Re-Sizing a Full-Size Template

I put this together off-forum a few weeks ago for SurferO using a Lis Fish (Kinstle, 1977) to illustrate.  Perhaps a few members/visitors might be interested.  Let us see what happens.

To re-size a full-size template, first divide desired length or width by original template length or width, respectively.  

[Dimensions shown in all figures below are those of the original figure (Kinstle, 1977) and have not been re-labeled to show the increases or decreases discussed.]

 

In this example, the objective is to increase template/surfboard width from 21" to 23".  Divide 23" by 21".   This gives you 1.095.  The measured widths of the Lis Fish template below (Figure 1) are at 12" intervals.  Now multiply each of these widths by 1.095 and you have the new widths for the same 12" intervals.  To decrease template width from 21" to 19", divide 19" by 21".  This gives you 0.905.  Now multiply the original widths by 0.905 and you have the new narrower widths for each of the original 12" intervals.  For either, plot and connect these new width points with a smooth, continuous curve.  You could use the old template or a flexible strip/batten to draw the curved outline.

Fig 1.  Lis Fish template: original width (21 inches), wider (23 inches) and narrower  (19 inches).  All three are the same length.

To increase template length from 65" to 74", divide the new length (74”) by the original template length (65"), 74/65 = 1.138.   The original template widths will now be placed at 13.66" intervals (1.138 x 12") instead of the original 12" intervals.  To decrease length to 56", divide the new length by the original length (56/65 = 0.862).  Now place the old template widths at 10.34" intervals (0.862 x 12") instead of the original 12" intervals.  Again, plot the points and connect them with a smooth, continuous curve (Figure 2).

Fig 2.  Lis Fish template: original length (65-inch), longer (74-inch) and shorter (56-inch).  All three are the same width.

To improve nose and tail shape resolution, you can measure template widths at 1-inch intervals for several inches (Figure 3) – starting measurements from the template tips (nose and tail).  Similarly, rather than using 12-inch intervals to measure the original template widths, you could use 2- to 3-inch intervals.  Whatever length interval suits you.  The closer together the width measurements are, the better the shape resolution will be overall.  These new intervals will be increased or reduced to increase or decrease template dimensions using the same method described above.

Figure 3.  Steve Lis Fish

Kinstle, J.F.  1977.  Surfboard Design and Construction, p. 61.  Published by Natural High Express Publishing, Long Beach, CA.

Makes perfect sense to me.  The modified diagrams look like everything works.  A far cry from one of the local shop rats who once told me, “DUDE… if you go longer, you gotta go narrower” or some such crap when I asked him about enlarging a shape proportionally.

Copy template onto blueprint paper.  Take paper to local blueprint shop.  Have them enlarge the sheet by your desired percentage.

Many ways to do it Mako.  I will assume blueprint shops can enlarge in a single dimension (x or y axis).

I live in a rural town., population 6,250.  Cannot buy blueprint paper in a 30-50 mile radius.  We have no blueprint shop.  We have a super WalMart but no office supply store.  But the local newspaper can do a little printing.  (I am being serious).

I frequently use computers to stretch and/or widen diagrams and graphics for quick comparisons.  Allows me to see/view the many possible combinations of length and width for any given template/shape.  Also facilitates effective communication of larger concepts, using figures and the written word.  I retired last year and still have limited access to a banner printer at a university research complex.  But I still draw templates by hand (using basic math).

This is an easy method that can be done with “simple” math – by hand – using a framing square, a pencil and a batten or a large curve.  It can be done on the floor of a porch, kitchen or garage.  You do not need a calculator to do the math.  

One good template can be used to generate many different combinations of length and width.  By varying one dimension at a time, it is fairly easy to evaluate the effect that changing one variable has on a shape’s performance.

 

Move to San Diego. Borrow old school fish from friend /co worker or next door neighbor… Make sure the fish you borrow is a top name brand. Rip a template…you’re not the first guy to do it… forget about the math…now you have a real template made from a top name brand…ha ha…I own a REAL fish template… …and…

Interesting use of basic math in a simple approach to tweaking dimensions on a board. This comes up periodically, so it’s useful to know.

Old news but worth repeating. Master shaper Greg Loehr taught me that for every inch in length adjusted you should adjust width by 1/8 inch. You otherwise end up with a poorly proportioned shape when you just enlarge or shrink based on percentage. You can usually adjust up to 6 inches either way using your base template. 

All the best

The ‘1/8" width for every 1" length’ rule-of-thumb helps to maintain the same outline curve or “proportion”. Notice the difference in outline curves when only width or length is change. The ‘1/8" rule’ will give you a width target, when desired length is known, but it doesn’t define the curve between the nose and the wide-point, or the tail and the wide-point. Neither answer is complete. In the end, it usually comes down to an approximation, derived from either a judicious application of an existing template, or a keen eye and a sharp pencil.

Another consideration when enlarging an existing shape is how to extrapolate rocker. Aku adds 1/32"-1/16" at both ends for every 1" of length added. Thickness flow or what some call “foil” is also to be considered. Aku will automatically change rocker, thickness, both or neither depending upon the designer’s choice.  

When hand-shaping in the bay, I have a general idea of where I want the “adjusted” end-point rocker to fall as I increase length, and I just “eye-ball it” from there. Likewise, with thickness I have a 12" up-from-tail  and 12" down-from-nose number I want to match up with “thickest point” number. Using those number from the exisitng board works well enough, unless I decide to increase thickness. In that case depending upon the new boards length and the amount of thikness added (or subtracted) I might adjust the end numbers a little. Its a ballpark reference, and here again, how it “looks” determines the final number.

Yep. Well said. All approximation. In the end you have to please the eye. 

All the best

Somebody might specifically want a 5’10" Takayama Scorpion shape that is 23" or 19.25" wide rather than 21.25" wide.  The proportional math method  maintains the basic shape when varying one dimension while the other dimension (width or length) remains fixed.  Curve “steepness/flatness” (slope) of the planshape outline is definitely altered by changing one dimension while the other remains fixed.  But the curve’s slope changes are still in the same position relative to the original shape/template.  It is a good way to see the effect that varying one dimension has on performance, while the other dimensions remain fixed.

“If” you want to maintain the same proportions for the rocker curve – tangent angles relative to given points; curve slope (steepness/flatness) – you have to multiply the rocker heights by the same proportion that the length changes.  Otherwise, the rocker curve slope becomes steeper or flatter; that is, bottom angle relative to the water surface changes and is different.  But, maintaining the same rocker curve slope may not be well suited for the new length.

““If” you want to maintain the same rocker curve – tangent angles relative to given points; curve slope (steepness/flatness) – you have to multiply the rocker heights by the same proportion the the length changes.  Otherwise the rocker curve slope become steeper or flatter.  But, maintaining the same rocker curve shape may not be well suited for the new length.”

Absolutely. 

Looks like you were responding John while I was editing.  The content/meaning was not changed – just an added bit for clarification.

Sorry about that…

For a directly proportional re-size, the new dimensions must vary precisely.  The aspect ratio must be locked.  Width changes by the same proportion as length.  With aspect ratio locked, a 20" x 72" shape that is re-sized to be directly proportional increases to 21.66" x 78" (figures below).  This represents a 0.277" increase of the maximum width for every 1.0" increase in length.  

However, length and width can be re-sized independently of one another to create a desired combination of length and width (aspect ratio unlocked).  If you plan to change both width and length for the new template, it is easier to calculate and plot the new length intervals first.

 

 

EDIT:  This post was re-worded to communicate the concepts intended.  The original post was poorly worded and incorrect.

All good.  Useful info and insights from all, lots of ways to manipulate shapes up, down, sideways, etc etc.  These are tools we should have in our computer, the one between our ears, to pull up when we need them.  Good thread.

When I worked as a ghostshaper, pre-CNC days, we all had multiple tempalates of each model.

1/4" of centre width per every 2" in length. So Greg’s 1/8" rule is viable.

Since I handshape all my boards, I route the outlines.

Have several templates for each model.

Cut them on bandsaw and clean 'em up with my beltsander.

10-15 min. to make 'em perfect.

Same math works.

As the length increases the nose and tail widen as well. Just not the solid 1/4" at centre.

Same outline, just scaled up or down.

Works perfect.

That’s how you end up with 200+ outline templates.

 

Barry, when you are making a new template (bigger/smaller) for the same model do you just use the template that you already have to connect the dots and create the curve of the new template?

No way you all sck. whet i dew is i git graf papera jest make bigger squates. so one little square equals a inch. i did my fis my 5.0 fish n skaled it up to a7.0fish. itendedup being7.0 bye 28 wide. itseuf jest fine for me.

…math works at a some point for this application; it is like BarrySnyder says.

-a fish should be less than 6; if not is not a fish, does not have that “fishness” in the design; would be only a big floaty board with a swallow tail. I remember a Rusty model long time ago; a dog.

Connect the dots

or do the math…

My password/pin is pi. (**π **the number)

3.14159265358979323846264338327950288419716939937510…