Foils in the 21st Century

A lot happened in the area of airfoil research, development and design that is applicable for use on surfboards long before shapers started carving surfboards out of foam and glassing them with fiberglass.  Decades ago, the late fin guru Curtis Hesselgrave started producing windsurfing fins with 1940’s era NACA airfoils carved into G-10 glass with aerospace wind tunnel model accuracy.  The performance improvements over mystery-foil fins were phenomenal and started a performance revolution.  Later, Curtis applied his talents to surfboard fins; however, most surfers are unaware of the technology behind some of Futures Fins offerings.

Much has happened since the development of the NACA airfoils, with little to none of it migrating to surfboards.  Surfboard fins operate in a transitional fluid dynamic regime where the molecules can have difficulty moving along the surface no matter how streamlined and precise that surface is manufactured.  Computation tools to accurately predict these transitional effects were not available until fairly recently - the most notable being Xfoil developed by M.I.T. engineers.  Xfoil analysis of foils commonly used on surfboard fins reveal problems, but ways to partially fix those problems have been known to the low Reynolds number aerodynamic community since surfboards were first carved out of foam and glassed with fiberglass.  Since Xfoil accurately predicts the transitional effects, it is possible to design foils that work well in these conditions, and it is even easier to just find ones that already fit our criteria.

I’ll be linking my discussion to pages on AirfoilTools.com.  All of the performance data are from Xfoil predictions.  Information and links about Xfoil can be found there.  When you go to the pages, uncheck the Rn 50,000 box and then click on the ‘Update plots’ box.  Few surfers will ever get to Rn 1,000,000, but it should stay for reference.  The angle-of-attack convention is the angle of the free stream flow relative to the chord line that extends from the trailing edge to the tip of the leading edge - that line is drawn on the airfoil plots.

Nearly everything here is about two dimensional flow.  Surfboard fins, especially thruster, twin fins and quad fins, have low aspect ratios.  Even our ‘high aspect ratio’ fins have relatively low aspect ratios.  And they have a lot of rake/sweep.  Both of these design characteristics result in significant three dimensional flow effects.  I will not be addressing 3D effects here.

The symmetric foiled back thruster fin commonly has a foil that is or resembles a NACA008 or NACA009 airfoil.  Taking a look at the lift coefficient (Cl) vs angle-of-attack (alpha) plot, you’ll see that the response at the lower Reynolds numbers is not linear - there is a partial dead-band on either side of zero/zero.  This results in poor centering response when we are first getting to our feet and during slower maneuvers.  Nor is it completely linear at our higher Reynolds numbers.  Note how more thickness of the NACA0012 foil results in significantly higher maximum lift coefficients while widening the drag bucket - drag coefficient (Cd) vs. Alpha plot.  A thicker foil can achieve lower drag at higher lift coefficients; however, don’t think that I’m suggesting we all start using fins with thicker foils - it’ll work well in some situations, not so well in other situations, and it has centering issues over a wider range of alpha.

Side fins with similar thickness distributions, but with the addition of camber, have the same problem.  Looking at the results for the NACA2408, depending on the Reynolds number, the zero lift angle of attack ranges from a wee bit less than 1° at Rn 100,000 to about -2.5° at Rn 500,000 and above.  If your fins are toed in properly for the higher Rn, then they are toed out too much at lower Rn.  Do take note how adding camber to the foil increases the maximum achievable lift coefficient at all Reynolds numbers.  Getting more grip out of our fins is a good thing!  

This centering deficiency results in surfboard handling difficulties and the creation of drag.  Throw in some manufacturing variations in the toe-in angles and the back fin alignment, and thrusters end up with three fins each going their own way a little bit different at different speeds.  Twin fins can get downright ugly with odd interactions between the direction each fin wants to go and what the rail/rocker wants to do.  Quads can have four fins going every which way at different speeds, and for Quads on narrow tails, the back fins move in and out of the slipstream and downwash of the front fins thus exacerbating the problem.  These characteristics are so prevalent on all surfboard fins, boards are designed to allow for it and all surfers are used to it and think it is normal.  

The fins commonly used decades ago had parallel sides and curved bevels on the leading and trailing edges.  Those foils resulted in numerous locations where the flow would separate from the shape of the foil resulting in non-linear Cl vs. Alpha response, excessive drag, as well as premature stall at fairly low angles of attack.  Those limitations undoubtedly drove both surfboard design and fin template design to compensate. 

Those were all WWII era foils that are employed on a limited number of fins offered for sale.  The foils on many fins are designed with a ’that looks about right’ process.

While trolling through the airfoil library of AirfoilTools.com, I came across several promising prospects for use on surfboard fins.  The Drela AG10 airfoil looks like a good foil candidate for all things ’side fin,’ and the Drela HT08 airfoil looks like a good foil candidate for thruster back fins and even single fins.  Both have good centering response and achieve decent maximum lift coefficients despite being very thin.  They’re similar to what is already being used, so improvements should be easily achieved without impacting 3D effects.  And I’d like to try the GOE 265 airfoil on a twin fin; it has good centering response, but that maximum lift coefficient of 1.7 or more has got to be tried.  Being radically different to what we’re currently using, 3D effects may change, so this is a totally experimental venture - a lot of CNC carving could result in a total flop.  With all that camber and a zero lift angle of attack of -5°, it won’t integrate into our boxes.  I’d also like to try the NACA0012 on a high aspect ratio longboard single fin - though I already have a couple fins with what look like NACA0010 foils CNC carved in them that I’ll soon be trying out.  I’ll see how they work first before thinking of going thicker.  And the NACA010 and NACA012 will both require modifications to make the molecules in the boundary layer think they are at a higher Rn.  Maybe trying a variations of parameter test with the HT08 and HT12 foils at 1.0x, 1.5x and 2.0x of the thickness ordinates to see how thicker foils with good pressure recovery work on single-fin setups.  

Characteristics of the foils that work best are: thin, max thickness around 20% chord, and straight taper shortly after the max thickness.  Do a survey of numerous production surfboard fins from several manufacturers, and you’ll find max thickness points as far back as 50% chord, and curved tapers into the trailing edge - such foils will have poor centering performance and reduced maximum lift coefficients due to separated flow.  Separated flow is not cavitation, nor is it a stall (complete separation), it is where the flow detaches from the surface and can even reattach further down the foil creating a bubble in between.  You’ll also find many fins that have the proper characteristics for good performance at our flow conditions.  Albeit many have leading edges that are too sharp for 2D flow - Futures calls their flat bottom foiled fins ’speed controlling.’  The sharp leading edges will cause early chordwise flow separation.  I presume the ‘speed controlling’ feature is because they’re easy to partially stall to slow down and stay in the tube.  By far the most advanced side fin is Futures’ Solus.  I wish they offered it in a twin-fin size.  I have one set that came with an AM1 back fin instead of the stock hatchet fin.  The second most advanced side fins are Futures’ V2 foiled fins that come in several template offerings.  Some of the worst offerings from all manufacturers are side bites and the trailer fins in twin + trailer setups.  The side bites operate at an Rn half of the center fin, and always are of poor design.  The trailer fins operate at 2/3 the Rn of the twin fins, and always are of poor design.  I have one with a foil that has the thickest part at 50% chord.  The purpose of the trailer fin is to provide some centering for the twin fins, but such a design will do little for that task.  Poorly foiled fins can be corrected to some extent with a sanding block.  I’ve gone so far as to extend the trailing edge with plate glass, thus making a template change, filled it in with epoxy filler, and sanded it to make a new foil.

So, what do we need to do to take advantage of this foil technology?  The same approach that Curtis Hesselgrave took decades ago: CNC machining of sheet glass or CNC machining of molds to make fins in.  We need to maintain tolerances around 1/1000 inch to fully take advantage of the foil characteristics.  Since Curtis started making fins, we now have more manufacturing tools, so 3D printing may be an option as well, either the fins themselves or molds to make fins in.  Or, just buy fins with foil characteristics as close as possible to what the data says works best and break out the sanding block if need be.

My own fins are highly modified from what everybody else uses, but I start with fins that have the best fundamental foil and template design to begin with.  I do have some fins that I plan to add plate glass and epoxy filler to the bottom of the leading edge to make them resemble a Drela AG08 airfoil - cause the rest of the foil looks nearly identical.  On some flat bottom foiled fins I’ve put very aggressive dropped leading edges onto them and initial results shows I’m getting more grip out of them - to get them to perform something like the GOE 265.  Another thing I plan to do is put an HT08 or HT12 foil onto a traditional longboard fin template to hide modern tech on something that looks like it’s from 1960.

Prototyping with a Drela airfoil is certainly Ok with them already being in the public domain, but production use might require some sort of licensing agreement.  But Mark is easy enough to find since he teaches at M.I.T…  The GOE airfoil is out of Germany from the 1920’s or so.

As you’ve looked at many of these foils, the thought may have crossed your mind that a portion of the curve looks a lot like a surfboard rocker.  And they do.  And that is something we have to be careful about not doing - making rockers that are upside down airfoils that actually suck the nose of the board into the wave.  But that is a different discussion. 

 

 

 

 

 

 

 

 

 

 

 

 

 

Interesting post.

Hi Scott,

XFoil has been around since the 80’s so not that recent :wink:

XFoil still uses potential flow theory, although with the addition of a boundary layer model, it is indeed a great tool to analyse foils for fins, especially since fins operate at low reynolds numbers and that current state of industry is extremely low.

You seem to pick some pretty decent foil profiles! Much better choices than the common so called “laminar” profiles with the thick point further back. The industry uses these because they are most common in maritime applications, which is wrong because these are optimised in the low AoA range.

So I fully agree with your analysis, I use other foil profiles but shape and simulation are very similar to the ht08. The GOE265, seems interesting indeed, however those assymetric side fins only make sense when pumping the board, IMO non-pumpers should ride the side fins symmetrical (or low camber) with little or no toe-in. That’s how we ride thrusters (2+1) on sailboards, symmetric sides and no toe-in.

Great writeup.

Is your take on the GOE 265 that it would provide lift for longer while turning? I’m imagining, somehow, as one turns the lift just keeps increasing and increasing.

The dead band idea is new to me. Could current fins/boards be utilizing this for the slower stalling types turns? Or the pivoty turns that occur with longboards? Could also bre a sort of “gear”. I’ve had lots of boards that all of a sudden take off once the wave gets just a scooch better.

I’m assuming the charts are for steady state flow. Any thoughts on how acceleration plays into this?

 

I’ve had the data about the dead band for decades, but it was just a couple weeks ago that I really looked into why I was feeling some things on some boards and not other boards.  Then came an ah-ha moment.

The first board I glassed had a fin with a foil that looked something like a NACA0008.  It was just something I grabbed off the shelf when I picked up my supplies.  It had about 1/4 inch thick fiberglass sandwiched between plywood.  Somebody with a grinder knew what they were doing in 1978 or so.  It was an absolutely magic board: fast, precise, and held well in turns and down the line.  After riding that board until it seriously needed extensive repair, I set about repairing it.  I also took off the fin, installed a fin box, and had the local fin shop make a fin with the identical template.  I got it back into the water, and the board was a complete dog.  All the magic was gone.  I had no clue what was going on.  The new fin had parallel sides, a rounded beveled leading edge and a slightly rounded beveled trailing edge.  The inferior foil ruined the board.  If all you knew was fins with parallel sides, and then you got on a board with a NACA0008 foil, you’d be amazed.  Going the other way was depressing.

So for a thruster, if all three fins have linear responses throughout the speed range that the board can be surfed, and they are all aligned precisely, all pointing the same direction in a fluid dynamic sense, the board will feel like it is on rails at all times.  You will have an amazing amount of mind/body/board/wave connectivity.  But, since pretty much everything being made today doesn’t have that, we don’t know what we are missing.

There are many factors that can contribute to a board working better in slightly faster conditions, but you may very well be feeling the fins getting out of their low Rn dead band rut.

f=ma

force = mass x acceleration

force = (fat-assed surfer) x acceleration

acceleration = force / fat-assed surfer

So acceleration is proportional to the amount of force that is available.  Fins that grip more allow us to turn harder - a turn is an acceleration.  And, if you can grab more water in the higher and steeper part of the wave, you’ll extract more energy from the wave and go faster.

I heard rumors decades ago of somebody swearing by highly cambered foils like the GOE 265, but nothing more than rumors.  Such foils are susceptible to numerous problems, but the GOE 265 is devoid of them.  Unfortunately for me, I don’t have the means to make a set and the guy I knew who could is gone to us.  It’s just a science project I want to try.  Maybe somebody with the means will make themselves a set and hopefully let the rest of us know how they worked.

Reading and rereading, and will read it again many more times.

 

I intend to no longer experiment with different boards, I’ve refined what I like, and they will last me until I can no longer physically surf.  I intend to experiment only with their fins, and I have been digging the lesser drag of high aspect ratio, along with a deeper more shark like pectoral fin, as my rail fins.

 

 

 

  The tail can fit higher in the wave face before stalling, and this feature opens up all sorts of new lines that can be taken, but there are times when it just turns in too short an arc and feels weird, which I’ve been mitigating somewhat with the size and rake of the thruster fin, but am now thinking I should be using less toe in with them instead.

 I know the Flat sides of rail fins make no hydrodynamic sense,  but…

I’ve yet to make/use any 80/20 or 70/30 fin which felt to give as much crisp responsive  projection when speed generation is desired, but they are certainly smooth and predictable in juicy conditions where generating speed is not an issue, controlling it is.

I’ve not yet really worked out how to accurately hand foil anything resembling the Drela AG10, but I imagine such a fin would be in a whole different realm when pushed hard through a bottom turn and this thread has me excited again to think about how to achieve it.  So thanks for starting this thread.

I was hoping MrMik would get interested, and design and 3d print some cambered foil rail fins, as his Turbucled G-whale large single fin in my traditional longboard was, and still is, a rather mind blowing improvement in a board I have 18 years of experience riding regularly.

 

I have used some of his 3d printed broken fins cut down to shortboard size,  re enforced in my shortboard’s proboxes, but never liked the feel of 50/50 foils as rail fins and never tried rail fins that are not toe’d in.  There were times when they felt very good andother times when it was drifty and very imprecise with the yellow turbucled fin as a rail fin. The turbucles can catch the leash and ruin waves. As a center fin trailer thruster fin they work quite well, unless they catch the leash. 

If I do attempt to handfoil something resembling a drega AG10, there should be less tow in, correct?

 

I wish Probox was more widely adopted, as it is the easiest to make the tabs for, in my opinion,  and the different inserts and fore/aft adjustment allow for some fine tuning, or at least seeing what changing those variables does in actual use.

 

I’ve tabled a project to make MrMik’s turbucled fins into center fins to fit Futures or FCS2. I own no boards with either of these fin systems, but wanted to see how some other, non closed minded surfers, react to their extremely forgiving and speedy  and loose qualities.

 

The super high aspect ratio means they need to sit further back in the board, and with FCS2, almost all the load is going to be on the rear tab of the box, and I don’t know if it( the box) can handle it, I’ll load the rear tab with as many  pultruded 5mm carbon bars as I can.

 

Not sure which foil MrMik used with his 3d printed fins, I’m hoping he sees this thread  and joins the conversation, and gets all excited to design and make them.

 

 

 

I can make a set on my CNC, the shape will be a nice test for my setup.

If you like, you can design it in finfoil, send me the .foil file and I’ll put them to the machine.

A feature I’ll be working on in finfoil, is the generation of hand router templates.

By routing some contours at the correct depth, you’re able to accurately foil a fin just using hand tools.

Also when working with fiberglass panels, the router can more efficiently be vacuumed for dust control.

Thanks, but I can’t load finfoil on my computer - security settings…  I have an older mac that I never updated the OS on, so maybe it’s time to get with it and set it up for non-critical tasks.  Even with me sitting around at home, it’s not going to get done anytime soon - too much like work.  As for surfing things, I’m concentrating of finishing shaping numerous blanks I’ve had sitting around for a year or two that were started and then neglected.

Pretty fins!

“I know the Flat sides of rail fins make no hydrodynamic sense,  but…… I’ve yet to make/use any 80/20 or 70/30 fin which felt to give as much crisp responsive  projection when speed generation is desired, but they are certainly smooth and predictable in juicy conditions where generating speed is not an issue, controlling it is.”

Changing camber changes where the zero lift line is pointed.  Zero lift for the GOE 265 is pointed at -5°, a symmetric foil is of course at 0° and the AG10 is around -1.4°.  When putting different fins with different cambers in box systems, we can inadvertently change the setup of our board.  Whatever fin is pointed straight ahead in a fluid dynamic sense will work the best.  Curtis Hesselgrave, Futures and the tow-in crew made this blunder when testing different fins with different cambers back in the early 00’s.

"If I do attempt to handfoil something resembling a drega AG10, there should be less tow in, correct?”

The data tells you exactly how much tow-in is required.  Go to “details” for the different Rn and you’ll find a table of the actual prediction data.  From that you can see what the angle of attack is for zero lift.  You may have to interpolate between a couple data points.  Of course, there is no way that we can maintain 0.1° alignment accuracy when installing/mounting fins.  Just do your best.

“The super high aspect ratio means they need to sit further back in the board, and with FCS2, almost all the load is going to be on the rear tab of the box, and I don’t know if it( the box) can handle it, I’ll load the rear tab with as many  pultruded 5mm carbon bars as I can."

The stiffness of the fin will transfer a lot of load to the front box.  Personally, I convert everything to mortise and tenon by splicing in G10 glass, router to the backside of the deck glass, and permanently install my fins.  Thirty years ago I went all the way through the board and glassed the fin’s tenon to both sides of the board.  I’ll start a new subject to detail how I do it one of these days.

 

 

You can use the online editor. I know it’s still a bit basic, but I uploaded a GOE265 profiled fin as a starting point.

Check the 3D render link below, you can start editing from there!

https://finfoil.io/3D/gtaaack

 

Wow, that’s incredible looking!  It just so happens I’ve brought a project forward due to SARS-CoV-2 mitigations.  I was going to work on some long overdue longboard projects this spring, but they require 50/50 rail research on the used rack at surf shops.  That ain’t gonna happen anytime soon!  Nor do I have enough resin/glass here to finish such a project.  But I did have a discount Marko blank sitting in storage…  Surprisingly this is my template for my magic 6’10” x 21 3/8” single fin, completely unaltered to yield this 6’0” x 20 3/4” fish.  Well, ten inches were cut off the tail.  Yeah, it looks like a Gary McNeill Torus Twin cause I used that as a guide.  No, I won’t be doing channels or anything of that sort.  

Apple doesn’t even recognize my old Mac as needing a software update.  I’ll have to take it to an Apple store - whenever that can happen.  Just for grins, I tried running your SW on it, it crashed.  That was done by copying the executable over to it.  Maybe I need to run the install program on it?  I kinda think that’ll fail too.  I tried to download from it, and that didn’t work either.

Having issues with my printer’s nozzles and cartridges, but I am going to lay my contour gauge along these various sized Ag10’s as a guide when foiling.

 

I did not realize how thin the Drela AG10 airfoil is. Glad the 3 layers of woven roving I used on this panel, are so stiff.

 

Gonna take me a while to finish this fin, and get some semblence of surf shape going on in order to test it under my toe side rail. Could of surfed this week, but the ocean smells like rotting brussel sprouts with the red tide, and beach  parking lots are still closed.

6" deep, 4 1/8" base and the drela Ag 10 will be about 1/4 inch thick.

I could not achieve the thin drela AG10 foil @100% thickness.

Having included cedar in my fin panel, it would have become too flexible.

 

This is pretty close to the drela AG10 @ 150% thickness.

Hi Scott, these GOE265 foil are on the way to you.

And a little video here: https://www.instagram.com/p/CBfRdtlF5Fc/

 

I hope shipment goes well!

Hans is there any performance or design reason why your GOE265 foil fils have a “break” between the fin and the base? Also did you glass one side before you cut them on the cnc? I’m trying to figure out how you cut the letters, specifically the floating bits of the O and 6.

Speaking of cutting fins with a cnc, have you had any issues cutting two sided fins? I’ve been doing test cuts the past couple weeks, but I keep having the same problem. When cutting the 2nd side the fin flexs due to lack of support and doesn’t get cut as well. I know I could increase the thickness to compensate, but that would leave more post cnc processing to sand it to the desired thickness. My other idea is to leave uncut support areas, similar to what they do for 3d printing. Again that would have to be something that gets cut off and sanded post cnc. Just wondering if you have any tips or tricks.

Also them fins are damn good looking.

I use the resin halo as a barrier for water intrusion this also give more room for reshaping the edges afterwards if needed.

No, I glass both sides at once in vacuum bag after CNC. All my fins are two sided cuts (no flat sides).

Ahaa, follow me on instagram and you’ll figure it out :wink:

Well I use custom code to generate my gcode. The trick is to leave a solid attachment just until the last finishing passes also a decent roughing step where the finishing step only finishes (like it is supposed to) significantly lowers the flex of the fin.

This is all baked into my custom gcode routines, I can generate them on request it’s not fully automatic yet, so I can’t make it available in the app yet.

You can contact me for more info: gcode_at_finfoil.io

Thanks!

Awesome, and the floating pieces in the “O” and the “6” is a cool trick, nice touch !

I rode my 150% thickness Drela AG10 twin sized   ‘batfin’ yesterday in mediocre conditions.

It did not suck.  I am excited to try it in better conditions.

I have a copy of vectric aspire that I use to generate the gcode. It seems to do a good job. I think my main problem with the flexing is lack of support tabs and overall thickness (<3/8" or <9.5mm). I had three tabs, but spaced in places that would be easy to remove not necessarily placed where they would provide the most support. Some of the flex can also be attributed to the plywood I was using. It was kinda junky, but free. For my next attempt I think I am going add some more tabs and to leave some small uncut “poles” on the first side. Hopefully when I flip it and cut the second side the pole will support the downward pressure of the machine and prevent the flexing. Only time will tell if it works.