molasses

Where were you two and a half years ago?    I'm at 95%+ done fabricating with only deck details left.  And I do not want to find that I have a bunch of mistakes to correct.
Seriously, Igor, I appreciate the offer but I really don't need them.

Where were you two and a half years ago?    I'm at 95%+ done fabricating with only deck details left.  And I do not want to find that I have a bunch of mistakes to correct.
Seriously, Igor, I appreciate the offer but I really don't need them.

If you liked that shot you'll love this one:
 

300 crew + 90 Chilean Navy midshipmen
 

If you liked that shot you'll love this one:
 

300 crew + 90 Chilean Navy midshipmen
 

I'm working almost exclusively from photos. I was unable to find any plans for Esmeralda or her sister ship Juan Sebastian Elcano. I did find a low-res drawing of the weather deck and the deck below which gave me my deck outline and helped with my best guess at the waterline shape. I also had the dimensions (OA length, WL length, beam, mast heights, etc.) which I used to extrapolate other dimensions from the scores of photos I collected from the web.
 

Deck plan used. The lower deck shows the hull shape at the lower deck which I used to make my best guess
at the waterline. The three images below are lower-res versions (for posting here) of high-res images (up to
6 MB) I found and used for my deck detailing.
 

 

 

 
I also used numerous on-deck photos and video tours of the weather deck for specific details and as checks against my extrapolated dimensions. In most cases I have two or more photos of a specific detail from different angles.
 
Spar dimensions were easy to derive; there are numerous photos of Esmeralda under sail in which the spars are parallel or near parallel to the plane of the camera lens. It's easy to compare that length to the WL length or the total mast height in several photos to come up with a number at E's scale (1/640).
 
That's my long answer. My short answer would have to be, "I'm eye-balling it."
 
Dave 

Welcome to resumption of work on Esmeralda.
I haven't worked on it in over a year so picking up where I left off was a bit difficult. I decided I have some re-work to do to incorporate some techniques I learned or improved upon since the last work done. I also have a lazy streak.
 
There are a large number of round port lights on most of the deck hatches that I didn't originally intend to reproduce but decided I can't leave off. I realized I could do them as decals on the eight hatches and prepared them on the computer and printed them on decal stock along with the yellow metal stern relief detail and name.
 

Hatch cover details and stern relief decals.
 

Before applying the decals to the tops of the white deck hatches.
 

After applying the decals. The difference is more noticeable when you click the image. There are 146 port lights in the hatch covers.
 
To do the stern detail I had to photo-shop the relief from a photo of it and adjust that image for the curvature of the stern to print it on a flat surface and for the distortion from the angle the photo was taken. I also had to do some color correction to make it look bright and shiny (the photo was taken on a dark, overcast day so the relief looked very dull). then reduce it for the model.
 

Esmeralda's stern relief. I have to wait about 24 hours before coating all the decals with a couple coats of artist's matte varnish.
 

The photo from which I extracted the stern relief.
 
I realized while I was making the decals that this was the way I should have done the port lights on the hull. Live and learn. If I ever decide to do Esmeralda's sister ship Juan Sebastian de Elcano (a topsail schooner while E is a barquentine), any of the Gorch Fock sisters and clones (nine of them) or any ship with lots of port lights, I'll know what to do.
 
 
Dave

I'm working almost exclusively from photos. I was unable to find any plans for Esmeralda or her sister ship Juan Sebastian Elcano. I did find a low-res drawing of the weather deck and the deck below which gave me my deck outline and helped with my best guess at the waterline shape. I also had the dimensions (OA length, WL length, beam, mast heights, etc.) which I used to extrapolate other dimensions from the scores of photos I collected from the web.
 

Deck plan used. The lower deck shows the hull shape at the lower deck which I used to make my best guess
at the waterline. The three images below are lower-res versions (for posting here) of high-res images (up to
6 MB) I found and used for my deck detailing.
 

 

 

 
I also used numerous on-deck photos and video tours of the weather deck for specific details and as checks against my extrapolated dimensions. In most cases I have two or more photos of a specific detail from different angles.
 
Spar dimensions were easy to derive; there are numerous photos of Esmeralda under sail in which the spars are parallel or near parallel to the plane of the camera lens. It's easy to compare that length to the WL length or the total mast height in several photos to come up with a number at E's scale (1/640).
 
That's my long answer. My short answer would have to be, "I'm eye-balling it."
 
Dave 

Welcome to resumption of work on Esmeralda.
I haven't worked on it in over a year so picking up where I left off was a bit difficult. I decided I have some re-work to do to incorporate some techniques I learned or improved upon since the last work done. I also have a lazy streak.
 
There are a large number of round port lights on most of the deck hatches that I didn't originally intend to reproduce but decided I can't leave off. I realized I could do them as decals on the eight hatches and prepared them on the computer and printed them on decal stock along with the yellow metal stern relief detail and name.
 

Hatch cover details and stern relief decals.
 

Before applying the decals to the tops of the white deck hatches.
 

After applying the decals. The difference is more noticeable when you click the image. There are 146 port lights in the hatch covers.
 
To do the stern detail I had to photo-shop the relief from a photo of it and adjust that image for the curvature of the stern to print it on a flat surface and for the distortion from the angle the photo was taken. I also had to do some color correction to make it look bright and shiny (the photo was taken on a dark, overcast day so the relief looked very dull). then reduce it for the model.
 

Esmeralda's stern relief. I have to wait about 24 hours before coating all the decals with a couple coats of artist's matte varnish.
 

The photo from which I extracted the stern relief.
 
I realized while I was making the decals that this was the way I should have done the port lights on the hull. Live and learn. If I ever decide to do Esmeralda's sister ship Juan Sebastian de Elcano (a topsail schooner while E is a barquentine), any of the Gorch Fock sisters and clones (nine of them) or any ship with lots of port lights, I'll know what to do.
 
 
Dave

Finished Ogallala and can now get back to work on Esmeralda. I hadn't intended to take this much time away from Esmeralda but Ogallala turned into a much more comprehensive project than I had originally planned. I did find some more, useful, photos of E for some details on deck and aloft and kept her in mind during my side journey. I learned a few things from O that I will apply to E and I have a bit of re-work to do and some details to add before I pick up where I left off. I'll post my progress soon.
 
I came across an interesting variation on a ship in a bottle with Esmeralda as the subject but the method is a bit too ephemeral for my taste. 
 

 
Dave

I'm working almost exclusively from photos. I was unable to find any plans for Esmeralda or her sister ship Juan Sebastian Elcano. I did find a low-res drawing of the weather deck and the deck below which gave me my deck outline and helped with my best guess at the waterline shape. I also had the dimensions (OA length, WL length, beam, mast heights, etc.) which I used to extrapolate other dimensions from the scores of photos I collected from the web.
 

Deck plan used. The lower deck shows the hull shape at the lower deck which I used to make my best guess
at the waterline. The three images below are lower-res versions (for posting here) of high-res images (up to
6 MB) I found and used for my deck detailing.
 

 

 

 
I also used numerous on-deck photos and video tours of the weather deck for specific details and as checks against my extrapolated dimensions. In most cases I have two or more photos of a specific detail from different angles.
 
Spar dimensions were easy to derive; there are numerous photos of Esmeralda under sail in which the spars are parallel or near parallel to the plane of the camera lens. It's easy to compare that length to the WL length or the total mast height in several photos to come up with a number at E's scale (1/640).
 
That's my long answer. My short answer would have to be, "I'm eye-balling it."
 
Dave 

I'm posting a follow-up on Ogallala.
 
Upon finishing her I submitted photos to Bottle Shipwright, the quarterly journal of the Ships in Bottles Association of America (of which I am a member), and quickly learned that Ogallala would be the featured SiB in the March issue which I received a few days ago. Here's the cover. I'm not going to bore you with posting the photos from the article because they have already been posted here.
 

 
As an aside, I mentioned a couple times in this build log that I can't really see my work for what it is. All I can see are the deficiencies no matter how minor. Ogallala has been sitting in a box on a shelf since I finished her in late October - until today. I'm now able to see her as a whole and I'll admit I was impressed and very pleased.
 
If you haven't seen them yet check out my two current projects listed in my signature.

Thanks for the very kind comments. Here's the best of the photos of the finished Ogallala.
 

 

 

 

 

 

 

 

 
Thank you to all who viewed this log and especially to those who contributed with 'likes' and with comments - you all pushed me to do my best possible work.
 
Now I can get back to work on Esmeralda. I learned a few things here I can apply to "The White Lady".
 

 
Dave
 
 

Thank you Keith and Omega and to all the 'likers', you're very kind. 
 
I started by wiping down the entire inside surface of the sphere to remove any debris from the base assembly. The rigging on many lines on the ship were slacked off, masts hinged down to the deck and the two big sails extended part way away from the ship and rolled to go in the bottle. The mainsail was rolled and started into the opening followed by the main masthead, fore masthead and then the transom of the hull. This was the tightest spot, then I had to adjust the fore topsail yard and crossjack as they slipped through.
 

Part way in and looking good, no problems yet, even though the rigging looks like a rats nest again but
it's more like organized chaos and will restore most of itself as the masts hinge back up.
 
When I got the ship a little further in I started raising the masts and pulled the fore mast into its position, tied off the forestay at the bowsprit and glued it, then did the same with the jibstay. I turned the ship over, tensioned the mainstay and glued it where it comes through the bottom of the hull.
 

Upside down, held in place so I can trim the two ends of the mainstay. Next step is to turn
the ship back over and glue it to the lower hull.
 

Upper hull turned over and glued to the lower hull. The crossed poles are bamboo skewers cut
off about 1/8 inch / 3 mm out the opening with a lead weight on the ends while the pva cures.
The three stays are the only lines that I've touched so far; more to go but the worst is over.
 
From here all that's left is several tedious hours of adjusting about twenty more lines, then securing and trimming them. One thing about working inside a sphere I like a lot is that I can turn the model so the part of the ship I need to work on is near the opening - something I can't do inside the usual, cylindrical bottle.
 

 
Dave

Welcome back.
 
Thank you, Jeff and Michael, you are both very generous. BTW, I proudly stole every one of "my" techniques - thank Donald McNarry and Lloyd McCaffery for many of them.
 
I started with the breast backstays, each rigged with a single and double block. After choosing the length of the tackle from outside to outside of the blocks I routed the thread between two pins in drilled holes in a scrap of wood and built the tackle using the discs as previously described then needle spliced an eye around each block and added the backstays to the model.
 
I rigged the main ratlines using a "harp" with ratlines spaced and glued to wooden frames. The frames were opposite hand, one for each side of the schooner. After pinning the frame in position with bug pins I glued all the intersections. I'm trying out artist's matte acrylic fixative instead of thinned white glue and am pleased with the results. The white glue sometimes goes milky while the fixative stays clear. I thinned it a little by using a wetted brush when I dipped it into the fixative.
 

"Harp" pinned in place with intersections fixed and frame ready to cut from the ratlines.
Starboard side has the harp removed but the excess on the ratlines un-trimmed.
 

Breast backstay and trimmed ratlines. The backstays go through the same hole where the
topmast shrouds were secured and are now tensioned with weights. The backstays need a
bit of adjustment before they glue. After this photo I rigged the futtock shrouds and
futtock staves then glued and trimmed the backstays.
 
 

Starboard view of the mainsail, spars and rigging. While assembling this I had a card stock sail
in place, tied to the boom, mast and gaff to hold them in their correct positions. This card stock
sail has been lengthened slightly along its leech (aft edge) to give the belly I want in the sail. The
aft ends of the boom and gaff have a thread to pull that belly in this temporary sail.
 

Larboard view. The card stock sail will be removed later to use as a pattern for making the sail.
The black line will be removed before the sail is installed. I have some cleats to make and install
where the leads from the blocks and tackles will tie off which explains some of the stray threads.
 

Close-up of main top showing the futtock shrouds and futtock staves. I used a needle to pass
each futtock shroud through a lower shroud, glued it (not ca) then passed the needle through
those intersections, sideways, for a third thread to be the futtock stave, glued it and trimmed.
 

Close-up of main top showing the boom and gaff halliards and gaff throat halliard. After making
and installing the throat halliard and installing the gaff I routed the throat halliard and the boom
and gaff halliards through eyebolts installed while building the masts. The blocks for boom
halliards were made in place. The block on the gaff bridle halliard at the eyebolt was glued
prior to rigging to act as a stop while rigging. All four lines continue to the deck.
 

Close-up of pin rail where the halliards tie off. I installed eyebolts earlier for the gaff halliards
to route through so that some of the tension on the pinrail is downward to balance the upward
pull by the boom halliards tied off there. Tensions are very light - just enough to hold the line
straight - but I felt it necessary to do this. You can also see where one of the breast backstays
tied off to the pinrail on the bulwark just below the shrouds. A white thread from the throat of
the boom, through the mast and out the bottle for pulling the boom into place is also visible
but mostly out of focus.
 

I almost forgot a size reference but recovered with a standard #11 X-Acto blade.
 
As always, click on a thumbnail shown here for the larger format photos.
 

 
Thank you for viewing this log.
Dave

I'm finally getting to my favorite part of a build when all the different mini-projects are installed and I can start tying them all together with rigging - in fact I enjoy doing rigging just on its own.
 
I took the method of making blocks (two discs with a thread sandwiched between then a thread tied around the block) and expanded it to double blocks.
 

This is part of the steering tackle with a double block. I still need to strop it. The white
discs are.040 in./1.0 mm in diameter, same as those used to make the double blocks.
 
After I finished making twelve deadeye assemblies with .060 in./1.5 mm deadeyes and eight assemblies with .040 in./1.0 mm deadeyes (plus a couple extras of each) I started rigging the smaller assemblies to the topmasts.
 

Main topmast while rigging the topmast shrouds. After the glue cured on the lower
deadeyes at the top, I routed each shroud through the same .020 in./0.5 mm hole in
the topmast and tensioned the shrouds with standard size clothespins as shown here.
 
I can't rely on "feel" for equalizing the tension in threads at this size so always use weights, usually clothespins - in this case two on each shroud until the lanyards looked right. I filled the hole in the topmast, which is .035 in./0.89 mm diameter at the hole, with ca. A spar gets seriously weakened at a hole and I avoid holes greater than 25-30% of the spar diameter unless the hole can be filled with ca to restore most of the spar's strength. After the ca cured I trimmed the excess thread flush then tied off the lanyards. I should have tied them off when I made the assemblies - something to remember next time.
 

Completed shrouds at topmast. I'll tie off the futtock shrouds to staves on the lower shrouds later.
 
Before I could step the mainmast I needed to make a heart and collar for the mainstay. I chose to split the mainstay around the foremast simply because putting the mainstay on the starboard side of the foremast messed with my sense of symmetry. After making them I needle spliced an eye around the heart and rigged a lanyard through the heart and around the collar. This stay will be used to erect this mast inside the bottle so I routed the two lower ends through holes drilled in the deck and out the bottom of the upper hull and stepped the mast and rigged a temporary pair of backstays. I needle spliced the upper end of the mainstay making sure that the mast was at the correct rake and the collar near the foremast would not interfere with it.
 

Completed mainstay heart, collar and lanyard. The .063 in./1.6 mm drill bit is standing in for the same diameter foremast.
 
With the mainstay in place I moved ahead with rigging the mainmast shrouds. After gluing the deadeyes to the channels and fixing the chains through holes in the hull I routed the shrouds through the top, attaching each to the opposite side shroud with needle splices. The middle shrouds have one shroud forward of the mast the other aft to form a cut splice around the mast. Not quite accurate to actual practice but pretty close - it's easier to get the mast vertical, side to side, this way than by duplicating the right way. It took a while to get all the tensions balanced out and all the needle splices done, but once I was satisfied that I had the rake correct and the mast was centered I glued the needle splices. After that glue dried (I use pva thinned 50/50 with water on my needle splices) I completed the second pass through of the needle splices, glued them, attached the shroud stretchers and tied off and glued the lanyards.
 

Completed shrouds with chains, stretcher and tied off lanyards. I still need to
 rig a breast backstay and touch up the paint on the edge of the channel.
 

Inboard view of deadeyes, stretcher and tied off lanyards.
 
Before I could do any of this work I had to cut holes in the bottom of the upper hull to get access behind the holes for the chains.
 

Another angle of the chains showing the access hole where the chains were
fed through a hole, bent to the sides and glued.
 

While the schooner was on its side I got a shot of the underside of the top
to show needle splices of the shrouds and the eye splice on the mainstay.
 
I found that the blackening on the chains flaked off from handling and bending (which left black smudges on the hull I need to touch up) and tried a permanent black marker on them which worked very well. I tried the marker earlier but it wouldn't take so must assume that the acid in the blackening product etched the copper enough for the ink from the marker to adhere. A bath in acetic acid (vinegar) will probably work just as well.
 
Sorry, I forgot to include visual size references in the photos. Next, the foremast rigging and finishing the incomplete work on the main.
 

Dave

Here's the first pair of .040 in./1.0 mm deadeyes with lanyard, posed with a bigger pair (.060 in./1.5 mm) from a few days ago.
 

Left, 1.5 mm deadeye assembly; right, 1.0 mm deadeyes.
 
 

Dave

I've been building ships in bottles for more than 35 years and every one of them has been a waterline model, most under full sail, with the wind direction and strength appropriate for the sails and correctly depicted (more or less, as my experience and knowledge grew) in the sea state and swell direction. However, there was always something that bothered me about my models, increasing as time went by, because I had no confidence that I could succeed in modeling something that everyone is extremely familiar with. I have finally gotten to the point where I can no longer leave out these details. In case you haven't guessed, I'm talking about putting people on my models. There's nothing like human figures to give a strong sense of the scale of a model and bring it to life - or ruin the project if they are poorly done.
 
I'm no artist, just a modeler, and I was always lousy at drawing people. Just the thought of sculpting human figures had me extremely intimidated so a couple months ago I started research and study of the human body from an artist's point of view and on ways to make very small figures. Ogallala seemed to me a perfect project to introduce people into my work since it is a rather large scale and has become an experimental test bed. It's basically a 2X scale Conestaga wagon mated to a 1/2 scale topsail schooner rig, then reduced to a scale appropriate to the bottle I initially selected. The result is roughly 1/96 scale, about 10% smaller than HO. I rejected using HO figures (which would be large anyway) because I couldn't learn anything by buying figures and adapting them.
 
Most methods I researched begin with wire armatures, so that's were I started.
 

My first armature with the arms not yet bent at the shoulder joint and arms and legs not trimmed to length.
 
I dropped - and thought had lost - that first armature, made a couple more, threw them away and finally got one that might work as a first try. I rejected using any kind of clay and settled on building up the armature with an acrylic gesso, mostly because Lloyd McCaffery uses gesso. [bTW, I found that first armature today while doing a bit of work area clean-up and photographed it. I'm convinced that a smooth, bare concrete floor is the best thing under a modeler's work bench/table, I hardly ever lose dropped objects on that floor except for the semi-microscopic bits I don't even bother to look for.]
 
So, here's my first try after three coats, applied at about 30-45 minute intervals, between drilling holes in groups of dead-eyes.
 

Front view. He's posed for steering, weight on left foot, left hip against the wheel stanchion, both hands on the wheel.
 

Left quarter view. I think I have the head and shoulders pretty close but he's still a bit thin elsewhere. The long left leg will be cut below the foot to glue into a hole in the deck.
 

Right quarter view. He's 5'-9"/1.75m tall to scale - 11/16 in./17.5mm. One or two more coats of gesso should have him ready for detailing.
 
I'm very surprised at how easy this method is, at least this far into it. If I was doing a larger crew, say 20 figures, at this scale (I'm only going to do three or four for Ogallala), I would be able to apply a coat of gesso to all, go back through them with successive coats and have them all ready for final detailing in one day, perhaps four hours or much less with practice or for a smaller scale. Making the 20 armatures and posing them in a wide variety of activities would take longer. The gesso, being a little thicker than Model Expo paint, builds up easily with a paint brush and sets quickly. Used as a base for painting, it's a brilliant white, so it's a bit difficult to see the shape of the growing figure - and very hard to photograph; my first photos were just white silhouettes with no shadows to give a discernible 3rd dimension. I'll add a bit of black paint to the gesso for the final coat to make it easier to see as I finalize the fleshing out and the drape of the clothing. I'll continue working on this figure to learn about the rest of the process before starting the two or three more needed. Don't be surprised if he winds up being a less cartoonish version of Windwagon Smith.
 
Looking back on my agonizing over doing people, I feel really stupid (a frequent feeling) and timid (very rare) for not trying this 20 years ago - it's not at all like drawing, it's really just modeling after all. I guess it took Lloyd McCaffery's work to make me see that I might be able to do it combined with the realization that I had no choice but to at least attempt it. As you can probably tell, I'm very pleased with what I have so far. He's off to a much better beginning than I expected and my confidence now is high, which is probably the most important result of this experiment.
 
I will greatly appreciate any comments, positive and negative. I need to learn quickly how to do these figures, and other eyes - and the intellects that go with them - may notice what I have missed.
 

Dave
 
 

I decided to drill holes in the dead eyes and to rig lanyards rather than to fake them by gluing segments of thread to both faces of the dead eyes. I don't fully trust ca (used to make the dead eyes) and paint, in particular, to withstand rigging stresses. By rigging them, even if the ca fails, the dead eyes are less likely to come apart with the lanyards to assist in holding them together. By faking the lanyards I would be trusting the paint-to-deadeye bond to not fail. In either case repairing failure(s) would be next to impossible inside the sphere. I realized I had no choice but to rig them for real.
 
I started by fabricating a fixture using a standard spring clothespin as a clamp to hold the dead eye for drilling through guides in the clamp. I used 1mm hobby plywood to make the parts added to the clothespin. The guide holes are .010 in./0.25mm.
 

The pieces on both sides of the clothespin keep the two wooden parts of the pin from shifting sideways.
 

Three bits of ply glued to one of the ply yaws to locate the dead eye and the three guide holes for drilling.
 
I also made a fixture to hold an upper and lower dead eye for reeving with a lanyard. I modified two mini clothespins to hold the dead eyes by the previously installed tails with a piece of the ply positioned between to sort of wedge the dead eyes against the clamps.
 

Dead eye reeving fixture, lower dead eye to the right.
 

Closer view of the fixture. The out-of-focus light thread is the tail from the knotted end of the lanyard.
 
I used some thinned white pva glue to secure the lanyard in the holes and on the knot. I will secure the other end of the lanyard after rigging the shrouds and attaching a stretcher just above the dead eyes. I know this all seems backwards to the method used on larger models but tying and seizing and reeving the lanyards on the model at this size seems to me next to impossible. It was hard enough with the parts held steady in clamps and fixtures. This way the only tying while on the model will be of the shrouds at the tops to form short splices or bight splices with a needle similar to the way I seized the dead eyes.
 
After letting the glue set I removed the shroud from the fixture and posed it in position in a channel on the model with an X-acto blade for size reference.
 

The lanyard is not tensioned at all. It has acquired a bit of a twist which comes out with a little tension. I have to
be careful when I reeve the lanyard so that the lay doesn't tighten as I pass it through each hole of the dead eye.
 
Now to get back to the other eleven pairs of dead eyes to drill holes and reeve lanyards, then go through the whole process again, starting with cutting discs, for the smaller topmast dead eyes and shrouds.
 
 

Dave
 
 
 

Did some hull detailing and bowsprit rigging, then encountered a problem due to a deficiency of prior planning. 

 
Two piece hull, separated. Note the gammoning at the bowsprit to the stem, and the bobstay and bowsprit shrouds.
 

 
Two piece hull, assembled. Bobstay and shrouds are rigged to eye bolts in the hull with .010 in. (0.25mm) diameter eyes made from 36 gauge (.0055 in./0.14mm) wire.
 

 
Here's my problem. I tried to pass the upper hull through the bottle neck knowing there would be contact with the inside of the neck at the top edges of the bulwarks and the lower corners - just enough contact to hold the hull in place for this photo. What I didn't see in my planning was that there will not be room for channels attached at the top edge of the lower black strake, as you can see here. The gap there, strake to bottle, might be as much as .020 in./0.5mm.
 
Fortunately, Ogallala is my own design and after considering my options I've decided to move the shrouds to the top of the bulwarks without channels as was done on boats. I think (hope?) this will solve my problem and not cause another later. We shall see. I know better than to plan for a "close fit" - but did it anyway. This will learn me.
 
Dave

I'm posting a follow-up on Ogallala.
 
Upon finishing her I submitted photos to Bottle Shipwright, the quarterly journal of the Ships in Bottles Association of America (of which I am a member), and quickly learned that Ogallala would be the featured SiB in the March issue which I received a few days ago. Here's the cover. I'm not going to bore you with posting the photos from the article because they have already been posted here.
 

 
As an aside, I mentioned a couple times in this build log that I can't really see my work for what it is. All I can see are the deficiencies no matter how minor. Ogallala has been sitting in a box on a shelf since I finished her in late October - until today. I'm now able to see her as a whole and I'll admit I was impressed and very pleased.
 
If you haven't seen them yet check out my two current projects listed in my signature.

Welcome to my reconstructed build log for ESMERALDA, the existing Chilean Navy Training Ship, built at 1/640 scale in a bottle.
 

Recent photo of Esmeralda - "The White Lady" - under sail.
 
History of Esmeralda
Esmeralda's construction started in 1946, under a different name, by Spain as the sister ship to Juan Sebastian Elcano, Spain's current school ship. Work was halted in 1947 after she was damaged in a ship yard explosion. In 1951 Chile acquired her as partial payment for Spanish debts and was launched, as Buque Escuela (school ship) Esmeralda in 1954. In the 1970's her rig was changed from a topsail schooner to a barquentine by replacing the fore gaff sail with two main stay sails. 
 
Since her commissioning, Esmeralda has been a training ship for the Chilean Navy. She has visited more than 300 ports worldwide acting as a floating embassy for Chile. She participated in Operation Sail in New York in 1964, 1976 and 1986, and the Osaka World Sail in 1983. She also participated in International Regattas of Sail in 1964, 1976, 1982 and 1990 winning the coveted Cutty Sark Trophy in the last two participations.  Esmeralda is the second longest and second tallest conventional sailing ship in the world.
 
Characteristics
Length:  113 meters Beam:  13.11 meters Mast height:  48.5 meters Sails:  21 total with a sail area of 2,870 m2 on four masts Crew:  300 crewmen and 90 midshipmen Top sail speed:  17.5 knots  
The Model
 
I acquired a 1.5 liter wine bottle, emptied it (which I enjoyed very much!), then cleaned and dried it. I measured the inside dimensions using a piece of paper attached to a rod, trimming the paper until it would just touch the bottle at the narrowest point.
 

Gauge for measuring the inside of the bottle.
 

Gauge in use.
 
I subtracted the planned sea thickness and clearance for the masts from the minimum inside dimension. I used that amount to determine the model length based on the ship's characteristics and printed a photo to match the model length and height then taped that photo to a rod to test how the model will look inside the bottle.
 

Photo of Esmeralda inside the bottle. A model this size will fill the bottle very well.
 
I also cut a strip of paper to the beam of the model to verify the bottle's neck inside diameter. This is important because most necks of long neck bottles have an inside dimension that is smaller than the mouth - something I learned the hard way years ago when I assumed the mouth of a bottle was the smallest part of the neck and my ship wouldn't fit through the neck even though it cleared the mouth easily.
 
I then generated a deck plan and waterline hull elevation and laid out the plan and elevation on a block of basswood.
 

Deck plan and hull elevation laid out.
 
I used a razor saw to make a series of cuts at about 1/16th inch (1.5 mm) intervals to the profile lines on three sides of the block. After getting the rough shape by breaking off the "fins" I sanded the hull to shape.
 

Hull cut out ready for more detailing.
 
Esmeralda has a partially open area between the main and mizzen masts which will be detailed with two blocks for the two structures in this "waist" (for lack of a better name) and three pieces of .5 mm ply for bulwarks and deck. I also decided to make and install the bowsprit.
 

Tapering the bowsprit with my "mini-lathe".
 

Waist deck houses, bulwarks, weather deck over the waist and bowsprit glued in place.
 
Esmeralda has a low bulwark (a little less than knee height) at the bow. I made this from a piece of paper slightly lapped over the hull, glued with cya to stiffen the paper then sanded to feather it in to the hull. It was trimmed to .025 inch high (approximately 15 inches to scale).
 

Completed low bulwark at bow. Note the two holes in the bowsprit for rigging.
 
Esmeralda has two half rounds running from the bow, along the sides and around the stern parallel to the deck. I used a scribing gauge to slightly cut grooves into the hull which were then deepened slightly with a triangular riffle. A wire or monofilament will be glued into these grooves later then painted over to replicate the half rounds.
 

Scribing tool.
 

Hull with scribed grooves mounted on my work board.
 
I started work on the masts and some of the deck details. The masts have a box hinge instead of the usual piece of wire through the base of the mast bent into a "U" with the two ends glued into holes in the deck. The box hinge will be nearly invisible whereas the bent wire would be obvious. I also made the four identical tops for the masts from .5mm ply sanded to remove one layer of ply (about .015 inch thickness). The skylights were made from basswood painted white glued to a base of .5mm ply painted green. Two hatches were made from the plywood and painted green.
 

Mast with box hinge and four tops. Drill is a #73
 

Skylights and hatches.
 
I finished the rigging plan I'd been working on since making the decision to build Esmeralda.
 

Rigging plan for the Esmeralda model.
 
Blue and green lines represent the standing and running rigging that will be completed outside the bottle and aren't used for moving the masts and spars into position inside the bottle. Red and magenta represent the standing and running rigging that will be set up but not glued in place outside the bottle and are used for locating the masts and spars inside the bottle then glued and trimmed.
 
I laid out the holes for the masts and rigging and started drilling them.
 

Drilling holes at the edge of the weather deck for shrouds and back stays with a #80 (.35mm) drill.
 
The holes at the edge of the deck need to be drilled at an angle so they don't come out the side of the hull. I had one hole out near the end of the fan tail that needed to be over 45 degrees from vertical. I didn't quite have the angle right and the hole came out the side of the hull but it was easily repaired, but redrilling the hole from the same starting hole was somewhat difficult. I also had to change to a larger diameter drill because the smaller drill wasn't long enough to drill all the way through to the bottom of the hull.
 

Close-up of holes for fore mast shrouds and back stays. Holes for main mast shrouds and back stay are visible next to the bulwark in the "waist" at upper right.
 
Work in general had progressed to the point that I felt getting the sea in the bottle was necessary. I used Fimo Classic modeling clay (a polymer based compound - Sculpey, Pendo and Plasticine are possible substitutes - but Fimo has a navy blue that comes very close to the color of deep water ocean). I formed it into a piece 7 inches by 2 inches by about 1/4 inch thick tapering to zero around the edges to fit into the cylidrical bottle and leave the top surface of the sea roughly level. I then put the hull on top of the clay in the position needed then pressed it firmly into the clay with a steel rod.
 

Hull pressed into the sea to leave an impression.
 
The steel rod is 3/16 inch diameter with the end ground, filed and sanded to a hemispherical tip then bent as needed. The gouge is a similar sized dowel with a slot cut in the end to hold a strip of aluminum soda can then wired tight. After determining the needed angle of the gouge by trial it was glued.
 
After the hull was removed I used the gouge to remove some of the clay from inside the impression of the hull. I used the removed clay to build up a wind swell running from near the near left corner to near the far right corner (replicating the sea with a wind from about 60 to 70 degrees off the starboard bow).
 

Using the gouge to remove clay from the hull impression.
 
I put the hull back in the sea, pressed it down, removed it and gouged out more clay until the hull fit snugly in the depression. After that was done I used the rounded tip of the steel rod to cover the surface of the sea with little overlapping dimples except for the ship's wake which was smoothed with the tip.
 

Sculpting the surface of the sea.
 
I made one more check of the hull in the depression for it in the sea then added some white Fimo to simulate the agitated white water at the bow, along the sides, in the wake and for a few wind blown white caps. I placed flecks of the white (or a partial mix of blue and white) on the tip of the steel rod to transfer them to the sea then feathered them into the blue.
 

Finished sea.
 
I completed basic mast assembly - lower masts with hinges, tops and upper masts - and set them into the appropriate holes in the hull.
 

Esmeralda in progress on the working board with some of the tools being used.
 
Then I rigged her with one pair of back stays on the fore and spanker masts, a single fore stay and the mast to mast stay that connects all four masts at the mast splices. This was done in order to do a trial fit of the hull and masts to verify that the completed assembly will fit in the sea with clearance for the masts to stand erect. I calculated my dimensions very closely with about 1/16th inch clearance between the mast tips and the inside of the bottle.
 

Esmeralda with minimal rigging. The rigging is tied off but not glued. It will be removed after the trial fit.
 
On a previous build I neglected to do a trial fit on a ship that I had also calculated very closely. I got that ship in the bottle and in place in the sea. I had pulled up the fore mast and positioned all the fore mast yards and had glued most of the fore stays and working lines to achieve this, then moved on to the main mast. The mast tip hit the inside of the bottle before it was in its correct position. I had to cut all the glued lines to get the fore mast to fold back down so I could remove the model. Even at that two fore mast yards snapped because I could not get them arranged the way they were when I put the ship in the bottle. I NEVER omit a trial fit after that experience no matter how confident I am with my calculations.
 

Esmeralda going into the bottle.
 

Esmeralda in place in the sea with masts all erect. There is about 1/16th inch clearance between the mast tips and the bottle.
 
This trial fit also gives me a chance to trouble shoot the insertion process. In this build the deep dimple in the base of the bottle prevents me from simply setting the model in place then erecting the masts. I'll have to start erecting the masts when the bowsprit is still partly in the neck of the bottle. Another good reason to do a trial fit with bare poles and minimal rigging.
 
Now that I know my masts are all correct it's time to detail them. I made some ring bolts out of 32 gauge (.008 inch) wire by bending the wire around the tip of a needle, forming the two ends next to each other then cutting the ring bolt from the length of wire. I had so much fun doing them I made 25 of them before I realized I only needed 12.
 

Ring bolts.
 
I made the fore mast yards, drilled the holes needed in all the masts and yards then painted them. I decided that I could make use of the extra ring bolts by linking two of them together, eye to eye, to use to connect the yards to the fore mast. I glued one ring bolt in a hole in the center of the yard then glued the other linked ring bolt in a hole in the mast. After that I rigged the topmast shrouds, three pairs on the fore mast, two pairs on the other three topmasts.
 

Rigging the fore topmast shrouds. Note the linked ring bolts connecting the yards to the mast.
 
I also added an extra ring bolt to the forward edge of the top on the fore mast to use to tie off the two fore stays that start there. After rigging the topmast shrouds I glued a crossbar of thread to the shrouds about 1 mm above the top. I painted the threads from the crossbar to the top white to simulate the turnbuckles used on Esmeralda instead of deadeyes and lanyards. The futtocks were painted to match the masts.
 

Masts with completed shrouds.
 
I then rigged the fore mast yards with one thread through each yard end, through the mast then through the other end of each yard. The two ends terminate with a thread block - in this case a bowline tied around the point of a needle then glued - to simulate a block used in the actual rigging of Esmeralda. The only place where the thread is glued is at the course (lowest) yard which allows the yards to turn so they are nearly parallel to the mast so the assembly (with sails in place) will pass through the bottle's neck.
 

Foremast with some of the running rigging. Note that the linked ring bolts merely locate each yard.
 
I worked on the hull as well by gluing 28 gauge wire in the previously cut grooves in the hull, installing four ring bolts on each side of the bow and rigging the bowsprit stays. Then I gave the hull the final coat of white and glued the deck details that were ready in place.
 

Some of the deck details installed.
 
I then painted the figure head - an Andean Condor. 
 

Close-up of figure head and bowsprit stays.
 
This brings us up to date on my progress reports on Esmeralda prior to the MSW shut down for the software change. In that time work continued on deck details.
 

Davits for three of six ship's boats, crane for the zodiac, the bridge, two catwalks, four winches and the anchor windlass and associated details in place.
 

Close-up of midship section.
 
Davits were made from 28 gauge wire bent to shape and glued in holes in the deck. Crane made from three pieces of wood and styrene. Bridge made from a .5 mm ply base and roof, wood between, with 32 gauge wire legs at the wings. Catwalks from .5 mm ply with eight 32 gauge wire legs glued into holes in the ply and deck. Winches made from two pieces of wood on a ply base.
 

Close-up of windlass, bits and hawser hole covers.
 
The anchor windlass was fabricated from eleven pieces of .010 and .020 inch styrene and three pieces of wood on a base made from paper stiffened with cya glue. The windlass isn't glued down yet. That will be done inside the bottle to cover a hole through which six lines pass after they are glued and cut. You may have noticed two threads coming from under the green covers. They will be used to pull the two chain ends into place during the windlass positioning. The chain is copper, 42 links per inch, blackened with the product from Model Expo. I know it's a bit over size (by about 50%) but I could not resist putting it in.
 
With davits on deck I need to make some boats: two longboats, two smaller boats similar to old whaling boats (pointed bow and stern, narrow beam), a zodiac and another power boat similar to a modern Boston whaler. I haven't seen this boat in any of the photos of Esmeralda I've found, I may just have to make my best guess and wing it. I can infer the length from the davits and the beam and length from Esmeralda's deck plan which shows the outline of this boat on deck but nothing more.
 
I started with the longboats. I made a plug from basswood, dipped it in warm paraffin wax (so glue wouldn't stick to it) and wiped away the excess wax. After attaching it to the tip of an Xacto knife I covered the plug with narrow (less than 1 mm) strips of wet newsprint laid lengthwise, edge to edge, like planks. I painted a layer of thinned wood glue over that layer then added a layer of wider strips laid across the plug, again edge to edge with another coat of thinned glue over it. After the glue thoroughly dried I lightly sanded the outside of the paper hull then trimmed the overhang flush to the plug. It popped right off the plug at that point. The second longboat hull is drying now.
 

Longboat plug and hull.
 
Now that I'm all up to date I'll end this report. I welcome any comments and questions.
 
 
Edited title of topic to reflect a change to my display name.
 
 
 
 
 
 
 
 
 
 

I'm posting a follow-up on Ogallala.
 
Upon finishing her I submitted photos to Bottle Shipwright, the quarterly journal of the Ships in Bottles Association of America (of which I am a member), and quickly learned that Ogallala would be the featured SiB in the March issue which I received a few days ago. Here's the cover. I'm not going to bore you with posting the photos from the article because they have already been posted here.
 

 
As an aside, I mentioned a couple times in this build log that I can't really see my work for what it is. All I can see are the deficiencies no matter how minor. Ogallala has been sitting in a box on a shelf since I finished her in late October - until today. I'm now able to see her as a whole and I'll admit I was impressed and very pleased.
 
If you haven't seen them yet check out my two current projects listed in my signature.

Thanks for the very kind comments. Here's the best of the photos of the finished Ogallala.
 

 

 

 

 

 

 

 

 
Thank you to all who viewed this log and especially to those who contributed with 'likes' and with comments - you all pushed me to do my best possible work.
 
Now I can get back to work on Esmeralda. I learned a few things here I can apply to "The White Lady".
 

 
Dave
 
 

Ogallala is finished. 
 
 
Tensioned and secured all of the lines but had to redo one that was done wrong. Had to cut it loose at one end and glue another thread to it end to end, then re-tension the line, glue and trim it inside the bottle. Don't bother looking for the splice - you won't find it.
 
Cleaned the inside glass again then located the prairie disc and ship using the four threads from the wheels taped to the outside of the opening. Once in position I epoxied the disc to the glass in three places. When the epoxy cured I painted the inside of the bottle below the soil line and re-checked the tension on the four threads. I had intended to simply epoxy the other medallion over the opening to trap the threads but decided this may not be secure enough. I cut a block of basswood to fit between the medallion and the bottom of the prairie disc and epoxied it to the medallion. When cured I epoxied that assembly to the disc and the opening.
 

The medallion with the basswood spacer after checking the fit. I epoxied the
assembly to the disc and the face of the opening to trap the lines as well.
 
After the epoxy cured and the outside of the bottle / sphere was cleaned I took lots of photos. Here's two full view photos.
 

 

 
I'm still going through them all and will post more tomorrow or the next day. Click on the photos for larger versions.
 

 
Dave
 

Finished Ogallala and can now get back to work on Esmeralda. I hadn't intended to take this much time away from Esmeralda but Ogallala turned into a much more comprehensive project than I had originally planned. I did find some more, useful, photos of E for some details on deck and aloft and kept her in mind during my side journey. I learned a few things from O that I will apply to E and I have a bit of re-work to do and some details to add before I pick up where I left off. I'll post my progress soon.
 
I came across an interesting variation on a ship in a bottle with Esmeralda as the subject but the method is a bit too ephemeral for my taste. 
 

 
Dave