How to calculate and display true wind from apparent wind and SOG?

[RhythmDoctor]

Hi all,

I've used OpenCPN since about 2011, and still rely heavily on it. A couple years ago I upgraded my boat, and now have a wind gauge. It's an old Raymarine unit, and I have a converter that puts all my Raymarine data onto Wifi in NMEA 0183 format for display on OpenCPN's dashboard plugin. It all works fine.

Unfortunately, my transducer will not read true wind direction/speed unless you hook up the paddlewheel STW transducer, which I don't like to use. I'd much rather use SOG from my GPS to calculate true wind direction/speed, but Raymarine's transducer refuses to use SOG for this. Since it's a rather simple calculation, I was wondering if there is a plugin for OpenCPN that can take VWR and SOG input stream and convert to VWT (or just display on the dashboard as True Wind).

I tried using the ancient PolarCOM 1.9 to make this conversion, but I cannot get it to accept my UDP instrument data.

[AedanC]

The tactics plugin can do it. By default it uses STW for the calculation but there is a place you can tick and it will use SOG instead.

[hoolie]

You can also use NMEA Converter described here in the manual.

It shows an example of how to achieve the conversion of apparent wind to true wind that you're looking for.

[Dockhead]

We discussed this recently.

Note that the differences between ground and water referenced data for calculating True Wind are profoundly important. True Wind is water-referenced and cannot be calculated with SOG (you can calculate something with SOG and heading, but it's not True Wind, nor is it Ground Wind -- it's a bastard hash). If you want Ground Wind, then you need to use COG as well as SOG -- you'll get a real hash if you mix ground and water referenced data in the same calculation.

[RhythmDoctor]

Originally Posted by Dockhead

We discussed this recently.

Note that the differences between ground and water referenced data for calculating True Wind are profoundly important. True Wind is water-referenced and cannot be calculated with SOG (you can calculate something with SOG and heading, but it's not True Wind, nor is it Ground Wind -- it's a bastard hash). If you want Ground Wind, then you need to use COG as well as SOG -- you'll get a real hash if you mix ground and water referenced data in the same calculation.

Sorry, I think I misspoke. I’m meant to say True Wind speed and angle (not direction). I agree that converting that to wind direction would require COG, which is readily available. It’s very helpful for comparing to weather reports, which are all referenced to ground without regard for set or drift. I don’t want to argue whether True Wind or Ground Wind is more helpful - I’m just interested in extracting whatever I can without having to use my paddlewheel.

[Dockhead]

Originally Posted by RhythmDoctor

Sorry, I think I misspoke. I’m meant to say True Wind speed and angle (not direction). I agree that converting that to wind direction would require COG, which is readily available. It’s very helpful for comparing to weather reports, which are all referenced to ground without regard for set or drift. I don’t want to argue whether True Wind or Ground Wind is more helpful - I’m just interested in extracting whatever I can without having to use my paddlewheel.

OK, I understand, but nevertheless, with any current running, you will get big distortions of both angle and speed if you use COG, distortions which can easily make the calculated data absurd. If you are using True Wind when running off to avoid a gybe, and all the more if you're using a pilot in wind mode, this issue can f you up big time, so this has very practical consequences.

This is one of the reasons why many use ultrasonic speed logs. To understand what true wind is doing. But one of your OpenCPN colleagues pointed out in another thread that even with accurate STW data, True Wind calculations will be distorted by leeway


A bit off topic and FWIW, there is nothing about Ground Wind which inherently links it to compass degrees as opposed to angle to your bow. It's just commonly displayed that way because we commonly want to know about Ground Wind for large scale planning, not to know what the wind is doing right now in relation to our boat. But if you had a steady wind blowing and you're about to get a tide shift, you could imagine wanting to know Ground Wind as an angle to your bow.

[RhythmDoctor]

I’d be happy to look at another thread for more discussion on this, but please post a link. I use other forums for general sailing info, and not familiar with CF aside from OpenCPN support from my “colleagues”.

I’m not sure why my question is so controversial. I’ll always use apparent wind for sail management, so Ground Wind is not going to affect gybing. Believe me, I’ve seen enough absurd results when calculating True Wind from a paddlewheel too. I just think Ground Wind would be nice to have for comparison with weather reports, and see no reason why that can’t be calculated from GPS data.

[Dockhead]

Originally Posted by RhythmDoctor

I’d be happy to look at another thread for more discussion on this, but please post a link. I use other forums for general sailing info, and not familiar with CF aside from OpenCPN support from my “colleagues”.

I’m not sure why my question is so controversial. I’ll always use apparent wind for sail management, so Ground Wind is not going to affect gybing. Believe me, I’ve seen enough absurd results when calculating True Wind from a paddlewheel too. I just think Ground Wind would be nice to have for comparison with weather reports, and see no reason why that can’t be calculated from GPS data.

Well, I don't think that there is anything really controversial about any of this. It's simple vector math.

Let's say you're sailing S to Cherbourg at 6 knots, with 3 knots of current running W -- typical Channel crossing (at springs can be more current). The wind is blowing from the SW at 12 knots in relation to the ground.

With simple vector math, you can calculate that Ground Wind of 12 knots and 225 degrees will be True Wind of 14.28 knots and 216.543 degrees. Angle from your bow will be Ground Wind 45 degrees, and True Wind 36.457 degrees.

As you see, it makes a big difference which could lead to a significant misjudgement of a tack -- that's just about the difference between sailing hard on the wind or not sailing. Or a significant misjudgement of when you are at risk of an accidental gybe. It would be downright dangerous to run DDW with your pilot set on wind mode, if you are using SOG as your speed data.


You are of course absolutely right that a dodgy paddlewheel throws a monkey wrench into True Wind calculations. But using data with the wrong reference is hardly an improvement. Accurate STW is right up there with accurate heading data, in my opinion, for what's important to get out of your instruments, if you are sailing keenly. I personally don't care nearly as much about position, for example.

[Dockhead]

Originally Posted by RhythmDoctor

. . . I’m not sure why my question is so controversial. I’ll always use apparent wind for sail management, so Ground Wind is not going to affect gybing. Believe me, I’ve seen enough absurd results when calculating True Wind from a paddlewheel too. I just think Ground Wind would be nice to have for comparison with weather reports, and see no reason why that can’t be calculated from GPS data.

Sorry, I didn't specifically answer this question.

If you don't look at True Wind for sail management then of course you won't miss not having it.

But do you really not?

Apparent wind is useless on a run, especially in light wind. It swings way back and forth according to small variations of course and is not useful in predicting gybe risk. Running wing on wing or with a kite, you really want to be looking at True Wind and ignoring Apparent Wind.

That's why autopilots switch to True Wind when in wind mode, when the wind gets a certain amount behind the beam. Using the pilot in wind mode is supremely useful on a long downhill run, for a shorthanded cruiser, and it doesn't work without decent True Wind calculations. Hell, I can't steer BY HAND DDW without True Wind.

True Wind is also very useful when sailing hard on the wind, and for tacking tactics -- only way to know where your course is going to be on the other tack over a variety of conditions and sea states.

Reaching of course you don't need True Wind at all -- Apparent tells you everything. Maybe you're one of those lucky guys who gets to reach all the time


And yes, I agree with you, Ground Wind is very useful for its own purposes. Understand what will happen to the wind when the tide changes is one; comparing with weather forecast is another -- just to name two.

I watch Ground Wind on the "Weather" screen on my Maretron DSM-250, which also graphs the trends in Ground Wind.

[RhythmDoctor]

None of this matters for how I intend to use this calculated value. It’s dangerous to give too much sailing tactic credence to any calculated value whether you use STW or SOG. You’ve got different response times, different nonlinearities, and additivity of variances for multiple inputs in any calculated value. But it’s good enough to get a sense of general weather conditions, and better than nothing, which is what I have now.

This sub-forum is for OpenCPN support, so further discussion does not belong here. Please post a link to the relevant discussions elsewhere inCF and I’ll have a look.

[Dockhead]

Originally Posted by RhythmDoctor

None of this matters for how I intend to use this calculated value. It’s dangerous to give too much sailing tactic credence to any calculated value whether you use STW or SOG. You’ve got different response times, different nonlinearities, and additivity of variances for multiple inputs in any calculated value. But it’s good enough to get a sense of general weather conditions, and better than nothing, which is what I have now.

This sub-forum is for OpenCPN support, so further discussion does not belong here. Please post a link to the relevant discussions elsewhere inCF and I’ll have a look.

Here's one ancient but still relevant discussion:

http://www.cruisersforum.com/forums/...ion-73563.html

You should also read this chapter in the B&G Essential Guide to Sailing Instruments, called "Why can’t I use SOG and COG for calculating
True Wind?", starting on Page 8 in this document: https://www.google.com/url?sa=t&rct=...1tWI6XEtbyBaqK

Good technical basis, useful for OpenCPN work:

https://www.google.com/url?sa=t&rct=...fhXx30iYb3SWLi


A very good article on why boat speed cannot be simply dispensed with:

Why Bother with Paddles - Ockam Sailing Instruments


A useful illustration:

[rgleason]

Rhythm Dr. we have had many such discussions in the Tactics_pi thread. I personally think it is a good healthy discussion worth having, but you may differ.
I am actually going to link to Dockhead's post under Tactics_pi FAQ as there are good links there. Tactics does offer the option, and it should work.

[RhythmDoctor]

If you’re fine with it, that’s great. For this thread, I just would prefer to focus on “How to do it,” and let the discussion of whether it’s a good thing to do happen in a different thread.

FWIW, I am aware of the extreme instability of apparent wind readings when heading DDW. But I’m unconvinced that using calculated true wind provides anything more than false confidence. I see it as GIGO. But I’ll continue to read the arguments as time allows.

[rgleason]

R Dr,
I think I neglected to mention use of NmeaConverter which just was released by Dirk.
There is an example that you can adjust which I started with the formulas and others got it working in nmeaconverter quite some time ago.

You could just adjust the formulas as needed for SOG
See Calculate True wind from apparent wind
PS this is using SOG

[RhythmDoctor]

Originally Posted by rgleason

R Dr,
I think I neglected to mention use of NmeaConverter which just was released by Dirk.
There is an example that you can adjust which I started with the formulas and others got it working in nmeaconverter quite some time ago.

You could just adjust the formulas as needed for SOG
See Calculate True wind from apparent wind
PS this is using SOG

Thanks, I was already working on it for the last few hours. I posted a glitch over on the thread.