Scratch building the Alfred Stern

[Dave Stevens (Lumberyard)]

Hello Dave, I have a question about the stepping of the frames to the deadwood. I'm sure that the stepping is more practicle in reality than the Hahn smooth bearding line. So the question: did you follow the Hahn lines? If yes, did you use the point of contact at the fore or aft end of the frame?
I intend to build the Warrior and will be using your thread to model the deadwood to frame look.
Thanks, Bill

i will be getting into this very thing a little later on. so stay tuned

 

[Dave Stevens (Lumberyard)]

it is a little difficult to see but Hahn used double frames both for the bends and for the filler frames. Because he built his model at 1/8 scale the filler frames would be very delicate and prone to break or not hold their shape. Even kiln dried wood will still move with changing conditions. double frames would be far more stable than a single frame at a small scale.
I increased the scale to 3/16 so i used single filler frames scarfed together at their ends.

 

[Dave Stevens (Lumberyard)]

Building off frame Z i clamped the first single frame to the deadwood and to Z. i had to use a C clamp at the deadwood because the fashion timber and frame Z was too wide for the spring clamp.




down towards the jig above the fashion timber i can use the spring clamp to clamp frame 51 to Z



To ensure the position of the foot of the frame i am using a spacer



a spacer is also used lower down

 

[Dave Stevens (Lumberyard)]

i did not want to rely solely on the jig or the stepping in the deadwood to position the frames at an equal distance, so i am using spacers.



the clamps i got from Model Expo are perfect for the job. They are small enough to fit between frames



the spacer at the foot is thinner than the space at the level of the jig. i did not glue the spacer at the foot, once the glue stuck i removed the spacer. Lower toward to jig i did glue in the spacer because that section of the hull will be covered with planking and the spacers will not be seen.

 

[Dave Stevens (Lumberyard)]

with all the frames square to the keel the space between frames would be the same from the keel to the top, Also the floors of the frames at the keel sit in a notch which will maintain an equal space and the top sits in the jig notch. At the stern and bow it is a different story the foot of the cant frames sits against the deadwood and there is nothing to maintain an equal spacing. Thus, the need for spacers.



The symmetry of the framing at the stern is the beauty of the framing as the space forms a nice, angled space between frames from top to bottom.

 

[Dave Stevens (Lumberyard)]

As i am building it is becoming noticeable for precision cutting of angles and placement of the framing. A little off goes a long way, The space between the frames is an angle, very slight but it adds up.



what i found out was clamping the frame together on the outer edge pulls the frames to more of a parallel to one another rather than following the slight angle. After a few frames it becomes noticeable between the location of the step in the deadwood and the notch in the jig. What i did was change my clamping to the inside of the frames.

 

[Dave Stevens (Lumberyard)]

Frame by frame the hull went together, notice the two frames 50 and 51 do not have a laser char, that is because i redid those frames. This should have been a warning sign that something was not quite right. Being a bit apprehensive i continued to assemble the hull, at this point i had built all the frames so i might as well carry on.

 

[Dave Stevens (Lumberyard)]

something seems to be way off the distance from one frame to the next looks to drastic. The tinted frame yellow, red and blue compared to the frames directly behind them look to short or the frames behind then are sticking to far out of a line



from the red frame to the yellow frame are to far out of line to get a smooth shape to the hull

 

[Dave Stevens (Lumberyard)]

this calls for an investigation and back tracking, starting with the CAD drawing i scanned the original frame drawings and traced them.



looking very close to the tracing there are two lines red and blue. You have to sort of stare at the image to see them. The dotted line shows the red tracing better than the others. The tracings of Hahn's original frame drawings are within 0.002 thousandths less than 1/3 the pencil line width.
Inaccurate tracing of the frames can be eliminated, that is not the problem.



moving on

 

[Dave Stevens (Lumberyard)]

having ruled out the tracing of the frames i looked at the positioning of the frames which are 3 points. Point one is the frame to the jig, the foot of the frame to the deadwood and the distance from center to the jig.

center to jig and keel notch to jig base

red is center top blue line is the jig and bottom green line is the keel notch.



that black arrow from center to the jig has to be right on



keel notch to jig, the way the frames were drawn in CAD the jig line and the keel notch line were drawn as parallel so there is no chance a frame would be to short or to long except where the frame start to rise toward the stern and that was taken into account.



no errors were found here

 

[Dave Stevens (Lumberyard)]

If no errors were found in the CAD drawings it leaves one last thing to look at and that is the hand of man.

sloppy workmanship!

if the angle at the foot of the frames were cut at the wrong angle or too much was removed at the foot, the frame would move in or out from center. If too much was sanded off the ends of the frame pieces the frames would be to short and they would float higher or lower in the jig.

 

[Dave Stevens (Lumberyard)]

The best way to test the sloppy workmanship is to build the model in 3D directly from the drawings and eliminate the hand of man.. With the actual build out of the equation the only thing left is what the files builds. CAD does not do sloppy work if something should be at an X,Y,Z coordinate that is where it will be with no + or -

lets take a look



now let's rotate the model and see what happens, ok see those frames at the right side next to Z i saw that and that's why i made new ones



WOW! the computer build the same thing i built and the frame are out of alinement. It was not my sloppy workmanship the plans i traced are not accurate.



how did this happen? Hahn drew the plans at 1/8 scale and i blew up the plans, so any slight errors became bigger errors.
what it comes down to is the Portia Takakjian method which is Portia was a top-notch model builder but when it came to drafting well not very accurate. What she did was to draw out the framing or parts with a + or - in mind and cut them oversize. After the hull was built, she sanded the hull to its final shape.
What Hahn did was to allow for minor errors and built the frames a little oversize and like Portia sanded the hull to its final shape. This works fine if you scratch building but when your drawing in CAD and laser cutting there is little room for error. It is what was drawn no fudge factor. Not as much oversizing if cutting parts by hand, there is an offset built into the drawings but not enough to compensate for bigger errors.

 

[Dave Stevens (Lumberyard)]

At this point in the build i finished the stern section so from this here on it is now a "lets see what happens if"
i am going to do a Portia on the hull and just keep sanding it and see what happens.

The overall shape of the hull allows me to sand the frames so they begin to line up, well not line up but sanding in a bevel so each frame flows into the next one giving a smooth surface.





the top section of the frames are not to bad, they are sanding out pretty even.

 

[Dave Stevens (Lumberyard)]

back in the day Harold told me when shaping a hull you have to 'look at it" from different angles. So here is my method of looking st it

it all starts where the frame sit in the jig the upper red line sand the frames so they all fall in line with each other
move up to where the frame begin to bend into the lower part of the hull.



the next section is where the hull goes from the top end to the bottom end so try to maintain a line so the frames blend into one another.



the top of the bend should be straight, you can see on the left i am still a little high with the double frame W and a little low with frames 43 and 44



looking towards the stern there is an arc in the foreground that gets smaller as it reaches the stern

 

[Dave Stevens (Lumberyard)]

by following the lines of the hull you will get a smooth hull with all the frames blending into one another



when done right, from different angles you get these graceful serpentine curves. Frame 43 right in the front is still a little out of wack, it is sitting low.



from where the hull started with all the frame out of alien it was still within working parameters to get a nice-looking hull. so the model should be good to go.

 

[Dave Stevens (Lumberyard)]

good to go? sure when you see the hull from the outside but here is the next problem. Out of alien or frames that are not quite to the right shape is still an issue on the inside. I sanded the outside until i had a proper shape but that took a lot of material off some frames. when it came time to shape the inside there just is not enough meat on those ribs. What will happen here is the hull will end up as thin as an egg shell before i can get the proper inside shape.

It is clear the blue frame W is way too far in and the frames before W and between W and the red frame are way too far out. The yellow frame is too far out of wack to fall in line with the frames before and after it.



it has been shown this happened because of small errors at 1/8 scale of hand drawn frames, when blowing up become bigger errors. When scratch building not leaving enough material to finalize the hull shape. Hand drawn frames when converted in CAD which has zero + or -
Plans drawn by Harold Hahn are by far very accurate or should i say well within scratch building tolerances. Any plans will have a certain amount of + or - it is up to you as the builder to take into account wiggle room is needed.
Model ship building has the same issues as real ship building. Frame shapes are taken off a half model or table of offsets, scaled up on the mold loft floor. The patterns are made and sent to the building site. Once at the site the patterns are used to cut out the frame parts and the frame assembled. Then the finished frame is lifted and set in place on the keel. All along this process there will be errors the way hulls were built is the final "dubbing" process which is the process of dressing the timbers with an adz using "battens" a thin strip of wood, used by ship’s carpenters to reproduce the curves from one frame to another.
Hahn's plans will bring you in close but not exact.

the model you are seeing so far in this build was scrapped and tossed out used to start a fire to smoke some salmon. But before it met its end this is what i did.

coming up Part 2

 

[Dave Stevens (Lumberyard)]

first thing i did is to dissect the hull by ripping off one side

​

here is what i found, by sanding the hull to the point of getting the shape right there was very little left of the frames. If i tried to finish the inside i would sand away what is left of the frames. Even at this point the outside was not perfect i could have gone a little more to refine the shape of the hull, but as you can see the frames would fall apart.

 

[Don Case]

Great write-up Dave. Couldn't you have seen that things weren't going right as you were gluing on the frames and corrected it then? Or were you trying to show us something?

 

[Dave Stevens (Lumberyard)]

redrawing the plans would be a big task here is the bodyplan on the Hahn drawing on the left, on the right is taken from the built model. You can see the absence of information on the original drawing when compared to what you actually need.



In CAD what i did was to take the frame drawings and place them as a bodyplan and the problem makes itself clear. Frame lines should not cross over one another nor should they be randomly spaced.
Keep in mind Harold drew these by hand at 1/8 scale, the lines look wide apart and thin lines BUT on paper with a pencil the lines would all run together and look solid.



what i am saying is the red dot is a pencil and the red line is how wide a pencil line would be on a drafting table drawing plans at 1/8 scale. So what Harold did was amazing he got as close as he did.



Looking at the same area of the drawing done in CAD you can refine the drawing down to the finest detail. i can go 0.0001 thousandths

 

[Dave Stevens (Lumberyard)]

Great write-up Dave. Couldn't you have seen that things weren't going right as you were gluing on the frames and corrected it then? Or were you trying to show us something?

i built all the frames first so at that point i did not know how things were going to line up. As soon as i had frame Z up and i saw the next 2 were out i remade them. Then i continued thinking ok this might work itself out. But it didn't.

looking at all the laser cut parts and having all the frames built i was committed to carry on. Besides i now need to find out how to do this. The project has become more of an experiment than just building a model.

my end game here is to produce a build project for the forum so i had to carry on and see what went wrong and how to fix it.

using drawings as is does to convert to CAD and laser cutting

it is not all that easy to draw in flat 2D and visualize the outcome in 3D.