Speer’s Bidirectional Foil-Sections: have they been tested?
Hey guys,
recently the blunt-edged bidirectional foils which Tom Speer designed have come up a fair bit. I’ve read the paper, and it all seems well and good, but there’s two things I want to call some attention to:
The blunt trailing edge sections analyzed in this paper barely met the assumptions behind XFOIL’s methodology. Therefore, the results should be treated with some caution until confirmed by experiment.
When it comes to fluid dynamics, all the simulation tools, XFOIL included, are approximations of reality, which are applicable to foils with certain general characteristics / operating regimes. Deviating from the operating assumptions of such a program can lead to very erroneous results. Having very blunt trailing edges is not in the least bit something which is common, so it’s not something which the program was designed for. Maybe it’s ok, but maybe it’s totally off. Tom Speer himself expressed some very serious doubt about his foil sections in this forum thread (post #3):
Yes, but I don’t have a lot of confidence in how those sections will actually perform. The computer thinks they will work OK, but Mother Nature may have different ideas. In any event, they will be very sensitive to small differences in shape near the edges.
If it was my boat, I think I’d go with sharp-edged sections.
Did I miss something, like proper water or wind tunnel testing being conducted on these sections to verify the accuracy of the results, or is it just being assumed that they are accurate?
Cheers,
Marco
Marco,
I built one as a leeboard for Kevin’s proa. It was in replacement of a smaller, not as deep, first design, which was not Speer profile. We used the Speer profile board on the Texas 200 (at least 4 or 5 years ago) and the board has been on Kevin’s boat since then.
I think that I can say on behalf of Kevin that he has been very happy with the performance of the board, compared to his first one.
That being said, obviously, this is not as scientific or accurate as actual comparative testing in a lab tank…
My 2 cents.
Laurent
I don’t think these have been exhaustively tested, Marco, or even built very often. I think most of us (I for one) are taking the information based on the author’s strength and background. Tom Speer took a leave of absence from Boeing Phantom Works to do aerodynamics design and testing for both Dogzilla (AC 17) and for the recently completed AC-72 regatta in San Francisco. He is credited with reducing drag from the massive Dogzilla mast section by some 5%, a Really Big Deal in the AC design world (section is so big you can climb up, inside the mast, right to the top!) They brought him back for the AC72’s regatta, but though I know him well, I don’t know his precise duties there.
Point is, he’s a pretty smart guy. These sections are completely unproven, but then again, they’re miles better pedigreed than whatever’s in second place. 😉
(BTW, there is no second place, to my knowledge. Nobody else has done any design work at all for asymmetric reversible foils, other than simple ogives, which are inherently bi-directional. To the best of my knowledge no shunter has either used them nor done any flume or wind tunnel work on them.)
Cheers,
Dave
Does someone know what Bieker has spec’ed out for his 32 ft proa? It does sport a reversible single daggerboard in the ama…
Is it a Speer profile? If not, what else?
Cheers,
Laurent
Well the L/D ratios of the boards are pretty important for performance (the course angle is the hydrodynamic drag angle + the aerodynamic drag angle after all). Jzerro seems to perform really well in general, but maybe we’re missing out on some speed potential, and the possibility to point even higher, as a result of poor foils. Without someone doing proper wind/water tunnel testing, I think I personally would rather take NACA-sections and use one board for each direction. Maybe I’ll end up with better performance—or maybe it won’t make a difference, but that way I can be sure I have something good. That’s just my nit-picky take on things though! 😉
The Jester board looks like a bidirectional one to me, what looks to me like the maximum chord line on the surface of the board, is right in the middle.
Cheers,
Marco
Well the L/D ratios of the boards are pretty important for performance (the course angle is the hydrodynamic drag angle + the aerodynamic drag angle after all). Jzerro seems to perform really well in general, but maybe we’re missing out on some speed potential, and the possibility to point even higher, as a result of poor foils. Without someone doing proper wind/water tunnel testing, I think I personally would rather take NACA-sections and use one board for each direction. Maybe I’ll end up with better performance—or maybe it won’t make a difference, but that way I can be sure I have something good. That’s just my nit-picky take on things though! 😉
The Jester board looks like a bidirectional one to me, what looks to me like the maximum chord line on the surface of the board, is right in the middle.
Cheers,
Marco
Russell Brown has worked with Paul Beiker for many years fabricating and developing foils for extreme performance I-14 dinghies (and others). His experience making high performance foils is at the top of the game - and he fabricates the foils Paul Beiker develops. I’d expect his own boat’s foils to be far beyond anything available. Beiker/Brown I-14 foils are held in the same esteem as Larry Tuttle’s (Waterrat) are in the 505 class. I doubt there is a lot of missing speed potential in Jzzero’s foils - he’s reported sustained hour+ runs of 22 knots - in a proven cruising design.
All these guys (Beiker, Brown and Speer) know each other and worked together on AC17 and AC72 design teams.
Whenever you consider individual elements in a complex design, everything has to be evaluated in context, not as an isolated component. Tweaking foils to the Nth degree may not be relevant if every other element is not optimized to the same level of refinement. The logical conclusion to refining a proa to optimal performance is Vestas Sailrocket 2 - a one-directional boat with foils only having to deal with keeping flow attached in one direction.
—
Bill in Ottawa
I used the same one on TP02 and TP03,
On TP02 they where used on two counteracting rudders - self balancing on each shunt
On TP03 a single moving leeboard
No issues with foils but testing was far from scientific
Tink
I wasn’t aware of Russell and Bieker’s background in that regard—those are certainly some very impressive credentials! 😊 That said, my question was if there’s hard data. 😉 I don’t mean to be disrespectful when I say that, but to my knowledge the foil sections on Jzerro are not published (I guess you could look at the Madness boards), and I haven’t seen any L/D polars of the sections either. Doing that properly is anything but easy, you need access to the right facilities, and it’s a LOT of work, and my impression is no one has done it yet.
For one-way foil-sections on the other hand you can flip open a textbook and get a relatively good idea of how good they are, but for the bidrectional ones that is, at least to my knowledge, an entirely unknown quantity. For all I know, the boards could be just as good as the unidirectional ones, or they could be a fair bit worse and no one has noticed.
One somewhat easier way to test it would be to create performance polars of the boats, using some data-logging equipment, and do that separately for several runs around a course with uni- and bidirectional foils respectively. Has anyone done it? Anyone want to try? I intend to set my boat up with the necessary electronics once it’s done (and yes I would share the polars), for me it’d be a cool project to learn more about working with µCs anyway, but my guess is it’s still going to take a year or two until Firstborne is in the water…
I wasn’t aware of Russell and Bieker’s background in that regard—those are certainly some very impressive credentials! 😊 That said, my question was if there’s hard data. 😉 I don’t mean to be disrespectful when I say that, but to my knowledge the foil sections on Jzerro are not published (I guess you could look at the Madness boards), and I haven’t seen any L/D polars of the sections either. Doing that properly is anything but easy, you need access to the right facilities, and it’s a LOT of work, and my impression is no one has done it yet.
For one-way foil-sections on the other hand you can flip open a textbook and get a relatively good idea of how good they are, but for the bidrectional ones that is, at least to my knowledge, an entirely unknown quantity. For all I know, the boards could be just as good as the unidirectional ones, or they could be a fair bit worse and no one has noticed.
One somewhat easier way to test it would be to create performance polars of the boats, using some data-logging equipment, and do that separately for several runs around a course with uni- and bidirectional foils respectively. Has anyone done it? Anyone want to try? I intend to set my boat up with the necessary electronics once it’s done (and yes I would share the polars), for me it’d be a cool project to learn more about working with µCs anyway, but my guess is it’s still going to take a year or two until Firstborne is in the water…
Paul Beiker has published polars for the Jester class on his blog - saw them about four months ago.
—
Bill in Ottawa
[
Paul Beiker has published polars for the Jester class on his blog - saw them about four months ago.
I tried to find these, Bill, but was unsuccessful. Could only find the study drawings, and on a different site, the VPP spreadsheet that Rob guided us to—but not the polars.
I suspect these are as-designed, pulled from the VPP, and not from measurements. So it will not only contains lots of assumptions (there not being very many proa data points in the record) but also the effect of boards versus sail plan versus conditions will be blended together in the results. Still be valuable to see the polars if anyone has a URL?
Thanks!
Dave
I suspect these are as-designed, pulled from the VPP, and not from measurements. So it will not only contains lots of assumptions (there not being very many proa data points in the record) but also the effect of boards versus sail plan versus conditions will be blended together in the results. Still be valuable to see the polars if anyone has a URL?
Thanks!
Dave
Given there wasn’t an example of the boat sailing at the time, I’d agree on your conclusions. I’ll try to take a look when a get a minutes.
—
Bill in Ottawa
Sorry to have not replied sooner, I’m trying to cut down on my web-surfing time.
The board Laurent built me is much, much better than anything I’d had on the boat before. Prior to this I’d had three different boards; one was the front of a 0006 board duplicated fore and aft, one was an ‘ogive’, just a section of a big circle on one side and flat on the other, and the third was the second cut down and reduced in thickness, and with the flat side rounded some. I’ve been sailing since I was a kid, so I thought I more or less knew what a foil should look like. They were smooth, fair, etc.
None of them were very good. In fact, none of them were much better than no board at all. The boat goes to windward ok even with no board down. The first two were 5’ x 1’, and putting them down was like dropping an anchor. You could feel the drag. They also fluttered badly at various speeds. The first one especially fluttered like mad at anything over about three knots, it was really bad.
The third, small board was the best. It didn’t flutter much, only at really high speeds, and it didn’t feel like an anchor when you dropped it. But as the rig got better it was clear that it was allowing for a lot of leeway. It was a little better than having no board at all, but not a lot.
Laurent’s implementation of the Speer section is great. It never flutters, and the boat goes to windward really well. I left it vertical and sailed down onto near a beam reach several times to see how it acted at higher speeds, and it never fluttered. I think the fastest I got it to was eight to ten knots of boatspeed. It does not, subjectively, feel like an anchor when you put the board down. It feels like it does when you drop the board in a Laser or other dinghy; a little more drag, but not like you just ran into something. In particular it works really wall at high angles of attack. The boat pinches up to windward really well now, and if I need to I can head up to well inside 45 degrees off the wind. This is not the fastest way to get to windward, but it’s sometimes useful.
Of course all this is subjective. I was planning to do some testing with a GPS and various boards, but never did. It would really be better to do it in a test tank, but I don’t have access to such a thing. But my feeling, based on sailing the boat for ten years with various boards on it, is that the Speer section is a significant improvement over the nice looking but intuitive boards I made before I had this one.
So they have been tested… subjectively. 😉 Thanks for sharing!
Is there any chance you could do a comparison between Laurent’s board and one or two known unidirectional boards with good performance? In principle all you’d need is a good number of runs on a triangle course. Do a couple of laps, and flip a coin after every lap on whether to swap the board or not (randomizing the samples helps avoid bias, which could be anything from wind/current effects to it taking a while to get into the groove of how to sail the boat fast with one board or another). Then you just need the GPS data, some regular wind measurements, and a timestamp of every event logged on a notepad or whatever, so we can stitch all the data back together again, and we’d get an idea of how much of a difference it actually makes.
I probably live way too far away to help with that little science experiment directly, but I’d definitely be interested in participating in the data evaluation! 😊
Cheers,
Marco
Hi
Sadly the proa is taken apart and in the garage right now, so I’m not going to be sailing it anytime soon.
But… I have a little sailing kayak/trimaran. It’s possible I could use the Speer foil in the slot of the trimaran as a daggerboard (flipping it when I tack, of course), and compare it to the 0012 foil. This would have to come after I make the 0012 foil, so not today or tomorrow, but maybe.
Another possibility is just to engage actual experts. These are a new set of foils, designed by someone who knows what he’s doing, which have no hard data on them at all. I know someone in the engineering department at TAMU Galveston, I just shot him an email asking if he knows anyone who would be interested in this. I seem to recall him saying that people with wind tunnels get lots of requests for analysis, but almost all of them are unpublishable, since they’re not really very unique problems. Maybe this would be unique enough to get someone interested in making a set of foils and testing them against known benchmarks.
So, anyone else? If you know someone in an engineering department, shoot them an email. A real paper from data produced under controlled conditions would be a lot better than me messing around on a lake in a trimaran, with random wind shifts and so on to try to sort out.