# real world measured drag coefficient, heeled and straight, of a common glass keelboat

Discussion in 'Sailboats' started by peterchech, Dec 9, 2011.

1. Joined: Aug 2010
Posts: 241
Likes: 9, Points: 0, Legacy Rep: 127
Location: new jersey

### peterchechSenior Member

So my younger brother is in engineering school, and for his hydrodynamics class/lab he designed an experiment for my 1981 Hunter 25. He wanted to calculate the drag coefficient of the hull, when upright and when heeled. With his lab partners (about 4 students) and his lab professor (who wanted to come along and observe), with myself at the helm of the boat and my girlfriend's father driving the dinghy, we took about 3 hours in conducting the experiment.

Using my dinghy, we tied off to the bow and after several failed attempts figured out how to tow the boat in a straight line consistently. We then towed the boat with the dinghy, measuring water speed, apparent wind speed/direction and force (in pounds) required to maintain that water speed at the time my brother yelled "mark". We repeated the experiment about 20 times, going into the wind and then with it to reduce/average out the effect of the 5-15 knot winds experienced in our harbor. The water was almost completely flat. The experiment was done both upright and with the boat heeled about 10-15 degrees (he measured the angle of heel as well on each mark). It was funny seeing a bunch of young engineering students hanging off the shrouds to heel the boat over. They didn't look too comfortable

Other than my brother lighting up a cigarette at a fuel dock everything went well and usable data was obtained.

The drag coefficient of this boat was calculated to be .34 upright, and only .22 heeled. So, its shape is 35% more efficient when heeled (at between 3-4 knots) than when upright.

Well he got an A, and his professor was impressed enough to submit him for a scholarship due to this experiment (he got the scholarship BTW), so academically it was a success. It seems few engineering students are so creative in designing experiments.

Since the boat has not been bottom painted in 2 years, nor power washed in 1 year, it has some seaweed and minor barnacle growth on the bottom. He will be repeating the experiment in the Spring after new bottom paint for comparison. I can't wait to see how much more efficient the hull is then!

2. Joined: Oct 2009
Posts: 4,949
Likes: 464, Points: 83, Legacy Rep: 1485
Location: Midcoast Maine

### DCockeySenior Member

Drag coefficient based on what area?

3. Joined: Dec 2003
Posts: 319
Likes: 46, Points: 38, Legacy Rep: 460
Location: Filia pulchra Lubecæ

### Olavnaval architect

And at which speed?

4. Joined: Aug 2010
Posts: 241
Likes: 9, Points: 0, Legacy Rep: 127
Location: new jersey

### peterchechSenior Member

I am not an engineer, and I do not know what the hull area was calculated to be.

The speed was about between 3-4 knots.

5. Joined: Oct 2009
Posts: 4,949
Likes: 464, Points: 83, Legacy Rep: 1485
Location: Midcoast Maine

### DCockeySenior Member

Question for your brother. Did you use upright wetted surface area, frontal area, or ?

6. Joined: Dec 2003
Posts: 319
Likes: 46, Points: 38, Legacy Rep: 460
Location: Filia pulchra Lubecæ

### Olavnaval architect

IMHO the magnitude of those drag coefficients indicates that frontal area (= main section area) was used.

peterchech, is it possible (with the permission of your brother, of course) to post a report of this experiment? I think this is an interesting topic and would love to read further details about it, e.g. how they towed the sailboat without exposing it to the wake and propeller stream of the towing dinghy etc. Thanks in advance!

7. ### Paul BPrevious Member

This is clearly GIGO. One year of growth on the bottom and none on the topsides, and someone is surprised there is less drag when some of the growth is removed (during heeling)?

A decent professor would have given the student an F for this effort.

This "experiment" is not controlled in any way that would provide useful information. It does not represent anything an engineering student should be congratulated for.

8. Joined: Oct 2007
Posts: 2,936
Likes: 148, Points: 63, Legacy Rep: 1593
Location: Arlington, WA-USA

### PetrosSenior Member

It also seems to me that if he used frontal area (which could be accurately determined from hull drawings), you could have different areas when up right and heeled. So did he normalize the coefficient for the heeled area?

Also, how do you accurately account for variation is drag due to wind? If you are only pulling at 3-4 knots, and you have a similar size wind that is varying, seems it could introduce a rather large error as well. wind drag on all the rigging, especially with a bunch of student hanging in it, is not insignificant. It would be better to do this in dead calm conditions if possible.

We would have to see the full report, and his methodology, to know if it resulted in any meaningful data. Clever idea for an experiment, but a bottom wash at least should have been done before the experiment.

9. Joined: Jan 2006
Posts: 3,147
Likes: 308, Points: 83, Legacy Rep: 1279
Location: Lakeland Fl USA

It is fairly common to have wetted surface diminish as heel angle is increased. A boat operating at the stated speeds will have wetted surface as the dominant factor. That is presuming that the boat is long enough to have been operating at a very small S/L ratio. Of course a foul bottom renders the whole experiment moot.

Nonethe less we can commend the student for his clever contrivances. Keep in mind that the student is a student. That implies that he is in a learning process and he can be forgiven for lack of detail knowledge.

Let us hope that he will never again light a cig at the gas dock. If the dockmaster had been around, the student would have been keelhauled.

10. Joined: May 2004
Posts: 2,051
Likes: 211, Points: 63, Legacy Rep: 611
Location: Michigan, USA

### sharpii2Senior Member

You do have a good point (which the author himself brought up), but don't you think you're being a bit severe?

Amongst the many young people and not so young people I have met, many with degrees, few of them have either the inclination or the imagination to design and set up their own experiment.

When you say the experiment wasn't controlled in any way, what do you mean?

I thought the test was to see if the boat had less drag when heeled than when up right. If we except that as the goal of the experiment, then the experiment was well executed and delivered credible results. The boat being towed up right could be considered the 'control'.

It did prove that that particular boat, bottom fouling included, had less drag when heeled than when up right.

Next year, after the bottom is scrubbed and repainted, the experiment could be repeated. If the results are similar, that will prove that that particular class of sailboat (hunter 25's) has less drag heeled than upright.

I think we have far too many people in this world with 'book' knowledge, knowledge based on other people's experience and insight. A lot of this 'book knowledge' includes actual experiments designed by somebody else. Nothing wrong with that, but, if that's as far as it ever goes, we in trouble.

Their are far fewer people who are able willing to test common assumptions (which are most likely true, but may not be) and have creativity and the analytical capability to design an experiment that actually tests what it is supposed to test.

I think that is the reason the professor heaped so much praise on the students. It was not so much the quality of the execution of the experiment as it was the quality of its design.

11. Joined: Oct 2007
Posts: 2,936
Likes: 148, Points: 63, Legacy Rep: 1593
Location: Arlington, WA-USA

### PetrosSenior Member

Another thing occured to me; is a situation like this total drag is the important issue, not the drag coefficent. The thrust from the sails has to over come the drag of the hull in the water (plus the drag of the deck and rigging through the air, but that is a separate issue), the total drag is all that is realvent. The drag coefficient is only used to normalize one shape against another, but on this particular boat, the total drag of the hull when heeled vs. upright is what was measured. And if most of the boats sailing life it has growth on the bottom of it, than the condition of the hull is exactly as a real world condition. Only racing boat are maintained to always be in top racing conditions, but recreational boats can be all over the map in terms of condition.

12. Joined: Aug 2010
Posts: 241
Likes: 9, Points: 0, Legacy Rep: 127
Location: new jersey

### peterchechSenior Member

Thanks Sharpii I think Paul B needs to get a life and stop being negative. My brother is not a naval architect, he is a civil engineering major and this was an experiment for his fluid dynamics class. Most of the students did a textbook experiment using pipes of differing materials and diameters. He actually got out into the field and took on a challenging experiment.
I think rudder drag was accounted for in the experiment, because as the boat heeled I had push the stick fairly hard to keep the boat straight. If not, it would turn hard and drag the dinghy with it.

Anyway, I spoke to my brother last night about the experiment, and clarified a few things. The average speed we got the boat to was 4 ft/sec. This is 2.7 mph, which is actually under half of hull speed.

The formula for force accounts for area twice. The coefficient of drag is the amount of drag created by the hull, between form drag and skin friction drag. But the force equation is as follows:

Force (drag) = 1/2 *p(density of water)*v(squared)*Cd*A(cross sectional area)

So while skin friction is accounted for in the Cd, cross sectional area at the widest/fattest point is another big thing to consider.

We theorized about the reasons the boat was less draggy heeled, and there are three theories:

1. Being somewhat closer to flat bottomed than to a true semi-circle, the waterline was narrower and more circular: lowering cross sectional area and also having a less draggy, narrower shape.
2. Again being somewhat flat bottomed, the immersed shape was closer to a perfect semi circle when heeled slightly, thus decreasing wetted surface and therefore skin friction.
3. Finally, by heeling the boat about 18 sq ft of clean surface area of the hull side was submerged, and 18 sq ft of somewhat barnacled surface area was no longer submerged on the other side, creating a cleaner hull overall and lowering skin friction.

The experiment will be repeated in the Spring, more out of curiosity than for a grade.

This was not a perfect experiment. But most factors were accounted for that could have seriously skewed the results. The wind was westerly, and the experiment was run over 12 times going both west and east. The apparent wind speed was also measured at each mark, and accounted for in the math. The original plan was to vary the speed of the boat, but it was so difficult to tow the boat with my dinghy that we only were able to get it to one speed. Weather helm and rudder drag were also somewhat unscientifically accounted for, because when the boat was heeled there was significant helm on the rudder and therefore drag. In fact, the hardest part of this experiment was towing the boat without the dinghy being pulled off to the side by the bigger boat. There were a few close calls! It only had a 2.5 horse ob, so that couldn't have helped lol...

I will post the actual powerpoint of the report when I get it from my brother.

Now I just need to install a canting keel for light air lol

13. Joined: Feb 2011
Posts: 3,172
Likes: 189, Points: 63, Legacy Rep: 579
Location: Ft. Worth, Tx, USA

### upchurchmrSenior Member

Peterchech,

I think your brother is to be congratulated for trying something in the real world.

There is an old book which would provide him with a useful perspective and testing methods which proved successful quite a long time ago. The Amature Yacht Society (AYRS) published a book "Design for Fast Sailing" which concentrated on sailing theory and practical testing which would allow you to understand and optimize your boat.

The testing included both sail and underwater keel, hull, and rudder testing which included the actual orientation of the boat thru the water.

The book is still available in hardback, and well worth the 10# and shipping. http://www.ayrs.org/ayrslist.htm look at line 82.

Marc

14. Joined: Aug 2002
Posts: 15,259
Likes: 954, Points: 123, Legacy Rep: 2031
Location: Milwaukee, WI

### gonzoSenior Member

Francis Herreshoff used that method to test models. He was not concerned with absolute values, but with comparisons.

15. Joined: Feb 2011
Posts: 3,172
Likes: 189, Points: 63, Legacy Rep: 579
Location: Ft. Worth, Tx, USA

### upchurchmrSenior Member

Gonzo,

Do you have a reference to how Herreshoff did his testing?

Design for fast sailing doesn't deal with models, it was based on full sized boats (small but one reference to 25').

Marc

Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.