TJM

That is a fantastic model!

I believe the wiki page just echoes the museum, but it will be very interesting to see if you have the display case text! Lex also calls it Christiania, referencing Orlogsbasen, Rigsarkivet: https://trap.lex.dk/Christiania_(1774) You may well be right here, and I agree that the drawing and the model are a match, but it is strange that current sources reference it as Christiania then! BR TJM

This is the section in 'Danske Orlogsskibe 1690-1860. It agrees with your description, however, like I mentioned, I wonder why it is then labelled as Christianias figurehead at the museum!? And whether some new information has appeared since the 70's that changed the identification of the model?

Hi Matthias, Thank you so much for your kind words and the very interesting comment! The figurehead is supposed to be that of Christiania, not the Printz Friderich and the model is identified as such in the museum. You will also find this on Wikipedia: https://commons.wikimedia.org/wiki/File:Galionsfigur_til_fregatten_Christiania_(model).JPG But I agree that it does look much like the coloured drawing of the mounting of the figurehead - I wonder if the same motif was re-used? It is a bit of a conundrum as the original plans does not show this exact motif, but rather a lion or another lady figure not holding a building. It does however seem like this figurehead model was previously identified as the one for Printz Friderich - it is mentioned as such in the 1979 book Danske Orlogsskibe 1690-1860. Perhaps it has later been re-identified as Christianias figurehead? Or perhaps the museum has it wrong! Unfortunately, my own photographs does not contain the annotation on the display case. There is another very similar figure just beside this one, could that be the Printz Friderich figurehead? What do you think? BR TJM

Log entry 31 - a new figurehead A while back, I made a figurehead for Christiania using Hero Forge. I was quite happy with it, as the pose and motif was faithful, though the details were somewhat different to the original. But now, as can be seen in Chucks recent post in the 3D modelling sub-forum, Meshy AI has become really good at making a 3D model from just one image. So I had to try to re-do the figurehead with this new tool. Here's the cleaned up image I fed to Meshy: Here i have removed headworks and the other details on the keel. I have taken this photo myself at Krigsmuseet in Copenhagen where the model of the figurehead is displayed. Meshy gives you 5 runs/tries when you submit a task. I submitted twice, and used 7-8 runs to get the version I went with. They were all very good, but I needed a version with space for the keel between the legs and with no weird stuff in the left hand. The Meshy models look great and can be printed as is, however I need a slot for the prow and it is really a pain to get a single solid with no topological issues. but after a few hours of trial and error, I managed. Here's a comparison of my Hero Forge figurehead and the new Meshy one: Honestly, the Meshy one is much better, as the style is correct. It is very close to the original model at the museum. Btw, I did the callipers myself, the AI could not get that right from that angle. Here's printed comparisons: I know there is a bit of a size difference here as well, but you can clearly see that the pose is also much better and more natural in the Meshy version. Finally here's some shots of the figurehead on the model - the fit is great and I have just 0.5 mm of filing on the prow to make it perfect. I am super happy about this upgrade to the figurehead! BR TJM

Ok, so I tried the reverse strategy for the painting. It may seem a bit foolish but there was a reason to try this, that I will get to. Here's how it went: The part was base coated, first with black, then with white. I use an airbrush for this. Then I used an Army Painter brown Speedpaint to get a dark yellow base. Speedpaints (like Citadel Contrast paints) are basically heavy pigment acrylic inks that will run into the recesses, but also leave a nice base colour on the raised parts. It is meant to be one coat and done. It looks super blotchy, as it is not so well suited for flat surfaces, but that is ok, it works where the yellow will be, particularly the friese. Next, I went owe the yellow with the ochre, mostly drybrushing: And then topped it off with a drybrush of beige for the highlights: After this I sealed it with an acrylic varnish and let it dry completely (a full day). So far, so good. Now comes the tricky part: tye recessed black parts. I did my best to mask off the yellow: But even then, the black seeped onto the yellow 😭. This was not really unexpected, but I had hoped masking it off would work better. I stopped here, as even if I could go over and clean up the yellow, the black edges are not straight enough and the point of doing it this way disappears if I have to re-do the yellow. So, the reason for trying this way is that drybrushing is very messy, especially when you have large, flat areas of a very different colour around the areas being drybrushed - as I do. Drybrushing the yellow gave a really nice result, the best I have achieved so far, but I can't get the black good enough. It is almost impossible to paint recesses without messing up the higher parts. Oh well. I will go back and do it the traditional way, building up the colours from lowest to highest. That just means I have to extremely precisely paint the yellow parts (of which there are many!) several times as I built up the colours... 😅. After white, yellow is the most difficult colour to get a good finish on. Anyways, back to painting the second version.... BR TJM

Thank you Matthias, that is very kind of you to say, but I think there are some improvements needed 😃. I have been painting Warhammer miniatures for a long time, but their 32 mm scale and fantasy/sci-fi style is much easier to paint compared to naturalistic 1:64 scale! 😅

Thanks for all the likes! And thank you @Ronald-V and @brunnels for your kind and encouraging words! BR TJM

Log entry 30 - final(?) stern version and a colour test After a few quite some iterations, I now have a fit that I am happy with. And I like the design. With an additional plank added below the stern and the lower parts of the side galleries made from wood (that will be a small challenge as well), I think this will be good. I made a colour test from the last version (the one that didn't fit as well as this newest one), as the only things I am 100% settled on is that the windows will be red and the friese and ornaments will be yellow ochre. So I did not aim for a super neat paint job here and have used mostly flat colours with minimal shadows and highlights. This is just to evaluate the colour palette: I am resonably happy. Or, I should say, I like this colour scheme, but I am a bit apprehensive about the prospect of actually trying to make a super neat paint job. It is extremely difficult! And even though I am out of training, I have been doing miniature painting for over 25 years and this is really some of the most difficult I have tried. The print is so crisp and nice and I feel I butcher all that lovely detail with globs of paint! Organic shapes are much easier to paint than flat surfaces or straight lines! This test took me at least 4-5 hours to paint, and I will have to go much slower, with more, thinner layers for the final version. I am considering if I should try a contrast paint method, basically a wash method, to get the basics colours in and then do a single pass of main colour and highlights for the yellow. I could perhaps also mask off and airbrush the red in after the initial wash. Or I could airbrush it all yellow, mask off and do the red and then add the black after. Hmmm, I will think about this while I continue with the planking .... BR TJM

I made the adjustments but it is really hard to see on the photos. The whole window section is now curving and I took the opportunity to add lower stern decorations. These are made with Meshy.ai, see the recent discussion on the 3D modelling sub-forum, and will be the first aI generated elements I use. The fit is much better, but there are still room for improvement. I also have a weird bend at the top of the side galleries, symmetrical on both sides. This is a printing issue, but I am not sure what happened. Something shifted or there is a lack of supports that make the model sag or compress. I also have a lot of support residues on the side galleries, a result from trying to support the bars to prevent bending during printing. But i changed the release film on the resin vat and am getting crisp prints across the build plate again, so that is nice! I have to look into the printing technicalities and also make another small adjustment to the angle of the side galleries to improve the fit further. But it is getting closer 🙂.

Log entry 29 - test print of stern I have tried to make a model of the stern that include everything in one part: windows, friese, side galleries, etc. I have a bit of issue with the print itself - I think I need to change the release film on the resin vat as I may have a tiny leak with a bit of cured resin underneath that blocks the light. Here is what it looks like: Overall, I am very excited to see this as it really shows what the stern will look like eventually, and I am very happy with how the different parts look. But... The fit is absolutely terrible 😅. Especially around the lower parts of the galleries. And the curvature of the rear window panel is also causing issues as the model is not a great fit there either. What I will try do do is this: re-model the rear window panel with a curvature and remove the lower parts of the galleries and make these separately or in wood to insure a tight fit. It will probably take a few more iterations to get it good enough, but I think I will get there eventually. BR TJM

It is not a problem to make a taper like that, it is just two mirrored planar cuts to the 'tool' part. You can see an example from the slicer below and that there is around a 1 degree taper in two directions: Then you can get something like this: I am sure this is not completely accurate either, reality was likely more complicated. The accuracy will of course depend on how good your model knee is, but the procedure for the slot remains the same. I think it should be possible to make something that is good enough for a model with an aproximate 'tool' shape.

Alright, here's a detailed walkthrough of how to make the slot in the figure using Onshape: 1: make the prow template I am sure there are many ways to do this, but this is how I do it. I draw the shape in QCAD, using the scaled drawing as reference: You can do this in any CAD program. Make sure all the endpoints touch - it needs to be a closed shape to enable the extrude operation. Then I make a new project en Onshape and create a 'Sketch': You have to pick the plane to sketch on (I pick the front, it dosn't matter) Then, before completing the sketch operation, click the DXF function at the top right: Import your prow 2D CAD file: click the tick mark on the sketch function to complete this: Choose the central shape (not the lines) and use the extrude function (top left): Here you pick the depth to extrude, the thickness of your keel, but give it a bit extra, othewise bleeding of the pixels during 3D printing will likely make the cut to shallow to fit. You must experiment with the correct thickness. Then export this 3D keel shape as an stl: Import this and your figure into the slicer (you can do it in onshape as well, but I am not good at moving things around in that program, so I prefer the slicer). Adjust the position of the parts to fit and combine the parts (at least in Lychee, you have to do this to export both in the correct relative positions): Export the object to an stl: 2: make the slot with the boolean function in Onshape Now we need to import this model into Onshape. From the main menu, click the create>import Import your stl - the progress will appear in the notifications tab (a bit counterintuitive): When it is done, you get a link to your imported file: Now we are redy to do the boolean. It is at the top bar, two intersection circles icon: As it says, you can do three types of boolean operations, union, subtraction and intersection - we need to subtract. The prow is the 'tool' and the lion is the target. You can have several tools operating on several targets, if needed. In my case, three lion parts appear - this is because I have two small, free floating bits of mane - normally you would just be left with one slotted figurehead part: Then you just export the part(s): I have seen instances where the boolean function fails - this happens if there are very extreme topologies as a result of the subtraction, but mostly, if you have so-called water-tight stl files, this just works like a charm. Let me know if there are any points that need clarification! BR TJM

If Chitubox can do this with non-perfect stl's, that might be easier! The repair was the biggest challenge for me. I will still post the step by step using Onshape though 🙂

Yes! I will do a step by step with more detailed screenshots. Stay tuned!

I also don't have any skills with the likes of Blender and Z-brush, but I have found a workaround that makes it easy to make the slot it the figurehead: I draw the shape of the prow and extrude it to the correct depth: I then import it into my slicer (Lychee) along with the figurehead and position them. I find it is much easier to work in the slicer than in a 3D CAD progam when manipulating the part positions: I then merge the parts and export a combined stl. The file actually still contain the seperate parts, which is detected by the next program, Onshape, a free online 3D CAD program. The prow is then subtracted from the figurehead with a boolean operation: and the final slotted figurehead can then be exported and re-imported into the slicer for printing: This all takes 10 min total - except that the original file produced by meshy has topological errors. They are fixable to a sufficient extend in the slicer for the purpose of printing, but that is not good enough for the high tolerances of the Onshape program, so it is nessecary to repair and re-mesh the original figurehead file before this simple workflow can be completed. And that took me a bit of time to work out! I used a free Autodesk program called Meshmixer to repair and re-mesh the model here. So provided that you have 'water-tight' stl files, making the simple geometric subtraction to get a slotted figure is relatively simple, but it took some work to get there! I have attached the slotted file here as well. BR TJM Grønland figurehead cut.stl

Sure! I ran it a few times, and they are all really good, but all have a few minor issues - some will be easier to fix than others. This one has a little thing sitting on the hands and it has the tail mirrorerd on both sides. But even with these small shortcommings, I am sure it is workable with a few tweaks! FYI, this is the figurehead for the Danish 50 gun Ship of the Line Grønland (Greenland) from 1756. Grønland Figurehead.stl

This is truly an incredible leap that has hapened in the last few moths with the latest AI models. I tried a few examples of using original construction drawings to get figurehead stl's and I am blown away. Heres a 'standard' lion: To me this is crazy! All I did was take 5 min in photohop to remove everything from the drawing that was not part of the lion, fed it to meshy and voila! And here is a non-symmetrical one where it had to extrapolate the left side: You can see that there is not really much to go for in terms of detail, especially in the face of the cupid, but look what I got from the model: For a 3 cm tall figure, this is going to be all fine. Basically, this was the last big nut to crack for me in my effort to design ships from the original plans. Lot's of learning and experience still to go, but the tools are here now. I am really amazed. Thank you chuck, for pointing out that AI have come to this point now! BR TJM

I have an Elegoo Mars 4 DLP. Small, but good enough for most parts we are interested in. And since it is a projector, as opposed to a LCD, I don't destroy the printer if I have a resin spill. But beware of the safety aspect of resin printing and the need for a dedicated space with good ventilation and room temperature!

Wow! This is fantastic! I have tried to do this with various apps, also the free version of Meshy, but were not able to get anything useful. I will definitely dive into this again! In any case it is just a question of (short) time before this will be possible from even worse reference materiel. Thank yo so much for exploring this and posting it here!

Ok, I made a print yesterday and in general, it turned out really nice. I printed two with grating and two without, but one of those with failed. A few observations: I should probably use slightly thicker planks. 0.4 mm is very thin (and see through!) 0.6-0.7 mm is probably better. There is a bit of cavity under the deck planks. Cavities with no drain is a big no no in resin printing. Technically these are not closed off, but it seems the openings are just a too tight here. You can see a bit of seepage around the openings. The solution is to fill out all the volume under the planks or make the gaps between planks a bit wider. I do think everything is so thin that I am able to cure everything, but it is not ideal as is. This is close to the largest object I can print. Including supports, this was 136 mm tall and I only have 150 mm total. Without rudder, I can go up to a boat of max 30 Danish feet. This was 2700 layers and it took a full 12 hours to print! However, each boat, including supports only consume 10 ml of resin 😄. If I can fit in wood grating, I think that will be the better option, though the 3D printed one is ok. All in all, I would say this experiment was a success and a good proof of concept for future ship boats! At some point I will make the 20 ft Jolle and a 23-28 ft Challoupe for Christiania as well.

Thank you @Thukydides! I think I got lucky and that my methodology just happens to 'work out'! I also think it helps that I am doing this on a plank by plank basis, as compared to making a surface that hugs the whole of the hull in one go. Here's an example of a plank of average complexity in terms of topology on my boat: Firstly, the planking runs are determined already in 2D CAD based on the lines from the original drawing (with my crude extrapolations on top): In hindsight, I should have been more carefull, especially around the upper 4-5 planks and made some smoother help lines. This would have avoided the slightly wonky run of those planks, but when I discovered it, I was quite far and it would have meant starting over, as all the frames would need to be altered. Here is a look at the surface that will form the 4th plank from the top: It is quite extreme and a lot of it will need to trimmed away. I found that the shape only worked when I had fixing points at the keel (the rabbet?). After this, I made two 'limiting' surfaces where the plank edges should be. I use the klinker shape of the frames to make the perpendicular surfaces: After extruding the two help surfaces, they are used as tools in the subtractive boolean operation, removing the unwanted part of the plank: This leaves just the plank in the correct shape, but it is quite tedious to do, and for some of the surfaces, I was getting close to the limit of computing power in Onshape, as it was getting quite slow! After repeating 12 times, I got the 12 planks, in the correct position, faired against the frames. I should mention that when thickening the frames, I extruded 0.1 mm outwards and 0.3 mm inwards. This means that non-faired nature of the frames gets covered in the plank thickness, and that any micro-protusions of the frames through the surface gets covers by the outward extrusion. I hope this makes some sort of sense! I basically have no clue what I am doing and I am lacking some vocabulary/teminology to properly describe the process, I just try until it seems to work!

Log entry 28 - a 25 ft 'Barkasse' Hi everyone, Wow, progress is slow for me at the moment 😅. Vacation, covid, other commitments, etc. has meant that I have not managed to progress with the actual ship since last time. But I found myself exploring a particular rabbit hole: would it be possible to make acurate ships boats for the model? Many of the drawing sets at the Danish National Archive includes drawings of ships boats, and there are two for Christiania, a 20 ft 'Jolle' (=Yawl) and and a 25 ft 'Barkasse' - I believe the english equivalent of this is the launch. I think it is likely the ship also carried a 'Challoupe' (Pinnace/barge), but there are no drawing of a challoupe specifically for Christiania, but they are the most prevalent class of ship boat in the archive. The Barkasse, is very box like ('kasse' = box - perhaps a coincidence, pahaps not...), more so than the English launches, so I decided to try to model this for 3D printing! a bit daunting, as I was just figuring out what worked as I went along, but I am happy with the end result. This is the original drawing (G312): There is also a sail drawing, but that is not so relevant for my use case. First, I drew the frames and keel acc. to the plans (at 1/64 scale): This was done in QCAD. I have prepared the frames for klinker planking - I know the English launches were caravel/cravel planked, and this might have been as well. However, all the drawings in the archive shows all boats, regardless of type and size as cravel planked, and I am pretty sure many were klinker planked, so I decided to do this as I like the detail that it gives on the hull. After this, the parts were brought into Onshape (free onling 3D CAD program) and extruded to the correct thicknesses. Then I aranged the frames on the keel and started making point-constrained surfaces using the plank section of the frames. That creates a surface thatcan then be extruded, but while it covers the plank in question, it is over-large and need to be 'trimmed' to the correct plank shap. This I did with two boundary surfaces which, when extruded to a 3D shape, can be used to trim the plank to shape (with a boolean operation). Then I repeat that for all 12 planks. This was the difficult part - planking the hull - and then it is easy to make the other elements in a similar fashion, drawing and extruding the parts and adding them to the model. This is the final result: The planking looks a but rough in a few places, but bear in mind that the model is really small: the overall lenght is 123 mm and the thickest part of the planks are just 2 mm wide! The grating holes are 1x1 mm. I have no direct evidence of how the grating and deck should look for this particular boat, so I have made a more or less generic interior. While the drawings do not show it in this case, there are many drawings of 'barkasser' in the archive that shows mountings for swiwel guns in connection with the heavy cross bracing fore and aft. To add a bit of interest, I have added this feature to my model as well. The drawing also does not show the rudder, but I have used a rudder shape from another (larger) barkasse from very close to the same year as Christiania - this is hopefully a resonable approximation. At the very least, I will make the seats out of 1 mm pear, but I may also make the grating out of pear. I think I will print a version with and without the grating and see what works best. The blue sections in the image below will as a minimum be made of wood: The very tiny oar 'pins' will not be printed, they are just to small at 0.2 mm thickness. I am not 100% sure this is actually what they looked like, or how they are supposed to work. Some drawings show what might be a small eye ar the top, so perhaps a loop of rope was attached to keep the oar in place? If anyone knows how this type of oar rest worked, please share! Now I look forward to seeing how this prints! And then I think it is high time I finished that planking so I can continue with all the interesting bits that comes next! BR TJM

This is so great! I really look forward to seeing them fully painted and rigged! When I have the time, I will start a thread on the 3D printing sub-forum where I will post stl files for Danish canon, in case someone else might want to go down this particular rabbit hole at some point 😅. I have a whole series from 3 to 36 pounders from the 1740's-1750's, but I am also looking into earlier designs from the 17th century and later ones from the end of the 18th century

Couldn't this be a consequence of 300+ years at the bottom of the sea? Do we know for sure that there was such a misalignment when the ship launched?