Contact   Imprint   Advertising   Guidelines

Wing profile info for backyard hydrofoil builders

Here you can exchange your experience and datas about your home build boards
thibkite
Rare Poster
Posts: 2
Joined: Thu Feb 22, 2018 10:25 am
Local Beach: AUCKLAND
Gear: Too much to list
Brand Affiliation: Ozone
Location: NZ
Has thanked: 0
Been thanked: 1 time

Re: Wing profile info for backyard hydrofoil builders

Postby thibkite » Mon May 24, 2021 4:25 am

PrfctChaos wrote:
Mon May 10, 2021 7:20 am
Damagedd wrote:
Wed May 05, 2021 3:34 am
Alright, couple of results below for you to have a look at.

750mm wingspan, 1350 cm^2 area, note for this one I have pushed the speed range up to around 17-40 km/hr (10-25 mph), from what I have seen the very-very low speeds (<10 mph) does not really occur when kiting. But happy to run for a lower speed range if you want. The induced drag at the lowest speeds will just become massive for a 750 mm wingspan.
Capture - Damagedd1350.PNG

And having a look at the wider range of wingspans and areas for this speed range, shows that a slightly lower area might also be worth considering. So have added results for a 1080cm^2 option below, to consider. It has a similar lowend, but lower drag around cruising speed of 30 km/hr.
Capture - Damagedd1350 range.PNG

750mm wingspan, 1080 cm^2 area, 17-40 km/hr option:
Capture - Damagedd1080.PNG

For the 1m wide wing, I have also pushed the speed range up just a bit to 12-35 km/hr.
1000mm wingspan, 1700cm^2 area, 12-35km/hr:
The combination of wingspan and area looked pretty good. Having a look through changing the surface area didn't show any real additional benefits.
Capture - Damagedd1700.PNG

FYI, this is now version 4 of the software. Same functionality as version 3, it just runs in a different way making it much faster. So can consider a few future detail optimisation options now that it is back to reasonable run speeds.

For example, on the graph there is a value called the "Lift distribution error" it shows how much the lift distribution on the proposed wing deviates from a elliptical lift distribution. In the future this could potentially be used to adjust wing shape and / or Wingtip washout angle to get the lift distribution as close to elliptical as possible over the speed range.
Cool tool, what Ncrit are you using when you generate these polar?
Last edited by thibkite on Mon May 24, 2021 7:20 am, edited 1 time in total.

User avatar
fluidity
Very Frequent Poster
Posts: 657
Joined: Sat Mar 19, 2016 11:20 pm
Kiting since: 2015
Weight: 115kg
Local Beach: Ngati Toa, Plimmerton, Titahi Bay, Waikanae, Petone, Seatoun, Lyall Bay, Eastbourne, Lake Wairarapa
Favorite Beaches: Plimmerton
Style: Wave, jump
Gear: Transitioned from Kiting to Wingsurfing late 2019. Building my own foils from my CAD designs and 3D prints, CNC machine.
Brand Affiliation: Designer of hydrofoils and many other things.
Location: Porirua New Zealand
Has thanked: 46 times
Been thanked: 91 times

Re: Wing profile info for backyard hydrofoil builders

Postby fluidity » Mon May 24, 2021 6:19 am

PrfctChaos wrote:
Mon May 10, 2021 7:20 am
For example, on the graph there is a value called the "Lift distribution error" it shows how much the lift distribution on the proposed wing deviates from a elliptical lift distribution. In the future this could potentially be used to adjust wing shape and / or Wingtip washout angle to get the lift distribution as close to elliptical as possible over the speed range.
I thought I'd mention Ludwig Prandtl's bell curve lift distribution that Albion Bower's team has been researching. (Albion has now retired as NASA's chief scientist)
Absolutely fascinating but not widely known, Ludwig Prandtl released an elegant 2 pages update paper about 13 years after the one that the aeronautics industry uses, and in French! It doesn't correct his previous paper that the aeronautical industry relies on, but it surpasses his eliptical solution in terms of explaining how to best create a high efficiency wing. Possibly this is the direction you are heading in with your discussion of washout?

Certainly it's what birds use and as Albion points out, evolution has no ongoing place for bad design, lift, efficiency for a bird is it's life!
Calories lost to poor evolutionary paths are never recovered and can be life ending.

Anyway, the washout seems to start at around 70% from centre to wing tip for regular use, torquing to a huge 11% upwash profile angle at wing tips. I suspect for pushing the max speed both lift angle and upwash angle would have to be moderated. Potential efficiency savings are less when compared against planes with rudders as the advantages of inherent banked yaw compensation are wasted on a Wingsurfer's kit but this is the way all serious wings are going to be heading. The general idea is to blend lift in the middle with no lift at the tips, vortices are generated near the start of the washout, the washout area generates thrust more efficiently than an upturned winglet.

Certainly it's not an eliptical solution though, that was in Prandtl's first paper which preceeded his refined solution by about 13 years.

Try this summary by Albion, took a couple of viewings for me to absorb it :D I took a bunch of notes and snap shots the second time!

I think Ludwig Prandtl is one of those rare genuises who had such a fundamentally honest and insightful look at a field that our understanding is forever after improved. I have massive respect for people like him who pushed amazing new theory in the absence of computerisation and electronic calculators.
These users thanked the author fluidity for the post:
Onda (Mon May 24, 2021 7:50 pm)
Rating: 3.03%

Onda
Frequent Poster
Posts: 478
Joined: Thu Nov 14, 2013 10:12 am
Kiting since: 2008
Local Beach: Baltic Sea / North Sea (Germany)
Favorite Beaches: Wijk aan Zee (NL)
Ouddorp (NL)
Norre Vorupor (DK)
Heiligenhafen (DE)
Style: strictly unstrapped, mainly foiling
Gear: Slingshot / Infexion / Alpine / F-One / CORE / Firewire / Duotone
Brand Affiliation: None
Location: Germany
Has thanked: 51 times
Been thanked: 99 times
Contact:

Re: Wing profile info for backyard hydrofoil builders

Postby Onda » Mon May 24, 2021 7:51 pm

Thanks for sharing this, fluidity. I understand maybe 5 % of the presentation, but that´s enough to be fascinated!

PrfctChaos
Frequent Poster
Posts: 317
Joined: Fri Apr 27, 2018 6:16 am
Local Beach: Perth
Style: Wave
Gear: Peak4's, Chrono V3, Skimboards, foils
Brand Affiliation: None
Has thanked: 36 times
Been thanked: 81 times

Re: Wing profile info for backyard hydrofoil builders

Postby PrfctChaos » Tue May 25, 2021 5:04 am

thibkite wrote:
Mon May 24, 2021 4:25 am
Cool tool, what Ncrit are you using when you generate these polar?
Thanks,
N_crit = 9, is used on the creation of these polars. An average, somewhere between clean and dirty ambient conditions. Here is a description from the Xfoil manual:

The e^n method has the user-specified
parameter "Ncrit", which is the log of the amplification factor
of the most-amplified frequency which triggers transition.
A suitable value of this parameter depends on the ambient
disturbance level in which the airfoil operates, and mimics
the effect of such disturbances on transition. Below are typical
values of Ncrit for various situations.

situation Ncrit
----------------- -----
sailplane 12-14
motorglider 11-13
clean wind tunnel 10-12
average wind tunnel 9 <= standard "e^9 method"
dirty wind tunnel 4-8

PrfctChaos
Frequent Poster
Posts: 317
Joined: Fri Apr 27, 2018 6:16 am
Local Beach: Perth
Style: Wave
Gear: Peak4's, Chrono V3, Skimboards, foils
Brand Affiliation: None
Has thanked: 36 times
Been thanked: 81 times

Re: Wing profile info for backyard hydrofoil builders

Postby PrfctChaos » Tue May 25, 2021 5:28 am

fluidity wrote:
Mon May 24, 2021 6:19 am
PrfctChaos wrote:
Mon May 10, 2021 7:20 am
For example, on the graph there is a value called the "Lift distribution error" it shows how much the lift distribution on the proposed wing deviates from a elliptical lift distribution. In the future this could potentially be used to adjust wing shape and / or Wingtip washout angle to get the lift distribution as close to elliptical as possible over the speed range.
Absolutely fascinating but not widely known, Ludwig Prandtl released an elegant 2 pages update paper about 13 years after the one that the aeronautics industry uses, and in French! It doesn't correct his previous paper that the aeronautical industry relies on, but it surpasses his eliptical solution in terms of explaining how to best create a high efficiency wing. Possibly this is the direction you are heading in with your discussion of washout?
Elliptical lift distribution is a exact analytical solution of lifting line theory, for minimising drag. My purpose for including "Lift distribution error", is to show the difference between the achieved lift distribution and the theoretically optimum elliptical lift distribution. It might be used in a later iteration to adjust the error to a minimum through outline or angle changes, but it is not actively used at the moment, only reported. The solution is a optimum for what I would call a 2.5D frame of reference.

To find a better optimum (full 3D frame of reference), one would need to go for one better level of detail. Which means CFD and more time and effort to optimise. CFD is based on the more general Navier-Stokes equations.

However, I feel like I need to say it again, the goal of my optimisation tool is to narrow PROFILE selection down to one or two of the most suitable profiles for the intended purpose. It is what I would call a 95% tool. It very quickly gets close to a good solution (and it is getting better along the way) and it will quickly eliminate thousands of unsuitable profile selections. LEaving a few that one might want to look at more closely.

If I was just going to freeride a foil, I would (and have) take that 95% result, built the wings and enjoyed them for years. Having a great time. Never wondering again about whether there is a extra 1 or 2 percent to gain in loss of drag, it is likely just not that important. Changes to the how the wing curves down, stab selection or fuse connection are likely way more important at that stage than a littlebit extra front wing efficiency.
If I was going to do competitive downwind or wing racing on one of these foils. Then I would definitely be taking that initial output and wringing the last 5% of performance out of it by spending a few weeks tweaking further and running hundreds of CFD runs (Also compromising practicality and turning for higher performance). (If I was going kite racing I would just buy a brand new Mike lab, Chubanga or KFA foil. They are already rippers.)

User avatar
fluidity
Very Frequent Poster
Posts: 657
Joined: Sat Mar 19, 2016 11:20 pm
Kiting since: 2015
Weight: 115kg
Local Beach: Ngati Toa, Plimmerton, Titahi Bay, Waikanae, Petone, Seatoun, Lyall Bay, Eastbourne, Lake Wairarapa
Favorite Beaches: Plimmerton
Style: Wave, jump
Gear: Transitioned from Kiting to Wingsurfing late 2019. Building my own foils from my CAD designs and 3D prints, CNC machine.
Brand Affiliation: Designer of hydrofoils and many other things.
Location: Porirua New Zealand
Has thanked: 46 times
Been thanked: 91 times

Re: Wing profile info for backyard hydrofoil builders

Postby fluidity » Tue May 25, 2021 8:57 am

PrfctChaos wrote:
Tue May 25, 2021 5:28 am

Elliptical lift distribution is a exact analytical solution of lifting line theory, for minimising drag. My purpose for including "Lift distribution error", is to show the difference between the achieved lift distribution and the theoretically optimum elliptical lift distribution. It might be used in a later iteration to adjust the error to a minimum through outline or angle changes, but it is not actively used at the moment, only reported. The solution is a optimum for what I would call a 2.5D frame of reference.

To find a better optimum (full 3D frame of reference), one would need to go for one better level of detail. Which means CFD and more time and effort to optimise. CFD is based on the more general Navier-Stokes equations.

However, I feel like I need to say it again, the goal of my optimisation tool is to narrow PROFILE selection down to one or two of the most suitable profiles for the intended purpose. It is what I would call a 95% tool. It very quickly gets close to a good solution (and it is getting better along the way) and it will quickly eliminate thousands of unsuitable profile selections. LEaving a few that one might want to look at more closely.

If I was just going to freeride a foil, I would (and have) take that 95% result, built the wings and enjoyed them for years. Having a great time. Never wondering again about whether there is a extra 1 or 2 percent to gain in loss of drag, it is likely just not that important. Changes to the how the wing curves down, stab selection or fuse connection are likely way more important at that stage than a littlebit extra front wing efficiency.
If I was going to do competitive downwind or wing racing on one of these foils. Then I would definitely be taking that initial output and wringing the last 5% of performance out of it by spending a few weeks tweaking further and running hundreds of CFD runs (Also compromising practicality and turning for higher performance). (If I was going kite racing I would just buy a brand new Mike lab, Chubanga or KFA foil. They are already rippers.)
Sure eliptical lift distribution is one tool for minimising drag but Ludwig Prandtl wasn't as satisfied with his solution as those in the aero industry who used it. I've read the disclaimers on one of the CFD tools though I've not used it yet. Interesting stuff, pity no one has written good software to model CFD the way it's done by the meterological industry to run on modern PC hardware, I'm sure that would point a lot of people in the right direction. I'm as you know pretty new to all this, only real advantage I have is objectivity, I'm not trained one way and then trying to determine whether all I believed and stake my profession on is now suspect! I'm an outsider to it all. That gives me objectivity. By being poor in preconceptions I can weigh relative advantages and disadvantages less emotionally. I recall Albion Bowers remarking in the video above that I posted(or another of his videos), that eliptical wing theory optimises drag relative to a fixed wing span. However I don't think it does even that properly, it does not properly address tip vortices. That's where Prandtl's later work shines, he was obviously not happy with his first solution and if you consider that it took him 13 years to come up with the better theory, in that time his first theory became embedded as a faith item for the aero industry, for aero engineers to mentally take in what their founder had moved on to was no small matter. I really suggest you watch the video a couple of times at least, we all take on an unavoidable level of brainwashing from the culture around us and it can take time to integrate new information into our world pictures when something surprising upsets our belief system. I got caught out a big one when I was about 13, with optics. I'd never really given it much consideration and felt that vision was an active process of sending intent out my eyeballs that just returned from objects to my brain. Luckily for me I was presented with a more coherent theory! Science is full of these little traps. My favorite is the shape theory of scent and how Luca Turin overturned it with his vibration theory of scent- which actually was a totally superior predictive model to boot! People are still arguing about that one because accepting it overturns careers of teaching and employment based on the old one.
These users thanked the author fluidity for the post:
direnc (Tue May 25, 2021 8:24 pm)
Rating: 3.03%

PrfctChaos
Frequent Poster
Posts: 317
Joined: Fri Apr 27, 2018 6:16 am
Local Beach: Perth
Style: Wave
Gear: Peak4's, Chrono V3, Skimboards, foils
Brand Affiliation: None
Has thanked: 36 times
Been thanked: 81 times

Re: Wing profile info for backyard hydrofoil builders

Postby PrfctChaos » Tue May 25, 2021 9:37 am

Oh my goodness, can this be a separate thread please?
Last edited by PrfctChaos on Tue Jul 20, 2021 12:13 am, edited 1 time in total.

direnc
Frequent Poster
Posts: 204
Joined: Sun Jul 22, 2012 11:50 pm
Kiting since: 2009
Local Beach: Alacati, Pirlanta, Ilica
Style: Freeride
Gear: Cult 7.5
Soul 10
Ride 10
Kestrel 12
Gin Spirit 15
Edge 8/9/11
Ten Velox 11
Zephyr
Mako 150, DIY foils and boards
Brand Affiliation: None
Has thanked: 47 times
Been thanked: 21 times

Re: Wing profile info for backyard hydrofoil builders

Postby direnc » Tue May 25, 2021 8:21 pm

PrfctChaos wrote:
Tue May 25, 2021 5:04 am
thibkite wrote:
Mon May 24, 2021 4:25 am
Cool tool, what Ncrit are you using when you generate these polar?
Thanks,
N_crit = 9, is used on the creation of these polars. An average, somewhere between clean and dirty ambient conditions. Here
Tom Speer recommends 1<Ncrit<3, even 1 kite/surf foils if I am not mistaken. I wonder how/if a lower Ncrit number will change the results.
These users thanked the author direnc for the post:
PrfctChaos (Wed May 26, 2021 3:25 am)
Rating: 3.03%

thibkite
Rare Poster
Posts: 2
Joined: Thu Feb 22, 2018 10:25 am
Local Beach: AUCKLAND
Gear: Too much to list
Brand Affiliation: Ozone
Location: NZ
Has thanked: 0
Been thanked: 1 time

Re: Wing profile info for backyard hydrofoil builders

Postby thibkite » Tue May 25, 2021 9:33 pm

direnc wrote:
Tue May 25, 2021 8:21 pm
PrfctChaos wrote:
Tue May 25, 2021 5:04 am
thibkite wrote:
Mon May 24, 2021 4:25 am
Cool tool, what Ncrit are you using when you generate these polar?
Thanks,
N_crit = 9, is used on the creation of these polars. An average, somewhere between clean and dirty ambient conditions. Here
Tom Speer recommends 1<Ncrit<3, even 1 kite/surf foils if I am not mistaken. I wonder how/if a lower Ncrit number will change the results.
Yep 9 is way too high, Ncrit should be really kept between 1 and 3 and it change things a lot. Some profil like s2055 are really good in clean flow but struggle when it get turbulent. In the other hand some hydrofoil dedicated profil have been designed to be less affected like e817 and H105 and perform better at low Ncrit than s2055 and other 4 digit NACA.
Attachments
2021-05-26_08h41_42.png
These users thanked the author thibkite for the post:
PrfctChaos (Tue May 25, 2021 11:14 pm)
Rating: 3.03%

PrfctChaos
Frequent Poster
Posts: 317
Joined: Fri Apr 27, 2018 6:16 am
Local Beach: Perth
Style: Wave
Gear: Peak4's, Chrono V3, Skimboards, foils
Brand Affiliation: None
Has thanked: 36 times
Been thanked: 81 times

Re: Wing profile info for backyard hydrofoil builders

Postby PrfctChaos » Tue May 25, 2021 11:16 pm

thibkite wrote:
Tue May 25, 2021 9:33 pm
direnc wrote:
Tue May 25, 2021 8:21 pm
PrfctChaos wrote:
Tue May 25, 2021 5:04 am


Thanks,
N_crit = 9, is used on the creation of these polars. An average, somewhere between clean and dirty ambient conditions. Here
Tom Speer recommends 1<Ncrit<3, even 1 kite/surf foils if I am not mistaken. I wonder how/if a lower Ncrit number will change the results.
Yep 9 is way too high, Ncrit should be really kept between 1 and 3 and it change things a lot. Some profil like s2055 are really good in clean flow but struggle when it get turbulent. In the other hand some hydrofoil dedicated profil have been designed to be less affected like e817 and H105 and perform better at low Ncrit than s2055 and other 4 digit NACA.
Good info, thanks guys. Will do a few runs at Ncrit of 2 and see which profiles come up tops at the new numbers. Cheers


Return to “Gear Builders”

Who is online

Users browsing this forum: sflinux and 126 guests