Re: fairing rudder (laminar, etc)
I'm not an expert on a broader issue like that, but I suspect it's NOT
the centerboard, since if I'm not mistaken lots of boats that go upwind
"ok" have this feature. Years ago I had one. But of course it doesn't
help. (and it wouldn't change the angle the boat pointed at, just the
angle made good.)
the centerboard, since if I'm not mistaken lots of boats that go upwind
"ok" have this feature. Years ago I had one. But of course it doesn't
help. (and it wouldn't change the angle the boat pointed at, just the
angle made good.)
>"Frank San Miguel" <sanmi@...>
>wrote:
>Lincoln,
>
>The Drascombe uses a flat 1/2" galvanized steel plate for ballast and
>centerboard (british call it a centre plate). I wonder if thats one
>of the reason's for my Drascombe Longboat's poor upwind performance
>(55-60 degrees to the wind is about all I can get)??? I've always
>thought it was the because of the gunter mainsail but maybe there's
>more to it than than.?
>
>Frank
>
A brief informational note. I am getting rid of duplicate issues of
MAIB on eBay.
95-96
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=2237&item=3561164647
97
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=2237&item=3561352812
98
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=2237&item=3561354334
--
Craig O'Donnell
Sinepuxent Ancestors & Boats
<http://freepages.genealogy.rootsweb.com/~fassitt/>
The Proa FAQ <http://boat-links.com/proafaq.html>
The Cheap Pages <http://www2.friend.ly.net/~dadadata/>
Sailing Canoes, Polytarp Sails, Bamboo, Chinese Junks,
American Proas, the Bolger Boat Honor Roll,
Plywood Boats, Bamboo Rafts, &c.
_________________________________
-- Professor of Boatology -- Junkomologist
-- Macintosh kinda guy
Friend of Wanda the Wonder Cat, 1991-1997.
_________________________________
MAIB on eBay.
95-96
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=2237&item=3561164647
97
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=2237&item=3561352812
98
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=2237&item=3561354334
--
Craig O'Donnell
Sinepuxent Ancestors & Boats
<http://freepages.genealogy.rootsweb.com/~fassitt/>
The Proa FAQ <http://boat-links.com/proafaq.html>
The Cheap Pages <http://www2.friend.ly.net/~dadadata/>
Sailing Canoes, Polytarp Sails, Bamboo, Chinese Junks,
American Proas, the Bolger Boat Honor Roll,
Plywood Boats, Bamboo Rafts, &c.
_________________________________
-- Professor of Boatology -- Junkomologist
-- Macintosh kinda guy
Friend of Wanda the Wonder Cat, 1991-1997.
_________________________________
Lincoln,
The Drascombe uses a flat 1/2" galvanized steel plate for ballast and
centerboard (british call it a centre plate). I wonder if thats one
of the reason's for my Drascombe Longboat's poor upwind performance
(55-60 degrees to the wind is about all I can get)??? I've always
thought it was the because of the gunter mainsail but maybe there's
more to it than than.?
Frank
The Drascombe uses a flat 1/2" galvanized steel plate for ballast and
centerboard (british call it a centre plate). I wonder if thats one
of the reason's for my Drascombe Longboat's poor upwind performance
(55-60 degrees to the wind is about all I can get)??? I've always
thought it was the because of the gunter mainsail but maybe there's
more to it than than.?
Frank
--- Inbolger@yahoogroups.com, Lincoln Ross <lincolnr@r...> wrote:
>
> I'm convinced that at less than hull speed, and especially when going
> upwind, foil shape will have a significant effect on overall drag. At
> higher speeds it's main perceptible effect will be on handling. On
> catamarans and really long, narrow hulls, the foil shape will probably
> have a more significant effect at speed because wavemaking is less
> significant.
>
Not sure where those guys got the data about flat plates, perhaps by
comparing flat plate to inappropriately fat foil at really low (<50k or
100k?) Reynolds numbers, though there are also some thinner sections
good in that regime. I looked at a couple of data plots from the
Princeton low Reynolds number wind tunnel tests done by Michael Selig
and co. and published in Soartech 8. If you really want to, you can get
the data from:
http://www.aae.uiuc.edu/m-selig/ads/princeton_tests.html
However, it is of limited value to us as most of the foils tested are
intended for wings and are not symmetrical. Not only that, but most of
the Reynolds numbers investigated are lower than ours. There were
further tests that appeared later which tested a limited number of other
foils. As I recall from the graphs (I didn't bother to check closer than
that), at Re of 300k, the Cd of the NACA 009 was arouind .007 and the Cd
of the flat plate was around .013. No point in getting closer as the
tests do have a significant margin of error, perhaps .002? The flat
plate's 0 lift drag stayed about the same as the Re increased to 300k,
but the foil's got better. If you really want to get obsessive, you can
investigate foil drag with Xfoil, which is supposed to be fairly
accurate. But I think it's hardly worth it for us, especially since
Xfoil doesn't handle really thin foils like a flat plate.
I agree with most of what you say about the impracticality of laminar
foils. However, the Reynolds number "window" is usually pretty wide, at
least when you go into Re higher than ours. I don't have a lot of data
so I'm not sure if it comes down into the range we use or not. The one
little bit I have shows some pretty low drag at 700k, so I suspect so.
That's well within our range, maybe something like a Chebacco at 5
knots? (I've been lazy and poking around the web at various Re
calculators to determine that, depending on water temperature, etc. the
Re of a 1 ft chord surface at 3.4mph might be something like 350k. WOuld
have been a lot faster to dig into my books if I wasn't a web addict)
Anyway, Xfoil would clear it up but there's no point since these foils
are so impractical. It's the angle of attack, or "laminar bucket" that
can be small on some laminar foils. If your canoe rudder is really at
low angles of attack, and you want to be obsessive, then this is the
application.
I'm convinced that at less than hull speed, and especially when going
upwind, foil shape will have a significant effect on overall drag. At
higher speeds it's main perceptible effect will be on handling. On
catamarans and really long, narrow hulls, the foil shape will probably
have a more significant effect at speed because wavemaking is less
significant.
Let's look at a hypothetical example: a 13' boat with 25 square feet of
bottom and 2.5 square feet (small) of board. Doesn't seem significant
yet, but the bottom is operating at 10X the Reynold's number which will
make for much less drag. Plus, though I'm not ENTIRELY sure here, it's a
better shape than a flat plate. Then you're asking the boards for some
lift, which on a flat plate GREATLY increases their drag. (at 300k in
that wind tunnel, a Cl of 0.35 about doubles the drag. A quick look at
Hoerner's "Fluid Dynamic Drag" indicates that we can probably expect a
Cd for the bottom in the range of .003 to .0045, depending on which
assumptions you make and which graph. So if we take an average, we can
expect at neutral helm with no side force that the boards will create
more than a third of the non-wavemaking drag from the hull. I'll bet
that's what Phil feels when rowing boats with boards. I'll further bet
that when just gusting along that's something like 1/4 or 1/5 of the
total drag.So, assuming 1/4 of the drag and given the same propulsive
force, you might go 12% faster by eliminating the boards. In reality,
eliminating intersection drag would probably make things even better.
Given the same power (as might be estimated for rowing) you'd still pick
up 8%. And that's compared to small boards. If you switched to NACA 009
and kept the intersections clean, I guess it might be 6% faster for
constant force and maybe 4% faster for constant power.
Upwind, the difference might be more, (keep in mind the awful
performance of flat plates when lifting) but now things are getting
complicated with stuff like induced drag.
Anyway, the reason I went thru the above, apart from curiousity and fun,
is that I have a bunch of paperwork to do, and I really should get to it.
P.S. If you put a flap on the back of your rudder and daggerboard you
could make them narrower and go faster. But what a pain in the neck.
comparing flat plate to inappropriately fat foil at really low (<50k or
100k?) Reynolds numbers, though there are also some thinner sections
good in that regime. I looked at a couple of data plots from the
Princeton low Reynolds number wind tunnel tests done by Michael Selig
and co. and published in Soartech 8. If you really want to, you can get
the data from:
http://www.aae.uiuc.edu/m-selig/ads/princeton_tests.html
However, it is of limited value to us as most of the foils tested are
intended for wings and are not symmetrical. Not only that, but most of
the Reynolds numbers investigated are lower than ours. There were
further tests that appeared later which tested a limited number of other
foils. As I recall from the graphs (I didn't bother to check closer than
that), at Re of 300k, the Cd of the NACA 009 was arouind .007 and the Cd
of the flat plate was around .013. No point in getting closer as the
tests do have a significant margin of error, perhaps .002? The flat
plate's 0 lift drag stayed about the same as the Re increased to 300k,
but the foil's got better. If you really want to get obsessive, you can
investigate foil drag with Xfoil, which is supposed to be fairly
accurate. But I think it's hardly worth it for us, especially since
Xfoil doesn't handle really thin foils like a flat plate.
I agree with most of what you say about the impracticality of laminar
foils. However, the Reynolds number "window" is usually pretty wide, at
least when you go into Re higher than ours. I don't have a lot of data
so I'm not sure if it comes down into the range we use or not. The one
little bit I have shows some pretty low drag at 700k, so I suspect so.
That's well within our range, maybe something like a Chebacco at 5
knots? (I've been lazy and poking around the web at various Re
calculators to determine that, depending on water temperature, etc. the
Re of a 1 ft chord surface at 3.4mph might be something like 350k. WOuld
have been a lot faster to dig into my books if I wasn't a web addict)
Anyway, Xfoil would clear it up but there's no point since these foils
are so impractical. It's the angle of attack, or "laminar bucket" that
can be small on some laminar foils. If your canoe rudder is really at
low angles of attack, and you want to be obsessive, then this is the
application.
I'm convinced that at less than hull speed, and especially when going
upwind, foil shape will have a significant effect on overall drag. At
higher speeds it's main perceptible effect will be on handling. On
catamarans and really long, narrow hulls, the foil shape will probably
have a more significant effect at speed because wavemaking is less
significant.
Let's look at a hypothetical example: a 13' boat with 25 square feet of
bottom and 2.5 square feet (small) of board. Doesn't seem significant
yet, but the bottom is operating at 10X the Reynold's number which will
make for much less drag. Plus, though I'm not ENTIRELY sure here, it's a
better shape than a flat plate. Then you're asking the boards for some
lift, which on a flat plate GREATLY increases their drag. (at 300k in
that wind tunnel, a Cl of 0.35 about doubles the drag. A quick look at
Hoerner's "Fluid Dynamic Drag" indicates that we can probably expect a
Cd for the bottom in the range of .003 to .0045, depending on which
assumptions you make and which graph. So if we take an average, we can
expect at neutral helm with no side force that the boards will create
more than a third of the non-wavemaking drag from the hull. I'll bet
that's what Phil feels when rowing boats with boards. I'll further bet
that when just gusting along that's something like 1/4 or 1/5 of the
total drag.So, assuming 1/4 of the drag and given the same propulsive
force, you might go 12% faster by eliminating the boards. In reality,
eliminating intersection drag would probably make things even better.
Given the same power (as might be estimated for rowing) you'd still pick
up 8%. And that's compared to small boards. If you switched to NACA 009
and kept the intersections clean, I guess it might be 6% faster for
constant force and maybe 4% faster for constant power.
Upwind, the difference might be more, (keep in mind the awful
performance of flat plates when lifting) but now things are getting
complicated with stuff like induced drag.
Anyway, the reason I went thru the above, apart from curiousity and fun,
is that I have a bunch of paperwork to do, and I really should get to it.
P.S. If you put a flap on the back of your rudder and daggerboard you
could make them narrower and go faster. But what a pain in the neck.
>Message: 12
> Date: Wed, 22 Oct 2003 19:39:54 -0400
> From: craig o'donnell <dadadata@...>
>Subject: Re: fairing rudder (laminar, etc)
>
>>don't have this problem. Here's a URL for a picture of a laminar section:
>>http://www.aae.uiuc.edu/m-selig/ads/afplots/naca66-018.gif
>OK, these are the "low-drag" NACA sections. They are low-drag *if*
>you can keep them in a laminar flow regime. They're typically limited
>to a certain Reynolds number "window" to get the low-drag
>characteristics. And low drag is meaningless to the average
>non-racing maniac, because a small boat generates a LOT more power
>under sail than the drag penalty of an unsophisticated foil...
>besides the fact that a sailboat cannot adjust its speed to maintain
>a narrow range of Reynolds number over a foil surface.
>
>>http://www.aae.uiuc.edu/m-selig/ads/afplots/naca0018.gif
>>Note that I've picked thicker sections than we might usually use in
>>order to exaggerate differences. I suspect going past 12% isn't worth
>>it, but that's just an impression.
>
>
>
>>For instance, I have
>>a book with some wind tunnel data for NACA 009 and flat plate at
>>Reynolds number of 300,000 (equivalent to a small, slow boat). At 0
>>lift (i.e. neutral helm on a rudder) the NACA 009 has about half the
>>drag of the flat plate.
>
>
>
>Mmmm. What book is this, Lincoln? Texts I've read concede that a flat
>plate has a poor lift/drag ratio outside of a couple degrees either
>way off axis, but has low drag (ie a benchmark value) at zero
>incidence. Hugh Horton and Meade Gougeon chose flat plate rudders for
>their sailing canoes for exactly this reason (they try to trim sail
>and leeboard so there's no rudder involved). If I'm wrong I'd like to
>know why I got this mistaken impression.
>
>don't have this problem. Here's a URL for a picture of a laminar section:OK, these are the "low-drag" NACA sections. They are low-drag *if*
>http://www.aae.uiuc.edu/m-selig/ads/afplots/naca66-018.gif
you can keep them in a laminar flow regime. They're typically limited
to a certain Reynolds number "window" to get the low-drag
characteristics. And low drag is meaningless to the average
non-racing maniac, because a small boat generates a LOT more power
under sail than the drag penalty of an unsophisticated foil...
besides the fact that a sailboat cannot adjust its speed to maintain
a narrow range of Reynolds number over a foil surface.
>http://www.aae.uiuc.edu/m-selig/ads/afplots/naca0018.gifMmmm. What book is this, Lincoln? Texts I've read concede that a flat
>Note that I've picked thicker sections than we might usually use in
>order to exaggerate differences. I suspect going past 12% isn't worth
>it, but that's just an impression.
>For instance, I have
>a book with some wind tunnel data for NACA 009 and flat plate at
>Reynolds number of 300,000 (equivalent to a small, slow boat). At 0
>lift (i.e. neutral helm on a rudder) the NACA 009 has about half the
>drag of the flat plate.
plate has a poor lift/drag ratio outside of a couple degrees either
way off axis, but has low drag (ie a benchmark value) at zero
incidence. Hugh Horton and Meade Gougeon chose flat plate rudders for
their sailing canoes for exactly this reason (they try to trim sail
and leeboard so there's no rudder involved). If I'm wrong I'd like to
know why I got this mistaken impression.
--
Craig O'Donnell
Sinepuxent Ancestors & Boats
<http://freepages.genealogy.rootsweb.com/~fassitt/>
The Proa FAQ <http://boat-links.com/proafaq.html>
The Cheap Pages <http://www2.friend.ly.net/~dadadata/>
Sailing Canoes, Polytarp Sails, Bamboo, Chinese Junks,
American Proas, the Bolger Boat Honor Roll,
Plywood Boats, Bamboo Rafts, &c.
_________________________________
-- Professor of Boatology -- Junkomologist
-- Macintosh kinda guy
Friend of Wanda the Wonder Cat, 1991-1997.
_________________________________
---
[This E-mail scanned for viruses by friend.ly.net.]
There is indeed such a thing as a laminar foil. These are meant to have
as much laminar flow as possible. They tend to have the thickest point
pretty far back and a relatively small leading edge radius. They might
have less favorable stall characteristics, although I recall some have
been designed not to. And they are very finicky creatures requiring more
accuracy and cleanliness than we might want to provide. Some of them
behave badly if forced to operate in turbulent flow, but some of them
don't have this problem. Here's a URL for a picture of a laminar section:
http://www.aae.uiuc.edu/m-selig/ads/afplots/naca66-018.gif
Here's a non-laminar foil:
http://www.aae.uiuc.edu/m-selig/ads/afplots/naca0018.gif
Note that I've picked thicker sections than we might usually use in
order to exaggerate differences. I suspect going past 12% isn't worth
it, but that's just an impression.
I think since the non-laminar foils are so much less sensitive than
laminar it makes sense to use them in boats. Realistically, no one but
an obsessed racer is going to wipe down the boards and fill the
scratches every time out. And few people are going to achieve the
required building precision.
I suspect that, while the differences between any decent foil and a
flat plate at our Reynolds numbers are probably very noticeable, the
differences between a regular foil and a laminar one probably aren't,
unless the laminar one starts to exhibit flaky behavior. One advantage
of a real foil is that you can probably go a little bit smaller on the
size of the surface without stalling any worse, at least for a deep,
skinny one (like on a Hobie). Another is that, perhaps contrary to
intuition, flat plates have considerably more drag. For instance, I have
a book with some wind tunnel data for NACA 009 and flat plate at
Reynolds number of 300,000 (equivalent to a small, slow boat). At 0
lift (i.e. neutral helm on a rudder) the NACA 009 has about half the
drag of the flat plate. At higher Reynolds numbers the difference will
probably be more. And the flat plate gets worse a lot faster than the
foil as you turn that rudder. And as we know from Bolger's comments
about rowing (I can't remember where they appear), daggerboards, etc.
create a large part of the boat's drag.
I once calculated Reynolds numbers for radio control model yacht keels,
getting 50,000 to 200,000 or so, so I think our boats probably range
from a few hundred thousand on up, a much more favorable range.
As far as making foils, I'm sure it's a pain. But maybe not as bad as we
think. At www.charlesriverrc.org in the section on the Apogee, it
describes how to sand an accurate foil into a slab of balsa wood. I've
done this. It's a pain, but I was trying for a very high level of
accuracy on a pretty small piece of wood. I think this technique would
be much easier on a nice big boat foil if it was adapted to use a plane.
Another trick is to make a hot wired foam core and cover it with glass
and epoxy, but that's probably a pain to figure out if you haven't done
it already, tho it only takes a wooden bow, some music wire, jumper
cables, four hands, and two car batteries (or a battery and a charger)!
You could make ribs and cover them with thin plywood, but this could be
fragile and laborious.
Compared to building another boat, real foils might be a worthwhile way
to get better handling and go a little bit faster.
as much laminar flow as possible. They tend to have the thickest point
pretty far back and a relatively small leading edge radius. They might
have less favorable stall characteristics, although I recall some have
been designed not to. And they are very finicky creatures requiring more
accuracy and cleanliness than we might want to provide. Some of them
behave badly if forced to operate in turbulent flow, but some of them
don't have this problem. Here's a URL for a picture of a laminar section:
http://www.aae.uiuc.edu/m-selig/ads/afplots/naca66-018.gif
Here's a non-laminar foil:
http://www.aae.uiuc.edu/m-selig/ads/afplots/naca0018.gif
Note that I've picked thicker sections than we might usually use in
order to exaggerate differences. I suspect going past 12% isn't worth
it, but that's just an impression.
I think since the non-laminar foils are so much less sensitive than
laminar it makes sense to use them in boats. Realistically, no one but
an obsessed racer is going to wipe down the boards and fill the
scratches every time out. And few people are going to achieve the
required building precision.
I suspect that, while the differences between any decent foil and a
flat plate at our Reynolds numbers are probably very noticeable, the
differences between a regular foil and a laminar one probably aren't,
unless the laminar one starts to exhibit flaky behavior. One advantage
of a real foil is that you can probably go a little bit smaller on the
size of the surface without stalling any worse, at least for a deep,
skinny one (like on a Hobie). Another is that, perhaps contrary to
intuition, flat plates have considerably more drag. For instance, I have
a book with some wind tunnel data for NACA 009 and flat plate at
Reynolds number of 300,000 (equivalent to a small, slow boat). At 0
lift (i.e. neutral helm on a rudder) the NACA 009 has about half the
drag of the flat plate. At higher Reynolds numbers the difference will
probably be more. And the flat plate gets worse a lot faster than the
foil as you turn that rudder. And as we know from Bolger's comments
about rowing (I can't remember where they appear), daggerboards, etc.
create a large part of the boat's drag.
I once calculated Reynolds numbers for radio control model yacht keels,
getting 50,000 to 200,000 or so, so I think our boats probably range
from a few hundred thousand on up, a much more favorable range.
As far as making foils, I'm sure it's a pain. But maybe not as bad as we
think. At www.charlesriverrc.org in the section on the Apogee, it
describes how to sand an accurate foil into a slab of balsa wood. I've
done this. It's a pain, but I was trying for a very high level of
accuracy on a pretty small piece of wood. I think this technique would
be much easier on a nice big boat foil if it was adapted to use a plane.
Another trick is to make a hot wired foam core and cover it with glass
and epoxy, but that's probably a pain to figure out if you haven't done
it already, tho it only takes a wooden bow, some music wire, jumper
cables, four hands, and two car batteries (or a battery and a charger)!
You could make ribs and cover them with thin plywood, but this could be
fragile and laborious.
Compared to building another boat, real foils might be a worthwhile way
to get better handling and go a little bit faster.
> laminar foil - have the thickest pointVery interesting post. Perhaps you have explained something that was
> pretty far back and a relatively small leading edge radius.
a mystery, at least to me. There was a story in Multihulls magazine
some years ago about someone in the NW USA who rehabilitated a Danish
trimaran than had come to grief there (due to failure of a wire, as I
recall). The new owner (it might even have been Kurt Hughs) found the
original rudder overly sensitive and remarked that the section was
thickest more than 50% back from the leading edge. A normal section
calmed down the steer quite a lot.
I always wondered why someone would have put that section on a rudder
in the first place. You've given a credible reason.
Peter