23ft Launch (1785) by Arthur Goulart

[Arthur Goulart]

So, I wanted a not too complicated to learn both 3D modelling and ship construction. I decided to go with a ship's boat, because you can't get much simpler than that for a first scratch project, and because the result might come in handy in the future for ships that, well, will be fitted with boats. Imagine the designer behind the plans had access to a CAD program, that is what I'm aiming to achieve. Thus, I'm looking at understanding how ship's lines were formed, so I can reconstruct this little launch from principle, not from merely tracing all the plan's lines. With that, I hope to get a more accurate model to the original designer's vision, by mitigating the limitations of the hand drawings and measurements of the time, of the 200+ years of wear the plans have been through, and of the distortion the scanning process might've added on top of it all.

 

On this topic, I'll be registering my progress, in a step by step manner. I insist that you guys point out where I've got something wrong, even if you're not sure of it yourself, we might learn something by looking into it! I'm expecting to make plenty mistakes, and the purpose to pick something simple is that I can start it all over again without pulling too many of my hairs off.

 

So far, my main reference has been The Elements and Practices of Naval Architecture, from '§ 2. OF THE CONSTRUCTION OF THE SHEER DRAUGHT OF THE EIGHTY GUN SHIP, FROM THE GIVEN DIMENSIONS' (p. 242) and onwards. Steel's book is aimed mostly at full on ships, but many instructions are aplicable to our purposes, many require a fair deal of interpretation - he is not the most clear and didactic, which is not helped by my still limited knowledge of ship building vocabulary, and by english not being my first language.

 

Why this 23ft launch was picked, specifically? I tried a couple different boats out, I made some sense of this one the faster. The stem's radii and those of the midship frame are really important dimensions that I struggled with on other plans, but that I fairly confidently figured out for this launch.

 

My take is that ship's boats were neglected subjects in their time, at least from a more technical stand point, and that today they likewise get less attention when compared to the more complex and exciting subjects, hopefully this will be a more thorough and complete reconstruction.

 

The next post, I'll take you back to the beggining of the model's construction, but, for now, I'll leave you with how the model is looking present-day:

 

- Arthur

Edited November 20 by Arthur Goulart

[Waldemar]

 

A very good and reasonable approach, so as not to say that the only truly satisfactory.

 

[Loracs]

On this topic, any good reference on modeling ship with CAD software.  I'm an utterly beginner planning to try using FreeCAD/blender, both of which I never really used.  They are freeware and quite comprehensive though.

[Arthur Goulart]

On 11/19/2024 at 5:15 PM, Loracs said:

On this topic, any good reference on modeling ship with CAD software.  I'm an utterly beginner planning to try using FreeCAD/blender, both of which I never really used.  They are freeware and quite comprehensive though.

Loracs, I'm using Rhinoceros, I found it to be rather intuitive, messing around with it a little and some quick google searchs have gone a long way for me. I have not tried something different, so I can't compare. Rhino seems to be what most folks use here too. And, part of the reason I started this topic is that I haven't found one that adresses my questions regarding redrawing plans, so I hope to crowd source some knowledge through here.

[3DShipWright]

@Arthur Goulart No - most people here use Blender, especially as of late.

 

@Loracs To each his own, but personally, I'd recommend Blender... Rhino may be easier to learn, but it doesn't come close to Blender in terms of features/functionality. I say that honestly and with no ego; it's simply a fact. Blenders' actual competitors are pro studio software like Maya or 3DS Max, but Blender is free. And as a newbie, don't you ultimately want a software you can grow into? Again, to each their own.

 

Best of luck whatever you decide,

-Nate

 

[3DShipWright]

@Arthur Goulart - BTW, I didn't mean to seem negative on Rhino.... still a good software, and I'm excited to follow your build.

[Loracs]

Thanks a lot all... now I need a good set of tutorials for blender, maybe one focusing on model ship modeling...  but that may be a lot to ask.  The learning curve seen very step though.

Edited November 21 by Loracs

[herask]

the problem with learning Blender is that it can do almost anything. it really is a jack of all trades. an honest advice I can give you, should you choose it as a main weapon, is to learn one thing at a time. you can quickly get overwhelmed with all the stuff it can do: hard surface modeling, sculpting, simulation (fire, water, gas...), rendering, compositing... that's the very mistake I made at the beginning: I wanted everything!! so it prolonged my learning experience. by staying focused (in your case, to hard surface modeling) you should progress much quicker.

 

I doubt you'll find quality tutorials focused specifically on ship building. but, going through some tutorials for cars, airplanes and weapons should give everything you need to build any ship you like. about naval terminology and specific maesurements, I'm affraid you're in for some pai... er... challenging experience

[Arthur Goulart]

A little disclaimer at first, when measuring distances off of the plans, I'm looking for the most logical values. That means, I don't think that when setting a line someone would go with the "7/64th of an inch" option, specially because, keep in mind, these plans were drawn at scale. When two possible values get close to the lenght of a given line, I'm picking the rounder one. As a general rule for this project, 1/2 is as far as I am going when dividing inches, unless there is good reason to think a more specific value is correct.

 

Onto the first steps.

 

The first view of the ship to be drawn is the sheer plan, that is, the side view. We start off with the base-line, which coincides with the upper edge of the rabbet of the keel, on ships, it ends perpendicular to the fore and aftmost edges of the lower deck. On boats, however, it stretches from the aftmost edge of the sheer strake to the foremost edge of the stem, and represents the lenght of the vessel, so, in our specific case, 23ft. From both ends of the lines, two perpendiculars are drawn, these are important references for other lines to be set.

 

Next up, to set the midship perpendicular, which not always is precisely amidships. Here it is 10'10-1/2" from the foremost perpendicular. From it, the other, evenly spaced, perpendiculars are drawn. For our launch, they are a weird 1'2-3/8" of one another. I came up with that for two reasons: that value aligns much better with the H and 9 perpendiculars (the outmost ones) of the original plans than a 1'2-1/2" spacing; and, because the plans for a 30ft launch (ZAZ7348), by the same author and date, show, again, the same 3/8" spacing, 1'3-3/8", in this latter case. I wonder the reason behind the odd choice.

The stem is mostly formed by arcs centered 3'4" up from the base-line and 3'10" aft from the foremost perpendicular. That point is reassured by its correspondence with a little dark spot on the original plans, that I assume is where the point of the draftsman's compass laid. From it, with a 3'5" radius the outer edge of the rabbet is set; with a 3'4" radius, the inner edge of the rabbet; and, 3'2-1/2", the inner edge of the stem. These latter two arcs, however, are fictional, they do not correspond with how the finished boat might've looked in the past, and will not stay this way on the finished 3D model. More on that later, they are good enough for now.

 

To the keel and outer edge of the stem. I previously thought that the base of the keel was a little too wide on the sheer plan: the body plan shows a narrower version (more like 6" from the base-line instead of the 6-1/2" shown in the sheer view), and that means that a 3'10" radius from the same center already discussed would intersect perfectly with both the foremost perpendicular and the bottom of the 6" keel. But, the neat result was a bit too off in relation to the plans to my likeing. One particular detail came to my attention, 3'3" from the foremost perpendicular there is a sharp edge, that suggests a not too perfect junction between the stem's outer edge and the keel's base, a bug or a feature? I came to the conclusion that it is, indeed, a feature. Many plans and models show such an edge (the Medway longboat for example), and notably, that 30ft launch (ZAZ7348) by the same author and date also features that edge, and there it looks even more intentional. So, I now have adopted a 6-1/2" keel width, the outer edge of the stem is made by an arc that is centered on the same height as the center of the other radii that form the stem - so they end at the same height too -, but that intersects the base of the keel 3'3" aft off the foremost perpendicular. That solution is much more in line with the plans.

- Arthur

Edited November 22 by Arthur Goulart

[Arthur Goulart]

On 11/20/2024 at 8:37 PM, 3DShipWright said:

@Arthur Goulart - BTW, I didn't mean to seem negative on Rhino.... still a good software, and I'm excited to follow your build.

By no means! You guys just gave me something new to fiddle with 🤣. I went with Rhino because I thought it was the best out there for our purposes, but nop. I'm sure glad I'm still just using a free trial, because Rhino is quite expensive. A free and better alternative? Count me in. By the way, those boats look awesome.

Edited November 22 by Arthur Goulart

[iMustBeCrazy]

15 hours ago, Arthur Goulart said:

A little disclaimer at first, when measuring distances off of the plans, I'm looking for the most logical values. That means, I don't think that when setting a line someone would go with the "7/64th of an inch" option, specially because, keep in mind, these plans were drawn at scale. When two possible values get close to the lenght of a given line, I'm picking the rounder one. As a general rule for this project, 1/2 is as far as I am going when dividing inches, unless there is good reason to think a more specific value is correct.

You might like to look at ZAZ7239, it gives some insight including dimensions down to the quarter inch and varying gaps between stations.

Edited November 22 by iMustBeCrazy

[Waldemar]

 

Nice. And what about trying to find the lengthwise position of the master frame in a proportional rather than just dimensional way? There are several possibilities to check – as a reference length you might eventually find the length between perpendiculars, the length between posts (if different from the previous one), the length of the waterline or the length of the keel (both tread and touch).

You could also check the ratio between rakes of the stem and stern posts. And their ratio to the total length as well, both individually and as the sum of the two.

 

 

 

Edited November 22 by Waldemar

[vaddoc]

Very nice drafts Arthur! Nice surfaces and offsetting.

You ve fallen down the rabbit hole of CAD - I suspect you ll never find the way out!

I think Rhino is fine for our purposes, which is basically lofting boats and getting patterns to cut frames etc and is very user friendly. But I think any CAD can do this. For advanced rendering maybe other options are better but to build a boat, this is not needed.

In the 18th century plans were not hugely accurate and thy did not to be - just shim any gap. However paper gets distorted after all these years and CAD is very unforgiving so some lofting is still needed but not too much - the wood will correct lots of tolerances.

 

Perhaps have a look at my longboat, maybe you ll avoid some of the mistakes I made. 

 

Best wishes

Vaddoc

[Waldemar]

 

Arthur, if you decide to investigate this at all, especially the lengthwise position of the master frame and the like, I'll just hint, just to be safe, that Rhino's command ‘divide curve by number of segments’, is very handy for this purpose.

 

[iMustBeCrazy]

7 hours ago, vaddoc said:

In the 18th century plans were not hugely accurate and thy did not to be - just shim any gap. However paper gets distorted after all these years and CAD is very unforgiving so some lofting is still needed

Yes, the drawing is just a pretty picture ( the shipwright would have built from the table of offsets which I haven't seen for a boat). On the drawing inaccurate setting of a compass would be very easy leading to incorrect arcs, a ruler not quite square, a pencil not sharp enough, heck, a cheap ruler v an expensive ruler. All these and more would lead to errors in the drawing.

 

8 hours ago, Waldemar said:

And what about trying to find the lengthwise position of ...

Yes, there must have been rules (or perhaps more like guidelines ). For masts we have formulae such as "Main-mast twice and a 1/4 the breadth of the boat" etc, there must have been others for, say, the thickness of timbers or the breadth of thwarts or .....

 

[Waldemar]

 

What is the origin of the tables of offsets? Not a previously produced design drawing? 🙂 

 

As a curiosity — in a work published in 1711, William Sutherland proposed a new way of designing. He cited as one of its major advantages the possibility of achieving greater precision in tracing the lines compared to the traditional way, and consequently less waste of time, material and energy on dubbing.

 

[Arthur Goulart]

Guys, thank you a lot for the posts. This little into the project, I already got some valuable insights!

 

15 hours ago, iMustBeCrazy said:

You might like to look at ZAZ7239, it gives some insight including dimensions down to the quarter inch and varying gaps between stations.

Sheeesh, those are really inconsistent gaps. Can you see some logic behind them?

 

14 hours ago, Waldemar said:

 

Nice. And what about trying to find the lengthwise position of the master frame in a proportional rather than just dimensional way? There are several possibilities to check – as a reference length you might eventually find the length between perpendiculars, the length between posts (if different from the previous one), the length of the waterline or the length of the keel (both tread and touch).

You could also check the ratio between rakes of the stem and stern posts. And their ratio to the total length as well, both individually and as the sum of the two.

 

 

 

I like the way that sounds, Waldemar! But, how do I know which proportions to look for? I mean, any division between two lenghts will give me a proportion, how would I know what is correct? I've had a look through The Elements and Practice, in the 'GENERAL OBSERVATIONS ON THE PROPORTIONAL DIMENSIONS' chapter, and I couldn't quite make out something aplicable. Maybe I can get somewhere by comparing different boat plans?

 

10 hours ago, Waldemar said:

 

Arthur, if you decide to investigate this at all, especially the lengthwise position of the master frame and the like, I'll just hint, just to be safe, that Rhino's command ‘divide curve by number of segments’, is very handy for this purpose.

 

Got it!

 

13 hours ago, vaddoc said:

Very nice drafts Arthur! Nice surfaces and offsetting.

You ve fallen down the rabbit hole of CAD - I suspect you ll never find the way out!

I think Rhino is fine for our purposes, which is basically lofting boats and getting patterns to cut frames etc and is very user friendly. But I think any CAD can do this. For advanced rendering maybe other options are better but to build a boat, this is not needed.

In the 18th century plans were not hugely accurate and thy did not to be - just shim any gap. However paper gets distorted after all these years and CAD is very unforgiving so some lofting is still needed but not too much - the wood will correct lots of tolerances.

 

Perhaps have a look at my longboat, maybe you ll avoid some of the mistakes I made. 

 

Best wishes

Vaddoc

Ohhh I will not indeed, Vaddoc! And drafting in CAD is quite addicting. Hours will go by and I won't notice it. I definetely will spend some time taking a good look at your longboat!!

[iMustBeCrazy]

2 hours ago, Arthur Goulart said:

Can you see some logic behind them?

No I can't. ZAZ7239 is certainly not typical but some variation is common. On ZAZ7349 (yours) I wouldn't be surprised if the station spacing was 1' 2 1/4" around the midship area and 1' 2 1/2" at the bow and stern, but total apparent randomness 😲

 

But with ZAZ7239 as none of the lines is where the numbers say they should be, I suspect the draftsman put the numbers on a rough sketch as a less confusing reference prior to doing the real drawing.

That might suggest that the table of offsets was calculated first but might have been 'corrected' as the drawing proceeded. At this stage I'm just guessing.

[Waldemar]

 

9 hours ago, Arthur Goulart said:

Maybe I can get somewhere by comparing different boat plans?

 

Here, please, are some examples with similar attempts. They relate to different periods, places and design methods. I have deliberately cited such a broad spectrum here to show the variety of solutions used by designers of the time. Within a narrower range, for example in terms of time or relating to the same design method, one can find regularities, but today, in practice, one still has to check the various possible variants each time when analysing such plans.

 

In general, when looking for such proportions, start with the smallest integers that divide the reference length. Increase these numbers until you get an adequate correspondence with the plan under examination. Usually the result is below 10, but in one of the examples shown it is as high as 20. But in fact the greatest difficulty is to find out the correct reference length to divide (or to compare, which, for example, applies to the posts rakes). Ideally, you simply need to check them all, as I mentioned in entry #12.

 

Note: the first three examples show designs sporting the so-called double master frame, which are not applicable to your particular project of a boat, however the other details are or may be analogous.

 

 

 

 

 

 

 

 

 

Edited November 23 by Waldemar

[bricklayer]

Arthur,

doesn`t the free trial of Rhino expire 30 days after 1st installation on your computer?

If you learned the basics and modeled  the boat`s hull within 30 days, then I really admire your skills.

Michael

 

[Arthur Goulart]

Following @Waldemar's suggestion, I took to measuring. Not only I looked for proportions on our 23ft launch, but I browsed around RMG's website and found a couple more launch plans dated 25th of April 1785 by Adam Hayes, namely ZAZ7353 (21ft), ZAZ7352 (22ft), ZAZ7351 (26ft), and ZAZ7348 (30ft). I figured, if there is some proportional rule to the construction of these boats, comparing the plans would reveal it. So, I did the unthinkable, created a spreadsheet. Here are the results:

 

I hope the measurement descriptions make sense. Everything is in inches. The "/lenght" columns refer to the previous measurement divided by the lenght of the boat.

 

I was also curious if this exercise could help me with the question for the height at which the lower points for forming the sternpost and the rudder are set, if at the height of the base-line, or at the height of the keel's base. An example will make it more clear, the sternpost's outer edge is a line that passes through the aftmost edge of the sheer strake, and through a second point down below, that is either at the height of the base-line, or at the height of the keel's base, I don't know which.

 

Although the spreadsheet gives pretty damn similar proportions between the different boats at both the base-line's and keel's base heights, I think those at the base-line's height make more sense, that is 1/30 of the lenght = sternpost width, and 6/100 of the lenght = rudder width, at least for the three longest boats, .

 

What do you guys make of it? Do you see any other proportions that also make sense?

 

-Arthur

Edited Saturday at 04:04 AM by Arthur Goulart

[Arthur Goulart]

On 11/23/2024 at 8:08 PM, bricklayer said:

Arthur,

doesn`t the free trial of Rhino expire 30 days after 1st installation on your computer?

If you learned the basics and modeled  the boat`s hull within 30 days, then I really admire your skills.

Michael

 

It's actually 3 months of free trial, Michael. My biggest struggle was more understanding all things shipbuilding related than the program itself 🙃

[Waldemar]

 

Oh my...! I feel a little guilty now, because it's like you're starting to follow my footsteps into the abyss of madness 🙂. Or, into a hitherto hardly explored space. I didn't expect you to approach it in such a meticulous way...

 

Anyway, one can see a lot of regularities on your numerical chart, which clearly suggest that the designer used the same procedure and set of proportions in most of the cases you measured. This kind of juxtaposition can help you to reveal the design logic of this particular boat, if that is your wish and intention.

 

However, as you are already aware, such a method and sequence of ships' boats design is described in both of Steel's works (The Elements and Practice of Naval Architecture 1805 and The Shipwright's Vade-Mecum 1805), but probably best and most comprehensively in Naval Architecture 1787 by Marmaduke Stalkartt. More and more publications are being made publicly available by archives and libraries these days, but if you can't find the latter work but still want to consult it, I'll do the scans for you. Just please check availability on the internet resources first, as it is quite a lot of scanning (a whole 28 pages of textual description plus a very large drawing sheet, which I would have to scan six or eight times and then stitch together). Please let me know.

 

 

Edited Saturday at 02:02 PM by Waldemar

[Thukydides]

Google books has stalkartt 1781 here:

https://books.google.ca/books/about/Naval_Architecture_Or_the_Rudiments_and.html?id=kdZLAAAAcAAJ&redir_esc=y

Edited Saturday at 05:50 PM by Thukydides

[iMustBeCrazy]

8 hours ago, Waldemar said:

but probably best and most comprehensively in Naval Architecture 1787 by Marmaduke Stalkartt

 

2 hours ago, Thukydides said:

Google books has stalkartt 1781 here:

Thankyou both, that looks like it's going to be an interesting read.

 

In just the first paragraph of chapter 1 we have for the stem "determine the shape of the stem; and find the centre (E), that shall sweep the lower part, at pleasure".

So no rule there.

 

And "And draw the line CF for the aft part of the stern-post, to what rake you please"

So no rule there either.

 

I believe this is plate 1.

Edited Saturday at 08:21 PM by iMustBeCrazy

[Waldemar]

 

Yes, on the one hand this is true. And there are many more such arbitrarily chosen design components in Stalkartt's description. And not without reason, as he describes things in a most valuable, universal way, that is, he lists all these design components, explains what they serve and how they are applied and in what order. Precisely consciously in the most universal way possible, independent of the specific parameters used by individual designers and for various types of boats.

 

Today, without this background knowledge, it is not at all easy to reproduce the design process even with the original plans. It is therefore usual to redraw the lines from the extant plans and then smooth them out by various means. This present-day alternative method is also good for specific goals, but has its own peculiarities and limitations.

 

Yes, plate I and pages 1–28 on whole-moulding.

 

[Arthur Goulart]

 

14 hours ago, Waldemar said:

 

Oh my...! I feel a little guilty now, because it's like you're starting to follow my footsteps into the abyss of madness 🙂. Or, into a hitherto hardly explored space. I didn't expect you to approach it in such a meticulous way...

 

Anyway, one can see a lot of regularities on your numerical chart, which clearly suggest that the designer used the same procedure and set of proportions in most of the cases you measured. This kind of juxtaposition can help you to reveal the design logic of this particular boat, if that is your wish and intention

 

 

 Yeah... The abyss has stared back at me, nothing I can do about it now 😆. I can already see some ways the comparison will be useful in the upcoming steps, I'll probably be able to improve, or be reassured of my design for the boat's rake, sternpost and rudder. Having those four other plans is quite the unique opportunity to understand what was the designer's process.

 

And on the topic of Marmaduke's Naval Architecture, I'm actually kinda surprised I had no idea that book even existed. I had a quick look through it, it seems to be just the thing for this project, much more complete than Steel's when it comes to whole moulding, and that plate 1 find was a great one, @iMustBeCrazy! 

 

And btw, do any of you have a pdf for The Shipbuilder's Repository? I have looked everywhere for that and found nothing but crazy expensive physical books.

[Waldemar]

 

There is quite a significant difference in the shaping of the hollowing/bottom curves in Steel's and Stalkartt's work. While the former uses a very traditional way, going at least back to the antiquity, by using a properly curved and scaled wooden template (related to non-graphical methods, i.e. without paper plans), the latter has a different way: simple straight lines for the central part of the hull, much more convenient to draw on paper plans, and the bends/frames at the extremities of the hull are even shaped in their entirety on the basis of previously drawn waterlines (such approach is strictly related to newer, graphical methods of designing). But it is already your job to recognise such specificities in the drawing you are examining 🙂.

 

 

 

Edited Sunday at 10:49 AM by Waldemar