Wow! Nice boards!
And yes, post results for everyone to see.
.
The amount of plastic debris that I have produced (in my attempt to avoid sanding back that first fin composite wood/fibre/epoxy laminate) can only be justified if it helps a lot of others to avoid the same mistakes.
.
So point out the mistakes, please.
Hi Mik, I got the fins today, the first thing my 14 year old son did was grab them and run away. We all dig the purple boomerang fin with the tubercles. They all feel good, look good, Have you thought o making and selling them ?
I’ll get the fins back from him tomorrow !
Haha, did he take them surfing or just run off?
The Gull-Whale fin may be a complete flop, I made up the foil rather than use the import function of finFoil. The foil is a sort of exaggerated Eppler 168 foil, with much more pronounced concaves aft. It would be a very tricky shape to sand manually. Put a ruler on it to see the concave.
He’s out surfing on it now before he goes to school so I suppose we will start with that fin.
The gull wing caught everyone’s attention, I’ve got older boys who immediately fought over it and not to detract from the other fins, there’s something about that fin that everyone likes. Even my wife had a good look at it and she doesn’t do too badly in the surf for a redhead.
We all noticed the special foil on it and the bit of tip flex. The overall finish and shape are great, it’s perfectly symmetrical and feels right. Holding it to the light gives an insight into the time and technical expertise you’ve put into it.
Hey Mik, had it out for the last two days with my sons and we like it. It’s a perfectly normal fin, ist got drive and pivot, the waves were only head high but it’s well suited to Cheynes board and we are giving it a blast over the weekend on a Longboard.
If you’re looking for anything wrong with it we couldn’t find it.
We saw the concave rear foil and maybe you might fill that in a bit but it didn’t detract from the performance at all.
The bad thing is that all of us want to own it now.
Yipeee! Not snapping off is a good start. The fin itself is easily changed if the UTFB is functional.
How is the fit in the box and how do you like the snap-in function?
I’ve already got one of those Japanese snap in Gull wing fins with the spring loaded ball bearings and yours is just the same. We snapped it in and it’s solid as.
We took all that extra stuff off the fin before we surfed it, so it’s just the fin in the box, no string , no swivel, no plate and screw.
Maybe 60 or 70 waves so far and It’s good Mik, it works.
One of the reasons that I got a bee in my bonnet about wanting to make my own fins was that I lost my only ‘Effect System’ Gullwing fin at the Pass in Byron, and then found out that they were wiped out by the tsunami in Japan and I could no longer buy a new one.
I hope the larger ball spring plungers in my version make the inserts (plastic cylinders of slightly different lengths) that the Effect System had unnecessary. The BSPs I use have a larger ball and can hopefully accommodate for a wider variety of fin box sizes than ‘smaller balled’ spring plungers.
Unfortunately I messed up and did not order more BSPs in time, now I have to wait a week or so to get new ones. I have also ordered a few that have plastic balls instead of stainless steel, maybe that is kinder to the plastic fin boxes. The larger BSPs should however cause less wear on the fin box than small balled BSPs, because the pressure is spread over a much wider area.
I will continue to print fins with 100% infill for now, even though none of the ones with solidly printed UTFB have ever snapped, yet.
It just takes too much time to adjust the 'Solidifying Micro-cylinders" to force the Slic3r software to print the UTFB solid.
.
At this stage of development, it saves ‘only’ about 11% plastic filament to print a solid UTFB fin (with 5% infill in the tip, like the ones Surffoil is testing at the moment), compared to a completely solid fin.
.
There is also the predominant ‘stiffer is better’ opinion about fins, and the likely infra-sound creation (felt through your feet rather than heard as a hum and mentioned by Thrailkill in another thread recently) by less stiff fins, due their lower resonance frequency, as pointed out by Hans recently.
.
It takes much longer to adapt hundreds of ‘Solidifying-micro-cylinders (SMCs)’ for a different fin shape, than to create a file that produces a completely solid print. I just spent about 3hrs just to transfer the latest UTFB changes (and rotation changes on the print bed) to the file for the smooth leading edge Gullwing fin (the smooth twin of the bumpy version that Surffoils is testing).
.
Eventually, if and when ride reports show that solid PLA fins do not snap for at least a year or two in actual use, I might put in the effort to reduce the amount of infill to the minimum required for mechanical strength. My guess is that 20% reduction is achievable with some effort.
.
An exception to this rule/plan are really fat fins. The fat portion just above the fin box does not contribute much to mechanical strength, and there is overall much more potential to integrate hollow spaces with fat fins. If positive buoyancy can be achieved, then its worth spending extra time developing partially hollow fins. As a rough guess, I think somewhere between 30-35mm max width (compared to 25mm in Finfoils fin) that point should be reached. It is also much less work to adapt the file for a wider fin, because the SMCs will not penetrate the outer shell. The penetrations occurring with slimmer fins are what causes most of the work to adapt the files, because it causes rough surface spots if the SMCs are not wholly inside the fin, and with sufficient distance to the external printing layers.
Printing the smooth leading edge version of Wanderfalke 2 fin:
WF2-SLEF2227HRx25x1.3mm_UTFB4070-hollow_X3DtrPLA0p2mm_20170728TLR.gcode
Some experimental changes to print settings;
- extruder fan speed reduced to 50% max (from 100% max). Except for bridges, i.e. the upper part of the round holes for the ball spring plungers, wher the fan speed ‘should’ remain at 100%. The intention is to reduce the creation of fairy fluff micro-PLA-particles.
Also using a different filament: X3D transparent PLA. Max nozle temp is 220C for this one, no idea why it is different from the ones I used previously (230C). I have save all the 220C max filament colours for last, and have largely run out of the hitherto used colours purple, blue and red.
Ooops, I just realised that I have not explained what I mean with ‘Solidifying micro cylinders’ . I was trying to write an explanation last week or so, but it took too long and I came up with the concept of printing solid fins instead.
.
So, “SMCs” are very small hollow tubes inside the CAD file. They get ‘differenced’ from the file to create a virtual hollow space.
However, because they are so small, there will not really be a hollow space in the 3D printed object. Rather, the Slic3r software will surround the SMCs with a certain number of perimeters to protect the hollow insides. With a radius of r=0.055mm, the diameter will be 0.11mm.
The default extrusion width is 0.45mm, meaning that the SMC diameter is only about 1/4 of the width of a single line of molten plastic as it gets deposited. In effect, the adjoining extrusions end up touching each other and no open space exists in the printed object. But, in the unsuccessful attempt of the printer to keep these micro cylinders open, solid material is deposited wherever SMCs are grouped together tight enough.
.
Code in OpenScad looks like this (but is much longer than this example):
union(){//solidifier cylinders for solid fin base adjusted for WG2-3362-BLEF fin and WG2-262_Winglets fin.
// use 1.5mm distance between centers
// use 0.1mm radius
// 20 units of 3x7 groups.
translate([42,0,0])
union(){// union of 7 x 3 micro-cylinders (25/25mm) length reduced to 20mm to not breach fin base
translate([-30,-25,-1])
rotate([-90,0,0])
union(){// group on starboard side
translate([-1.5,0,0]) cylinder (r=0.055,h=20);
translate([-3,0,0]) cylinder (r=0.055,h=20);
translate([-4.5,0,0]) cylinder (r=0.055,h=20);
translate([-6,0,0]) cylinder (r=0.055,h=20);
translate([-7.5,0,0]) cylinder (r=0.055,h=20);
translate([-9,0,0]) cylinder (r=0.055,h=20);
translate([-10.5,0,0]) cylinder (r=0.055,h=20);
}
//
translate([-30,-25,0])
rotate([-90,0,0])
union(){//group in centre, 14mm longer
translate([-1.5,0,0]) cylinder (r=0.055,h=20);
translate([-3,0,0]) cylinder (r=0.055,h=20);
translate([-4.5,0,0]) cylinder (r=0.055,h=20);
translate([-6,0,0]) cylinder (r=0.055,h=20);
translate([-7.5,0,0]) cylinder (r=0.055,h=20);
translate([-9,0,0]) cylinder (r=0.055,h=20);
translate([-10.5,0,0]) cylinder (r=0.055,h=20);
}
//
translate([-30,-25,1])
rotate([-90,0,0])
union(){// group on port side
translate([-1.5,0,0]) cylinder (r=0.055,h=20);
translate([-3,0,0]) cylinder (r=0.055,h=20);
translate([-4.5,0,0]) cylinder (r=0.055,h=20);
translate([-6,0,0]) cylinder (r=0.055,h=20);
translate([-7.5,0,0]) cylinder (r=0.055,h=20);
translate([-9,0,0]) cylinder (r=0.055,h=20);
translate([-10.5,0,0]) cylinder (r=0.055,h=20);
}
//
}
// End union of 7 x 3 micro-cylinders 1
//
translate([31.5,0,0])
union(){// union of 7 x 3 micro-cylinders (34/29mm) length reduced to 20mm to not breach fin base
translate([-30,-25,-1])
rotate([-90,0,0])
union(){// group on starboard side
translate([-1.5,0,0]) cylinder (r=0.055,h=20);
translate([-3,0,0]) cylinder (r=0.055,h=20);
translate([-4.5,0,0]) cylinder (r=0.055,h=20);
translate([-6,0,0]) cylinder (r=0.055,h=20);
translate([-7.5,0,0]) cylinder (r=0.055,h=20);
translate([-9,0,0]) cylinder (r=0.055,h=20);
translate([-10.5,0,0]) cylinder (r=0.055,h=20);
}
//
translate([-30,-25,0])
rotate([-90,0,0])
union(){//group in centre, 14mm longer
translate([-1.5,0,0]) cylinder (r=0.055,h=20);
translate([-3,0,0]) cylinder (r=0.055,h=20);
translate([-4.5,0,0]) cylinder (r=0.055,h=20);
translate([-6,0,0]) cylinder (r=0.055,h=20);
translate([-7.5,0,0]) cylinder (r=0.055,h=20);
translate([-9,0,0]) cylinder (r=0.055,h=20);
translate([-10.5,0,0]) cylinder (r=0.055,h=20);
}
//
translate([-30,-25,1])
rotate([-90,0,0])
union(){// group on port side
translate([-1.5,0,0]) cylinder (r=0.055,h=20);
translate([-3,0,0]) cylinder (r=0.055,h=20);
translate([-4.5,0,0]) cylinder (r=0.055,h=20);
translate([-6,0,0]) cylinder (r=0.055,h=20);
translate([-7.5,0,0]) cylinder (r=0.055,h=20);
translate([-9,0,0]) cylinder (r=0.055,h=20);
translate([-10.5,0,0]) cylinder (r=0.055,h=20);
Printing solid fins with maximum strength is achieved by basically two methods:
.
- Print fin standing on leading edge or trailing edge, so that layers of filament will be deposited in parallel along the lines of maximum bending force, not perpendicular to them. Just like you would orientate the wood grain if you want to shape a fin from a single solid piece of wood.
.
2): In order to achieve equal stability against bending towards port or starboard, and to get the longest straight lines of filament deposited in the fin, you rotate the angle of infill in the Slice3r software so that each second layer runs straight from the base to the fin tip. Each other layer of infill will of course run at 90degrees to the longitudinal layers. I have not figured out a way to change this, yet. And maybe that is not a good idea anyway. But, I think it would possibly be better to print 2 layers longitudinally, then one layer perpendicular, then again 2 layers longitudinally. That would increase the number of layers orientated in the optimal direction from 50% to 66.6%, while every longitudinal layer still has contact with one perpendicular layer to hold it together sideways.
Anyway, I’ll probably leave that one in the ‘too hard basket’ and just print alternating layers of 100% infill (Rectilinear).
The clear PLA print looks good so far. Now 16h30min and 83% finished.
I think I’ll try to increase the layer height from 0.2mm to 0.3mm for the next attempt, that should get it finished in about 2/3 of the print time, and possibly increase mechanical strength a bit further.
Only apparent problem in this print so far is that the widest part of the fin skillet is being printed messily. I hope it will be an easy fix by sanding after printing. The problem is caused by choosing the "Detect thin walls’ feature in the Slic3r software. It realises that the edge is thinner than 2 perimeters, and so attaches a single extrusion line at the thinnest part. This single line will not attach firmly, and as you can see in the pictures, some are even being extruded into thin air.
The setting ‘Detect thin walls’ is however required, because without it, the thin trailing edge of the foil would be printed messily.
The effect of ‘Detect thin walls’ on the aft thin part of the fin is caused by cramping too much extruded filament into several lines parallel to the trailing edge.
Without thin wall detection, some of the inner perimeters end in a point, which means that almost double as much filament ends up at that point. Note the pointy end of the inner perimeter of three perimeters (the red lines) vs. the blunt end of the same perimeter when ‘Detect thin walls’ was used for the slicing process.
The pictures show the centre of the fin at the level where the proximal and distal parts of the trailing edge unite for the first time.
‘Detect thin walls’ effect shown close up: The inner perimeter does not end in a point, but as a blunted end.
Without this, with each successive layer, the small area of over-extrusion is shifted by a fraction of a millimeter, and the end result is a raised line parallel to the trailing edge, but a few millimeters away from it. With some fins, several lines appear at different distances from the trailing edge.
These lines are messy to sand back, because it requires sanding the foiled surfaces, so they loose their shine.
.
The (hopefully) no longer needed carbon rods have found a new use as archiving system for failed and good prints. It helps immensely to be able to go back and check the prints. Of course, good records of print settings etc is also a must.
I threw out a lot of failed prints before I came up with this handy way of archiving them.
The easy way to make a fin…
.
Funny how after a lot of effort and learning the solution turns out to be easy!
.
I started my attempts to make a fin about 13 months ago.
.
I bought the 3D printer 10 months ago and started to learn a bit of OpenScad designing and to learn how the printer works.
.
So here it is in a nutshell (and then I’ll get on with my life!):
.
- Use the easiest print material: PLA. Use transparent varieties to allow non-destructive evaluation of the prints.
- Select all the ‘Quality (slower slicing)’ options in Slic3r, namely: Extra perimeters if needed; Ensure vertical shell thickness; Avoid crossing perimeters; Detect thin walls; Detect bridging parameters.
- Print external perimeters first, with seam position ‘Rear’.
- Print with 100% rectilinear infill, aligned at 0deg and 90deg to the longitudinal axis of the fin.
- Design a support system to allow to print the fin with trailing edge up or down, to increase strength (inter-layer adhesion is the weakest link in FFD printing).
- Print in an enclosure and keep it closed for the entire print. Even small temperature changes cause minute layer shifts that are very obvious on a fin.
- Use a silicone sock for the hotend. Otherwise, fine PLA debris will accumulate in molten form on the hotend, until it drips off and ruins your print.
- Change the Start-G-code and the End-G-code so that the extruder gets parked at z 200 before and after each print. This results in very consistent temperature of the PINDA probe during the pre-print calibration process, and the first layer turns out perfect every time. Poor first layer will ruin your fin print potentially after much plastic has been wasted.
.
Much thanks to Hans, the developer of finFoil. Without finFoil, this would not have worked; and to the Swaylocks members and Prusa printer forum members for all the help and support. And thanks to all the busy beavers contributing to all the Open Source stuff that is required to pull this off: OpenScad, Linux / Ubuntu, RepRap printer developers and so on.
20h 15min to print.
Still came out with plenty of over-extrusions all over the place, not sure why. But that is much better than under-extrusion, because it is very easily sanded.
Spreading the load on the PEI sheet by using three alternative placements for the Wanderfalke2 fins.
.
This allows the bulges which form in the PEI sheet (due to warping forces) to settle down while printing on the alternative locations.
Hi Mik, any more thoughts about trying to print molds versus fins? A day long printing cycle for one fin seems like an investment.
PS They are nice fins!
Mould printing would work, but probably take longer than fin printing.
I don’t think you would get many laminations out of a PLA mould, and other material would warp too much to print a large fin mould.
So the way to go would be to print a mould-mould, then cast a silicone mould into the PLA mould-mould, and then use the silicone mould to make multiple resin / cloth laminated fins.
But the devil may be in the detail, I’m not sure you could laminate nicely around the ball spring plunger holes. But if you want to make standard fin bases and then drill holes, that would be more likely to succeed.
.
However, I don’t think it’s very feasible, unless you have a fin that will sell many times for a good price. Making a silicone mould is not cheap and takes several days.
.
The usefulness of the PLA solid printing method (assuming they prove to be strong enough in the longer turn) is that you can make a fin with less than 24hr turnaround, and each fin can be subtly different, or totally different, or different in just one aspect of it’s design.
.
And don’t forget that the long print time is time that the machine does, not you. You can run a printer farm, or just go surfing while it completes the job, although it’s not recommended to leave experimental technology with heaters and flammable materials alone for long. Probably good to have smoke sensors and be home to save the day in case anything ever goes badly wrong. I’m considering moving the printer to the shed if/when I have files that print well and I just want to ‘print another on of those’. If the shed burns down that’s not as bad as the house burning down. I could crank out one fin per day while working full time in another job. Not too bad.