Jib question

My personal experience at work is exactly the same. 1 in 10 successes is a very good record.
Rutan gave himself a chance to succeed by dismissing a large number of assumptions in current design. Sometimes those were well founded assumptions, but as you said, it was one way to ask a different question and see where it went.

 

Yes Rutan has broken down a lot of technical barriers. He's designed some very sweet yachts, planes & space craft. I worked with him on a prototype jet aircraft project to refine the lift on it.

He works with his own brains and validates his info using digital wind tunnels, then proceeds to build prototypes. If the up-front analysis is done right that saves a lot of the prototype/tooling costs. If you've got a concept for a wind-based product and you want to optimize it...he's the man to call. By the time he's done with it you'll likely have a world record setting airfoil on your hands.

http://www.scaled.com/projects/

For this case I would run a sample model of any scale by an aerodynamicist. The variables can be entered (wind speed, wind direction, boat heading, boat draft, etc. ), and insert some surfaces to represent the variable locations of the sails. Do we have anybody on the forums who have CFD tools for sail aerodynamic analysis?

It would also be good to know the software used as I would be interested in learning it myself. I did some quick checking and found a tool called MacSail, but can't seem to track down who sells it, what it can do, price, etc.

 

Of course we do, and he has been actively partecipating to this thread from the beginning.
His name is Mikko Brummer, and he makes his living with stuff like this: http://www.boatdesign.net/forums/hydrodynamics-aerodynamics/ultimate-validation-44809.html

 

Very cool. I would toss this one by Mikko.

FYI, Burt Rutan used to work at McDonnell Douglas years ago before they merged with Boeing. Burt used to stick airfoils out the top of his sun roof to see what they would do as he drove to/from work. This was before the days of CFD software of course. For subsonic testing like this such practical testing is certainly an option if any working models are available.

OK, signing out.

 

No comment..

 

If Mikko gets some spare time to post some results "quickly" I think that would do the job.

There's an old saying "It's time to shoot the engineer and build it." Conversely, it's time to analyze it and shoot the engineer.

In all cases the engineers are cannon fodder so all is good

 

Interesting story. Nothing in the biography of Burt Rutan on Wikipedia shows him working for McDonnell Douglas. http://en.wikipedia.org/wiki/Burt_Rutan Perhaps you are thinking of someone else. And CFD has been around since the 1960's or even earlier depending on what you classify as CFD.

 

I agree: even when CFD is available it is sometimes useful to just build a model and test it.
I bet Rutan didn't get the strange looks we did with our outdoor laboratory

 

Yes, Burt & a well known aircraft & friend of Burts named Jim Bede used to work at McDonnell Aircraft when they were young chaps. Jim told me the story himself. More on Jim below...he's famous for his varioius BD aircraft.

http://en.wikipedia.org/wiki/Jim_Bede

He designed/build the jet in that James Bond Octopussy film.

http://en.wikipedia.org/wiki/File:Acrostar.jpg

 

Whatever works Leo! I do like that outdoor lab. You ought to apply for a position on Myth Busters!

 

Rutan told me that he worked on the F-4 for McDonnel. Never said what he did. This was while he was under contract to show us how to revolutionize the design of the F-35.

Last Rutan comment - we are getting far from the thread. I don't mind personally, but.....

 

In few words, Chuck Norris of aeronautics...

 

"From 1965 to 1972 Rutan worked for the U.S. Air Force at Edwards Air Force Base as a flight test project engineer, working on nine separate projects including fighter spin tests and the LTV XC-142 VSTOL transport. Shortly after, he became director of the Bede Test Center for Bede Aircraft, in Newton, Kansas, a position he held until 1974." from http://www.spaceref.com/news/viewpr.html?pid=25222

Wikipedia entry says he worked on the F4 during that part of his career. So looks like he was an Air Force employee and one project involved the McDonnell F4.

Irrelevant to the topic of this thread though.

 

OK, I finally have had an opportunity to review the WU report. I wonder if anyone else posting here has?


The first thing to note is the fact that nothing in this report indicates a boat will be able to sail one degree higher for each degree the tack is moved to leeward.


A second observation worth mentioning is a comment in the introduction. It states the tests were supervised by the client. I think this is a significant comment.


There is a comment about a photo album accompanying the report. I do not see it. Seeing the photos would be helpful with analysis. A data sheet for the data used to produce the graphs would be helpful as well.

The scale of the model being tested is not identified.

The description of the mast/sails used in the testing raises a large red flag. It is mentioned that the wing mast was deforming under some conditions. However, this does not seem to be noted in the data. Actually, with no data sheet to accompany the charts it makes the report rather difficult to verify.


It is also mentioned the mainsail had issues initially, but these seem to have been solved to the satisfaction of the participants. However, it does not sound as if the setup was optimal.


Of all the configurations tested I will limit comments to the “Wingmast, Main, and Jib” configurations.

The first thing to note is the comment regarding the poor cut of the jib. There are comments about the inappropriate cut at “low apparent wind angles” and panting at “high apparent wind angles”. Sadly the terms “low” and “high” are not quantified.

I have considerable experience in laboratory testing, including using some fairly sophisticated Taguchi Arrays in experiments. I can say that given a known dependent variable was not optimal or even representative I would have stopped such an experiment at the start and re-started when I had the appropriate test model. Carrying on a test with this type of issue is a waste of time and money.


The report mentions that measurements were taken at 2 or 3 angles for each sail configuration. However, the data seems to indicate many more points were measured. Sadly there was not a rigorous method for testing like numbers of angles for the configurations.

The report cites the lift/drag (Cl/Cd) as 5.8 when the tack was set at the bow and on the CL. The report does not indicate the sheeting angle, but does mention the sheet tension was “increased”. There is no mention of why the sheet tension was increased, and if this increase produced an optimal sail shape.

The report then cites that with the jib tack moved to the leeward deck the L/D increased to 6.1. There is no mention of how many degrees the tack was moved. There is no mention of the sheeting angle with the tack to leeward. There is no mention of the amount of sheet tension used. All of these variables need to be available in order to ensure we are looking at apples and apples.


The tack was then moved aft 6”. Do we assume that is an actual 6”, not a scale 6”? If actual, how far is that to scale? What was the jib overlap before the move, and after?

With the tack on CL and 6” aft the L/D is slightly higher, but probably within the noise. It is noted the leeward setup again provided a small increase in L/D. Again, there is no data available regarding the amount the tack is moved to leeward, the sheeting angles at each position, and the amount of sheet tension required.

Now these L/D numbers seem to be from the data point at 25 degrees AWA (“upwind sailing”). Seeing as how this report is from 1986 I think that 25 degrees apparent wind would have been a pretty aggressive angle. Considering the boat has a WL length of 40 feet the Theoretical Hull Speed would have been about 8.5 knots. So with a TWA of 40 degrees, a TWS of 17 knots, and a Boat Speed of 8.5 knots the AWA would be about 30 degrees. To get an AWA of 25 degrees would mean tacking through less than 70 degrees while sailing at hull speed. That’s IACC type performance, and pretty much unheard of back in 1986. In fact, through the haze of the years I seem to recall seeing AWAs in the 27 or 28 degree range when racing back then on some pretty “Grand Prix” type boats.


Regardless, let’s look at the data available.

In Table 2 we are interested in Runs 3.2, 3.3, 3.4, and 3.6. Run 3.5 is a repeat of run 3.4 with the jib set “wider”. Can we assume this means the sheeting angle? What was the original and what was the “wider” setting?

One thing to note is the AWA used for runs 3.4 and 3.6. At 23 and 22 degrees we are talking about very close winded sailing. Given the description of the jib having a draft-forward issue I can’t imagine this is optimal. I think maybe this is a typo, since in Fig. 12 it appears the data points are at 20, then 25 degrees.


Let’s now look at Fig 4. This graphically shows the results of runs 3.2 and 3.3.

One thing I wonder about is the mast rotation setting. It seems to me that if the rig is tested with the tack of the headsail on CL and again with it “to leeward” the optimal setting of the mast rotation should be different for the two runs. So if 16 degrees is optimal for the CL tack, then it would be sub-optimal for the “leeward” tack position, and vice versa.

Another thing to note is the claim the points are annotated with AWA. It appears only the endpoint is annotated. So it does make it a bit difficult to datamine, due to the resolution of the report as presented. I would guess the graphics for the report were probably done in Lotus 123 back in 1986, and it could be a pain in the a$$ to work with at times.

Looking at the graph we can see that at the 25 degree markers the two configurations are nearly atop the other. So we have to wonder a couple of things. Have the settings been optimized? What is the accuracy of the test equipment at this scale? I know that in 1986 with small scale models the accuracy of hydro testing was outside of the delta percentage shown here.


On to Fig. 5, showing runs 3.4 to 3.6.

The “upwind” data point for the jib on CL run is clearly an outlier. It is so far off the curve compared to all other points on all other runs that it must be questioned.

If we disregard the point and look instead at the curve we see the configuration with the tack on CL is far superior to the tack to leeward in the 25 to 30 degree range of “upwind” sailing. In fact, the 30 degree point shows the highest Cl of any of the 4 configurations I am discussing. The Cl/Cd at both 25 degrees and 30 degrees appears to be better with the tack on CL in this 6” aft configuration. It is still not as good as the 6.1 claimed for the Run 3.3 configuration. But are any of these setups optimized?


Finally, when checking the Cl/Cd from Fig 4 and Fig 5 I cannot come up with the 6.1, 5.8, or 5.9 numbers stated in the report. Perhaps someone can have a look at the graphs and show me where I have erred (data sheet attached).


To conclude, I suppose someone could read this report and believe the Conclusions. Others who know something about test methodology might not be so sure of the claims.

 

Eric said a few pages back that his point was "to spark everyone’s thinking that, just because boats are built symmetrically about a fore/aft centerline, and that we stick the tack of the jib on the bow, it does not mean that that is the best place to put it. Maybe we should move it around. Not once, in all of the documented studies that I am aware of, has anyone attempted to test whether the centerline of the boat is the best place to put the jib. Why should it always be on the centerline? Why should the jib always be well forward of and in line with the main? Why can’t it be off to one side to open up the slot more? Very few people have tried it, and although those that have reported improved performance, no one has studied this concept scientifically—that is, in a properly designed and executed study. If they have, I would really like to know about it.

I put it to you and all of the readers here that we should experiment with putting the jib tack in different places to generate more force from a given sail area and selection of sails by improved slot shape and improved air circulation around the rig. That is what I mean when I say “thinking outside the box”—the box is our collective mental inertia and reluctance to change the rig outside of normal racing rules—the major racing rules don’t allow you to move the jib tack off the boat’s centerline, but that is a political rule, not an aerodynamic or scientific rule. From what I have seen in my experience, there is a real benefit to putting the jib in different places, as I have described."

Some dude called Bruce had a "gybing bowsprit" that moved his forestay to leeward. This was in a 12 Foot Skiff in the '60s. He raced about the same time as some engineer called Russell like that sailed the same boats. Both were Kiwis.

I seem to recall that Bruce and Russell designed a boat or two. But the gybing bowsprit idea was abandoned even in 12s, where it was allowed.

Of course, the idea of moving headsail tacks around was known at the time, because it was regularly done with tallboy staysails. Moving headsail tacks to leeward is also done on multis. In a slightly different but similar area, mizzen staysails were tacked to windward in boats like Stienlager IIRC. Plenty of racing sailors DON'T suffer from mental inertia, which is well proven by the fact that these things have been tried and rejected by people as well known as Farr.

The belief that "improvements in boat design are overly restricted by rules and conservatism" is enormously over-stated. IMHO the belief itself is conservative, IMHO. It's something that has been said so many times that people believe it.

Arguably, the reality is that there are many, many areas in which free-thinking design has lots of room to play. If one actually goes to the effort to study what has been done in these classes, one finds that most of the "revolutionary improvements" supposedly prevented by rules and conservatism have in fact been tried and have in fact failed.

The problem is that some people are so conservative that they can't bring themselves to study what happens in the classes and sailing cultures where (almost) free reign is allowed, and they can't bring themselves to examine whether the "improvements in boat design are overly restricted by rules and conservatism" theory is true or not. Looking hard into these areas shows that most mooted ideas have already been tried and have already been found not to work, or to be too much trouble in other ways.

It's not that the rules and conservatism stop all these things, it's just that most of them don't really work well enough to make them worthwhile when you take into account other objective criteria like complication and obsolescence. And that's coming from someone who spent last season racing shorthanded multis offshore and will be doing sailing what it arguably the least-restricted boat class in the world this season.