Jump to content

Boats and foils comparison


Recommended Posts

On 12/29/2020 at 3:20 PM, MaxHugen said:

As said by others, the Mast Rotation Point is defined.

I think NZ is doing a better job of lowering the CE, by increasing twist & reducing camber in the upper sections, and increasing camber in the lower section. Thus, more drive, less drag, and lower heeling moment.

And they still have scope to power up if they need to.........

  • Like 2
Link to post
Share on other sites
  • Replies 4.1k
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

Just a few interesting bits of the straight-line performances from today: Upwind /Downwind VMGs - race 1: Upwind /Downwind VMGs - race 2: Same story in both races actually.

Thanks to weta27's pics I have created an approximation of NZ's "BFB v2" foil. Main points: Foil area is almost the same, possibly even a smidge larger. Flaps have increased in area as

OK, it sounds like there's some interest in this topic, so here goes.   Any engineering effort starts by defining the requirements.  From this figure, it looks like the average foil area is 1.64

Posted Images

On 12/27/2020 at 10:13 PM, nroose said:

I thought they can only use publicly available weather data... Probably they can interpret it differently, though. Still so much of a guess and they need to make that risk/reward judgement.

It would be interesting to know how they police that.

Link to post
Share on other sites
On 12/29/2020 at 8:45 AM, dorox said:

Ratio means VMG divided by TWS. This gets you a number that is fairly independent from the way the boat is sailed on the course and the conditions on the course. This is a good measure of the efficiency of both sail layout, boat trim and underwater appendages. According to Bethwaite's book, the AC75 is as efficient as a wind powered vehicle, but not as good as iceboat, see the image below ( ignore total drag angle on y axis). This plot is for boat speed, not VMG at broad reach, but the data above is for either upwind or downwind VMG, but still comparable.

 640px-Speed_ratio_graph_for_high-performance_sailing_craft.png

https://en.wikipedia.org/wiki/High-performance_sailing

Just so we can compare apples to apples. Here is the boat speed / TWS ratio at 135 deg TWA from the model:

BS-TWS.thumb.png.d8edcefade4e16a6eed7eac74c0d7621.png

The average of the BS/TWS ratio is pretty much where the "Sailing hydrofoil" arrow in Bethwaite 's figure points (~2.7 - 2.8). However, in the lower wind range, the ratio is closer to or even above the "wind-powered vehicle" point. I remember he made a point somewhere in the book that there's a lot of room for improvement for boats to catch up with the performance of iceboats. Even though there's a long way to go, I'm sure he'd be thrilled to see these AC75s.

 

  • Like 2
Link to post
Share on other sites
4 hours ago, dullers said:

Now here is a question. Will a boat race with different foil configurations?

Hard to imagine so Dullers.

I would have thought the motivation behind doing such a thing would be to take advantage of a prevailing wind/tide condition on either tack/gybe and whilst I am woefully lacking in knowledge about the Waitemata Harbour, I would have thought the fact that there are a number of different tracks to choose from would preclude this as a strategy since the foils are declared so far in advance.

Compare this with San Fran where the Ozzie's holiday home (Alcatraz :D) gave reasonably predictable effects or in Bermuda where you could greatly enhance the possibility of getting to mark 1 first by putting on a single higher speed foil configuration and I just don't see how it's workable with an upwind rather than reaching start.

  • Like 3
Link to post
Share on other sites

Following on from my earlier post about the analysis of the speed and VMG graphs from the various boats by the use of the first and second derivatives of their speed- which stimulated a lot of discussion! I have been thinking more about these first and second derivatives of the speed, as you do. The first derivative of the speed is obviously the rate of change of the speed which is the acceleration or deceleration of the boat- quite a useful parameter to get your head around. The second derivative , is the rate of change of the rate of change of the speed, ( still with me ?) . This has actually a name , which is " Jerk"- I kid you not, google it. So if we got a measure of this for the various boats we could assess the Jerk factor, who is the biggest Jerk? Ben, Jimmy, Deano, - I wouldn't like to prejudge that one!

Link to post
Share on other sites

Seems more important to pay attention to the opposite of the derivative, which is the integral. The area underneath the curve. Or, in the case of speed, the distance traveled over time. Having the highest (VMG) distance traveled over time means you win!

7 hours ago, oatsandbeans said:

Following on from my earlier post about the analysis of the speed and VMG graphs from the various boats by the use of the first and second derivatives of their speed- which stimulated a lot of discussion! I have been thinking more about these first and second derivatives of the speed, as you do. The first derivative of the speed is obviously the rate of change of the speed which is the acceleration or deceleration of the boat- quite a useful parameter to get your head around. The second derivative , is the rate of change of the rate of change of the speed, ( still with me ?) . This has actually a name , which is " Jerk"- I kid you not, google it. So if we got a measure of this for the various boats we could assess the Jerk factor, who is the biggest Jerk? Ben, Jimmy, Deano, - I wouldn't like to prejudge that one!

 

Link to post
Share on other sites
20 minutes ago, Kiwing said:

@amc don't they have to stay inside the measurement pyramid all the time?

How would the measurers know what they did when loaded?

Link to post
Share on other sites
7 minutes ago, amc said:

How would the measurers know what they did when loaded?

I remember in the 2003 cup they did undertake some projections to determine if the hula would touch the hull under load, but I don't think that this sort of projection is applicable here.

However, I would image a foil capable of flexing would play mary hell with turbulence.

Link to post
Share on other sites
1 hour ago, jaysper said:

I remember in the 2003 cup they did undertake some projections to determine if the hula would touch the hull under load, but I don't think that this sort of projection is applicable here.

However, I would image a foil capable of flexing would play mary hell with turbulence.

Could do but there will be flex, just how much and it can be controlled by the fap actuator rod. 

Link to post
Share on other sites
On 12/29/2020 at 1:19 PM, MaxHugen said:

I haven't found any definition of exactly where the "end" of the foil arm is, so I've presumed "zero" degrees arm cant to be as per the diagram in the rules. This would then define the angle between the foil arm and the foil base plane (bottom of the "box") to be 48°. Thus, the foil cant from the horizontal is (reported FCS cant - 36)°

image.png.23924dc373a1d81aff918362a1558234.png

Eg, arm at 60° - 36° = 24° Foil cant:

image.png.cfde8489ce8c6db2d43a87b247eeac72.png

Note that this makes no difference to foil cant between T and Y foils, since it's using the foil base plane for reference.

Perhaps I'm missing something.

  1. The internal angles of a triangle add to 180°
  2. 60 + 48 = 108 so the missing angle is 72°
  3. If the black axes are at 90°, then the angle marked 24° must be 18° (the compliment of 72°)

What is the relevance of the 60° and 24° angles?

How do you arrive at a cant angle of 36° for the foil to be horizontal?

If the fully down position is 0° and you are working off the schematic where the base of the foil wing region is at an angle of 48° to vertical, then surely the foil wing is horizontal when the cant angle is 90 - 48, i.e. at 42° (the compliment of 48°)?

Link to post
Share on other sites
16 hours ago, amc said:

Could do but there will be flex, just how much and it can be controlled by the fap actuator rod. 

13.9  Except for the permitted rotation and twist of foil flaps, no device shall be used to induce deformation in the foil; any deformation may only be the result of external forces and reactions by the FCS.

Link to post
Share on other sites
On 12/23/2020 at 3:40 PM, erdb said:

As I've speculated before, the difference may be that with the big anhedral foils, they can alter the angle of the combined foil force vector by engaging the foil flaps on the two wing halves differently, while ETNZ and AM keep lift between the two wing halves symmetric and cant the foil arm instead to adapt (or react to changing conditions mostly by sail trim). Since both foil wing halves lift equally, they can reduce the area of the foil!

Flat foils do not necessarily use the same angle on both flaps. By having a bigger angle (and presumably more lift) on the outer flap, the centre of lift is moved outboard, increasing RM.

Of course I have no idea whether teams with flat foils actually do that, but I don't think you can dismiss it out of hand based on the foil geometry.

Link to post
Share on other sites

Separately controlled flaps on either side of the main foil would be very helpful to reduce the load trying to break the foil off the foil arm when the outer foil tip is exposed.  A friend of mine broke a similar foil on a small board when turning hard with the tip exposed through the surface of the water. 
 

The outer flap could be set to more lift to balance the inner flap that has its full length working for it.  Perhaps the outer flap length can also be set to be below the surface. 
 

I suspect it is critical to largely balance the lifting forces on the cant arm, though maybe the foils with all their weight are so robust they can deal with only one side providing lift incase of a flight error?  Or maybe that is not possible as the hull will touch down before the most extreme unbalanced loads are possible?  I don’t know but I’m sure the engineers on the teams do.  

 

 


 

 

Link to post
Share on other sites
4 hours ago, RobG said:

13.9  Except for the permitted rotation and twist of foil flaps, no device shall be used to induce deformation in the foil; any deformation may only be the result of external forces and reactions by the FCS.

I don’t think that rule prevents them from preventing any deformation. So for example, flap retracted =bendy, flap not retracted = stiffer. In other words the rule prevents a competitor deforming a foil but does not prevent them from preventing said deformation. The words would have said Controlling rather than Inducing. 

Link to post
Share on other sites
2 hours ago, amc said:

I don’t think that rule prevents them from preventing any deformation. So for example, flap retracted =bendy, flap not retracted = stiffer. In other words the rule prevents a competitor deforming a foil but does not prevent them from preventing said deformation. The words would have said Controlling rather than Inducing. 

Can you explain "retracted"? Flaps can only rotate around a single hinge point.

I don't think a foil with varying flex characteristics based on flap angle is a desirable feature. Surely flaps should only used to modify lift (and consequently drag)?

Link to post
Share on other sites
1 hour ago, RobG said:

Can you explain "retracted"? Flaps can only rotate around a single hinge point.

I don't think a foil with varying flex characteristics based on flap angle is a desirable feature. Surely flaps should only used to modify lift (and consequently drag)?

Okay firstly the single hinge point can be anywhere, well almost, it doesn’t have to be coincident with a point on the actual flap. Think Fowler flap mechanism. 
Secondly a flap will have very different flexibility characteristics depending on its orientation. By design or accident it will modify the overall stiffness of the entire foil unless quite complicated hinge mechanisms are used. 
Retracted in my description means not deflected. 
Hope that adds some information. 

Link to post
Share on other sites
On 1/3/2021 at 1:13 AM, RobG said:

Perhaps I'm missing something.

  1. The internal angles of a triangle add to 180°
  2. 60 + 48 = 108 so the missing angle is 72°
  3. If the black axes are at 90°, then the angle marked 24° must be 18° (the compliment of 72°)

What is the relevance of the 60° and 24° angles?

How do you arrive at a cant angle of 36° for the foil to be horizontal?

If the fully down position is 0° and you are working off the schematic where the base of the foil wing region is at an angle of 48° to vertical, then surely the foil wing is horizontal when the cant angle is 90 - 48, i.e. at 42° (the compliment of 48°)?

How embarrassing! Of course you're correct. I scribbled this down a a bit of paper well after "beer o'clock" -  must stop doing that.

Glad you can't see my very red face. :unsure:

Link to post
Share on other sites

Dan B I think it was was speaking of deformations of flaps, trim tabs at the time, and emphasising the importance of video confirmation of what your think the flap does underwater. This bit leaves quite a bit of leeway :blink: on flap and foil flex I reckon.

 any deformation may only be the result of external forces 

Link to post
Share on other sites

This is just a quick analysis, more to follow. Used the data @dorox posted. Wanted to see how much distance is lost during tacks and how the boats compare. So far, only looked at the 1st ACWS race between LR and NZ. I clipped segments from 1 min before to 1 min after each tack. Calculated VMG relative to TWA, so this is VMG relative to the wind, not to the course. Then used the area under the VMG curve to look at distance gained.

Here are all the "clean" tacks from both boats that had a long enough segment before and after tack (black line is where the boats pass head to wind):

nz-ita1.png.e012e85d2dd68b9a56b880a63528ce8b.png

NZ has a huge advantage, but it's not really the tack, they just have a higher VMG. Here are the averages for each team:

nz-ita2.png.a13e9b6c8d56d24cda1da92644fc087d.png

What's remarkable is that they barely lose any distance through the tacks. I haven't calculated it yet, but I'm planning to compare distance gained to the gain without a tack.

Something like this - it's only hand-drawn at this point, the straight lines show how much they'd gained without a tack:

nz-ita3.thumb.png.df607264235eff571d9a08c5de3f62b0.png

While ETNZ has a better overall VMG, LR actually loses less during the tack relative to their straight-line VMG. Probably showing the pros and cons of small vs large foils. The wind was pretty good during this race ~ 15 kts, I'm sure the light wind races will show much bigger drops in VMG during tacks.

 

 

  • Like 7
Link to post
Share on other sites

My glance at the graphs suggested that the biggest difference was in the speed build after the tack.  This was particularly evident in the US-NZ match 1 ( Race 4). I think the obvious difference was the fact that NZ had a bigger jib set, and ThI s probably results is more power coming out of the tack, and a more rapid speed build.  I see that since the regatta, Patriot has not sailed with a small jib even in conditions  they certainly would have set one earlier in the season.  I view this as confirmation.

This is the kind of thing you discover when you put your solo testing/ tuning into competition for the first time.

SHC

  • Like 1
Link to post
Share on other sites
On 12/29/2020 at 4:33 PM, dorox said:

Boats efficiency as measured by dividing VMG with wind strength.
First comparison by race number: (ignore INEOS race 1,2 , as their data seems corrupted)

image.png.6f93f05aa5d3d5553e430e882c24d264.pngimage.png.ae470b03994828b4efae69a6816cfa36.png

More interestingly to see how it depends on wind speed: seems like each boat is optimised for a different wind speed, while USA is good all rounder. NZL dominates in 11 knots. Anything more than 15 kn and everyone's performance converges. 

image.png.159390464b9541d9a350e095798478f1.pngimage.png.0ca28ea4f7ea8b86aa453897d04ae6a5.png

PS: added VMG/TWS in https://ac36.herokuapp.com/stats_app 

Sorry, I'm late to this party.

This is really interesting, thank you. You note a convergence of performance in the upper wind range, but I'm unsure if a simple ratio gives you a true picture.  We saw the boats doing ~30knts in light airs (sub 10knts?). As the wind speed doubles you wouldn't expect to see VMG double.  Certainly, when it's blowing 30knots you wouldn't expect to see them sailing at 90knts.

I'm unsure of your exact methodology, but do you think this is accounted for?

Link to post
Share on other sites
22 minutes ago, Ex-yachtie said:

Sorry, I'm late to this party.

This is really interesting, thank you. You note a convergence of performance in the upper wind range, but I'm unsure if a simple ratio gives you a true picture.  We saw the boats doing ~30knts in light airs (sub 10knts?). As the wind speed doubles you wouldn't expect to see VMG double.  Certainly, when it's blowing 30knots you wouldn't expect to see them sailing at 90knts.

I'm unsure of your exact methodology, but do you think this is accounted for?

Light air is tricky, because you have to sail lower angles to stay on foils. So while speed is good, vmg is poor - relatively. In strong winds, you can point high, but speed is limited and the VMG/TWS ratio start to go down again. This was from my model, which seems to track real-life data pretty well:

vmg2.thumb.png.94dd814502d36ba7df7ebf691058d5fc.png

It seems the highest VMG/TWS ratios are achieved in ~ 10 kts of wind both up- and downwind. Even if the ratio is not constant over a range of TWS, it still varies much less than VMG alone, which makes comparisons between boats easier.

  • Like 3
Link to post
Share on other sites
1 hour ago, Steve Clark said:

My glance at the graphs suggested that the biggest difference was in the speed build after the tack.  This was particularly evident in the US-NZ match 1 ( Race 4). I think the obvious difference was the fact that NZ had a bigger jib set, and ThI s probably results is more power coming out of the tack, and a more rapid speed build.  I see that since the regatta, Patriot has not sailed with a small jib even in conditions  they certainly would have set one earlier in the season.  I view this as confirmation.

This is the kind of thing you discover when you put your solo testing/ tuning into competition for the first time.

SHC

Here is that race between AM and ETNZ - very interesting. I cut the segments shorter, only 30 sec before and after tack, because this way I can include more tacks. Seems to be long enough.

Looking at all the tacks, both AM and ETNZ had some outliers - good and bad, and if I include them all, the averages are very similar between the two boats. AM actually comes out on top:

usa-nz-r4-1all.png.6d2bfc19d909cb2f51ab621622b6f039.pngusa-nz-r4-2all.png.3ec1fc2f6759cc24c5de5f147920b12c.png

If I remove the outliers, NZ comes out on top:

usa-nz-r4-1.png.e4efa20712b765e2fe60a0e6165114f9.pngusa-nz-r4.thumb.png.cc47b4c2eb3e4437ddf0ad260c2165b8.png

From this, I don't see better VMG increase from ETNZ after the tack. If anything, they seem to be slower in getting their VMG back up. Weirdly, the big difference seems to be before the tack. Maybe the difference is that AM foots off before the tack and comes out with a better speed, while ETNZ goes straight into the tack (higher VMG before the tack), but has to foot off more after the tack.

  • Like 2
Link to post
Share on other sites

Just one more (instead of doing my real job...:D)

TWA and speed from race 4 ETNZ vs AM. I flipped TWA so it always starts negative and goes positive after tack. This way I can summarize all the tacks regardless of direction (maybe for LR it would be fun to compare port vs starboard tacks to evaluate JS vs Bruni :))

Individual and average TWA graphs - it seems on average, ETNZ points higher both before and after tack:

usa-nz-r4-twa1.png.3060f77fb886dea79ffc1e91afc5de83.pngusa-nz-r4-twa2.png.337d9f5c517fe3be5a4803ba7a4e8433.png

 

Speed graphs - ETNZ goes in faster, but comes out slower and also accelerates slower after tack.

usa-nz-r4-speed1.png.f78e812172a040442d61bb47b1a00a20.pngusa-nz-r4-speed2.png.1b342bdd821a7746992a36cf8f7b3928.png

Altogether they are very close. ETNZ seems to gain a bit going into the tack with higher speed and pointing higher. AM gains back a bit by coming out with more speed after the tack although pointing lower. 

Disclaimer - as pointed out by others, TWA and therefore VMG numbers are all subject to various errors that come from converting AWA and AWS measured on the boats back into TWS and TWA. Plus, if there is any significant current on the race course, that would mess up all this as well. Let's just hope all these errors are similar on every boat.

  • Like 3
Link to post
Share on other sites

Alright, I'm back with more data and some improvements in the graphs. Tried to get data for all four teams. These graphs represent the average of 6-10 tacks for each boat in the race. 30 sec before and 30 sec after tack.

Race 1, ITA vs NZL, wind speed ~15kts:

Speed and TWA:

bs2.png.490a7c9a0dde564816a2b116b4cd9f6d.png  twa2.png.c2890824b65020db799e5dc76fd952e8.png

VMG and distance gained:

vmg2.png.2e716c725c270796c9bcd2b49c558d25.png  vms2.png.581688f232a763c54c0508f74e6dc7d7.png

Loss indicates how much distance is lost due to tacking compared to straight line upwind of the same boat.

The angles are quite similar, but ETNZ was faster in this race producing higher VMG. I'd say acceleration out of the tack was similar between the two.

 

Race 4, USA - NZL, wind speed ~ 15kts:

Speed and TWA:

bs2.png.e8f4277fc216f2e466c169dcf3cad96a.png  twa2.png.be9f971acdfc388d65c6ed309a4edb88.png

VMG and distance gained:

vmg2.png.44e1e2e506358a05209a9021e01a64f7.png  vmgs2.png.379a3e77c7c82818d9d17b851006602a.png

As discussed above, the major differences are that ETNZ sails higher than AM and goes into the tack with more speed.  Acceleration after the tack is actually lower for ETNZ, but because of the better angle, VMG is higher.

Race 7, USA - ITA, wind speed ~ 14kts:

Speed and TWA:

bs2.png.981c1d0488db013bab91057fbf5b8376.png  twa2.png.6c3f1fd6fd86662f75333a73b4205cf1.png

VMG and distance gained:

vmg2.png.e9359a3ed205dc50d07361e7fcdff4ac.png  vmgs2.png.7fb36278ee7b029f79fa5c8190e48ed5.png

Not much in it. Very similar angles, speed and VMG by these two teams. There was some kind of glitch in the reported wind direction when boats tacked causing those weird peaks.

 

Finally, Race 8, NZL- GBR, wind speed ~ 12kts:

This was probably the best race from INEOS.

Speed and TWA:

bs2.png.81f4346f08ae97775633bb6c0a1151f3.png  twa2.png.c1c0ea1c99b17919cbbbcc5f8c23d272.png

VMG and distance gained:

vmg2.png.21ebc7823b63b98dfd5205e5a096b9e4.png  vmgs2.png.b1db9f93d43c0c8226b95662c92161da.png 

TWAs going into the tack are similar, but INEOS is much slower. Acceleration after the tack is similar to ETNZ, but INEOS needs to foot off a lot more so VMG suffers.

Overall it seems a tack costs about 30 - 70m depending on conditions. I'll look at jibes at some other time.

 

  • Like 11
Link to post
Share on other sites

I thought this may also be interesting - how the teams drop and raise foils during tack. Foil1 is the active foil going in, Foil2 is the active foil coming out of the tack. Again these are averages from 6-10 tacks by each boat in each race.

Race 1, ITA vs NZL, wind speed ~15kts:

f-cant.png.5f231a5f6a1e89cd35270b0ada3f0f5f.png

Teams usually hold the new foil deeper for some time after tack during the acceleration phase. LR  holds it in this position longer during acceleration than ETNZ, and interestingly also keeps the foil deeper (less cant) by an average of 2 degrees sailing straight upwind. They were also sailing much slower than ETNZ in this race (see previous post), so maybe they were underpowered having the wrong jib up or something.

Race 4, USA - NZL, wind speed ~ 15kts:

f-cant.png.35f779cd994f22abc686bce2617fc74d.png

Overall quite similar approach, although AM lowers the old foil deeper for a few sec before the tack.

Race 7, USA - ITA, wind speed ~ 14kts:

f-cant.png.a136c036e9de514cb28ea256bdd956b0.png

Interesting to see that AM was trying something different here. Didn't lower Foil 1 deeper before the tack, but dropped Foil2 ~ 4 degrees deeper than in race 4 during acceleration. Again, LR holds the new foil deeper much longer after the tack.

Race 8, NZL- GBR, wind speed ~ 12kts:

f-cant.png.7b06c25a94220b423767600c118fb692.png

INEOS is different from all the others. They drop the new foil into a higher angle after tack, and only move the new foil into the "acceleration position" 15 sec later.:blink: 

  • Like 6
Link to post
Share on other sites
On 1/6/2021 at 1:15 PM, erdb said:

I thought this may also be interesting - how the teams drop and raise foils during tack. Foil1 is the active foil going in, Foil2 is the active foil coming out of the tack. Again these are averages from 6-10 tacks by each boat in each race.

....

These are great and very insightful. Where did you get the raw data? I'd be curious to look at how consistent teams are with how they tack, and if they change things up, what's faster.

Link to post
Share on other sites
On 1/5/2021 at 2:24 AM, erdb said:

This is just a quick analysis, more to follow. Used the data @dorox...

nz-ita2.png.a13e9b6c8d56d24cda1da92644fc087d.png

What's remarkable is that they barely lose any distance through the tacks. I haven't calculated it yet, but I'm planning to compare distance gained to the gain without a tack.

Something like this - it's only hand-drawn at this point, the straight lines show how much they'd gained without a tack:

nz-ita3.thumb.png.df607264235eff571d9a08c5de3f62b0.png

While ETNZ has a better overall VMG, LR actually loses less during the tack relative to their straight-line VMG. Probably showing the pros and cons of small vs large foils. The wind was pretty good during this race ~ 15 kts, I'm sure the light wind races will show much bigger drops in VMG during tacks.

 

 

If you took the derivative the effects would be more obvious.  Can you just plot vmg versus time for us?  This way your straight lines will be horizontal and that more horizontal jog at head to wind will show as a dip and recover.  As a bonus it may have a nice seagull wing look too it, and the blokes on these forums love seagulls.

  • Like 1
Link to post
Share on other sites
1 hour ago, WindySurfer said:

If you took the derivative the effects would be more obvious.  Can you just plot vmg versus time for us?  This way your straight lines will be horizontal and that more horizontal jog at head to wind will show as a dip and recover.  As a bonus it may have a nice seagull wing look too it, and the blokes on these forums love seagulls.

VMG vs time graphs are there, too in  post #2738.

Link to post
Share on other sites
2 hours ago, sailer99 said:

These are great and very insightful. Where did you get the raw data? I'd be curious to look at how consistent teams are with how they tack, and if they change things up, what's faster.

The data is originally from https://www.americascup.com/en/advanced-dashboard@dorox made a tool to look at the data https://ac36.herokuapp.com/stats_app and I'm using his data trying to average tacks / jibes to answer the same kind of questions you're asking:))

As for how consistent they are, this is how the VMG data looks like before averaging. Each line is one tack.

Race 1 and  Race 4:

vmg1.png.7068d72299ed2fd01e52e4f742113e52.png   vmg1.png.d2a6d1ee0ac8756085786ae0b6dbf719.png

Race 7 and 8 (with some glitching of TWA and VMG right when the boats go through the tack):

vmg1.thumb.png.9e28470c039c4869830032affc0ec2cb.png vmg1.png.04f2188763e95ca18cc25aae1a14f9e1.png

  • Like 5
Link to post
Share on other sites

Same analysis as above, but for jibes:

Race 1, ITA vs NZL, wind speed ~15kts:

Speed and TWA:

bs2.png.bb69d6810dbc7930165f769395351ac6.png   twa2.png.e7c23d89bfc3ef86b36e75b97068704e.png

Angles pretty similar, but ETNZ much faster.

 

VMG and distance gained:

vmg2.png.c55ae3c8fb691bcf8a07a000d1ee4c71.png  vmgs2.png.708e0e09553c0ba31c1653008c813b1b.png

ETNZ had much better VMG overall, but compared to their excellent straight line speed, they actually lost more distance during a jibe than LR. I'd point out that it's much harder to find "clean" jibes, there are much fewer jibes than tacks in a race, and the variability in speed and angles is much higher as well.

Foil cant angles:

cant.png.840f8ec5953b534d711d08a67e352f04.png

It's the opposite to what happens during a tack. After a tack, teams drop the foil deeper while accelerating, during a jibe, they run it higher while accelerating before dropping it lower for max VMG. Makes sense - the higher they point, the more they cant the foil out. However, the cant angle difference between LR and ETNZ is pretty big despite sailing very similar angles. This could be a sign of how the anhedral foils are used vs mostly or completely straight foils. LR also keeps both foils down longer than ETNZ.

 

Race 4, USA - NZL, wind speed ~ 15kts:

Speed and TWA:

bs2.png.4f5d52b098a534279294bd6041aae9df.pngtwa2.png.bb4aca3864c4dc6230cef8d60f18d968.png

VMG and distance gained:

vmg2.png.9d9fbab5dc7ad69016ffe3acf5f69d82.pngvmgs2.png.f77cb7e8d0af8d5cc1abef3ff695651f.png

Foil cant:

cant.png.5fdf90a3db829e5b1c0e1c604e1e404e.png

Overall very similar performances by these two; AM is faster, and ETNZ sails a tiny bit lower. However, AM still cants the foils higher than ETNZ although not as high as LR. Interestingly their anhedral angle also falls between ETNZ and LR.

 

Race 7, USA - ITA, wind speed ~ 14kts:

Speed and TWA:

bs2.png.f8f96f1479dfef3e4bd44e992b7bc24d.pngtwa2.png.0869759d170a845d8ebe0b8a87afcf79.png

 

VMG and distance gained:

vmg2.png.8486cf94c2e0623033c32c84e5c45dd9.pngvmgs2.png.730db099f3fb8f7c8228ba05428b8d2b.png

Foil cant:

cant.png.a8d0215a566dae66ac25e6e93537e9c9.png

Interestingly both team used lower cant angles than in their previous races, but the difference between AM and LR remained. LR sails with the foils canted out the most, and they stay on two foils longer than AM.

 

Race 8, NZL- GBR, wind speed ~ 12kts:

Speed and TWA:

bs2.png.7d3d245d644faa99161b1e4e9c78554c.png  twa2.png.44427b4dae6b2083acf58140a7e9d24c.png

VMG and distance gained:

vmg2.png.7e4ea3e3582dd9ca176149d85b0fb84e.png   vmgs2.png.498bdea862b19ef995ba822467389855.png

Foil cant:

cant.png.3d570b73cad32a653973c58dab48c355.png

Quite obvious differences in performance. This race was in slightly lighter wind, and the distance lost during a jibe is pretty big even for ETNZ.

  • Like 4
Link to post
Share on other sites
1 hour ago, Varan said:

The twin skin thread has a link to Sailing World's article on sails. Here is a link to their article on hulls.

Some 'interesting' gems in that article...

"The skeg of INEOS Team UK’s Britannia II, is considered the most extreme of the fleet, with the primary purpose of reducing drag at takeoff speeds." - caption to Rita pic. 

I didn't know that for minimum [wetted] surface area of a given volume, you need a shoe-box.

"... there’s no point in having any righting moment from the hull. The whole righting moment comes from the foil ..." - Benjamin Muyl , ITUK 'architect'.

Glaring at my RM diagram, I'm wondering why I thought RM was due to boat mass ± stabilator force.

"... we went for this very squared bustle to try to create a vortex off the sharp edge that would effectively seal [the gap].” - Muyl. 

Having seen diagrams of how vortices are used in F1 cars, did I interpret that backwards?  Plus it's adding another useful trailing vortex... 

“I think as usual, [ETNZ] did a good job, but I don’t think they...well, I hope they won’t be superior, and I don’t think they will be superior. I think it will be pretty tight racing.” - Martin Fischer,  LR co-design coordinator.

OK, I'm hoping for tight racing too. :)

Link to post
Share on other sites
1 hour ago, MaxHugen said:

"... we went for this very squared bustle to try to create a vortex off the sharp edge that would effectively seal [the gap].” - Muyl. 

Having seen diagrams of how vortices are used in F1 cars, did I interpret that backwards?  Plus it's adding another useful trailing vortex... 

Interesting point.  F1 is essentially a niche field of aerodynamics that focuses on vortex management. 

In F1 they use vortices to seal the floor to the track, especially around the rear tires where they have tire squirt. The better you seal the sides of the floor, the better the diffuser works to create downforce.

I think what they are saying here is that creating a vortex on the skeg will provide a better seal than other designs when the gap from the skeg to the water isn't zero, which is most of the time. And that will create a better endplate effect. 

AM seems to be going for a physical seal, which is probably better when it's achieved and worse when there's a gap.

Link to post
Share on other sites
15 minutes ago, Thistle3841 said:

Interesting point.  F1 is essentially a niche field of aerodynamics that focuses on vortex management. 

In F1 they use vortices to seal the floor to the track, especially around the rear tires where they have tire squirt. The better you seal the sides of the floor, the better the diffuser works to create downforce.

I think what they are saying here is that creating a vortex on the skeg will provide a better seal than other designs when the gap from the skeg to the water isn't zero, which is most of the time. And that will create a better endplate effect. 

AM seems to be going for a physical seal, which is probably better when it's achieved and worse when there's a gap.

Yes, that is one of the uses of vortices in F1.  From front-on, the 'bottom' of the vortices on either side of the floor pan are acting to keep the airflow 'in' - ie under the car.

But I visualise a vortex created by Rita's skeg to be rotating the wrong way, and accelerating airflow under the skeg towards the leeward side?

Link to post
Share on other sites
1 minute ago, MaxHugen said:

Yes, that is one of the uses of vortices in F1.  From front-on, the 'bottom' of the vortices on either side of the floor pan are acting to keep the airflow 'in' - ie under the car.

But I visualise a vortex created by Rita's skeg to be rotating the wrong way, and accelerating airflow under the skeg towards the leeward side?

Ah, I see what you mean. Although my understanding is that a vortex can work as a seal rotating either way. One way is just more effective. No?

I'm visualizing the vortex flowing down next to the skeg, which I think is opposite the way you see it, then returning air away from the skeg at the water to form the vortex. It wouldn't be a perfect seal and would leak plenty of air under the skeg. 

  • Like 1
Link to post
Share on other sites
10 minutes ago, Thistle3841 said:

Ah, I see what you mean. Although my understanding is that a vortex can work as a seal rotating either way. One way is just more effective. No?

I'm visualizing the vortex flowing down next to the skeg, which I think is opposite the way you see it, then returning air away from the skeg at the water to form the vortex. It wouldn't be a perfect seal and would leak plenty of air under the skeg. 

You may well be right, my grey-matter CFD program can't compute this! :wacko:

Yes, I did visualise the vortex being created at the sharp edge and beyond (ie to leeward). Mainly because that's where I see a low pressure area forming, which is the precursor to a vortex.

Pity we don't have a CFD guru here that could whip up some models and run a few scenarios through it.

  • Like 1
Link to post
Share on other sites
8 minutes ago, MaxHugen said:

You may well be right, my grey-matter CFD program can't compute this! :wacko:

Yes, I did visualise the vortex being created at the sharp edge and beyond (ie to leeward). Mainly because that's where I see a low pressure area forming, which is the precursor to a vortex.

Pity we don't have a CFD guru here that could whip up some models and run a few scenarios through it.

I'm just an armchair doing this in my head for fun so I could easily be wrong. But I view the air flow as front to back primarily and secondarily from top to bottom. 

Link to post
Share on other sites

Does anyone know how to calculate the Resistance Force of a keel (in a keelboat), or in this case, the leeward foil, when you only know the Leeward Force? I'm trying to derive the Leeward Angle theta ( θ ) :

image.png.d5f8fef878ff77e0aa7aca3bbb50bf8d.png

Link to post
Share on other sites
8 minutes ago, MaxHugen said:

Does anyone know how to calculate the Resistance Force of a keel (in a keelboat), or in this case, the leeward foil, when you only know the Leeward Force? I'm trying to derive the Leeward Angle theta ( θ ) :

image.png.d5f8fef878ff77e0aa7aca3bbb50bf8d.png

If the boat is in balance, 'Leeway force" = "Resistance force" (assuming that the rudder doesn't exert any significant lateral force). Leeway angle is whatever AOA you need on the foil to achieve that force for a given boat speed. In case of the AC75, you also have to consider that the foil wing is not vertical, so only a certain portion of the lift provided by the foil wing produces lateral force. The rest of the foil lift produces vertical (up) force.

  • Like 1
Link to post
Share on other sites

I'm still going through the data from the ACWS. After looking at tacks and jibes, this time, I analyzed straight line performance. Now all the algorithms are done - hopefully we'll have data  for the rest of the races. It'd be fun to follow their progress.

For straight line performance, I filtered out all segments from the data where the boat is turning or changing speed rapidly, and only kept segments where speed and direction are fairly steady. There are a lot of parameters to look at, among the  most obvious: boat speed, TWA, VMG. For these, I'm using histograms - the parameter is on the horizontal axis, and the vertical axis shows how often a certain value happens. The higher the bar for a certain boat speed for example, the more often a boat sails at that speed. The dotted lines show the average of the parameter for each boat.

For example, upwind boat speed and VMG in race 7, LR vs AM:

2004772354_uwbs.png.97cacdcf30a90541f867022ece948513.png  1487192770_uwvmg.png.f6350cb0786d26f5c44f77302cfeb1ec.png

Downwind boat speed and VMG in race 7, LR vs AM:

1134386194_dwbs.png.5a9a3f66cdfb59c6627fd35c2d268ce9.png  290435726_dwvmg.png.f1f20f41a764dd81577c27c2a8709d52.png

Upwind, advantage LR; downwind, advantage AM.

Some other fun stuff: heel and pitch:

Upwind:

257831033_uwheel.png.8d85db92a5fad945e431d55ad2832e5b.png  254502351_uwpitch.png.026c11450e4e3cc5cfbb364710abc023.png

Downwind:

1697625613_dwheel.png.d8c2f2b9045657a3df39d0af7bb0f3bb.png  11468258_dwpitch.png.a4b0b53df23d7eb51a0a835b1581329d.png

AM heels more to windward and pitches more forward both upwind and downwind.

Up- / downwind foil cant angles:

788743555_uwfcant.png.a59fe4ab23525bb4619cd3a33f0ee30f.png   62521715_dwfcant.png.0f1d7f0992c9b95a2f3c067023ffbd35.png

This was obvious from tacks and jibes, too - LR uses less cant upwind than the others and more cant downwind than AM and ETNZ.

Finally, a look at tactical choices. What I tried to do here is that I looked at whether a boat is on the right or wrong side of a wind shift. Wind shift is the difference of wind direction at a certain time compared to average wind direction over the whole race. For every second a boat is on a lifted tack or a headed jibe, the shift in degrees is added to its "shift total". For every second a boat is on a headed tack or a lifted jibe, the shift in degrees is subtracted from its "shift total". If a boat is on the favored tack or jibe, this line goes up and vice versa. Only straight line segments are included, tacks and jibes are cut out.

What I found is that AM was really good at getting the shifts right, while ETNZ wasn't that interested in being on the favored side (maybe aiming for more pressure/speed instead or sandbagging ;)).

For example, Race 4 NZ-USA and Race 7 ITA-USA:

favtack.png.7d59a45a9757b52c1b553acb0a7d089a.png  favtack.png.9cc6009375919455629bc88cfdd0f432.png

Race 10 NZ - USA, Race 12 ITA - NZ:

favtack.png.3df36b25496589b350eddb04c9aca7c4.png  favtack.png.4eb91cfbbd5aa61a080c53836426d0a2.png

I don't know if and how teams analyze this. If anyone knows a better way to quantify tactical choices, let me know.

  • Like 6
Link to post
Share on other sites
48 minutes ago, erdb said:

I'm still going through the data from the ACWS. After looking at tacks and jibes, this time, I analyzed straight line performance. Now all the algorithms are done - hopefully we'll have data  for the rest of the races. It'd be fun to follow their progress.

For straight line performance, I filtered out all segments from the data where the boat is turning or changing speed rapidly, and only kept segments where speed and direction are fairly steady. There are a lot of parameters to look at, among the  most obvious: boat speed, TWA, VMG. For these, I'm using histograms - the parameter is on the horizontal axis, and the vertical axis shows how often a certain value happens. The higher the bar for a certain boat speed for example, the more often a boat sails at that speed. The dotted lines show the average of the parameter for each boat.

For example, upwind boat speed and VMG in race 7, LR vs AM:

2004772354_uwbs.png.97cacdcf30a90541f867022ece948513.png  1487192770_uwvmg.png.f6350cb0786d26f5c44f77302cfeb1ec.png

Downwind boat speed and VMG in race 7, LR vs AM:

1134386194_dwbs.png.5a9a3f66cdfb59c6627fd35c2d268ce9.png  290435726_dwvmg.png.f1f20f41a764dd81577c27c2a8709d52.png

Upwind, advantage LR; downwind, advantage AM.

Some other fun stuff: heel and pitch:

Upwind:

257831033_uwheel.png.8d85db92a5fad945e431d55ad2832e5b.png  254502351_uwpitch.png.026c11450e4e3cc5cfbb364710abc023.png

Downwind:

1697625613_dwheel.png.d8c2f2b9045657a3df39d0af7bb0f3bb.png  11468258_dwpitch.png.a4b0b53df23d7eb51a0a835b1581329d.png

AM heels more to windward and pitches more forward both upwind and downwind.

Up- / downwind foil cant angles:

788743555_uwfcant.png.a59fe4ab23525bb4619cd3a33f0ee30f.png   62521715_dwfcant.png.0f1d7f0992c9b95a2f3c067023ffbd35.png

This was obvious from tacks and jibes, too - LR uses less cant upwind than the others and more cant downwind than AM and ETNZ.

Finally, a look at tactical choices. What I tried to do here is that I looked at whether a boat is on the right or wrong side of a wind shift. Wind shift is the difference of wind direction at a certain time compared to average wind direction over the whole race. For every second a boat is on a lifted tack or a headed jibe, the shift in degrees is added to its "shift total". For every second a boat is on a headed tack or a lifted jibe, the shift in degrees is subtracted from its "shift total". If a boat is on the favored tack or jibe, this line goes up and vice versa. Only straight line segments are included, tacks and jibes are cut out.

What I found is that AM was really good at getting the shifts right, while ETNZ wasn't that interested in being on the favored side (maybe aiming for more pressure/speed instead or sandbagging ;)).

For example, Race 4 NZ-USA and Race 7 ITA-USA:

favtack.png.7d59a45a9757b52c1b553acb0a7d089a.png  favtack.png.9cc6009375919455629bc88cfdd0f432.png

Race 10 NZ - USA, Race 12 ITA - NZ:

favtack.png.3df36b25496589b350eddb04c9aca7c4.png  favtack.png.4eb91cfbbd5aa61a080c53836426d0a2.png

I don't know if and how teams analyze this. If anyone knows a better way to quantify tactical choices, let me know.

Awesome work, really enjoy your analysis. Gives a great picture of what might be going on. Thanks!

  • Like 1
Link to post
Share on other sites
16 minutes ago, buckdouger said:

 

What I found is that AM was really good at getting the shifts right, while ETNZ wasn't that interested in being on the favored side (maybe aiming for more pressure/speed instead or sandbagging ;)).

For example, Race 4 NZ-USA and Race 7 ITA-USA:

I think for apparent wind sailing pressure is king. 

Link to post
Share on other sites
1 hour ago, erdb said:

I'm still going through the data from the ACWS. After looking at tacks and jibes, this time, I analyzed straight line performance. Now all the algorithms are done - hopefully we'll have data  for the rest of the races. It'd be fun to follow their progress.

....

I don't know if and how teams analyze this. If anyone knows a better way to quantify tactical choices, let me know.

Great stuff. The downwind boat speeds matching so closely is interesting, I guess they is a certain optimum boat speed for a certain wind speed, and you sail at that target speed at as low of an angle as possible. How does this line up with the theoretical boat model?

Assuming the TWA is somewhat accurate, it would be interesting to see if you can find points in time where ETNZ uses their high mode. Visually it looks like a killer mode, but probably is very hard for them to hold.

Link to post
Share on other sites
2 hours ago, buckdouger said:

Awesome work, really enjoy your analysis. Gives a great picture of what might be going on. Thanks!

Your welcome, I'm having fun with it, too. There's a lot more, I'm just lazy to post every graph here, but here is a zip with data from all the races I've analyzed so far. There's a bunch of image files with graphs in it. 

acws stats.zip

  • Like 3
Link to post
Share on other sites
2 hours ago, amc said:

I think for apparent wind sailing pressure is king. 

Downwind certainly. Upwind, a good shift still helps a lot. What we can't see from these graphs is whether there was actually more pressure on the negative tack or not. The windspeed data is from the boats, so you don't know what the wind speed on the other side of the course would be. We need a cluster of weather stations reporting live from the course. Grant if you are listening...:D

Link to post
Share on other sites
1 hour ago, sailer99 said:

Great stuff. The downwind boat speeds matching so closely is interesting, I guess they is a certain optimum boat speed for a certain wind speed, and you sail at that target speed at as low of an angle as possible. How does this line up with the theoretical boat model?

Assuming the TWA is somewhat accurate, it would be interesting to see if you can find points in time where ETNZ uses their high mode. Visually it looks like a killer mode, but probably is very hard for them to hold.

Yes, it's probably easier to improve downwind vmg by going deeper than going faster, but there are some differences in speed as well between boats.

As for the theoretical model, I'm very happy how accurate it turned out to be. Usually it's within a few knots of speed and a few degrees of angle (except for ETNZ, which always overperforms it by a little). Even the foil cant angle is within +/- 1 degree of the predicted.

That race 7 between AM and LR was in 14 kts for example. I didn't run my model for every possible wind speed, but for 15 kts of wind, this is the predicted downwind speed and vmg. I zoomed in for the downwind angles. Best predicted TWA is 145, speed is 38.2, VMG is ~31.2 kts.

 vmgcr_LI.jpg.fb3e82b23ad6549d90c13cd3188ca3bd.jpg

This is the actual data from the boats. Speed ~38.2 kts, TWA ~ 145, VMG ~32 kts ;)

243430508_dwbs.png.b3566e4ccbe7e9553e6a51531d7aa448.png   73821165_dwtwa.png.ad8679f48125e9b8f343b161bff6be1e.png  372710960_dwvmg.png.c21816cd6d31d35511fbfcb2eabb89f0.png

Considering how simplistic some of the drag calculations are, I think it's pretty close.

  • Like 2
Link to post
Share on other sites
2 hours ago, sailer99 said:

Assuming the TWA is somewhat accurate, it would be interesting to see if you can find points in time where ETNZ uses their high mode. Visually it looks like a killer mode, but probably is very hard for them to hold.

This is summarized data from all legs of a race, so it's not good for finding those moments where ETNZ goes into high mode, but maybe these summary histograms also tell the story. This is from race 4 ETNZ vs AM. Upwind TWA and boat speed.

715224149_uwtwa.png.0ffda99f9a1470826c8a645bcffdf46a.png   175799266_uwbs.png.b1c2940694de48f99b7c5af8738e669a.png

Overall ETNZ's TWA is much better, but there was a small percentage of data points in the 30 - 35 degree range, which was way beyond what AM could do. There are also some low speed points for ETNZ in the 25-30 kts range, so maybe that's when they go into high mode.

BTW, this is why I disagree with those who say that upwind starts in this AC are much more important than in the last one, and PB's lack of experience in match racing can hurt ETNZ. First, he might as well win all the starts again when it counts, but even if he starts close second, he can easily get out of the wind shadow of the other boat. It's so easy to escape by going a few degrees lower or higher. If they have just a few % VMG advantage, they'll grind the other boat down. Steering for maximum VMG is much more important than starting well.

  • Like 6
Link to post
Share on other sites

@enigmatically2 I think I need help from your trig skills again pls!   I'm trying to determine the AoA on a canted foil for a given Leeward angle. For the sake of simplicity, the 'foil' is a flat plank, canted at 20°, with 0° initial AoA.    At 0° Leeway, the flow presents as 0° AoA to the foil.   At 90° Leeway, it presents as 20°.

So what AoA is presented to the foil if Leeway is, say, 10°?

image.png.a92b336ec8abc1caec3642ffd8b61b31.png

It's tempting to consider that since 0° Leeway = 0° AoA, and 90° Leeway = Cant Angle (20°), then

at 10° Leeway = 10/90 * Cant Angle  = 2.2° AoA.

Am I missing something, or is it just that simple?

Link to post
Share on other sites
1 hour ago, erdb said:

This is summarized data from all legs of a race, so it's not good for finding those moments where ETNZ goes into high mode, but maybe these summary histograms also tell the story. This is from race 4 ETNZ vs AM. Upwind TWA and boat speed.

715224149_uwtwa.png.0ffda99f9a1470826c8a645bcffdf46a.png   175799266_uwbs.png.b1c2940694de48f99b7c5af8738e669a.png

Overall ETNZ's TWA is much better, but there was a small percentage of data points in the 30 - 35 degree range, which was way beyond what AM could do. There are also some low speed points for ETNZ in the 25-30 kts range, so maybe that's when they go into high mode.

BTW, this is why I disagree with those who say that upwind starts in this AC are much more important than in the last one, and PB's lack of experience in match racing can hurt ETNZ. First, he might as well win all the starts again when it counts, but even if he starts close second, he can easily get out of the wind shadow of the other boat. It's so easy to escape by going a few degrees lower or higher. If they have just a few % VMG advantage, they'll grind the other boat down. Steering for maximum VMG is much more important than starting well.

This did this when they were (often) forced to in AC35, and the result was pretty amazing. The times they were in front, the results were even more emphatic.

Link to post
Share on other sites
6 hours ago, erdb said:

I'm still going through the data from the ACWS. After looking at tacks and jibes, this time, I analyzed straight line performance. Now all the algorithms are done - hopefully we'll have data  for the rest of the races. It'd be fun to follow their progress.

For straight line performance, I filtered out all segments from the data where the boat is turning or changing speed rapidly, and only kept segments where speed and direction are fairly steady. There are a lot of parameters to look at, among the  most obvious: boat speed, TWA, VMG. For these, I'm using histograms - the parameter is on the horizontal axis, and the vertical axis shows how often a certain value happens. The higher the bar for a certain boat speed for example, the more often a boat sails at that speed. The dotted lines show the average of the parameter for each boat.

For example, upwind boat speed and VMG in race 7, LR vs AM:

2004772354_uwbs.png.97cacdcf30a90541f867022ece948513.png  1487192770_uwvmg.png.f6350cb0786d26f5c44f77302cfeb1ec.png

Downwind boat speed and VMG in race 7, LR vs AM:

1134386194_dwbs.png.5a9a3f66cdfb59c6627fd35c2d268ce9.png  290435726_dwvmg.png.f1f20f41a764dd81577c27c2a8709d52.png

Upwind, advantage LR; downwind, advantage AM.

Some other fun stuff: heel and pitch:

Upwind:

257831033_uwheel.png.8d85db92a5fad945e431d55ad2832e5b.png  254502351_uwpitch.png.026c11450e4e3cc5cfbb364710abc023.png

Downwind:

1697625613_dwheel.png.d8c2f2b9045657a3df39d0af7bb0f3bb.png  11468258_dwpitch.png.a4b0b53df23d7eb51a0a835b1581329d.png

AM heels more to windward and pitches more forward both upwind and downwind.

Up- / downwind foil cant angles:

788743555_uwfcant.png.a59fe4ab23525bb4619cd3a33f0ee30f.png   62521715_dwfcant.png.0f1d7f0992c9b95a2f3c067023ffbd35.png

This was obvious from tacks and jibes, too - LR uses less cant upwind than the others and more cant downwind than AM and ETNZ.

Finally, a look at tactical choices. What I tried to do here is that I looked at whether a boat is on the right or wrong side of a wind shift. Wind shift is the difference of wind direction at a certain time compared to average wind direction over the whole race. For every second a boat is on a lifted tack or a headed jibe, the shift in degrees is added to its "shift total". For every second a boat is on a headed tack or a lifted jibe, the shift in degrees is subtracted from its "shift total". If a boat is on the favored tack or jibe, this line goes up and vice versa. Only straight line segments are included, tacks and jibes are cut out.

What I found is that AM was really good at getting the shifts right, while ETNZ wasn't that interested in being on the favored side (maybe aiming for more pressure/speed instead or sandbagging ;)).

For example, Race 4 NZ-USA and Race 7 ITA-USA:

favtack.png.7d59a45a9757b52c1b553acb0a7d089a.png  favtack.png.9cc6009375919455629bc88cfdd0f432.png

Race 10 NZ - USA, Race 12 ITA - NZ:

favtack.png.3df36b25496589b350eddb04c9aca7c4.png  favtack.png.4eb91cfbbd5aa61a080c53836426d0a2.png

I don't know if and how teams analyze this. If anyone knows a better way to quantify tactical choices, let me know.

I think the tacticians are more worried about number of tacks and gybes. 

Depending on the length and width of the course will determine the 'par' number of tacks or gybes. You can then manipulate your entry to each leg (which gate or which end of the start line) if it's  tight call between X and X+1 manoeuvres. Finally, there is the call on how high to sail above ideal VMG angle to reduce tacks, and when is best to do this. 

Also, it's very unlikely that each leg will be perfectly X number of manoeuvres (i.e. you bounce off every boundary and the final boundary bounce puts you exactly on the lay), unless you have pushed things hard with sailing a high mode. A which point, you might have several 100 meters of 'free sailing'. This is distance you can tack before a course boundary without it resulting in you doing extra manoeuvres. So there are tactical decisions of when to take this 'pre-boundary' tack, like to clear your air, or take a shift, or bounce over to a gain feature on the course. Or, save this free tack until the layline to afford you more wriggle room. For instance you may tack out early at the start of a beat to clear air, but then hit the boundary slightly before layline at the top as a result and have to sail a high mode, or go to the less favoured mark and do an extra tack. So you have to way that risk up against the damage of being in dirty air. 

All in all, simply the time spent on the lifted tack will be nice, but I think it only really should be coming in to play after the factors I describe above. 

  • Like 1
Link to post
Share on other sites
6 hours ago, amc said:

I think for apparent wind sailing pressure is king. 

I think once at maximum righting moment, then you don't want much extra pressure. If you're at maximum cant, and the sails are already flat (or inverted), then extra pressure will just be more drag.

This depends on which sails they have on, but I would imagine they would be a little over canvassed upwind to allow deeper angles downwind and earlier / quicker take off. So in most conditions pressure upwind might not be as critical. 

Link to post
Share on other sites
2 hours ago, MaxHugen said:

@enigmatically2 I think I need help from your trig skills again pls!   I'm trying to determine the AoA on a canted foil for a given Leeward angle. For the sake of simplicity, the 'foil' is a flat plank, canted at 20°, with 0° initial AoA.    At 0° Leeway, the flow presents as 0° AoA to the foil.   At 90° Leeway, it presents as 20°.

So what AoA is presented to the foil if Leeway is, say, 10°?

image.png.a92b336ec8abc1caec3642ffd8b61b31.png

It's tempting to consider that since 0° Leeway = 0° AoA, and 90° Leeway = Cant Angle (20°), then

at 10° Leeway = 10/90 * Cant Angle  = 2.2° AoA.

Am I missing something, or is it just that simple?

The distance travelled sideways is proportional to speedxSin(Leeway), so the amount of elevation of the foil that water flows past in a unit of time is

speed x  Sin(Leeway) x Sin (Cant)

In the same unit of time the water flowing over the foil in a fwd direction is speed x Cos (leeway)

Cos^2 + Sin^2 always equals 1. So the total water flow travel is proportional to speed

Thus the AOA is

asin ((speed x Sin(leeway) x(Sin Cant) /speed) = asin(sin(leeway) * sin(cant))

Thsu for example at 10 degree leeway and 20 degree cant you get

AOA = 3.4 degrees

 

 

  • Like 1
Link to post
Share on other sites
7 hours ago, Mozzy Sails said:

I think the tacticians are more worried about number of tacks and gybes. 

Depending on the length and width of the course will determine the 'par' number of tacks or gybes. You can then manipulate your entry to each leg (which gate or which end of the start line) if it's  tight call between X and X+1 manoeuvres. Finally, there is the call on how high to sail above ideal VMG angle to reduce tacks, and when is best to do this. 

Also, it's very unlikely that each leg will be perfectly X number of manoeuvres (i.e. you bounce off every boundary and the final boundary bounce puts you exactly on the lay), unless you have pushed things hard with sailing a high mode. A which point, you might have several 100 meters of 'free sailing'. This is distance you can tack before a course boundary without it resulting in you doing extra manoeuvres. So there are tactical decisions of when to take this 'pre-boundary' tack, like to clear your air, or take a shift, or bounce over to a gain feature on the course. Or, save this free tack until the layline to afford you more wriggle room. For instance you may tack out early at the start of a beat to clear air, but then hit the boundary slightly before layline at the top as a result and have to sail a high mode, or go to the less favoured mark and do an extra tack. So you have to way that risk up against the damage of being in dirty air. 

All in all, simply the time spent on the lifted tack will be nice, but I think it only really should be coming in to play after the factors I describe above. 

All good points. I guess the differences we see in the time spent on favored tack could be an indicator of how heavily a team weighs the importance on being lifted/headed in the calculations you outlined (or could just be luck). There are also permanent shifts on the course. Dean referred to it as "banana" in one race as the wind bent around Rangitoto. So again, there may be a difference in the game plans of the teams - how important they think it is to hit that shift vs minimizing the number of tacks. AM was definitely the most willing to throw in an extra tack to cover their opponent. Either way, it's noteworthy that part of the solid performance AM showed may have been good tactics and not purely boat performance.

Do you think their little monitors display all that info you mentioned? How many tacks are expected on the next leg, how much "free sailing" distance they have to play with? It must be fun to take all that information in while approaching the bottom mark at 45 kts of speed trying to pull off a jibe without flipping the boat over.

Link to post
Share on other sites
52 minutes ago, erdb said:

All good points. I guess the differences we see in the time spent on favored tack could be an indicator of how heavily a team weighs the importance on being lifted/headed in the calculations you outlined (or could just be luck). There are also permanent shifts on the course. Dean referred to it as "banana" in one race as the wind bent around Rangitoto. So again, there may be a difference in the game plans of the teams - how important they think it is to hit that shift vs minimizing the number of tacks. AM was definitely the most willing to throw in an extra tack to cover their opponent. Either way, it's noteworthy that part of the solid performance AM showed may have been good tactics and not purely boat performance.

Do you think their little monitors display all that info you mentioned? How many tacks are expected on the next leg, how much "free sailing" distance they have to play with? It must be fun to take all that information in while approaching the bottom mark at 45 kts of speed trying to pull off a jibe without flipping the boat over.

I think if there is a left wind bend as you go up the course (for instance), then you'd look to approach the windward marks on port and start the leg on starboard. But, again, this will be quite a game of geometry to get right. However, all the tacks you do in the middle are just filler, you are limited on course leverage and have to make those tacks regardless. 

I think they will have pretty good modelling of the pay offs of sailing high and slow with one less tack, or sailing best VMG and doing one more. So I think the helms / tactician will know that crib sheet off by heart. I.e. 500m in high mode = 1 tack. So they can quickly weigh up the pros and cons. However, what you can't put in to a model, and why I don't think they have a real time 'minimum tack solution' produced on screen as it will all be wrecked by changes in the wind across the course. They also get the course information very late and it's often incorrect, so I am unsure they would rely on it. 

Link to post
Share on other sites
11 hours ago, erdb said:

This is summarized data from all legs of a race, so it's not good for finding those moments where ETNZ goes into high mode, but maybe these summary histograms also tell the story. This is from race 4 ETNZ vs AM. Upwind TWA and boat speed.

715224149_uwtwa.png.0ffda99f9a1470826c8a645bcffdf46a.png   175799266_uwbs.png.b1c2940694de48f99b7c5af8738e669a.png

Overall ETNZ's TWA is much better, but there was a small percentage of data points in the 30 - 35 degree range, which was way beyond what AM could do. There are also some low speed points for ETNZ in the 25-30 kts range, so maybe that's when they go into high mode.

BTW, this is why I disagree with those who say that upwind starts in this AC are much more important than in the last one, and PB's lack of experience in match racing can hurt ETNZ. First, he might as well win all the starts again when it counts, but even if he starts close second, he can easily get out of the wind shadow of the other boat. It's so easy to escape by going a few degrees lower or higher. If they have just a few % VMG advantage, they'll grind the other boat down. Steering for maximum VMG is much more important than starting well.

Do you know if the yaw of the boats is similar? I don't know if that data is included but some yaw from NZ would explain some of the angle difference, although not all of a difference that large. 

Link to post
Share on other sites
3 minutes ago, Thistle3841 said:

Do you know if the yaw of the boats is similar? I don't know if that data is included but some yaw from NZ would explain some of the angle difference, although not all of a difference that large. 

There is no yaw data unfortunately. There is a geometric relationship between foil cant angle, pitch and yaw, so in theory, one could estimate yaw from the cant angle and pitch data. In general, as the boat pitches forward, it also turns bow to lee. The problem is that the relationship depends on how the foil wing is angled on the foil arm and what flap settings they use. It comes out nicely from my model, but I doubt the teams will share any of these specifics.

Link to post
Share on other sites
1 hour ago, erdb said:

AM was definitely the most willing to throw in an extra tack to cover their opponent. Either way, it's noteworthy that part of the solid performance AM showed may have been good tactics and not purely boat performance.

That was certainly true in Race 10 (NZ v USA), can’t speak to AM’s other races. They emphasized close cover several times. 

Link to post
Share on other sites
2 hours ago, Stingray~ said:

That was certainly true in Race 10 (NZ v USA), can’t speak to AM’s other races. They emphasized close cover several times. 

Unfortunately they seemed to lose a lot of ground in the couple of tacking duels they have had with ETNZ. They may need to rethink that unless they can get more efficient.

Link to post
Share on other sites
41 minutes ago, uflux said:

Unfortunately they seemed to lose a lot of ground in the couple of tacking duels they have had with ETNZ. They may need to rethink that unless they can get more efficient.

Agreed, they did pay for choosing the tight cover tactics. They did win that race against NZ, but barely. 

Link to post
Share on other sites
17 hours ago, enigmatically2 said:

The distance travelled sideways is proportional to speedxSin(Leeway), so the amount of elevation of the foil that water flows past in a unit of time is

speed x  Sin(Leeway) x Sin (Cant)

In the same unit of time the water flowing over the foil in a fwd direction is speed x Cos (leeway)

Cos^2 + Sin^2 always equals 1. So the total water flow travel is proportional to speed

Thus the AOA is

asin ((speed x Sin(leeway) x(Sin Cant) /speed) = asin(sin(leeway) * sin(cant))

Thsu for example at 10 degree leeway and 20 degree cant you get

AOA = 3.4 degrees

Thanks!  I had sort of got part way down that path, before I thoroughly confused myself. :rolleyes:

Link to post
Share on other sites
On 1/12/2021 at 11:03 PM, MaxHugen said:

@enigmatically2 I think I need help from your trig skills again pls!   I'm trying to determine the AoA on a canted foil for a given Leeward angle. ...

The way I like to determine angle of attack is to resolve the velocity vector into the foil's axis system.  That gives me the three components, Uf, Vf, Wf in the foil's X, Y, Z axes.  (X being in the chordwise direction, Y in the spanwise direction and Z perpendicular to the chord plane)  The angle of attack is arctan(Wf/Uf).  Getting the three velocity components is a matter of applying one direction cosine matrix after another, to resolve the velocity vector into boat axes and then through cant, dihedral, and foil incidence/twist angles to get to the foil coordinate system of interest.

  • Like 1
Link to post
Share on other sites