Bench Construction

I've started work the the construction of a 4 x 1 m bench for building the long board. Good fun, although the challenge is going to be to get the top of the bench very level; particularly given that the floor of the garage is not flat. However, I think this will be achievable by adding an appropriate number of cross members to the bench in order to keep things straight.

The image above shows my sketch of the bench- truly back of the envelope stuff! Also shown on the left and below is the basic starting point with the materials and the construction of one side of the bench.

One another matter, I've been trying to source some Phantom 380 parts- a centreboard and a mast track. Turns out that such parts are extremely difficult to obtain in Australia- certainly if the lead time is short. My next option is to try the Net. I'll also be visiting the south of France in the near future- perhaps this may represent an opportunity to order such items.

Merry Christmas!

Rocker and Rise - 3D modelling

The plot below shows a side-on view of the longboard design and is based on the rocker and rise as  discussed in previous posts. Recall that this represents the total thickness and does not currently include any rounding and/or contouring of the top of the board.

As shown the design is not quite right with a small reduction in the thickness of the board apparent at about three quarters along from the stern. In this regard, what I have realised is that the top of the board should be flat along the centreline for a distance which covers the length of the mast track and the front of the centreboard (duh!). I'll build this into the design and will have another go at building up a 3-D model of the design over the next few days.

On another topic, the garage clearing has gone well and I'll shortly begin building the board construction table. I also have some news regarding a rig for the board and will likely discuss this in the next post.
Bye for now.

Rocker and Rise

Just a short post to log the progress made over the last few days. As discussed in a previous post, the intent is to marry a plan shape and rocker based on the Pan Am with a rise and rail system based on a Phantom 380. So a chunkier looking Pan Am shape. I think this will suit my lighter weight and will also have the aesthetics and high wind performance of a slightly narrower board.

I've been scaling the Phantom 380 rise (rail heights) from one of the images shown in an earlier post and have plotted it here added on to the model of the Phantom 380 rocker, also shown in an earlier post. Because I've scaled the rise by hand from an image, my estimates are a little noisy and so I've done a least squares fit of the data using a 4th order polynomial to provide a smoother representation. Just need to go and check some of the numbers now against a real board to ensure that they look reasonable.

The rise shown here should be the total thickness of the board prior to any rounding at the top of the rails. As such, the functions used to generate this image would then be used to construct a couple of templates which would be pegged to the side of an uncut polystyrene block (see image on the right from Emsworth Custom Boards) which would in turn be hot-wire cut to the template. Of course the functions describing the rise would also have to first be adjusted to account for the thickness of the high density foam outer layer.

Cheers

Look Ma, Multi-tasking!

It can be seen from the posts below that the board design is progressing reasonably well. Given this, a secondary task is now to set up a board construction area. Our house garage has been allocated as this site, however, given that it is currently crammed from floor to ceiling with junk, the site preparation is a bit of a challenge.
 

Once the garage has been re-arranged (to make a space roughly 4 m x 2 m for board construction), it will be time to put together a work bench.

Bye for now.

Schematic of Board Shape

Next step is to build in the rise. As noted in the previous post, the intent is to add in some nice chunky rails.

Board Shapes and Aspect Ratios

This last month I've been thinking a little about the best board shape (plan as against rocker discussed in previous posts) to adopt for this longboard construction process. It's been an interesting process because it comes on the back of the recent (well within the last few years) replacement of the IMCO with the Neil Pride RS-X as the board of choice for the Olympics. So we have gone from a traditional longboard with a good all-round performance to a hybrid which bridges the gap between the longboard and the Formula class boards. This transition has been nicely summarised on LBWS.

So I've tried to be open minded and have consider these different designs. In the end I've decided to stick with the longboard design simply because it is a good all-round performer, and because aethestically it looks more attractive, certainly compared to the somewhat stubby looking formula board designs; and also because that is what I raced with much enjoyment in the '90s. Also, the hybrid board reminds me of my old Tiga 320 (a somewhat pancaked version), where the only reason for having a centreboard is to make the thing track in a straight line in light wind conditions.

I've also decided to stick with the traditional surf-board type of design rather than look to an approach which might yield improved performance in light wind displacement conditions. Such hybrid displacement-planning boards currently appears to force board shapes, which in the extreme, look like the Exocet Open 310 which doesn't sit well with my concept of a longboard shape. Also that shape must be a real killer downwind in a chop in 15 knts + .


Given this background, it now remains to finalise the overall board shape. To help with this, I've been plotting up normalised length to width profiles of various longboard shapes in order to get a better understanding of how the board shapes are evolving. For example, shown below are the board shapes of the Phantom 380 and the 'Equipe-III'.


A plot of the normalised length to width ratio for the two boards is given below and illustrates that the two boards have very similar shapes, with a maximum normalised width of 0.2 occurring at about half the distance from the tail to the nose. The plot also demonstrates that the Phantom 380 has a marginally wider tail and nose compared to the Equipe-III.

I've generated these normalised aspect ratio plots for the IMCO, Pan-Am (thanks Richard for the images), the 380 and Equipe-III (as shown above), and for the Tiga 320 and have posted these below. These plots have helped me to make a decision regarding a good overall board shape to use with the longboard construction project.

Following on from these comments let's compare the 380 and the IMCO. In so doing we are essentially comparing the old guard with the new kid on the block. It can be seen that the Phantom 380 is significantly wider- being up to 30% wider towards the tail and 20% wider towards the nose and 12% wider at the point of maximum width.


Now lets include the data on the Mistral Equipe-II Pan Am. This board is one of the last of the previous generation of longboards and probably represents the top of the evolutionary tree within that generation of boards. Looking at the aspect ratio plot below it can be seen that the shape of the Pan Am board lies between the Phantom 380 and the IMCO towards the nose and tail of the board, and matches the aspect ratio of the IMCO at the point of maximum width.


Finally, and for completeness, shown below is a plan view and the normalised aspect ratio plot of my old faithful Tiga. You can see that the Tiga is quite a bit smaller than the 380, although it has a larger width to length ratio at the point of maximum width when compared to the 380. I guess this is a consequence of having to maintain a reasonable board width simply from a point of view of board handling and board volume. Formula- and hybrid boards take this requirement to a much greater extreme and presumably would have maximum aspect ratios which are considerably greater than 0.2.

Following an examination of the information shown above, plus a review of comments on the boards shapes available from the web, I've decided to base the longboard design on a slightly bulked up version of the Mistral Pan Am. I like the shape of this board, as do others who have or would like to obtain a Pan Am. Also there is that whole intangible nostalgia thing about racing against this board in the '90s. This class of board also recently won the world longboard championships.

As it stands, the Pan Am has a length of 375 cm vs. the 380 cm of the Phantom. The Pan Am board width is 66 cm vs. 70 cm for the Phantom; and the Pan Am volume is 258 litres vs 295 litres of the Phantom. So the intent is to maintain the Pan Am shape, but to increase the length to 380 cm. The volume will be increased 1/ through the increase in length and, 2/ through adding boxier rails and greater 'rise' as per the Phantom 380. And of course the kick arse Pan Am rocker will be maintained.

See ya.

IMCO image:
(http://www.albaria.com/mistral/2005/onedesign_2005/onedesign2005.htm)
Exocet Open 310 image:
http://www.boardseekermag.com/special_features/olymic_board_2/olympic_board2.htm

Where is the plan Luke?

Now that an adaptable rocker design has been formulated, it is time to work through the plan shape. This is the fun bit because it is here that you can make the board look really good (or really bad!). Also there is quite a bit of information available on the web- certainly more so that in the case of the longboard rocker. Note that some discussion of this was given in an earlier posting which can now be extended with additional information.

A really good starting point is the discussion on racing board design at LBWS and the reference in there to the Online Windsurfing Mag feature on Olympic board selection. Additionally, check out the board comparison chart here and have a look at the discussion about current and historical board designs on UK Raceboard (select the 'Choose a Raceboard' link).

Just for the Record

Apologies for this stuff- gotta store it somewhere.

Rocker mathematics.......

The fat an' thin of it

Hi Blogosphere- I thought it would be fun to periodically show where the longboard construction project is at. So here we go- finally figured out the rocker. Next stop the horizontal form. Here the choice appears to be- Old Style: slim and elegant; New Style: fat (and early planing). hmm

Longboard Rocker and the Universal Numbers

I have compared the rockers of four different longboard designs of varying vintages (Phantom 380, IMCO, Tiga 330 and [an old] Wayler). Guess what? All of the rockers can be described by a single set of simple mathematical functions which can be related to the board characteristics for sailing upwind with the sailor in the forward footstraps and the mast step fully forward; and planning downwind with the sailor in the back footstraps with the mast track fully back. The functions are fully consistent with the discussion in an early post regarding the need for a three-part rocker to correctly transition the board handling characteristics between these two extremes of upwind and downwind sailing.

The following figure (thanks German Windsurfing Association for all of the photos in this posting) shows the typical distribution of the centre of mass and centre of effort for upwind sailing in approximately 12 knots of wind with the board and rig powered up.


As you would expect, the centre of effort is approximately located at the point of maximum chord on the centreboard and this will line up with the point of maximum depth in the sail (which probably lies about 30% back from the mast. The centre of effort in the sail and the centreboard have to line up in order for the sailboard to track in a straight line (ignoring leeward drift) and not round up to windward or bear away to leeward. Note that the centre of effort lies about 40% forward from the tail of the board on the Phantom 380 (for a fully rotated centreboard).

Now I would expect that the centre of mass would lie forward of the centre of effort under these conditions. Visualise yourself being on the water in 12 knts of wind. Everything is fully powered up, and if its not too rough then you are sailing with legs and hands close together as shown in the figure above (sans the hand). However, in order to balance the weight of the wind on the sail you are not only leading outward, but also forward (at least that is what I do). Under these conditions, the centre of mass is probably sitting about mid-way along the board (50%).

Lets relate this back to the rocker. Under these conditions, it would be logical to locate the point of maximum rocker under the centre of mass because this is the point about which the board will pivot in a fore and aft position in response to the chop when sailing to windward. So a schematic diagram of how this looks is as follows.

What you want in this situation is a rocker shape that is locally symmetric about this pivot point, so that the board rocks easily backwards and forwards without 'slapping' and generating extra drag in the chop.

So this is the first, and probably the most important component of the 3-part rocker on a longboard.

Now lets consider the situation in which you are powered up and sailing downwind on a reach. You're on a longboard of course (what else!) and so the centreboard is fully retracted, the mast base is at the back of the mast track, you're hooked on to 9.5 m^2 of sail with the feet in the back straps and with the board skipping along on the first two metres of the tail, desperately hoping that the water flow around the fin stays laminar and doesn't detach and spin you out.

.So in this situation, the only component of the board which is relevant to the rocker shape is located within 1-2 metres of the tail (barring a catastrophic nose dive in which case it is the first metre at the nose of the board which is important!).

For powered-up reaching conditions, it would be expected that a pretty flat, to a slightly curved rocker shape would be optimal. Flat would be fastest under optimal conditions, however a small amount of curvature is probably required in order to aid with gybing, to provide a smooth fore-and-aft response to the chop and also to reduce drag when the board is sailing upwind. Care must also be taken to ensure that the tail rocker transitions smoothly into the mid-point rocker as described above.

The front part of the rocker provides a transition between the mid-point rocker and that safety valve- the nose rocker; and also generates lift for planning under marginal conditions. Here the aim is to minimise the rocker as much as practical in order to maximise the waterline for windward sailing (particularly when railing). However, eventually the rocker has to sweep up to the bow and hence provide some protection against the nose dive.

All of this information can be represented in the following schematic diagram.


Our mid-point rocker is represented by a curve which goes from x1 to x2 and has the deepest point at xc. Our tail rocker goes from x0 to x1 and starts with a value of r0 and a slope of r0'. Finally our nose rocker goes from x2 to x3 and finishes with a value of r3 and r3' at the nose of the board.

The curves that we use to describe the rockers have a quadratic shape for the tail and the mid-point rocker and a cubic shape for the more complicated nose rocker. The curves are constrained to match the local rocker and rocker gradient at the transition points.

So all we need to do to fully describe the rocker for our longboard is to prescribe the tail and nose rocker and the points of transition between the tail, centre and nose rocker. Now I've tested this model against a set of best estimates of the actual rocker for an IMCO (thank you for the pictures John), Phantom 380, Tiga 330 and a Wayler and the result of this testing is shown below.

So you can see from the images that the three part conceptual rocker model matches the actual rocker very well. What is particularly useful is that in all cases, the transition from the tail to the mid-point rocker occurs about 40% forward of the tail, the point of maximum rocker occurs about 50% forward of the tail and the transition from the mid-point rocker to the nose rocker occurs 60% forward of the tail.

The slope of the nose rocker varies between 20% (380) and 35% (IMCO; Tiga) and the slope of the tail rocker varies between 1% (IMCO), 5% (Tiga; 380) and 7.5% (Wayler). This seems consistent with the knowledge that the IMCO is pretty flat towards the tail (see an earlier post on this), and that the Wayler, being an older, displacement type board has the largest rocker in the tail. Note also that the 380 has the smallest nose rocker which may be an attempt to maximise the water line.

The result of this work is a simple and self consistent mathematical description of the longboard rocker which can be used for the design of a longboard. And the universal numbers? 40:50:60.

Bye for now.

Moving right along

Back again. A couple of good things have happened since I last posted. The first is that my dear partner has decided that she will support (tolerate actually) my longboard construction project. It took a little bit of convincing for this to happen (and hence the length of time since the last posting!), but at last we have came to an agreement (painting part of the house probably helped). The other useful thing is that one Malcolm Jones (of Melbourne too would you believe) posted a very nice step-by-step account of a similar construction project (thanks LBWS). Malcolm has, to my mind, solved a number of the engineering problems, which leaves me free to concentrate more on the design and aesthetics of the longboard.

Some encouragement has also come from a three part video released by Witchcraft Sailboards which shows how the experts design and construct a sailboard using state-of-the-art CAD design and shaping equipment. Very nice if you have the equipment. However for me its still gonna be the hand saw, the rasp, some templates and careful use of "the eye".

So back to the design process. In the last posting there was some discussion about what should constitue the correct rocker shape for a longboard, and following the scaling of some longboard photos using the "match new photo" option in Sketchup, a suitable composite profile was developed. However, I'm now having some doubts as to the accuracy of the profile. The problem is that none of the photos give a true side-on view of the boards, and so it is necessary to first correct the photos to remove perspective effects. This process of correction can lead to errors in the measurement of the rocker.

The magnitude of this error was tested using an old Tiga 330 which a friend has loaned to me for the summer. By directly measuring the rocker, and comparing it with the rocker derived from photos, it is possible to determine the magnitude of the error due to perspective effects in the photos. Shown below are three examples of how an image may look together with the an indication of how the rocker is scaled off the photos after correcting for perspective effects.

The rocker is the most difficult to scale and measure in the image on the left because the board does not line up with any straight edges in the image and also because the nose of the board is closer to the camera and so a larger correction has to be made for perspective effects. The middle image is better because the board is lined up with a straight edge and so the perspective correction can be more accurately determined. The right hand image is the best because the board has been photographed from a side on view and so the need to correct for perspective is minimised.

The magnitude of the errors in measuring the rocker is shown in the figure below, where the derived rocker is compared with the measured rocker for the three examples shown above.

You can see that the best match between the measured rocker and the derived rocker occurs for the right hand image where the perspective correction is small. So this means that I should go hunting for some side-on images of longboards in order to be confident that the rocker is correctly measured. Alternatively if any of you kind longboarders out there in the blogosphere are willing to take and email in a good photo, I'll be eternally grateful. In the interim, it's time to check out some plan views of the various longboards out there and to come up with a good overall boardshape.

Bye for now.