The Ship Model Forum

The Ship Modelers Source
It is currently Thu Mar 28, 2024 4:34 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 14 posts ] 
Author Message
PostPosted: Sun Nov 08, 2015 1:14 pm 
Offline
User avatar

Joined: Fri Dec 15, 2006 6:53 pm
Posts: 232
Location: Calgary, Alberta
Attachment:
Figure 1 HMS Caroline 1915 in 3D.jpg
Figure 1 HMS Caroline 1915 in 3D.jpg [ 153.29 KiB | Viewed 6503 times ]


I have always been intrigued by HMS Caroline built by Messrs. Cammel Laird’s at Birkenhead in only 10 months (a record), commissioned December 17, 1914, served at Jutland and is the only such ship afloat, she provided base support for trawlers and drifters in WWII, after the war was headquarters for the Ulster Division R.N.V.R. and decommissioned April 7, 2011 the second longest ship in commission in the Royal Navy (after HMS Victory) for service as a museum ship. In 1946 she was refit by Harland and Wolf’s yard where office space was installed in the boiler rooms (removed in 1926) and false funnels installed. It is Captain Shillington her then commanding officer who worked magic within the Admiralty to get the funnels, austerity was to remove them. This converted her into the ship known in Belfast as HMS Never Budge that we see today. My interest is to model her in her Jutland configuration of 1915.

Reference Material
Reference materials for HMS Caroline in 1915 are all in conflict. My primary photo came from DK Brown’s The Grand Fleet. On page 66 there is the Profile Drawing and a photo of HMS Caroline in ‘original’ configuration, though in fact she is altered from her 1914 state, both of these were scanned at high resolution. Alan Raven and John Roberts British Cruisers of WW2 provided details of modifications and Mark numbers of equipment fits. There are also great photos of early C Class cruisers. So going to Norman Friedman and British Cruisers page 43 has a clear as completed plan drawing by John Dominy. This was scanned has a high resolution grey scale in TIFF format. In the back of HMS Caroline by R.S. Allison there is a sketch drawing both profile and plan that also has the sections. This was also scanned in high resolution colour as the drawing had yellowed. Both of the plan drawings where resized and printed at 1/700 scale, this allowed them to be compared by holding them up to the light. It is interesting to note that they are different in details but that the hull lines perfectly overlap. Having visited Belfast and taken ~100 photos of HMS Caroline, the current ship varies from both drawings for the bridge and funnels. You can however see in the photos what is original and what was added over the years. Still more interesting the NMM profile drawing shows 5 washroom portholes at the foredeck break where there are 6 on the actual ship. Looking closely at the ship, all six port holes are original construction and not part of the later additions. Amgram Navel Architects were asked to do a stability study in case HMS Caroline had to be moved. This resulted in outstanding sections lines and a 3D hull shot being made available on the internet. NMM has on the web two small photos of the 1/48 scale builders model of HMS Caroline (SLR0029). David Hobbs in Warships of the Great War Era has taken full sized photos and spread them across 2 pages in his book. When scanned in high resolution and viewed on a monitor; all sorts of details emerge. Here are 4 photos I used and the similarly positioned 3D image of HMS Caroline about 1915.

Attachment:
Figure 2 HMS Caroline As Launched DK Brown.jpg
Figure 2 HMS Caroline As Launched DK Brown.jpg [ 152 KiB | Viewed 6503 times ]


Attachment:
Figure 3 HMS Caroline As Launched.jpg
Figure 3 HMS Caroline As Launched.jpg [ 80.56 KiB | Viewed 6503 times ]


Attachment:
Figure 4 HMS Caroline as built.jpg
Figure 4 HMS Caroline as built.jpg [ 146.47 KiB | Viewed 6503 times ]


Attachment:
Figure 5 HMS Caroline as built.jpg
Figure 5 HMS Caroline as built.jpg [ 120.14 KiB | Viewed 6503 times ]


Attachment:
Figure 6 HMS Caroline As Built RMG SLR0029.jpg
Figure 6 HMS Caroline As Built RMG SLR0029.jpg [ 110.19 KiB | Viewed 6503 times ]


Attachment:
Figure 7 HMS Caroline As Built RMG.jpg
Figure 7 HMS Caroline As Built RMG.jpg [ 89.48 KiB | Viewed 6503 times ]


Attachment:
Figure 8 HMS Cleopatra Completing.jpg
Figure 8 HMS Cleopatra Completing.jpg [ 142.67 KiB | Viewed 6503 times ]


Attachment:
Figure 9 HMS Caroline Stern.jpg
Figure 9 HMS Caroline Stern.jpg [ 147.61 KiB | Viewed 6503 times ]


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Sun Nov 08, 2015 1:32 pm 
Offline
User avatar

Joined: Fri Dec 15, 2006 6:53 pm
Posts: 232
Location: Calgary, Alberta
Software

I have drawn with computers in 2D since 1987 starting with In*a*vision, a vector based program. I had no experience with CAD programs, 3D printers or making Photo Etch. So my first task was to trial CAD programs. I looked at Autodesk 123D, CorelCAD and Design CAD 3D max and attempted to draw a hull. This took me to looking at MeshLab and NetFabb to correct the STL files. I came to the conclusion that I wanted a program that was:
1. Vector Based
2. Could import JPEG’s
3. Rotate JPEG’s
4. Scale JPEG’s
5. Have a GetDistance command

I chose to go with CorelCAD and NetFabb, not the cheapest programs around but they meet my criteria for usage.

I also made full use of MS Excel to convert from feet and inches to mm in 1/700 scale. Using the GetDistance command on photos scaled to 1/700 you could measure something you know as a check and then measure a new item nearby. By using screen measurements with a plastic ruler of a known and of an unknown item, in Excel you could then ratio the two items and so get the size of the unknown bit. I kept Excel running constantly in the back ground. I also used Excel to go from mm in 1/700 scale to actual feet for sanity checks.

Attachment:
Figure 10 Excel for Sizing Calculations.jpg
Figure 10 Excel for Sizing Calculations.jpg [ 142.19 KiB | Viewed 6500 times ]


For all measurements I went to 1/1000 of a mm. If you think of Photo Etch, 0.1 mm is visible to the eye. By going to .000 mm any measurement error is invisible.

Test 1

Not knowing the software or the capabilities of 3D printers I thought a simple test to establish some normals would be in order. So I made my first complete hull drawing, the boiler casing, conning tower, placed bollards and fairleads on the hull and a series of cowl ventilators. CorelCAD allows you to save as dwg files but to export stl files that the 3D printers use. The stl files were corrected in NetFabb prior to printing. As this was a test I did not want to use a commercial 3D print service such as Shapeways.

Finding a local printer turned into quite the search. I would up on a Dutch site that has coordinated many 3D printers on a worldwide basis by city. The printers are called Hubs and vary in the type of printer used, fees, accuracy and layer thickness. Find a 3D printer near you at http://www.3dhubs.com/ and have a look around.

I wanted a quick test and so choose a local printer that used a Ultimaker 1 printer, PLA material with 60 micron layers and a 0.4 mm nozzle. PLA is Poly Lactic Acid and is derived from plant material, it is clear or coloured can detail well and the parts tend not to bend. 60 micron layers is medium resolution now a days. 16 or 20 micron is the ultra high resolution that some hubs can provide. But a successful print needs more than material and layer resolution. Turns out the nozzle diameter controls what size of object can be printed. If your part thickness is less than 0.4 mm, in this case, then the part either does not print or gets smeared. PLA is a thermoplastic, you heat it and it flows into position, cools and you get your part. If it cannot cool fast enough then you get smeared parts or slumps. So here is what the Ultimaker 1 can do.

Attachment:
Figure 11 Test 1 in CorelCAD.jpg
Figure 11 Test 1 in CorelCAD.jpg [ 42.97 KiB | Viewed 6500 times ]


Attachment:
Figure 12 Test 1 from a Ultimaker 1 at 60 microns.jpg
Figure 12 Test 1 from a Ultimaker 1 at 60 microns.jpg [ 91.37 KiB | Viewed 6500 times ]


So overall the shapes are correct but there is no detail. The midships bollard and fairleads did not even print.

Attachment:
Figure 13 Test 1 Boiler Casing in CorelCAD and at 60 Microns.JPG
Figure 13 Test 1 Boiler Casing in CorelCAD and at 60 Microns.JPG [ 55.89 KiB | Viewed 6500 times ]


The skylights and funnel bases are not too bad but the portholes are missing and the doorways bleed upwards. The intakes are also smeared due to the nozzle size and layering. So this is OK for a fit check but 60 microns would create a lot of extra work to get a useable part.

Test 2

Test 2 is from a second 3D printing Hub near Toronto, Rapid-Fab's Hub, https://www.3dhubs.com/toronto/hubs/rapid-fab, that has a Projet 3510 with XHD mode using VisiJet Crystal material; this results in 16 micron layers and 0.025 – 0.05 mm part accuracy.

Printed and delivered with Canada Post shipping across Canada in 8 days, not bad service at all.

Attachment:
Figure 14 Test 2 from a Projet 3510 at 16 Microns.jpg
Figure 14 Test 2 from a Projet 3510 at 16 Microns.jpg [ 61.25 KiB | Viewed 6500 times ]


Shapes are correct with hard edges. The layers can just be felt on the deck and the edges are slightly rough to the touch. Simple wet sanding will result in smooth deck surface, with a coat of varnish this should cast. The details are there as well with the bollards and fairleads accurately printed.

Attachment:
Figure 15 Test 2 Boiler Casing at 16 Microns.jpg
Figure 15 Test 2 Boiler Casing at 16 Microns.jpg [ 78.69 KiB | Viewed 6500 times ]


The doorways are correct and the portholes are also now present and correct. The boiler air intakes are now correct and the cover supports tried to print. These need to be thickened slightly. This part also fit perfectly into position in the hull and is ready for casting with minimal cleanup.

Test 2 learning’s confirm that 3D printers have the printing resolution required for 1/700 scale projects. Not everything can be printed as shown by the supports but the cowl vents turned out OK so thicknesses of 0.05 mm should print. Looking at the hull I got the shear completely wrong. At this point the Amgram Navel Architects hull sections became available and so I decided to start over from scratch. As the VisiJet Crystal material seems quite brittle, I also decided to add pour stubs to the parts so that I could recast them in resin.

3D Drawing

Ship Boats
So to get further practice drawing I decided to make the ships boats next. This resulted in another learn curve but the results allowed me to redo the hull applying the lessons learned.

Attachment:
Figure 16 Ship Boats Rendered.jpg
Figure 16 Ship Boats Rendered.jpg [ 149.29 KiB | Viewed 6500 times ]


These I drew on pour plugs and also created outlines suitable for PE.

Attachment:
Figure 17 34 Ft Steam Cutter Rendered with PE Positioned.jpg
Figure 17 34 Ft Steam Cutter Rendered with PE Positioned.jpg [ 121.01 KiB | Viewed 6500 times ]


Here is an animated rendering with the PE mounted on the steam cutter. This to scale is about 15 mm long, theoretically it should print.

The Hull
The sections drawing was imported into CorelCAD and rotated so that it was perfectly horizontal and scaled so that full size is 1/700 in mm. The half sections where then redrawn and converted into full sections.

Two sections had to be generated by measurement and referenced to photos; the break to the main deck and just ahead of this where the bow flare ends.

In a second drawing the keel line that was imported, rotated and scaled to 1/700 mm. The sections were then Copied from the first drawing and pasted into the second drawing. The sections where then roughly positioned along the ‘keel’ line. Sections where then rotated 90° counterclockwise along there Y axis. Construction lines where drawn for each section along the keel to mark there exact location. The bottom middle of the sections where shifted into position along the keel line using the Move command.

Attachment:
Figure 18 Hull Lines.jpg
Figure 18 Hull Lines.jpg [ 65.93 KiB | Viewed 6500 times ]


After much trial and mostly error, lessons were learned. One, Polylines should only be used to draw sections. Two, Arcs should be used to join the sections along a horizontal plain. Three, Splines should be used to join sections that curve through 3D space.

Attachment:
Figure 19 Hull Loft.jpg
Figure 19 Hull Loft.jpg [ 79.78 KiB | Viewed 6500 times ]


By doing this you can loft between the Sections into a 3D solid. If the space is complex, then many small sections will have to be created so that the shape when lofted is correct, an example of this being the bow and stern. The sections where calculated by using construction lines to identify points in space. These were then formed into sections and connected with arcs. Individual sections where then lofted and the resulting solids joined with a union command.


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Sun Nov 08, 2015 1:40 pm 
Offline
User avatar

Joined: Fri Dec 15, 2006 6:53 pm
Posts: 232
Location: Calgary, Alberta
It was important to constantly refer to the photos I had taken of HMS Caroline at Belfast to fill in the missing pieces that the drawings simply could not give.

Attachment:
Figure 20 Details Came From Photos.JPG
Figure 20 Details Came From Photos.JPG [ 136.24 KiB | Viewed 6500 times ]


The basic hull loft was continually added to and subtracted from using the Union and Subtract commands. Often I used a negative shape subtracted from a positive shape to create the required shape. All of this creates tiny imperfections in the hull loft that would cause failure of the 3D print. So once exported as a stl file I used NetFabb to automatically correct the imperfections. When first imported all of these errors show as grey or orange sections in the drawing. After you go through the correction routines the final shape is green for good. NetFabb was easy to learn and painless to use. All of the 3D parts created in CorelCAD were corrected in NetFabb prior to being successfully printing.

Attachment:
Figure 21 STL File Corrected by NetFabb.jpg
Figure 21 STL File Corrected by NetFabb.jpg [ 150.85 KiB | Viewed 6500 times ]


3D Printing

When working with a Hub there is communication forwards and backwards. This allows you to optimize your parts for 3D printing using the experience from the Hub. I had all of the subparts on wafers. This would not allow for optimal printing. So I removed the wafers so that the Hub could rotate the parts for best output. I originally had the masts and davits on standoffs but these were lowered so that the parts where flush with the wafer at one end. This is very much NOT mass production but custom printing for your needs. So here is what the parts came together as.

Attachment:
Figure 22 The Parts in CorelCAD.jpg
Figure 22 The Parts in CorelCAD.jpg [ 142.59 KiB | Viewed 6500 times ]


The corrected STL files where uploaded to Rapid-Fab's Hub, that uses a Projet 3510 with XHD mode using VisiJet Crystal material. The Projet 3510 uses a soft wax as a support material for overhangs. Once the VisiJet is UV hardened the part is placed into an oven where the support wax is mostly melted away. The part is now placed in hot oil, like dissolves like, so the oil removes the remaining support wax. The part in finally placed into Dawn dishwashing soap to remove the oil. The results are very clean, ready to use 3D parts. Here is the output.

Attachment:
Figure 23 The 3D Printed Parts.jpg
Figure 23 The 3D Printed Parts.jpg [ 121.74 KiB | Viewed 6500 times ]


Photo Etch

You do not know, what it is, you do not know. So taking advice from Timmy C and EJFoeth at http://www.shipmodels.info/mws_forum/viewtopic.php?f=2&t=154011#p599047 sent me to the Hauler site at http://www.etchworks.eu/. On the site I read and downloaded everything I could. Two documents are critical for success; Hauler Etching Formats.pdf and Hauler Design limits schema.pdf. Two nasty little details did surface. First, do not make your drawing the true A4 or A5 dimension. Instead go to http://www.etchworks.eu/dimension-of-sheets/ and for the dimension of the sheet use the Values in Brackets. So for A4 not 300 x 200 but make your sheet 290 x 190. The second detail is that CorelCAD does not output drawings in CorelDRAW format that are useful for making PE, there is insufficient resolution. Also Hauler needs your drawing to be in CorelDRAW X3 format or lower. So the work around became to go into CorelCAD and scale the drawing for PE x1000 and then to export it to CorelDRAW. Note: the PE drawings were drawn without lines so the actual size could be measured. I then purchased a copy of CorelDRAW opened the drawing from CorelCAD and cleaned it up. I then scaled it x 0.001 and saved into X3 format resulting in the correctly sized PE. Hauler is indeed a pleasure to work with and they are fast. From emailing ‘2 sets please’, to receiving a registered package from the mailman, took 9 days from the Czech Republic to Western Canada.

I have had the pleasure of using a lot of Peter Hall designed PE from White Ensign Models. I took mentorship from his PE designs when designing my own. When designing to Hauler specifications for a A4 0.1mm Positive Brass sheet the minimum dimension is 0.10 mm. I drew HMS Caroline to actual scale in 1/700. So the life lines, at 0.015mm, where about 10 times to narrow to successfully generate PE. This meant that all the PE parts had to be redrawn at printable widths. One other thing, to make 2D parts visible in a 3D render; in CorelCAD you convert the Line or Polylines into a Region. Regions cannot be imported into CorelDRAW, so I got into the habit of saving many versions of the same drawing so that I always had the raw source to use elsewhere.

One of the benefits of drawing a vessel in 3D is that you can also rig the vessel in 3D, this I did. This gives you exact details of wireless aerials and signal flag halyards along with all of the exact life rail measurements. So these along with everything else that did not print in 3D became 2D as PE. Hauler stated that the design minimum is 0.1mm but I chose to experiment so I also did the set at 0.05mm width to see what would happen. Here is the output in brass.

Attachment:
Figure 24 HMS Caroline 1915 PE.jpg
Figure 24 HMS Caroline 1915 PE.jpg [ 133.14 KiB | Viewed 6500 times ]


The 0.05mm drawing width is at the top of the photo with the 0.1mm width at the bottom. (In the middle is a row of experimental PE for a Canadian Patrol Frigate Stack Grill for a different project.) Yes, it all printed but is the 0.05 mm actually useable or is it simply to fine to have structural integrity? Only the actual build will answer this question.

So I found that it is completely possible to 3D print 1/700 scale ship models and that the 3D model greatly aids PE production.

Cheers,
George


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Sun Nov 08, 2015 10:38 pm 
Offline

Joined: Thu Apr 28, 2005 8:46 pm
Posts: 201
Wow. Beautiful work, George, congratulations on it. Beautiful ship as well! Michael


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Sun Dec 27, 2015 6:36 am 
Offline

Joined: Tue Jun 14, 2005 6:23 am
Posts: 3688
Location: Bonn
Very interesting! :thumbs_up_1: I should look more often in this part of the board!

Will there be a kit of HMS Caroline soon? I would be very interested!

And the hull can be used for Calliope, Cambrian and Centaur class - each early and late versions!

_________________
Image


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Sun Dec 27, 2015 9:59 am 
Offline
Model Monkey
Model Monkey

Joined: Sat Aug 20, 2005 9:27 pm
Posts: 3952
Location: USA
Brilliant!

_________________
Have fun, Monkey around.™

-Steve L.

Complete catalog: - https://www.model-monkey.com/
Follow Model Monkey® on Facebook: - https://www.facebook.com/modelmonkeybookandhobby


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Sun Dec 27, 2015 1:22 pm 
Offline
User avatar

Joined: Fri Dec 15, 2006 6:53 pm
Posts: 232
Location: Calgary, Alberta
Hello,

Thanks for the comments.

This was done to satisfy my own curiosity in the CAD and 3D printing realms; so there are no follow on kits planned.

The later C Class light cruisers where actually a different hull form and they were both longer and with greater beam.

Cheers,
George


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Mon Dec 28, 2015 1:32 am 
Offline

Joined: Tue Jun 14, 2005 6:23 am
Posts: 3688
Location: Bonn
Why no kit? You have already made all parts.

Perhaps you can offer it on Shapeways?

I have checked my books: Caroline, Calliope, Cambrian, and Centaur class had the same hull. Only some mention a very small increase of beam in case of the Centaur class.

_________________
Image


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Mon Dec 28, 2015 11:47 pm 
Offline
User avatar

Joined: Fri Dec 15, 2006 6:53 pm
Posts: 232
Location: Calgary, Alberta
Hi Maxim,

This is my first go at 3D modelling. I included a lot of redundancy between printed parts and PE. Also the parts are not suitable for Shapeways in that there are many separate parts and they are on mold blocks (this adds cost). So the parts would have to be reworked. So I'm not saying never, its just that I have a lot to learn and debug before she is ready.

I'm not sure what reference books you referred too but look at the bows. Early C Class have scalloped bows. Later C class have straight bows and some rather than flare have trawler bows. This gives a clue that the hulls are actually not the same at all.

Cheers,
George


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Tue Dec 29, 2015 2:15 am 
Offline

Joined: Tue Jun 14, 2005 6:23 am
Posts: 3688
Location: Bonn
Hi George,
Gwylan Models prints their masters also on sprues and cast them in resin afterwards.

I referred only to the first four classes of the C-class (Caroline, Calliope, Cambrian, and Centaur), not the late ones (Caledon, Ceres, Capetown). The later three for sure had a different hull. But there are anyway kits of the Ceres and Capetown classes by HP Model, i.e. the hull of the last three classes is available. But not the hull of the first four classes.

But the first four classes alone give many variants: early versions with 6 in and 4 in guns for earlier ships, late versions with only 6 in guns, even later versions of the 1920s with modified superstructures...

Cheers,
Lars

_________________
Image


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Tue Dec 29, 2015 9:48 pm 
Offline
User avatar

Joined: Fri Dec 15, 2006 6:53 pm
Posts: 232
Location: Calgary, Alberta
Hi Lars,

Oops, my error. You are correct.

Cheers,
George


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Fri Apr 29, 2016 4:52 pm 
Offline
User avatar

Joined: Fri Jun 22, 2012 12:02 pm
Posts: 2350
Location: Herndon, VA
I was referred to this post after posting a thread on Caroline on Facebook.

You may count me as a potential customer for this historic ship. I'm, frankly, floored that no kit exists of Caroline!

_________________
- Chris

1/700 Saratoga w/Pontos (Needs paint)
1/700 Potato w/Kurama (On hold)
1/700 Murdertorpedoboat Ooi


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Wed Jun 29, 2016 7:50 am 
Offline

Joined: Sat Sep 17, 2005 9:19 am
Posts: 1479
There was a difference between the Caroline class and the other C's. When the two forward funnels were trunked into one the break in the forecastle went aft. I learned this the hard way when trying to build Canterbury using Caroline's lines so this hull is now slightly redundant.
viewtopic.php?f=4&t=157837
Cheers,
Pieter

maxim wrote:
Hi George,
Gwylan Models prints their masters also on sprues and cast them in resin afterwards.

I referred only to the first four classes of the C-class (Caroline, Calliope, Cambrian, and Centaur), not the late ones (Caledon, Ceres, Capetown). The later three for sure had a different hull. But there are anyway kits of the Ceres and Capetown classes by HP Model, i.e. the hull of the last three classes is available. But not the hull of the first four classes.

But the first four classes alone give many variants: early versions with 6 in and 4 in guns for earlier ships, late versions with only 6 in guns, even later versions of the 1920s with modified superstructures...

Cheers,
Lars


Report this post
Top
 Profile  
Reply with quote  
PostPosted: Wed Jun 29, 2016 7:59 am 
Offline

Joined: Sat Sep 17, 2005 9:19 am
Posts: 1479
Oh, and as you can guess from my post I'm not interested in a 3d printed hull as I'm a fan of old fashioned balsa/styrene for hulls but a good early C class PE set is most welcome. Is it possible you'll be selling some of these PE sets?


Report this post
Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 14 posts ] 

All times are UTC - 6 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 15 guests


You can post new topics in this forum
You can reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
Powered by phpBB® Forum Software © phpBB Group