Tuesday, June 28, 2011

BoardOff In Action

Dean in WA had started to plan out his board project when BoardOff first came out and so he picked it up and has been using it from pretty early in the piece. Now his new board well underway and its looking great. Looks like a great workshop setup too with a heat box being the latest addition.

Dean's blog is at: http://downunderbloghr.blogspot.com

Dean sent me the details of his next board which is an amazing looking lightwind board with parabolic rails and cut away tips. Bring it on!!!!

Free cad software - A9CAD

I've  been spending a bit more time working with dxf files and came across a free cad program that seems pretty solid. Allows you to import DXF and DWG files, scale and edit and export as PDF.

To get a 1:1 scale print out of the dxf file you need to first export it as a pdf file on a large format for that the 1:1 version fits on a single page (I use CutePDF or PDFCreator and select a large page size - both freeware programs that install as a printer and allows you to print any document from any program as a pdf). Then you need to get some program will chop up the oversized pdf into multiple A4's. Apparently Photoshop and Illustrator will do this as well as a number of other freeware programs though i haven't tracked any down yet that take a pdf and create multiple pdfs.

The freeware is called A9CAD and you can download it from tucows


 I'm sure there are other but this one seems to do what is needed.


Friday, June 24, 2011

More BoardOff Functions

I've been tweaking BoardOff some more and added a compositing tool so that it is easier to create multiple deck outlines or compare different outlines to examine the real differences closely. A few other features I've added are asymmetrical corners, convex tips and ability to the latest north-style waists on the upper deck and added a function that merges dxf files so that the multideck design can be exported into a single file for printing.

The updates haven't been uploaded yet but will do over the weekend. In the meantime here are some snapshots of the multi-deckboard designed on BoardOff show a host of the new controls over the outline.

Composited board template showing inverted corners and waist at mid-section.

Pickle-fork type top deck using convex tip option

 The mid-section 'waist' and the 'inverted corners' on the  top deck highlight the 4 sections of the outline that BoardOff lets you control: the tip, corner, mezz section and mid section (waist).

The pickle-fork top deck highlights a shortcoming of using cubic splines instead of higher order polynomials to create the curves. it was nearly impossible to avoid the sharp corner created on the mid line of the tips as the curve could not get vertical enough to have a smooth line there. In practice you probably would try to avoid this kink as it is likely to concentrate the stresses and be a candidate for cracking. One remedy for this short coming in BoardOff is to change from cubic splines to using Bexier curves which give you control over the slope of the lines at the end points rather than only the location of the end points.......

Sunday, June 12, 2011

DXF Export Function Added to BoardOff

Version 1.3.3 of BoardOff had the DXF file format export ( document exchange format used by many popular CAD programs such as AutoCAD) but the export function had not been wired up.

Last night I had a chance to implement the DXF export function for the Board Outline and it was remarkably smooth sailing. Here's a snapshot of the export function with annotations being viewed in SolidWorks DXF viewer.

This will go up on the web in version 1.3.4 after another few functions are refined.

Obituary for Board #2

I gave board #2 a punishing workout at both the cable park in Cairns and on the choppy water at Port Douglas and I am very sad to report that it was not up to the challenge and cracked under the footpad after an out of control landing though not from any great height.

Although this very clearly sucks a great deal, it does give a great opportunity that I would not have otherwise been happy to do - destructive testing on a board to see where the faults are so that I can continue to improve it.

Analysis of the failure

The board failed catastrophically right under my trailing foot heal after a crash landing. The crash was not so big that it could have been solely responsible for the fracture and so my guess is that the failure started earlier and perhaps at the cable park where I really loaded heavily for a couple of hours.

A few things are noticeable:

A) It was a compression failure - the crack appeared on the top of the board on landing when the top surface was subject to strong compression. There was a small crack on the underside of the board ( about 1cm starting at the rail) but I think this was probably a secondary effect of the failure of the top surface leading to the laminate being bent beyond some critical radius and breaking. The crack on the top extends half way across the board.

This is the most expected failure as laminate has very little strength in compression. The glass reinforcement adds nothing to the compressive strength, unlike tension where its very large E (c. 70 GPa) drastically improves the properties. In compression the resin bears all the compressive load and the mode of failure tends to be rupture in the laminate due to buckling and delamination. The strength then depends on the compressive strength of the resin and the strength of the adhesion to the core material. I suspect that the rupture location would be at some irregularity in the surface and difficult to predict.

B) The top laminate delaminated from rail - I think that this may be the most significant issue. When one surface is free to move relative to the other ( this is the case when the laminate is not bonded to the rails) a huge amount of strength is lost.

When the laminate is bonded to the rails the cross section of the board form a closed rectangular section. The vertical walls of laminate create 'I' beams on either side of the cross-section which add enormously to the strength because of the thickness squared contribution of material to the cores strength. When these sides are removed the top surface and the bottom surface can move much more freely relative to each other the strength of the board depends largely on the elastic modulus of the material which for foam cores is very small and its likely the the shear forces ( the force parallel to the surface of the board) are no longer negligible as its possible for the radius of curvature can get quite small.

Close inspection of the rail where it delaminated suggests a few things:

i) The rail material is quite smooth where its delaminated. This suggests that maybe I did not rough the surface up enough to get a good mechanical bond ( surface roughness creates 'grab handles' for the epoxy).

ii) The chemical bond was not good. I have read in many places that the bond between resin and ABS plastic is not good and hence the need for flaming the plastic first. Flaming it ( literally running it through flames and stopping short of it burning) oxidises the surface and so that the chemical composition of the plastic bonds better to the resin. I did heat treat the plastic in the course of bending it into shape. There was lots of heat applied to the plastic pieces that protect the tips but less so on along the sides where it probably needed it the most. Perhaps these side rail pieces need more flaming that i gave it.

In Board #1 I fully encapsulated the rails in resin by leaving a1-2mm layer of resin on the outside vertical part of the rail. I did this because i was concerned about delamination. Board #1 has taken a lot of punishment including hitting a rock at full speed on a 20 knot day. If the resin on its own is tough enough to do the job it raises a question mark over the value of using ABS in DIY boards. (in production boards I understand it speeds the process up).

iii) The rail material cracked about 10cm away from the heal point. This has got me a bit puzzled. I've tried to snap ABS plastic by bending it repeatedly to fatigue it and even after bending it in 2 and working it still did not break. Interesting the crack is on the top half of the rail which means that it happened after my heal went through the deck as this crack could only happen by the top half being under tension. To it to crack like this suggests that the bond with the laminate to the right of the crack is very good and so after the laminate to the left separated, the section to the right stayed fixed and so concentrated the stresses in the rail sufficiently to crack. However, this doesn't seem complete. To get this very ductile material to fail like that seems very strange. Either the material was also more brittle at that point for some reason or the dynamic behaviour of the material under impulse loads rather than slow bending is very different than expected.

The faulty bond with the rail seems to be where all the evidence points.

D) The top laminate cracked half the width of the board. I was surprised to see the laminate crack. I was under the impression that the resin may have failed under compression but that the strength of the glass would have meant that it would have remained intact. Obviously as the board began to fail the radius of curvature at the point where it buckled and delaminated it reach a critical radius that caused the glass to fail. In e-glass the strain to failure is 1.5% so it must have been that this limit was reached. 

E) I did not provide any extra reinforcement under the  footpad as I was thinking that the plastic plate under the foam footpad would have spread the load. Obviously not the case. Footpad reinforcement goes back on the must have list.

Having a broken board means that I can not run some tests on the board to see what the actual strength is as maybe use this to calibrate the BoardOff's flexmodel.......to be continued.

Wednesday, June 8, 2011

New rockerline idea

Today while standing on four mile beach waiting for wind I had a good chat with fellow Sydney winter escapist Yuko who rode his own boards for a lot of years until he turned his hand to catamarans and parallel programming the matrix of processors found in graphics cards!!!

He stopped making his own boards when he came across the Nobile NHP which seemed to tick all his boxes. I had a good chance to take a close up look at the board and there are a couple of things that are great food for thought.

One of the biggies is the rocker line. It's curved through just a portion of the middle section and then dead straight right through to the tip. Overall there is a good amount of rocker but this is attributable to the curve in the middle section and not the tips. This is the opposite of convention 3-stage rockers which are flatter through the middle section and curved through the tips. His contention is that conventional 3 stage rocker is inefficient because the shape of the line that runs through the water when you're edging is kinked and curved along the part that sits deepest in the water which means it doesn't plane efficiently and creates too much drag. Whereas with a short curved section followed by a long straight line the section cutting through the water is dominated by the straight line and so planes early.

If you try to visualise the shape of the board section being pushed through the water, the NHP line is probably a bit more like an inverted aerofoil with the maximum chord toward the direction you are travelling. With this setup you maximize the flat planning area at the back of the board which naturally sits deeper in the water whereas the conventional 3 stage would have the maximum flat planning area in the middle, much of which will be out of the water when edging. I guess this adds up to earlier planning.

Just thinking about the trade offs, I suspect that the board may not carve as well as a continuous rocker because of the flat section you're dragging through the water. However, maybe having more flexible tips so that when you do edge really hard the board will flex and follow the curve you carve along. The tip flex many also overcome the need to have bigger fins or channels in the rails to make them grippy which is seems to be what is needed with very flat boards (my theory here is that with flat boards, when you edge hard the water has to pass under the board and so lifting the edge up leading to loosing your edge. With boards with lots of rocker, the water can more easily pass around it and so you keep your edge in the water and increase the drag to help load up.)

The latest version of Board-Off Design tool that I uploaded today has a version of the rocker profile sales as RockerTemplate #1. I think I'll give this one a try in board #3.

Sunday, June 5, 2011

Musing about board #3

I've started thinking about board #3 and working out what the goal will be.

Without much wind around at the moment it's a bit hard to set performance improvement goals for the board and son I think for board #3 the goals will be around improving the production techniques. In particular speeding the proces up and improving the finish quality.

The three features I want to change are potential working with a full woodcore, adding channels and pouring the rails.

Sources of delays

i) Waiting for resin to cure for inserts, building up the core and setting the rails in place at a minimum this adds a day to the process and with my set up ended up adding several days as I did each one seperately. it would be ideal to be able to create a 'kit' of the insert pieces that could then be put in place at the same time as doing the lay up. Using ABS plastic disks for all inserts cut perfectly to size seems to be the best way to do it other than using 5 minute epoxy if cost permits. ABS washers about 15mm would do for the footstrap inserts if 6mm flange nuts are are being used and they would be more than enough for the fin inserts. Handle inserts....we say no to handles!!!!
ii) Laying up the top and the bottom in seperate passes. This was partly due to my concern that I  would not be fast enough to do both sides because I was using 3 layers of cloth each side and I wanted to really work on reducing air bubbles. Ideally this would all be done in one pass and possibly using 2x8 oz layers of glass rather than 3x6oz each side. The limitation there is it won't be possible to do one of the layers at 45/45 degrees to make sure it is torsional stiffness......
iii) Cutting and forming the rails was another slow process but the issue was more the potential for having a poor rail, core bond. Pouring either epoxy or polyurethane would address this but would introduce waiting time so the tradeoff is not great. Using ABS ( or PVC) rails and then being able to clamp the rails to the core at the same time as laminating would be ideal. So long as only one side board has 3-d shaping then this would be doable by using a template of, roughly, the negative of the board cut down the centre and then using it to clamp the rails agains the core ( brokite style).
iv) Doing a seperate bagging to get the smooth finish on the board. This added an overnight wait for the curing. I suspect it would be faster to finish it by hand.