist below.

 

But, Siggi pointed me to Falconer's Dictionary of the Marine, which clearly labels the top rail as the Waist Rail:

 

 

And then I looked at the Contract for the Bombay Castle, which cannot make any sense if one assumes that the top Rail nailed on top of Sheer Strake is named the Sheer Rail. It clearly refers to this top rail as the Waist Rail.

 

"Sheer Strake: To have a Sheer Strake wrought in two breadths that the upper Edge of the upper Strake may be agreable to the upper Edge of the Waist Rail in the Waist and one Inch higher from thence forward and from the Waist aft to be in two Breadths 12 Ins and 4 Ins thick and of English Plank in Wake of the Channels".

 

(both models of the Bellona, BTW,  show just one strake for the Sheer Strake, not two.)

 

Further in the contract, the Rails are defined as follows:

"The Sheer Rail to be 8 inches broad and 3 1/2 thick the Waist Rail 7 inches broad and 2 3/4 thick the Rails in the Drift to be 6 inches broad and 2 1/2 in thick. Since the rails in the Bellona sheer drawing appear to be widest below and narrowing as they climb higher, this suggests the Sheer rail is the lowest, then the Waist Rail, then the Drift rails above.

 

Interesting, when our primary sources of Steel and Falconer disagree, while the contract agrees with Falconer as best I can see here. Maybe they didn't mind naming the rail over the Sheer Strake the waist rail, and the sheer rail is the one down below! Perhaps they tripped up apprentices with this one....

 

Best wishes,

 

Mark

 

 

 

 

 

 

[druxey]

Welcome back, Mark! I hope that you feel fully recovered.

 

Steel's text (page 54, Naval Architecture) reads:

 

RAILS. The long narrow pieces of fir or oak, with mouldings struck on them, which are  fastened, or sometimes wrought from solid plank, as ornaments to the ship's sides, and also at the head or stern. The principal are as follows: The lower rail on the side, named the Waist-rail; and the next above it, the Sheer-rail, which are generally placed well with the sheer or top-timber line; the rails next above the Sheer-rail are called Drift-Rails, and the rails above the plank-sheer the Fife-rails.... 

 

In the list of errata at the back of the book (unnumbered page) there is no mention of any correction to this entry.

 

[Mark P]

3 hours ago, SJSoane said:

"Sheer Strake: To have a Sheer Strake wrought in two breadths that the upper Edge of the upper Strake may be agreable to the upper Edge of the Waist Rail in the Waist

Good Evening Mark;

 

I interpret this phrase to mean that the sheer strake is to be parallel to the waist rail (or concentric to it; virtually the same thing) Once this is applied, there is no trouble with definitions of what rail is what in the contract. Falconer is still in opposition, but I believe that he is in error; firstly because he is outnumbered by other contemporary sources; but secondly because of the following:

 

William Keltridge, in his book giving many dimensions for ships of the latter part of the 17th century, lists the 'plansheer' rail. Since the 'plansheer' is what we know as the planksheer, which is the moulded capping plank laid over the tops of the timbers in the waist, it would seem a fair deduction that the 'sheer' rail was once called the  'plansheer' or 'planksheer' rail, and is fixed directly below the plansheer/planksheer. 

 

The waist rail is, as most sources indicate, and as you first described, the moulding which is interrupted by the gunports.

 

It may also be important that Keltridge was quite prolific in his output and was a renowned ship's carpenter, ending his career as carpenter of the first Royal Sovereign in the 1690s, which would make him the Navy's most senior ship's carpenter (he was not in this post by the time she burnt at her moorings) and lived much longer than Falconer, who was lost at sea on a voyage to India, whilst still only 37 years of age. Falconer served as a seaman, and a purser; and briefly as a midshipman. He was also well fairly well-known as a poet. On balance, Keltridge's experience-derived knowledge will have been much more fitted to enable him to correctly name the parts of a ship. 

 

All the best,

 

Mark P

Edited March 7, 2022 by Mark P

[SJSoane]

Hi druxey and Mark,

 

It is nice to be back. Not quite over the covid, but fast getting there. I didn't see any warnings on my power tools not to use while suffering from Covid, so onwards I go!

 

It really does seem irrational that the rail on the face of the Sheer Strake is not called the sheer rail. But digging into this, I have now discovered three contemporary sources from about mid 18th century, all seemingly calling this the waist or waste rail.

 

There is Falconer, sometime before 1769 calling the top rail a waist rail; the HMS Berwick Contract 1768 I quoted earlier, and the Marlborough Contract, 1763, which says:

 

"To have a Sheer Strake Wro't fore & Aft, the upper Edge of which to be the Upper edge of the Waste Rail in Midships & 1 Ins higher from the Drift Fore & Aft, to be in Breadth 12 Ins and in thickness 4 Ins this, that the Channels be placed on the lower part thereof..."

 

I liked Mark's interpretation that the Berwick contract was really talking about parallel rails, not necessarily the location, but the Marlborough Contract doesn't seem to allow for this. It has the top edges of the Sheer Strake and the Waist Rail aligned with each other.

 

And yet, by the time we get to the Ship Builder's Repository, 1788,  page 270, we have "Sheer Strakes, 2 in number, each in breadth 12", the upper edge of the lower sheer strake to be agreeable to the upper edge of the sheer rail." Now all is well, and this is later restated by Steel who also has the upper rail as the sheer rail.

 

I wonder if there was a transition at mid century, swopping the names to be a little more rational.

 

And I can't  stand the irrationality of a waist rail over a sheer strake; whatever really happened here, I will call my upper moulding a sheer rail, to match the sheer strake under it!

 

Best wishes,

 

Mark

 

 

[Mark P]

Good Morning Mark;

 

I believe that the Marlborough specification is missing a couple of words; the writer is near to the bottom of the page and may well have been trying to condense what he said as much as possible, leaving out a word or two in his haste. The sentence does not really read very smoothly at all as it stands.

 

All the best,

 

Mark 

[SJSoane]

Hi Mark,

 

Yes, I can see how ambiguous these old contracts really are, especially when they run one idea into the next without punctuation, leaving the reader unclear whether the next phrase refers to the previous one or is a new thought. It is also interesting to see how different contracts try to say the same thing but with different phrases; you would have thought that there would have been some standard "boilerplate" text for a lot of this...

 

Best wishes,

 

Mark

 

 

 

[druxey]

In the late 18th century, contracts were indeed 'boilerplate'; printed forms with spaces to insert specific dimensions and details.

[Mark P]

Good Evening Mark;

 

Druxey is quite correct; however, one thing which is perhaps surprising is that many contracts contain large amounts of very similar wording, even as far back as the mid seventeenth century. It would appear that it was quite customary that every time a new build needed a contract to be drawn up, the first thing to do was to reach for the previous contract and copy it; perhaps adding a few extra words or paragraphs here and there. Contracts grew gradually in length. The oldest one I have found is from the 1580s, and is three pages long. The ones dating from the decades before the French Revolutionary Wars began frequently run to thirty or more pages; and those pages are quite large, A3 size. 

 

All the best,

 

Mark P

Edited March 8, 2022 by Mark P

[SJSoane]

With the nomenclature conversation taken care of, the moulded railings needed to be made.

 

I refined my moulding cutter setup to work more efficiently.

 

I cut the moulding profile as before in a Lie-Nielsen tool steel blank. This steel is soft enough to shape with files, but keeps its edge for long enough for a run of one set of mouldings. It is also very thick, and so it does not chatter like thinner metals I tried.  I mounted it in a wooden clamp at what I determined to be the best angle of attack, with the clamp also forming a fence at right angles to the tool steel.

 

In my refined method, I built a jig that could be secured in my tail vise, and then three clamps hold the moulding blank against one side of the jig. I clamped the blank, projecting the exact thickness of the finished moulding above the jig, using a temporary spacer for accuracy. (In the photo below, you can see the profile previously used for a different moulding; I will just keep sliding the tool steel along as I need more profiles in its edge, until I don't have enough left to clamp in the wooden block.)

 

I have read and viewed reports from others using a scratch stock in a more free-hand manner, but my freehand efforts resulted in wavy surfaces up and down. I don't know if it is a lack of skill on my part, or I have unusually stiff wood, but I found that I really need the scratch stock to be guided with this fence arrangement to get satisfactory mouldings.

 

 

 

It was then a simple matter of pulling the cutter along the projected edge, using the fence to keep the cutter parallel to the blank, and cutting until the flat part of the cutter holder hit the top surface of the jig.

 

As before, I then cut off the moulding on the Byrnes saw, using a strip of paper against the fence to avoid discoloring the moulding face while rubbing against the aluminum fence:

 

And now perfect mouldings along the top of the waist, on top of the wider sheer strake: 

 

 

Best wishes,

 

Mark

 

 

[jdbondy]

Mark that is a really robust-looking setup. I am going to keep that in mind when it comes time for me to cut moldings. Thanks!

[SJSoane]

Hi JD,

 

I am glad it might help you. This is a case where I had never made mouldings before, I researched how everyone had done this, and tried their techniques without success. I saw that their methods worked for them, I just couldn't get them to work for me. I had to experiment until I got something to work.

 

I have suspected over the years that my wood--which I thought was boxwood when I purchased it over 30 years ago--is something related to boxwood but much stiffer. For example, it does not submit to bending techniques others have used; I have to steam it for longer and then clamp over a former with exaggerated curves because of so much springback. And when I freehanded scraping cutters for mouldings, the cutter waved up and down even though the grain is very fine and not the obvious cause of the waviness. Only a very stiff cutter, a fence along the cut, and a solid surface to bottom out on did the trick.

 

Best wishes,

 

Mark

[Gaetan Bordeleau]

4 hours ago, SJSoane said:

I have suspected over the years that my wood--which I thought was boxwood when I purchased it over 30 years ago-

I bought  what I think is the same wood also about 30 years ago. I still have a few pieces. I took a photo of one near boxwood.

Palo blanco was the name if I remember well.

 

Edited March 11, 2022 by Gaetan Bordeleau

[SJSoane]

Hi Gaetan,

 

Yes, that looks like my wood. It was labeled S.A. Boxwood at Woodcraft stores, which I presumed to be South American Boxwood. I didn't know better at the time.

As it turned out, it is still a lovely wood, with a fine grain, and it darkens over time to a golden honey patina, like something one would expect to see in an antique ship model.

 

I just needed to learn how to work with it!

 

Mark

[Gaetan Bordeleau]

South American or castelo boxwood (Calycophyllum multiflorum) is also known as palo blanco or ivorywood. 

[druxey]

My 'trick' - if trick there be to it - is to rubber cement a blank strip of the maximum thickness of the molding onto a carrier sheet of hardwood. This carrier acts as a depth stop when the molding is fully developed. The profile is also wave-free.

[AON]

There are always multiple ways to skin the proverbial cat, and so, that is why we need people like Druxey following... to highlight the forest for the trees.

[bruce d]

On 3/9/2022 at 11:44 PM, SJSoane said:

As before, I then cut off the moulding on the Byrnes saw, using a strip of paper against the fence to avoid discoloring the moulding face while rubbing against the aluminum fence:

Very nice setup and very good timing: I have been playing around with fixtures to make a  moulding and will now go straight to your method, thanks.

One question, if you don't mind? Did you try reversing the arrangement so the cut-off moulding was not between the workpiece and the guide? 

[AON]

my thin strip jig or thickness guide setup on my Byrnes tablesaw

(with thanks to Kurt Van Dam and John Garnish)

I set the point distance from the blade with setup blocks

Test cut and mic the thickness

I can fine tune the thickness cut with the Byrnes micrometer setup

Edited March 12, 2022 by AON

[bruce d]

Thanks, Alan, I see a path forward ... 

👍

[SJSoane]

Gaetan, I looked up the species name you gave me, and discovered that this wood is described as "very resistant to flexing", which might explain why I have found this to be so challenging when steaming and bending. Thanks for the information!

 

druxey, I will try a piece using your method, and the cutter I already made. I'll see what happens.

 

Bruce, I cut with the moulding face against the fence to ensure that the piece would be a uniform thickness. The scraping process sometimes seemed to end up cutting things at a slight slant relative to the original face, and therefore at a slant to the face at the opposite side of the blank. Running the blank through with the finished part on the side opposite the fence would have resulted in a moulding thinner at one end than the other.

 

I always work with a blank wide enough to keep my fingers well away from the blade, using a pencil with a rubber eraser as a push stick. I hold the push stick very close to the blade so the thin piece never has a chance to ride up on the saw blade, and I push in short moves so I can reposition the pencil often. I am super aware of possible kickback with a thin piece trapped between the fence and the blade, and I stand to the side and wear a full face shield. But keeping the wood firmly down on the table and against the fence with the eraser tipped pencil seems to keep everything tame. So far.

 

Alan, I also use a blank of the thickness I desire to set the fence to the blade, although this still needs some adjustment after trial cuts are made.

 

Alan, I like your "fingerboard". I have been away from the website so long, did you show any details on how you made it? I have long used one that our Model Ship World tool wizard Michael Mott helped me design years ago. The rod can slide in and out for rough distance, and then the thumb screw in the end can be adjusted for fine movements. Most of the time, I use this when I can cut the finish part on the side of the saw opposite the fence, for repeatable cuts to the same width. I just push the blank up to the jig, slide the fence up snuggly, clamp down and cut. Repeat until the blank gets too narrow for safety. Thanks again, Michael, it has worked great for years!

 

Mark

 

 

 

 

 

 

 

[AON]

Does the thumb screw on the end of the post move at all with the passing of multiple strips over time?  Your and Michael's design has some merit to it.

 

My posting #1277 (8 JAN 22) in my build log shows an image with drawings I made.  I'd be happy to send you a PDF copy.  Kurt says the NRG are looking at selling a simplified version (mine was a combination of his and John's versions) sometime soon.... he is getting pricing for the bits and pieces.  If I were to do mine again I'd likely make it at least 1/4" wider.

 

 

[SJSoane]

Hi Alan, Thanks!

No, the screw doesn't move over time. If anything, it is a tiny bit stiff to turn for my increasingly aging grip!

Mark

[mtaylor]

Mark,

Instead of a pencil, give some thought to using something chopsticks (wood, not plastic).   I used the pencil method for a bit until things went messy and the metal ring holding the eraser to the pencil got hit by the blade.  I think if I had been standing inline with blede, I would ended up in the hospital for surgery.  Instead, I went and sat down to stop shaking.