Material for keel

[mikegr]

I am going to start my first scratch built project. Length is 110 cm.

What is the suggested material thinkness for the keel and  frames?

Apart MDF and plywood I am thinking of using a synthetic material which i believe would be easier to work with as I have not worked with wood so far. So I have found the following:

-ABS

-PVC

-Styrene

-Forex

If anyone was worked with the above materials would like to know his experience.

 

[Vladimir_Wairoa]

Hello Mike, thickness pf the keel is according to the plan/ scale. i can imagine dense grain wood like cherry pear probably best suited in regard of comfort of eork etc...stay far away from hardwoods like oak etc...frames if you mean bulkheads choose something you can later push pin into etc...and if you are going to fill reat of skeleton woth balsa etc..you can have fine base fo gluing and wood wont work if future...

[Jaager]

At first, I thought you really meant "keel".   I am seeing that you are likely asking about a center spine and the moulds for a POB build. 

Consider 1/4" Baltic Birch plywood  -  if you are US based, HomeDepot has convenient to handle sizes - on the final model,  none of it is displayed.

Go for quality.  A-A  is ideal - B-BB will probably serve. 

The synthetic stuff - among other factors, may provide a poor bond with hull planking and any space filling material you might use.

[mikegr]

Thanks for the replies. Here is the plan I have to get a better idea. Scale is 1/144.

 

[allanyed]

Mike,

As this is a metal hull you could consider making a wooden plug of the hull, then using that as a mold to form a plastic hull.    I recall seeing some fantastic hulls and other parts for that matter, of freighters and warships built in this way by a couple members of the NJ Ship Model Society 5 or 10 years ago.  Hopefully there are some members here that can give you some information. 

 

I have made plugs, then used the plug to make a mold, then once cured, apply a mold release, then gelcoat, then fiberglass and resin to make the hulls.  But for a one-off, it does not seem worth the time and expense compared to a plug former and hull made from a thermoplastic sheet.

Allan

[Roger Pellett]

For a modern steel built vessel, I’d just carve the hull.  I make two half models from laminations shaped from buttock lines.  A ship like this does not have a pronounced keel like a sailing vessel but with this type of construction you can sandwich a brass or if you must, a plastic center plate between the two half hulls.  This helps to preserve the bow and stern profiles.

 

The benefit of carving two half hulls is that you now have a flat surface, the vessel’s centerline that can be laid on a flat surface to check contours with templates.  I drill holes for locator pins in each half hull block prior to shaping.  Once the half hulls are shaped these automatically ensure accurate alignment.

 

If you try to build this as a planked POB model there are some areas like the sharp bilge radii, the bow, and the stern where planking will be difficult.  Furthermore, once your planking is done you’re going to have to hide it by filling the seams with some kind of goop, with concerns about them later opening up or by covering the whole thing with fiberglass, a messy job.

 

Roger

[Jaager]

Mike,

I am without doubt obsessed with the method and this vessel is far more recent than my eras of focus.  If your plans include a complete Body plan with delineation of each of the stations shown above, The Station Sandwich Method would get you a hull. 

How I would approach it =

I would use clear Pine sliced from framing 2x4.  No thicker than 1/4" ,  but it is easier and less open to mistakes if the sum of the thicknesses is an exact match to the distance between each station.

The interior of the hull is not of much interest so a solid hull is possible.  I would still make it somewhat hollow.   The moulded dimension would be enough to encompass any bevel that is between each pair of station lines.  I would not shape the inside.  It would be horizontal above the floor and vertical at the sides.  It saves on lofting time.

Rather than cutting each of the layers as a single piece,  I would use a rough version of wooden ship framing.  Do it at the first step as  a pair of layers with overlapping butts. 

One of the pair would be three timbers - a full "floor" and a "2nd futtock" on each side that extended to the deck.

The other would be four pieces -  two "1st futtock"s  that butt at the centerline and go beyond the floor/2nd futt join.  And two "3rd futtock"s that extend to the deck.

The smaller pieces are easier to scroll cut.  there will be no cross grain.  The butt line of the "1st futt" provide an easy way to locate where the keel is.

The  lofting process can be done easily using a drawing program.  The plans provide the precise outside shape.  Connect the dot straight lines define the inside shape.

Each piece has a pattern rubber cemented to it.  When the layers between two stations are all glued together, the pattern is left on the "frame" at each end.   Because of the locators. the patterns on each end are in precise alignment.  The bevel will be correct.

 

Locator points and the lofting of them:

On a wooden hull sailing ship, where the inside shape is important and relatively narrow,  a perpendicular locator to position two frames, much less the series of them between two stations,  will not go thru the actual body of more than one frame.  This starts to happen when you get much beyond the middle of a ship.  I solved this by placing my locators outside the actual frame.  It makes for extra wood for each timber and for more wood to remove when getting the final shape for each frame sandwich assembly.

This is not a factor with your hull. The thickness of each frame sandwich would be wide enough that an inside perpendicular would work as a locator.

You would need 3 sets of locators.

1 - to align the timbers of each pair.

2 - to align each of the pairs into the sandwich of layers between two stations

3 - to mate the two sandwiches that meet at each station.

For  #1  I would use  the 1.75" long steel quilters pins  they are #73 wire gauge and #70 hole is close without being difficult to remove after the glue has set.

For #2 and #3  I would use off the shelf bamboo skewers.  Measure the gauge and buy a few bits that are close enough not to wobble, are a push fit,  but no not need a hammer.

Glue in the bamboo.

 

The assembly method would be 

scroll cut the timbers  -  no need to get too close to the pattern line -   A good hand fretsaw would get your there,  a scroll saw if you have one,  I use an 1/8" blade with a Carter Stabilizer on a 9" benchtop bandsaw.

Join and glue up the "frame" pairs. 

Assemble and glue up all of the pairs in a station sandwich.

Using a sanding drum, shape the near outside shape and do the bevel for the sandwich.

Join pairs of sandwiches and fine tune the transition by sanding.

Starting from the middle and going to each end,  join the pairs of sandwiches to the whole and fare those transitions.

The bow and stern build and shaping are a different challenge.  I would use a buttock dimension series of layers there.

 

Pine is relatively inexpensive and readily available.  It is easy to work.

Worse comes to worst, the hull can be the subject of a serious sealing and undercoating.  With a good final coat of paint, it should look metal.  If you wish the metal plates to be hinted at, rectangles of paper can be glued to the hull before sealing.  Things like bilge keels - parts needing to be glued to the hull - mask the glue area before sealing.

 

Anyway,  here is an alternate method that breaks the hull shaping process into smaller and more manageable sub assemblies.    

 

 

 

[mikegr]

Found this method also. The square balsa pieces looks ideal for planking a model of a modern ship.

 https://youtu.be/8qlNx_6wD8Q

[mikegr]

What about the thickness of the frames? 3 or 4 mm would be enough?

[Jaager]

Mike,

 

If your question is about my presentation,  yes, ~ 1/8" +/-   would work just as well.   I kinda jumped the description to a thickness that - to me - is about the max as far as stock thickness for reasonable ease of cutting ( ~1/4").   I do not know the dimensions that you will be working with,  but if 1/8" x 8-10 layers or less does the job, you are where I am with the hulls that I build.   If it required 20 or so layers, I would go thicker.  If you go with Pine, the cutting and shaping will be fast work.   

Right now,  I am fighting with the different bonding required for the spaces next to the room for USS Vincennes (1825).  La Renommee was a bit of a flyer in that I built it with all room.  All room, which is what I suggested for you, is much easier and faster.  And even less work on pesky details, your ship has no gun ports to worry about. 

[Jaager]

I do not know if I addressed this before,  but one factor to defend against with a multilayer build technique is  error creep. 

When each pair of overlapping frames is glued up,  it will be strong and stand up to sanding.   I remove the pattern and rub it on a sheet of 220 grit sandpaper on a 12" x 18" piece of tempered plate glass.  There should be no difference in thickness at each butt.  The thickness at the rail should be the same as that at the keel.  The pattern on the two end frames of a sandwich should not be removed now,  Once the station sandwich is a single unit and it has been shaped, then the pattern is removed and those two face sanded on the glass surface.  Measure the thickness at the rail and keel for the sandwich.   

When the build gets beyond this, it is difficult to get accurate thickness measurements.  As the sandwiches are joined together, the key  control is to make sure the keel is dead flat. 

If the top is off, it can be shimmed or sanded to get  things right.  I was remiss in measuring with Marseilles.  It is a first rate and has a lot of upper works.  I built the fore half as a single unit and the aft half as a single unit and joined them at the middle.  With a flat keel, there is a 1/4" gap at the rail - right in the middle.  I can shim it and it will be planked over anyway,  but that taught me to start the joining in the middle and work to each end,  Any error creep is much easier to fix.

[Jaager]

Here are some fine tuning points, that I am being reminded of as I assemble the hull of Vincennes:

It was an early and transitional development stage.  I milled the stock, arranged the patterns, scroll cut the timbers, used a disc sander to set the butt joints,  assembled the timbers into frame pairs (bends) using pins - to check for butt joins that were too fat, and disassembled and stored each sandwich in a multi compartment storage box. 

Now it is over 5 years later. The Black Cherry has darkened nicely.  but it also reminds me of things that I now do a better way.

Four pin locations are better for each timber.  The points being two at each far end.  The points not being too close to the pattern line.

 

I was using a 4 point Ariel Black lower case letter "o".  It is round in the center. The area is close to a #70 drill bit hole.

The wall is thin and guessing where - how far out to place each one - is tedious - having to decide for each. 

My solution is to make a new letter.  It is a 7 point "o" merged over a 4 point "o".  I saved the layer and have lots of copies.  The hole is a precise size and the wall is thick enough that if I just kiss the pattern line, it is far enough out that the hole does not mar the face of the timber.  The wall is thick enough that by staying outside it when scroll cutting, The hole is not cut into and made useless. 

( I loft at 1:48 and reduce the patterns 80% to get 1:60,  I place the locator points on the 1:60 - the final size.)   Placing the points and then reducing is not a good idea.

With this idiot proof construct, now the main "be careful" is to make sure the locators are inside the butt lines at the end and are not sanded into when the butt lines are sanded.

A model where all of the locators are inside the timbers will avoid most of these problems.  

 

One suggestion.  more points than you need is better than not enough.  It just costs a few seconds when lofting to place a point, and if you do not need it just do not use it.  If you do not have enough, it is too late once the timber has been isolated. It is actually too late after the pattern has been printed out.

 

For the floors, a really wide line to sand to is a difficult task to do precisely.   The butt of the floor with futt 2 is better done if it is above the inside moulded line of the floor.  

It is also easier to disc sand if it is an angle . higher on the inside and sloping down to the outside line.

For POF,  I have gotten better at disc sanding exactly to the butt lines than I was 5 years ago.  Too fat and it is back to the sander, so that the assembly works.  Too much off, and there is an ugly gap.  This is not the best look for POF with visible frames.  Wood flour in PVA fixes small gaps, thin shims fix larger ones.  For a solid hull, that is sealed and painted,  precision at the butts is not so important. Wood flour in PVA will fill and paint will hide. 

 

For the lofting -  

When importing a scanned plan into the drawing program.

Adjust the change in scale that most every scanner will do.  It is different for each scanner, but is constant for each one.

After the scale is repaired, set a large, but not too large sensitivity range for the magic wand,  click on the background, CUT

The layer should now only have the desired lines and be otherwise transparent.

Now adjust the rotation to get vertical to a vertical and horizontal background.

Lock the layer, duplicate it.  Clean up the junk using CUT on the duplicate,  lock, and duplicate this clean one. Unless you like repeating work, only make changes to duplicates

When you add something new, do it on a new layer.

Layers are your friend.  The only limit on the number of layers is in how large the drawing program will let a file size become before it gets squirrely and or adds artifacts.

Backup often  -   

Large hulls may require more than one file.  At 1:48, a frigate requires two files.  A 74 requires even more.  I really fear what the Pennsylvania will require.

[Srodbro]

Mike Bryan:

I’m just curious ... perhaps I missed something. 
 

Given the plan you show, how do you go from that to drawings of the contours of the hull?