„Święty Jerzy” („Sankt Georg”) 1627 – reconstructing an opponent of „Vasa”

[Kranck]

I'd go with a diagram, sir !

 

Seems to me that you  define the rabbet location beforehand, which is fine, but what about its shape? Since the rabbet shape is sure to be different along the queel

[Waldemar]

 

The original intention was to show the entire process and in different variations, but there would have to be a manual longer than this thread and it would take much more time than I can devote to it.

 

So the idea itself:

 

1-2. create the keel-stempost assembly as one entity, define the master surface inside the planking;
3. divide the master surface into separate strakes, offset them as solids to the plank thickness, find the intersection of the outer surface with the keel (red line);
4. explode the garboard strake and delete all resulting surfaces except upper and outer;
5. loft lacking surfaces using existing edges and the red line, combine them into the closed polysurface;
6. this is another possible variant featuring trapezoidal garboard strake and rectangular regular strakes (in contrast to bow like) in cross-section. For this, lower the red line, explode all strakes, delete their non-touching surfaces, loft new surfaces using existing edges, then combine them into separate strakes. This variant, while more realistic, may be also a blessing if you intend to unroll the strakes into the flat surface for later CNC machining,
7. Boolean cut the keel assembly.

 

And good luck!

 

Edit: in step 3-4 you can still simplify the task by lofting the surface using both rabbet lines, inner and lower (red line). Then just use this surface to Boolean cut off the garboard strake to its final shape.

 

 

 

 

Edited December 21, 2022 by Waldemar

[Waldemar]

 

Yet another variant, if your model would be large, or you seek for still more realism or just prefer to have shallower rabbet. Loft the cutting surface using the upper rabbet line (red line) and the second line, placed about midway of the garboard strake thickness (another red line). This is all shown below.

 

 

 

[Kranck]

That is a very precious post. 

Thank you so much. 

It will solve a lot of issues, CNC milling wise. 

I'm struggling with surface lofting at the present, and quite happy about it (gotta sweat a little to get something nice in this life !). Your previous explanation will come very handy when limiting the loft to the queel (that I just finished drawing). 

 

Now, time to remember to the joys of gradients and curvatures. 

 

By the way, I reviewed your previous posts and I can confirm that I am also thinking about drawing the INNER master surface for the full delineation of the frames. However, since there is no direct 2D projection of those curves, I was thinking about cheating a little and using the variableoffsetsurf fonction: i,e, offsetting the OUTER master surface with a variable thickness to account for the thicker base of the frames. Not sure it will be that accurate though. 

 

 

Best regards,

 

Nicolas

 

Edited December 22, 2022 by Kranck

[Waldemar]

 

15 hours ago, Kranck said:

However, since there is no direct 2D projection of those curves, I was thinking about cheating a little and using the variableoffsetsurf fonction: i,e, offsetting the OUTER master surface with a variable thickness to account for the thicker base of the frames. Not sure it will be that accurate though. 

 

Well, perhaps no good news here, as I guess you may simply waste your time with this apparent shortcut. Most probably it just won't work as you would wish with that shape. Rather, consider using the blueprints of the frames and your ingenuity to get the right form by „manual” means. Initially, it may seem more labour-intensive approach, yet, in the end, you may save a lot of time by avoiding later corrections, re-doings and even starting the whole again from scratch.

 

 

[SZKUTNIK]

Hi Waldemar,

 

Could you show the decorative elements for the model St. George?

 

 

[Waldemar]

 

Hi Christopher,

 

Although I already have a general vision of the decoration, I have left its detailed design for the end, according to the order requested by modellers already making a wooden model of the ship. But maybe this order will still change. Anyway, I will show it as soon as there is something to show.

 

🙂

 

 

[Waldemar]

 

A complete anchor set for the ship. According to the 1629 inventory, there should be „four heavy anchors” (vier schwere Ancker) and „one kedge anchor” (ein Wurff Ancker). For some ships, grapnels (Draggen) are also listed as boat equipment. These were also useful in boarding fight. 

 

All six anchors are of different sizes.

 

 

 

 

 

Edited January 5, 2023 by Waldemar

[Waldemar]

 

Thanks to the extraordinary kindness of Danish museum curator Peter Kristiansen, I can now also exploit the full potential of an extremely important model of the Danish ship Norske Løve (ship built 1634, model built 1654), especially for reconstructing the masting and rigging. The masts themselves are almost complete.

 

To give an idea of the appearance of this unique model, I include below a photo taken from the website of The Royal Danish Collection in Rosenborg Castle:

 

 

 

[Martes]

3 hours ago, Waldemar said:

Thanks to the extraordinary kindness of Danish museum curator Peter Kristiansen, I can now also exploit the full potential of an extremely important model of the Danish ship Norske Løve

Lucky you.

 

When I asked Danish Kriegsmuseum about the model of Christian VII's stern, they told me it's not in their collection and they don't know where it is. It should have been transferred from the former Maritime museum, but apparently wasn't, or wasn't indexed, anyway, they couldn't find it.

[Waldemar]

 

@Martes

 

You have probably already found something on the web about the model of Christian VII. If not, and if the matter is still relevant, I can offer you to make scans of the below illustrations from the Danske Orlogsskibe 1690-1860. Here only photos, because of the large format.

 

Anyway, when asking for help of this kind, it is essential to show your work. This is always effective, and the Scandinavians are particularly kind and helpful, as I have personally experienced many times.

 

 

 

 

 

 

Edited January 16, 2023 by Waldemar

[Martes]

Oh, I did! I even explained what I needed exactly, and they found me a photo of that model (numbered 149, the black one) that was not anywhere on the net (I never encountered it, at least - see below), but they still don't know where the model is and if it exists at all. They were extremely nice and helpful, it's the disappearance of the model as a physical object that made me envy you The plans are in the Danish archives and I used them all already

 

 

Thing is I needed a very specific view of that model (top projection, to verify the rail run on the roundhouse and top tier of the quarter galleries - it is very awkwardly positioned, as if stepping from the edge roundhouse onto the quarter gallery, and how exactly it is done is not drawn anywhere), and without the model it is absolutely impossible.

 

I just wanted to see this curve from above, and the corresponding place on deck - where the roof of the roundhouse ends and where the gallery cover begins.

 

 

And the fact it could be just lost because they merged two museums is very saddening. Just hope it would turn up during next inventory.

 

I apologize for intruding here, of course, you can remove all three posts if you want to clear the thread.

Edited January 17, 2023 by Martes

[Waldemar]

 

@Martes

 

Ah, it's as much your space as mine. And you in particular are always welcome, as after all I am still in your debt especially for pointing out good copies of the originals of early 17th century ship plans in the Russian Hermitage (those presumably by David Balfour or his contemporary).

 

Anyway, now you have no choice but to make the inevitable detective visit to Europe, specifically Copenhagen next summer 🙂. That's how I've had to go to Stockholm and a few other places on more difficult cases in the past, too. Until then... however, I decided to scan the one photo shown above, which may not be in the Danish online archives. Indeed some bending can be seen there...

 

 

 

Compressed JPG file (6804x5037 pixels):

 

model of the Christian VII's stern.zip

 

 

[Martes]

14 minutes ago, Waldemar said:

now you have no choice but to make the inevitable detective visit to Europe, specifically Copenhagen next summer

I'll give them some time to go through the storage, but if I'd ever have (or get) to do it myself, it'd be fun.

 

Thanks for the image, I don't have it in that resolution

And here is what they sent me - the image above in original and a set of photos of Prince Christian Frederick, which has different config, and is a modern interpretation, but still it's something.

 

And just to note that I do watch your project with great interest.

Prins Christian Frederik.zip ChrVII(1803)agterspejl-285-2000.zip

[Waldemar]

 

Thanks a lot Martes!

 

But wait a minute! After all, these are two different photographs and we can even play the popular game "find 10 differences in two pictures". In one there is no rudder chain, in the other there is no boom support, they are at a slightly different angle, maybe something else....

 

 

[Blas de Lezo]

Hello Waldemar,

 

Your project is really inspiring. Thanks a lot for sharing it. Let me make a couple of questions, if you don't mind it. If I understood right, your model is made with surfaces, is that right? That means, based on plans and diagrams, you draw curves (splines basically), and created surfaces between these curves. Correct? If so, that means, for example, one mast is formed by three surfaces (cylinder), or a simplified beam (straight or curved) by six surfaces. Is that correct?

 

If it's so, to make STL or OBJ files from the surfaces you have to transform the surfaces sets to objects (polygons). Do you experience issues in this transformation or with Rhino is straightforward?

 

Thank you very much for helping me to understand properly.

 

V

 

[Martes]

2 hours ago, Waldemar said:

But wait a minute! After all, these are two different photographs and we can even play the popular game "find 10 differences in two pictures". In one there is no rudder chain, in the other there is no boom support, they are at a slightly different angle, maybe something else....

Wow. I didn't pay enough attention, thinking that two images taken from (almost) same point are the same. And they are not.

There are two other differences, the removed boom support enables us to see the horizontal positioning of the last netting support, and it is indeed on the quarter gallery (however, there are some questions as to how turn of the deck is executed there), and the image without the starboard chain shows rails on the gallery balcony that go to the aftermost gunport (they are obscured by the crown on the second image, but they are visible on the photos from the book below).

 

And there are two more images of this model, on page 35 of Holck's catalog (1939) (PDF), one of them is side-view.

 

 

 

 

Edited January 17, 2023 by Martes

[Waldemar]

 

@Martes

 

Martes, I saw in your thread that you were creating attractive images of 3D models of ships from this era, but you have now just made me aware of your diligent approach to this project. Few, maybe even no designer for movie, animation or video game purposes except you, would care about such details, especially after encountering any difficulties. The secret of the appeal of your models is also explained – they are not sloppily based on the designer's imagination, but on real, historical data. It is something like this that distinguishes a reliable professional/researcher from others and creates his authority. Respect is due.

 

 

Edited January 17, 2023 by Waldemar

[Waldemar]

 

@Blas de Lezo

 

Blas, your question is a very good one, as it will allow me to clarify the general mechanism of my reconstruction.

 

This reconstruction is not based on any plans, because such plans are non-existent. I am also not basing it on any modern interpretations or today's reconstruction/modelling plans of other ships (unless for results comparisons), but solely on the primary sources from the period. In fact, I am creating this 3D model from geometrical-visual scratch in order to later produce 2D documentation for modellers on this basis.

 

The example of the main mast illustrates this reconstruction process very well:

 

From the notarial fleet inventory of 1629 (Inventarium uber Ihre Königl. Maytt. zu Polen undt Schweden etc. Erlogs Schiffe, so den 8 February A[nn]o etc. 1629 zu Wismar glücklich einkommen, undt daselbst besichtiget worden. Wismar den 3/13 February 1629) it is known that the larger ships of the fleet had a main mast of four parts, as it lists three sails for this mast, and flags: main course (ein Schon Vor Segel mit seim Bonnett), topsail (ein gross Marss Segel) and topgallant sail (ein gross Bramm Segel).

 

Then, the dimensions, shape, position, rake, accessories and arrangement of the masts are taken from the two 17th century Dutch manuals I mentioned earlier in this thread: Evenredige... (c. 1650) and Witsen (1670), and in addition from iconography, shipwrecks and period models such as Peller-model (1603), Tyresö-model (early 17th century), Vasa (1628), Norske Løve (1634/1654), all of which are Baltic ships models.

 

Now, back to your question:

 

Generally, masts taper at the top between 2/3 and 3/4 of their largest diameter (at the partners), and the length of their tops (i.e. above the circular fighting top) is 1/10 the length of the mast (taken from the works above). Masts are shaped like a fragment of a circle rotated around the mast axis. But that's not all, the cross-sections of the mast change along its length. They can be circular, elliptical, quadrangular, octagonal, be it regular or irregular. Below is the appearance of my mast at the fighting top height (please note its variable cross-section and the side cheeks supporting the trestletrees).

 

 

I personally have no need in this particular case to convert NURBS geometry to a mesh (i.e. to STL or OBJ format), but Rhino does a good job at such an export. You can also convert NURBS geometry to a meshes and then edit them inside Rhino, as Rhino has advanced tools for such mesh editing. I love this program, it only has one drawback, acutely important to me: it doesn't always do a good job of converting 3D shapes to 2D projections, and as a result, it often has to be done manually. But this actually may not apply to you...

 

Feel free to ask any other questions, preferably ones that I can answer. 🙂

 

 

Edited January 17, 2023 by Waldemar

[Blas de Lezo]

Hello,

 

Thanks a lot for your kindly explanation. Well, I can say you have my admiration, because what you do is research, it doesn't matter if the tools and/or the result is a academic paper, an exposition or a 3D model. What is important is the collection of information through original sources, and the process to create or recreate the different elements, as you explained with the mast. That is history investigation, and you have my admiration.

 

Time ago I worked professionally with Maya, later I "converted" myself to Catia, also for professional reasons, and now as a hobby I work with Fusion360. The problem with Maya, at least time ago, to do a model like yours in NURBS was all the time I needed to group the different surfaces that conform an object (e.g. cube, cone, etc.). In Maya I used to model Modernism and Art Deco buildings, and with NURBS was incredible easy to work with.

 

For your problem with the projections, maybe Fusion360 or SolidWorks could be a solution, try versions could help you to experiment and check if it works better than Rhino. Another solution maybe is to import the .OBJ in 3DMax and export the object as .DWG

 

Kind regards,

 

😄

 

 

[Waldemar]

 

Fighting tops. Five fighting tops.

 

 

 

 

 

[Chapman]

I am pleased that the Sankt Georg is continuing here.
The reconstruction looks very convincing. A cardboard model of the ship would be nice.

[Waldemar]

 

Thank you Chapman. Yes, it should, as I am making every effort to base this reconstruction solely on extant period sources. For now the priority is to complete the plans for the museum model in wood and at the moment I cannot promise a cardboard kit soon. However, thank you for the idea. I would only add that for a commercial kit alone it would probably not be possible to devote even 1/50th of the time consumed so far on this reconstruction.

 

 

Edited February 26, 2023 by Waldemar

[Waldemar]

 

 

 

 

 

 

Edited February 27, 2023 by Waldemar

[Waldemar]

 

Blocks are a great feature of CAD software as they allow multiple detailed assemblies to be inserted into the main file without size penalty, which in this case has already grown to almost half a gigabyte. This is shown here using the artillery example.

 

 

 

 

 

 

 

 

 

 

 

Edited February 28, 2023 by Waldemar

[scrubbyj427]

Incredible modeling! I would love to build this.

[Waldemar]

 

Thanks a lot, Scrubby. Are you more interested in building POB or POF models?

 

The modellers here are already at an quite advanced stage of hull construction. They claim that the frame station shapes are ideal in the sense that it is very comfortable to lay the planks and get nicely profiled body straight away. In just one layer. But they also had the frames (bulkheads) cut with a precision CNC milling machine. As an aside, the hull shape was developed by the historically correct hauling down/pulling up futtock method, without spoiling it with later design waterlines or diagonals.

 

I 'only' still have the rigging and decoration to work out.

 

 

Edited March 1, 2023 by Waldemar

[scrubbyj427]

9 hours ago, Waldemar said:

 

Thanks a lot, Scrubby. Are you more interested in building POB or POF models?

 

The modellers here are already at an quite advanced stage of hull construction. They claim that the frame station shapes are ideal in the sense that it is very comfortable to lay the planks and get nicely profiled body straight away. In just one layer. But they also had the frames (bulkheads) cut with a precision CNC milling machine. As an aside, the hull shape was developed by the historically correct pulling up/pushing down futtock method, without deforming it with later design waterlines or diagonals.

 

I 'only' still have the rigging and decoration to work out.

 

 

I would build it any way you release it.  But I would want the main gun deck built completely without dummy barrels or anything hidden. 
Are there any photos of prototype models being built?

[Metaspace]

Amazing work!

I'm a bit surprised to see the gun ports of the aft cannos seemingly to be smaller then the rest. Do the guns have a smaller calibre?
With the Papegojan of the same period, with assumedly a similar setup of aft cannons being used either in the rear, or to add to the broadside, were of the biggest cailbre, slightly larger than the rest of the main gun deck.

As I am confident you have done thorough research, would you have an explanation for this? Or am I simply mistaken because of the angle of the drawing?

[Waldemar]

 

@scrubbyj427

 

Scrubby, yes, there are, I even made a few during my last visit to the modelling studio, but I didn't get permission to publish them for publicity and advertising reasons. I am unwilling and unable to disobey this explicit request. Perhaps in a private post just for you...

 

 

@Metaspace

 

Many thanks, Roman,

 

The size of the aft gun ports on the stern and the broadside gun ports on the lower (artillery) deck are the same, as the calibre of the guns is identical (6 lbs). Please don't take into account the last smaller port on the side as it is a cargo port (of the mail type etc). But the gun barrels themselves are not and cannot be the same. The broadside guns have to be short and the aft guns long for structural reasons, otherwise the gun barrel muzzles would be inside the ship. Please take a look at the attached renders to see that shown by graphical means.

 

Instead of the iron 6-pounder guns, theoretically the bronze 3-pounder guns (also mentioned in the gun 1628 inventory for this ship) could be inserted into the constable's room (gunroom), but this would not be in accordance with the quite detailed description of the battle.

 

Also for the same structural reasons the bow guns had to have very long barrels. You will also see this in the renders attached below, and these bow guns just cannot be moved forward any further. 

 

By the way, it must be stressed that during this period of dominant boarding tactics, the offensive bow guns were the most important artillery asset in the ship and these usually were of the largest calibre. This is not only evident from Dutch or English sources, but is also confirmed by the very description of the battle itself in which the Sankt Georg took part (in this particular case two 9-pounder guns mentioned in the gun inventory).