Fitting scarf joints for wales

[glennb17]

Hi everyone,

I'm having some problem with the scarf joints on the wales of my HMS Alert from Trident. The stock is 3/32" cherry and the pattern of the scarf is slightly complex. I scraped off the char and filed until I got a snug, mostly "no daylight" fit. On my first shot of affixing to the hull, I bent the wood with water but ended up with these awful, malformed, splayed joints. I tried to finesse it, but they just looked terrible. That first scarf joint sits at a spot with a respectable turn radius and even a little curvature from the top to bottom.

I had to laser cut a few more pieces (I don't trust myself to get things right the first time with kits, so I always scan parts first). I got a tight fit again off model. This time I put up one segment of the wale at a time and let constant tension slowly bend the scarf flat to the hull after a long soak, banding up, and drying overnight. I did the same for the next segment with the first segment in place. They both look like they took the form properly, but the joint, overall, looks splayed outward.

Here's my question. Do I file along the scarf joints at an angle through the thickness to get a match? Or did the train leave the station when I fitted the two pieces off model and I need to start again and fit once bent? I know there are plenty of modellers that make fantastic scarf joints and anchor joints in their wales, I could use some pro tips.

A word about bending the wood. I found wetting the 3/32" stock with water, adding newly heated water every 5 minutes for a total of 20min gives you something bendable but you have to let the tension do the work on the scarf joint overnight to flatten it completely while it is drying. I was hoping to get more pliability and ammonia was recommended, but that seems to discolor/bleach the wood and the dried form is a bit brittle. If you think the bending step is the problem, I could use some proven techniques to get it right. Everything but a soldering iron. The curves are too complex for a 1D heat curve (the hull at the joint sites are curved both along it's length and width).

Thanks for any help you can offer.

 

[dockattner]

Hi Glenn,

I have used two solutions. I have found success by pre-bending the stock and only then cutting the scarf (tricky, but doable). But my most recent solution was to keep the strake full length and put in a faux scarf joint with a sharp blade (and that includes the top and bottom, not just the outward face). The problem with this approach is the grain pattern gives away the fake joinery. Then again, if the wale is painted...

 

[glennb17]

Hi Glenn,

I have used two solutions. I have found success by pre-bending the stock and only then cutting the scarf (tricky, but doable). But my most recent solution was to keep the strake full length and put in a faux scarf joint with a sharp blade (and that includes the top and bottom, not just the outward face). The problem with this approach is the grain pattern gives away the fake joinery. Then again, if the wale is painted...

Thanks Paul, it's appreciated

 

[Dave Stevens (Lumberyard)]

the wale scarf was actually talked about in the school here is a link

School for Shipmodel Building - School for model ship building

Charlie, A comment relative to your question but I’ll leave exactly modeling techniques to others. The great weakness of wooden ships was their longitudinal strength. In the days before modern waterproof adhesives ships were a collection of pegged together wooden parts each one trying to move...

shipsofscale.com

 

[Bob Cleek]

the wale scarf was actually talked about in the school here is a link

School for Shipmodel Building - School for model ship building

Charlie, A comment relative to your question but I’ll leave exactly modeling techniques to others. The great weakness of wooden ships was their longitudinal strength. In the days before modern waterproof adhesives ships were a collection of pegged together wooden parts each one trying to move...

shipsofscale.com

The "search thingie" is our friend. Before asking a question, it's best to check it. More often than not, there's been a long discussion posted previously which will answer the question and then some. This forum has been around long enough that there aren't that many possible questions which can arise that haven't been answered already.

 

[Bob Cleek]

A word about bending the wood. I found wetting the 3/32" stock with water, adding newly heated water every 5 minutes for a total of 20min gives you something bendable but you have to let the tension do the work on the scarf joint overnight to flatten it completely while it is drying. I was hoping to get more pliability and ammonia was recommended, but that seems to discolor/bleach the wood and the dried form is a bit brittle. If you think the bending step is the problem, I could use some proven techniques to get it right. Everything but a soldering iron. The curves are too complex for a 1D heat curve

It is heat that softens lignin and causes wood to bend easily when heated and to hold that bent shape when cooled. Not water! Not amonia! Not steam! Heat. Just plain dry heat. I really don't know why this message just doesn't compute with so many modelers.

In large sized stock, steam is frequently used to permit "steam bending." (Actually, "heat bending.") This is because steam is an efficient vehicle for distributing the heat through the wood under slight pressure in a "steam box." The steam heat is applied for a minimum of one hour for each inch of thickness of the piece's smallest dimension. However, with small pieces used for ship modeling, a clothes iron, electric plank bender, soldering iron, or an electric hair drier or heat gun are all far more efficient and far less messy ways of heating wood in order to bend it.

If "the hull at the joint sites are curved both along its length and width," it's no surprise you're having problems bending wood to fit. What you're describing is like trying to teach a fish to ride a bicycle. It's not in wood's repertoire to bend in two directions simultaneously, AKA "compound curves," isn't something God made wood wanting to do naturally. The required solution is to shape the piece to create the curves, rather than trying to bend it. To get the flat of a plank to lay tightly against a curve, the faying surface on the plank must be shaped by "backing out" if a concave curved plank shape is required, or the reverse if a convex curved plank shape is required. Cut your wale a bit over size as required and plane to fit. Trying to incorporate a scarf in the middle of it all is truly crazymaking. Use a solid piece of wood and forget the scarf. If you feel the need to show the scarf (something that would never have been intentionally visible on the real ship,) then draw it in with a sharp hard-leaded pencil and straight edge after it has been shaped.

See:

 

[AndyA]

Use a solid piece of wood and forget the scarf. If you feel the need to show the scarf (something that would never have been intentionally visible on the real ship,) then draw it in with a sharp hard-leaded pencil and straight edge after it has been shaped.

I agree, but this brings us back to a question that keeps popping up in modeling: Is the modeler's goal to recreate the construction of the prototype or to make an accurate visual representation? Further, what would an accurate visual representation show? As Cap'n Cleek suggests, would one be able to see the joint on the prototype except from a couple of feet away? Each modeler must choose, Fair winds!

 

[rtwpsom2]

I was thinking about this myself for an upcoming project and I had this idea. It's only an idea and it's untested, so take it with a grain of salt. Make a cardboard template of the entire wale that fits the model precisely. Draw scarfs in where you want them to be then lay it flat and cut your wood strips to be just a little wider. Then cut your scarfs to match those drawn on the template. Glue it all up and let it dry thoroughly. Then trim back the edges to match the template. You should be able to then just bend the wale like a normal strip and glue it up to the model like one. The one thing I can't be sure of is if the strength of the glue bond will withstand the bending. Maybe I should do some tests.

 

[AndyA]

I was thinking about this myself for an upcoming project and I had this idea. It's only an idea and it's untested, so take it with a grain of salt. Make a cardboard template of the entire wale that fits the model precisely. Draw scarfs in where you want them to be then lay it flat and cut your wood strips to be just a little wider. Then cut your scarfs to match those drawn on the template. Glue it all up and let it dry thoroughly. Then trim back the edges to match the template. You should be able to then just bend the wale like a normal strip and glue it up to the model like one. The one thing I can't be sure of is if the strength of the glue bond will withstand the bending. Maybe I should do some tests.

Interesting. To recreate a prototype construction technique (scarfing planks in place), you have proposed a different construction technique (scarfing stock before shaping the plank). That doesn't exactly meet the criteria of recreating prototype construction. It is very, very difficult to do that perfectly. For example, most modelers use glue in lieu of bolts, spikes, and treenails. Fair winds!

 

[AllanKP69]

It is heat that softens lignin and causes wood to bend easily when heated and to hold that bent shape when cooled. Not water! Not amonia! Not steam! Heat. Just plain dry heat. I really don't know why this message just doesn't compute with so many modelers.

Another take on this. If the piece is not too thick I can edge bend a piece around a forming jig that allows me to bend it and prevent lift from the edge bending. I can then heat it while dry and that works, but the thickness versus breadth of a plank for instance gives limitations in my experience. I find water works much better, for me at least, for some species such as castello, Swiss pear, and holly. I soak the piece for minutes to hours depending on the thickness of the piece. I can then form it to the shape I want with the home jig fixtures. I can either heat it with a hot air gun or let it dry overnight with the same results. I have no idea of the chemistry of lignin and such but I do know that a piece of wood that I saturate with water bends a LOT easier than a dry piece. From a quick search: lignin softens when wet as moisture acts as a plasticizer, lowering its glass transition temperature (Tg). Dry lignin requires high temperatures to soften, but the presence of water enables it to soften at a much lower temperature, making it pliable and allowing it to be deformed for applications like wood bending and hot-pressing paper. If there is any doubt, take two identical pieces and see the maximum edge bend on the dry piece and then compare to the max bend on one that is thoroughly soaked and see which is easier.
Allan

 

[Bob Cleek]

I have no idea of the chemistry of lignin and such but I do know that a piece of wood that I saturate with water bends a LOT easier than a dry piece. From a quick search: lignin softens when wet as moisture acts as a plasticizer, lowering its glass transition temperature (Tg). Dry lignin requires high temperatures to soften, but the presence of water enables it to soften at a much lower temperature, making it pliable and allowing it to be deformed for applications like wood bending and hot-pressing paper.

However, dry lignin does not require water to soften.

"... lignin softens when wet as moisture acts as a plasticizer, lowering its glass transition temperature (Tg). Dry lignin requires high temperatures to soften, but the presence of water enables it to soften at a much lower temperature" was a new one on me, so I had to do some research. Chemistry was never my long suit. I learned this:

The glass–liquid transition, or glass transition, is the gradual and reversible transition in amorphous materials (or in amorphous regions within semicrystalline materials) from a hard and relatively brittle "glassy" state into a viscous or rubbery state as the temperature is increased. An amorphous solid that exhibits a glass transition is called a glass. The reverse transition, achieved by supercooling a viscous liquid into the glass state, is called vitrification. https://en.wikipedia.org/wiki/Glass_transition
, that explains the confusion. Lignin is, for the purposes of this discussion, a "glass-like" material in so far as it softens and ultimately liquifies as it is progressively heated and it hardens again when it cools. According to published laboratory studies, when water is absorbed or adsorbed by lignin, the softening temperature of lignin is lowered, in one study from 332.6 degrees Fahrenheit to as low as 129.2 degrees Fahrenheit. https://bioresources.cnr.ncsu.edu/r...itions-in-lignin-hemicellulose-and-cellulose/ (The boiling point of water is 212 degrees Fahrenheit.)

It's always been recognized by boatbuilders that the favored bending stock is "green" wood, i.e. wood that has not been seasoned and has a relatively high moisture content. (Hence the market designation, "bending oak.") The high moisture content has long been recognized as serving to accelerate the transfer of heat throughout the timber. (The Rule of Thumb for heat transfer in a steam box is an hour for every inch of thickness of the piece to be bent.) Heat transfers far faster through moisture than through dry cellulose. This fact is relevant to how long a piece must be heated for the lignin that is holding its celulose fibers together to be sufficiently softened that the fibers will not tear when bent, but rather "slide" past one another such that the tensions created by the bending can be redistributed evenly throughout the piece. Think of the piece to be bent as if it were a bundle of long paper soda straws held together by hot-melt glue. If the bundle is bent "cold," the straws on the outside of the bend will tear and the straws on the inside of the bend will crush. When the hot-melt glue is heated to the point it becomes soft, the bundle of straws can be bent without the outside straws tearing and the inside straws crushing because the tension between the inside and outside straws is released when the glue holding them togethe softens so they can move independently of one another.

Scientists, it seems, have discovered that not only does moisture transfer heat more efficiently through the cellulose-lignin matrix, but it does actually also contribute to lowering the temperature at which the lignin begins to soften. But, for the woodworker, there's a catch. In order for the piece to be bent without breaking and to stay as bent when cooled, like the bundle of soda straws, the wood has to be heated all the way through. Simply heating the outside edges, while it may to some small degree make it easier to bend, is insufficient to evenly distribute the stresses, which results in "springback" when the wood dries, if not breakage upon bending. Just placing a stick in water, even if boiling, isn't enough to raise the overall moisture content of the piece. The moisture content must be increased and reach an equilibrium for both the effect of the moisture's ability to increase the transfer of sufficient heat evenly throughout the piece and for the moisture to have the effect of reducing the temperature at which the lignin softens.

In summary, while the moisture content will make it easier to distribute heat through the workpiece to be bent, and, according to the laboratory tests, will lower the amount of heat required to soften the lignin, the ultimate prerequisite for bending success is sufficient heat, not moisture content.

Now, the moisture content of freshly cut wood can be as high as 150% of the wood by weight. Dried wood suitable for modeling has a moisture content of between six and eight percent. The optimum moisture content of wood used for heat-bending is between fifteen and twenty percent. Most all commercially sourced modeling wood has been kiln dried. Kiln drying, basically oven drying, is fast and economical, and fine for most purposes, but kiln drying changes the chemical properties of the wood, making it stiffer and more prone to cracking when bent, as well as impairing the wood's ability to absorb sufficient moisture to regain a moisture content sufficient for optimum heat-bending. Therefore, we must conclude that no wood sold for ship modeling is optimally suitable for heat-bending to begin with. Moreover, the fact that it is all kiln dried precludes the efficacy of wetting it to increase its moisture content to anywhere near the fifteen or twenty percent that is optimal for heat-bending. In other words, in practice, there's no point in wetting it. In terms of wetting, there is also the problem of thorough saturation. Apply water to the flat side of a piece of wood and then look at the end grain to see how far the water penetrates into the wood! Not much, if any.

This brings us back to the point: One can wet their modeling wood before bending it if it makes them feel good, and if the wood is soaked long enough, maybe it may have a small positive effect, although to my way of thinking, what is benefits are perceived may well be in whole or in part just a "placebo effect." (Granted, though, that very small-dimensioned pieces may "soak" better than larger ones.) What it comes down to is if one wants to bend wood and have it stay put, they must heat it sufficiently throughout and hold it in that bent shape until it cools, and there's not much one can do with water that is going to make much practical difference. What we do know, however, is that regardless of whether its kiln dried, or whatever its moisture content may be, with enough heat, lignin will soften and when it does, the wood can be successfully bent and, when cooled, and it will hold the shape formed when hot. While I won't argue with other's contrary anecdotal evidence, I will continue to maintain that water alone will not soften lignin sufficiently to have any effect on heat-bending.

All other things being equal, if the wood won't bend well, it needs more heat, not more water. In the wider universe of ship modeling, it must also be recognized that not all woods are equal when it comes to bending. Some can be heat-bent easily and dependably. Others are so-so. And others, basswood being a prime example, are close to impossible to heat-bend at all.

Notwithstanding, if anybody has a way that works for them, by all means, they should feel free to use it!

 

[Bob Cleek]

I agree, but this brings us back to a question that keeps popping up in modeling: Is the modeler's goal to recreate the construction of the prototype or to make an accurate visual representation? Further, what would an accurate visual representation show? As Cap'n Cleek suggests, would one be able to see the joint on the prototype except from a couple of feet away? Each modeler must choose, Fair winds!

Accurately recreating period scale scarfs can be a wonderous thing to do, and I wouldn't dissuade anyone so disposed from attempting it, except that there are some things that simply don't translate practically in scale and the mechanics of some scarfs can sometimes be one of these. The mechanical strengths and properties of wooden parts aren't linear in relation to scaling. In such instances, you've got to "fake it to make it."

 

[Bob Cleek]

Interesting. To recreate a prototype construction technique (scarfing planks in place), you have proposed a different construction technique (scarfing stock before shaping the plank). That doesn't exactly meet the criteria of recreating prototype construction. It is very, very difficult to do that perfectly. For example, most modelers use glue in lieu of bolts, spikes, and treenails. Fair winds!

The short answer is that it is sometimes, if not always, not possible to "recreate a prototype construction technique" in a scale model. Moreover, although building a model that is finished bright to show every construction detail is a most impressive undertaking, the fact is that nobody knows how a particular period vessel was actually constructed in every detail because we have virtually no records of period ships which record such details. They were left to the shipwrights to determine as they went along. The positioning of a particular scarf depended more upon the length of the timber available at the moment than anything else.

 

[DAWarden-(Jaager)]

A question:

Would water migrate into the center of a piece of wood any faster after it has seasoned (reached an equilibrium with atmosphere water concentration -humidity) than it originally took to migrate out? Once in, does the water migrate back out any faster than it did during seasoning?

I bet a hyperbaric chamber might drive water in faster. And perhaps a vac chamber could speed extraction.
A pressure cooker would do it - but it would also probably cook the wood and change its molecular structure partially destroying it.

Ammonia does solubilize lignin. But it is not the form of ammonia that a civilian has access to. It has to be pure ammonia - anhydrous ammonia. A gas at room temp. An industrial chemical that was also used for commercial refrigeration. There were catastrophic explosions and fires. It is also a deadly poison. The household cleaner ammonia solution just damages the surface of wood.