Cruisers Forum
 


Join CruisersForum Today

Reply
 
Thread Tools Rate Thread Display Modes
Old 16-01-2015, 12:27   #31
Registered User
 
Seasick Steve's Avatar

Join Date: Jan 2015
Location: England
Boat: Westerly Centaur, 26
Posts: 32
Re: Planning a short journey to windward, with a backing wind

I repeat my earlier comment:

I think any attempt to try and calculate precisely beforehand what you're going to do in a sailing boat is doomed to failure. There are so many variables, some partly predictable, most less so and they interact in so many ways, not all of them understood, that it's just too complicated.

Apart from wind you've got tidal stream, currents, waves, traffic, direction of sun and no doubt other factors all have an effect. If you can take account all of those in your algorithm, good luck!
__________________

__________________
Seasick Steve is offline   Reply With Quote
Old 16-01-2015, 13:44   #32
Moderator
 
JPA Cate's Avatar

Join Date: Nov 2011
Location: aboard, cruising in Australia
Boat: Sayer 46' Solent rig sloop
Posts: 10,657
Re: Planning a short journey to windward, with a backing wind

Good one, Steve, I'm with you, but it I think it is a way Matt has of playing when he can't go sailing and try it out!

When you're out there sailing, you just have to make adjustments for unanticipated variables, and there are too many to include in a mathematical model, for normal mortals, anyways.

Ann
__________________

__________________
Ann, with Jim, aboard US s/v Insatiable II, in Oz, very long term cruisers
JPA Cate is offline   Reply With Quote
Old 18-01-2015, 00:15   #33
Bailing as fast as I can.
 
GILow's Avatar

Join Date: Sep 2008
Location: Adelaide, South Australia
Boat: Swanson 42
Posts: 3,585
Re: Planning a short journey to windward, with a backing wind

Quote:
Originally Posted by DefinitelyMe View Post
Well, here's another completely un-tried and un-tested theory! The curve will be a y=x^2+mx+c function......... Maybe if you model it in vector form, then you could define limits when the vector angle to the destination is greater than 75d (the point where one tack becomes favoured over the other), and then have another x^2 function. Would that work?



I didn't realise when i wrote my first 'solution' that it isn't a gradual wind change but rather a shift that takes place over an hour or so. In that case most of your course will be a straight line and you just need a bit in the middle, at whatever time you expect the wind shift, to model that section. If it shifts in the manner you described then the rate of change follows a logistic curve. The maths for this has already been done - it's just the same as a tidal change but condensed into the period over which the shift takes place....... I seem to remember learning about finding optimum values when i was at school...... I think you define what your optimum value is and then integrate the function to figure out where it lies on the curve....... or something. It's been too long!


Close, but no cigar. The equation is not a regular quadratic. I wish it were, it would be much easier.

Nope, (simplistically) boat speed is a function of two variable, wind angle and wind velocity. (I can ignore currents for my model, thank goodness, they are negligible on this journey.). Even if I keep wind velocity constant for the sake of simplicity, the angle, and thus the boat speed, changes with time.

For the model where you start on the starboard tack and then change to a port tack, you need to solve for two equations such that t ( time) is minimised. This may not necessarily coincide with distance being minimised either, just ask any of the clever sailors who told me to tack downwind.

This is where I get really frustrated, as in theory that's no great drama. But for the life of me I cannot get the two equations in a form where I can solve for the minimum t.

For the other model, where you stay on the starboard tack, the variable I want to solve for is still t where t is the time at which the wind shift commences. I.e. If the wind shifts at time t, will you need to tack at the end of the journey?

If I still lived in the town where I went to university I would shamelessly go back to the uni and attempt to bribe my lecturers with trips on the boat if they would help me build the equation for even the simplest model of constant velocity wind. Alas they are 800 km away.

The brute force computational method is more likely to be in my grasp, but writing the code feels a bit too much like a day at the office.

And yes, I concede the point about the unpredictable nature of these things. This morning I set off for home, 30 miles due east of my position into what was forecast to be 10 knot Easterly for the first two hours. As it was I spent the first two hours with the MPS poled out running pretty much before the wind, before an abrupt SE shift came in and I pointed the whole way home in a straight line. The track on the plotter looks like I used a ruler.

Go figure.

Matt


Sent from my iPad using Cruisers Sailing Forum
__________________
Very funny Scotty, now beam down my clothes.
http://www.swansonsailor.id.au
GILow is offline   Reply With Quote
Old 18-01-2015, 00:43   #34
Senior Cruiser
 
Jim Cate's Avatar

Cruisers Forum Supporter

Join Date: May 2008
Location: cruising SW Pacific
Boat: Jon Sayer 1-off 46 ft fract rig sloop strip plank in W Red Cedar
Posts: 11,440
Re: Planning a short journey to windward, with a backing wind

Matt, your experience today exemplifies why one shouldn't waste too much time on modeling these sorts of adventures. The ability to provide good values for the variables is just not there, and I reckon you are wasting your time and getting frustrated to boot. A lose/lose conundrum IMO.

To me, the art of sailing successfully is to have a grasp of the fundamentals (ie relationship between AWA, AWS and boat speed) and the ability to fold that into a developing situation as you travel towards your destination. Trying to fine tune it before departure... well, good luck!

It is interesting to get the general picture of the prospective voyage, and then to watch for visual and instrumental inputs that provide some more or less factual data as you travel. Things like watching for build up of uplift clouds over the land signaling the developing sea breeze, a swell coming in from a new direction... the things sailors have always watched.

I'm sure that your analytical mind can do a better ad hoc job of this in real time than some contrived formulaic approach done with poor input data ahead of time. When you add the ideas of liking to do the hard parts early in the day, the maths fly out the window.

I admire your enthusiasm for the number crunching, BTW. Don't think that I am putting your efforts down... I just don't think they will produce a useful model.

Sail on, and enjoy... just think how good it will feel when you get it right using intuition!

Jim
__________________
Jim and Ann s/v Insatiable II , lying Port Cygnet, Tasmania once again
Jim Cate is offline   Reply With Quote
Old 18-01-2015, 03:06   #35
Registered User
 
Seasick Steve's Avatar

Join Date: Jan 2015
Location: England
Boat: Westerly Centaur, 26
Posts: 32
Re: Planning a short journey to windward, with a backing wind

Quite.

And dare I mention that the boat sails in its apparent wind, not the true wind that we've been talking about?
__________________
Seasick Steve is offline   Reply With Quote
Old 18-01-2015, 04:41   #36
Eternal Member
 
monte's Avatar

Cruisers Forum Supporter

Join Date: Jul 2011
Location: Australia
Boat: Lagoon 400
Posts: 3,650
Images: 1
Re: Planning a short journey to windward, with a backing wind

I would start on a starboard tack to make the most to weather before it starts to turn. When it backs a bit, maybe 10 degrees, tack to port. When back at the rhumb line tack to keep to weather. If it backs again, tack to port. Plan to be hard on the wind for all of the trip for the fastest time overall. Beating then beam reaching won't be the fastest (otherwise we would do that when racing).
The dotted lines are for a sudden wind shift at 1400, and the solid line is for the more likely slowly backing till 1400. Tacking about every hour should work and save 30 mins or so on the passage. Click image for larger version

Name:	ImageUploadedByCruisers Sailing Forum1421581241.467303.jpg
Views:	91
Size:	46.1 KB
ID:	95638
__________________
monte is offline   Reply With Quote
Old 18-01-2015, 06:09   #37
Senior Cruiser
 
delmarrey's Avatar

Cruisers Forum Supporter

Join Date: Mar 2003
Location: Now in Blaine, WA
Boat: Modified Choate 40
Posts: 10,702
Images: 122
Re: Planning a short journey to windward, with a backing wind

Have you considered motorsailing through the transition. Then go on the port tach once the shift is over.

Motorsailing can point you farther into the wind. It's cheating but you're on a schedule, right.
__________________
Faithful are the Wounds of a Friend, but the Kisses of the Enemy are Deceitful! ........
A nation of sheep breeds a government of wolves!

Unprepared boaters, end up as floatsum!.......
delmarrey is offline   Reply With Quote
Old 18-01-2015, 13:52   #38
Bailing as fast as I can.
 
GILow's Avatar

Join Date: Sep 2008
Location: Adelaide, South Australia
Boat: Swanson 42
Posts: 3,585
Re: Planning a short journey to windward, with a backing wind

Again, in no particular order... Monte, that's a terrific paper analysis, thank you. I will very interested to see how it compares to my overly complicated solution.

Steve, yep, I know boats sail in relative wind, but they create that relative wind from their motion, and that motion is a product of the wind speed and the boats sailing capability. So from a modelling perspective it easier to treat the trip as a plane with two x and y coordinates and superimpose the wind direction and speed on the plane. Then any boat's potential speed and direction can easily be calculated at any point on the plane in time.

Jim, yesterday morning I woke at 6 am, staggered up on deck to have a look around, only to discover that the wind was blowing exactly 180 degrees from the forecast direction, towards home, not away from it. So Intuition told me to skip breakfast and make hay while the sun shone. Hence a delightful hour or so of MPS run instead of three hours beating waiting for the forecast wind shift, which, as it turns out, never eventuated either. So yeah, intuition probably wins over dumb plans every time. But I still think it is a valid excercise, so long as one does not ignore ones eyes and ears. (And in my case, one ignores ones stomach as I did not have breakfast till a bit later than I'd have liked, but I suppose breakfast was all the nicer for the presence of the MPS.)

Delmarry, your motor sailing option is, of course, totally logical, unless you have met our engine. It is horrendous, noisy and shakes the boat. So I try to use it as little as possible. ( but that's a whole separate project at the moment anyway.)

Part of me thinks I am totally over analysing this, but part of me thinks that this is a simple problem with two solutions, one of which may be significantly better and it would be nice to set off with a plan. (Maybe only to watch the plan go to crap). But given there seems to be much disparity of views here on CF about the best approach, it seems only logical to see what maths and logic come up with. Even if their opinion is subsequently ignored by everyone including me.

Matt


Sent from my iPad using Cruisers Sailing Forum
__________________
Very funny Scotty, now beam down my clothes.
http://www.swansonsailor.id.au
GILow is offline   Reply With Quote
Old 18-01-2015, 13:59   #39
Bailing as fast as I can.
 
GILow's Avatar

Join Date: Sep 2008
Location: Adelaide, South Australia
Boat: Swanson 42
Posts: 3,585
Re: Planning a short journey to windward, with a backing wind

Quote:
Originally Posted by monte View Post
I would start on a starboard tack to make the most to weather before it starts to turn. When it backs a bit, maybe 10 degrees, tack to port. When back at the rhumb line tack to keep to weather. If it backs again, tack to port. Plan to be hard on the wind for all of the trip for the fastest time overall. Beating then beam reaching won't be the fastest (otherwise we would do that when racing).
The dotted lines are for a sudden wind shift at 1400, and the solid line is for the more likely slowly backing till 1400. Tacking about every hour should work and save 30 mins or so on the passage. Attachment 95638

Monte, I reread this, as it was interesting at a number of levels. All that you write seems right to me, but I should add a caveat that tacking is a bit of a negative for me. Firstly, I am usually solo, so it's a bit of work dragging the Yankee around the staysail etc, and secondly our 18 tons takes a while to get moving again. We sure 'ain't no round the cans day racer. So I maybe should have added that requirement to my original problem description.

Secondly, you make me realise that I need to better understand our boat characteristics. For instance how much better are we when reaching, or at least not sailing hard on the wind... My gut feeling after two years of sailing the boat is that there is a big enough difference to be important, but the extra distance required may negate the performance improvement anyway.

Still, more to consider, thank you.

Matt


Sent from my iPad using Cruisers Sailing Forum
__________________
Very funny Scotty, now beam down my clothes.
http://www.swansonsailor.id.au
GILow is offline   Reply With Quote
Old 19-01-2015, 09:02   #40
Registered User

Join Date: May 2013
Location: Mangareva, French Polynesia
Boat: Heritage West Indies 36
Posts: 513
Re: Planning a short journey to windward, with a backing wind

Do any of you use a computer program called 'visual passage planner'? It's route optimisation software which is supposed to tell you the optimum route to take from A to B. You input your basic boat characteristics, tell it where you want to go and hit go. It uses wind and current data (presumably the same data that is used to make pilot charts) and models a load of different scenarios to give you an optimum route. I have found it to be, on the whole, useless, however the mathematics should be sound in theory and i don't see why they couldn't be applied to shorter passages. I know nothing about computer code so forgive my naivety if this is a non-suggestion, but if you're a coder could you take a program like this apart and apply their coding to what you're trying to do? I'm not suggesting it would be any more useful than the visual passage planner thingy but it would be an interesting academic undertaking!
__________________
DefinitelyMe is offline   Reply With Quote
Old 19-01-2015, 17:04   #41
Bailing as fast as I can.
 
GILow's Avatar

Join Date: Sep 2008
Location: Adelaide, South Australia
Boat: Swanson 42
Posts: 3,585
Re: Planning a short journey to windward, with a backing wind

Well, given the fact that the application has proved to be "on the whole, useless.." I am not sure there would be a great advantage in trying to decompile it for the underlying code logic. Except maybe as an indication of what NOT to do.

In answer to your question though, yes, in some cases it is possible to reverse engineer software and figure out how it was written, provided you have the time and energy. It is supposed to be getting harder now with various code obfuscation tools, but I do not know how widely they are used. I have always give my source code to my clients, or worked with open source code, so it has never been a big issue for me professionally.

I think the point made here a couple of times is that such pure models do not map so well to the real world, but I love a challenge, so I am going to keep thinking and sketching on this one and see what I come up with.

I am a bit relieved to hear there is software for this sort of thing, regardless of how well it works, it reassures me that I am not completely mad for considering such an idea. (Or maybe the creators of such software were also nuts... )

Matt
__________________
Very funny Scotty, now beam down my clothes.
http://www.swansonsailor.id.au
GILow is offline   Reply With Quote
Old 19-01-2015, 17:11   #42
Bailing as fast as I can.
 
GILow's Avatar

Join Date: Sep 2008
Location: Adelaide, South Australia
Boat: Swanson 42
Posts: 3,585
Re: Planning a short journey to windward, with a backing wind

OK, that officially freaks me out.

I just looked up that software, and found their web site. I had look through the various product pages, and stumble on my very own journey, there, on the site.

Wierd.

M
Attached Thumbnails
Click image for larger version

Name:	Untitled.png
Views:	91
Size:	40.0 KB
ID:	95734  
__________________
Very funny Scotty, now beam down my clothes.
http://www.swansonsailor.id.au
GILow is offline   Reply With Quote
Old 20-01-2015, 09:08   #43
Registered User

Join Date: May 2013
Location: Mangareva, French Polynesia
Boat: Heritage West Indies 36
Posts: 513
Re: Planning a short journey to windward, with a backing wind

Quote:
Originally Posted by GILow View Post
OK, that officially freaks me out.

I just looked up that software, and found their web site. I had look through the various product pages, and stumble on my very own journey, there, on the site.

Wierd.

M
And how does it's recommended route compare to your experiences?

I might have been a bit harsh when i said it was 'completely useless'! I usually figure out what i'm going to do, then get this program to model it. If they agree then all is well. If they don't agree then i try to figure out if i'm missing something important. 9 times out of ten though, whatever route it picks is very optimistic. It also doesn't take localised effects into place, or the positions of actual weather systems at the time. So, it's good for getting an idea of the kind of thing you should be doing but not much use on the day since your actual decisions will be based on more immediately relevant weather data, not the average of several years which the model is based on.

I see no reason however why the same algorithms couldn't be used with any data set, including one generated by a short-term weather forecast.

Such software exists. When i did the Newport-Bermuda race last year we had some software that cost an exorbitant amount of money. GRIB files downloaded on a sat phone including wind as well as gulf stream data went straight to the laptop and the software then spat out our optimum route. it had been pre-programmed with the entire sail inventory, sailing angles for each sail and target boat speed for each sail at any given wind angle and wind speed. I'm kind of an old-school sailor (I started out in tallships) so i was kind of pleased when this fancy new system that the navigator and tactician had been hounding on about for the last few months packed up on day two and we had to figure out the rest by ourselves. As it turned out, a high-pressure to the west was larger than the software had predicted. All the other boats went west because the software told them to, we went east because we felt like it and we sailed around the entire fleet and won lots of prizes

Anyway i digress. The point is, that software is stupidly expensive but still very fallible. After all, a model - however complex, and however good - is only as good as the data that goes into it, and weather data is notoriously imperfect. But, if you took the same kind of coding that is used in a simpler model, like the one used by Visual Passage Planner, and applied it to whatever data set someone wanted to input.......... and made it available at no cost; well, i think you'd have a lot of very grateful sailors! It might not be perfect, but it would still be a great tool As far as i know there isn't currently any open-source, freeware that does this. Is there?
__________________
DefinitelyMe is offline   Reply With Quote
Old 20-01-2015, 14:13   #44
Moderator
 
JPA Cate's Avatar

Join Date: Nov 2011
Location: aboard, cruising in Australia
Boat: Sayer 46' Solent rig sloop
Posts: 10,657
Re: Planning a short journey to windward, with a backing wind

Well, Matt, the gauntlet is on the table now!
__________________
Ann, with Jim, aboard US s/v Insatiable II, in Oz, very long term cruisers
JPA Cate is offline   Reply With Quote
Old 20-01-2015, 15:17   #45
Bailing as fast as I can.
 
GILow's Avatar

Join Date: Sep 2008
Location: Adelaide, South Australia
Boat: Swanson 42
Posts: 3,585
Re: Planning a short journey to windward, with a backing wind

I should have stayed quiet...

In no particular order... the picture from the software site only shows the Gulf St Vincent with Port Adelaide (my home port) and Port Vincent (my destination port). Sadly, they don't tell you the best way to get there. I still reckon, with 70% of the earth covered in water, it's a pretty strange coincidence they chose that particular screen shot.

So, just to step back a bit, my original question was what I THOUGHT was a very simple one, go left or go right?

So far the very clever, capable sailors here have made it totally clear that:

a) There are actually THREE options. Go left. Go right. Start the motor and go right down the middle. Great. I started with two options, now I have three.

b) There's absolutely no point planning on anything anyway since the wind will blow from the wrong direction, the motor probably won't start when it matters, and anchoring in the middle won't help because I have the wrong sort of anchor anyway.

At this point a fourth option is starting to appeal, stay home. I confess, that as I plodded across the Gulf St V on Saturday, feeling slightly sea-sick from the horrid sea-state left from a Friday night blow, I did wonder why I was doing all this anyway, I could have read the same book I was reading on my sofa.

BUT, I am mathematician and a programmer, so I think I can still answer what was a slightly theoretical question anyway. I am going back to the left vs right question, and I will solve it, if only for the pleasure of having done so.

Out of this process will come a piece of very accurate, useable software that I WILL make readily available to all users of CF, with no restriction, at no cost, most probably via my web site. This piece of software may be freely used (once written) by ANY CF member who wishes to know whether it is theoretically better to go left or go right when leaving Port Adelaide for Port Vincent with a forecast Westerly backing to a forecast South Westerly in a big, heavy old cruising yacht that can do about 50 degrees from the true wind direction. Aforesaid CF members are cordially invited to the Gulf St V (if they do not reside here already) to test the validity of the assertions made by the software, and in return I will do my utmost to ensure they get one of the better visitors berths at our yacht club (at the normal very reasonable rates) and will also assist them in choosing from the local beer selections, based on my considerable wealth of experience and impeccable good taste.

Matt
__________________

__________________
Very funny Scotty, now beam down my clothes.
http://www.swansonsailor.id.au
GILow is offline   Reply With Quote
Reply

Tags
wind

Thread Tools
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Crew Available: Punta Gorda FL - Short Day or Short Trips Mediaguy Crew Archives 0 13-12-2014 16:41
Calamity! ICW Journey Log #1 Alexei Sailor Logs & Cruising Plans 2 08-06-2009 07:13
ICW Journey Alexei Sailor Logs & Cruising Plans 21 08-03-2009 17:11
East Med Journey swagman Europe & Mediterranean 8 18-02-2006 00:33
Already in the Caribbean: Journey of Makai sv_makai Sailor Logs & Cruising Plans 1 20-07-2005 02:09



Copyright 2002- Social Knowledge, LLC All Rights Reserved.

All times are GMT -7. The time now is 06:25.


Google+
Powered by vBulletin® Version 3.8.8 Beta 1
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
Social Knowledge Networks
Powered by vBulletin® Version 3.8.8 Beta 1
Copyright ©2000 - 2017, vBulletin Solutions, Inc.

ShowCase vBulletin Plugins by Drive Thru Online, Inc.