True and Apparent Wind

I have to agree with Talbot here - the ship's heading is only required if you want to know the true wind direction in relation to the ground, rather than the true wind angle (which is all most instruments give you).

I could prove this mathematically but it should be obvious if you ask yourself the question "Why does the heading of the boat make a difference to the relationship between true and apparent (relative) wind directions - all other things being constant?"
 
refer to the original query - true wind direction
for that one needs a reference to north (or other fixed bearing)
boats heading is needed if moving etc ...
 
golly what a lengthly thread to a simple question! I would say that you only need apparent wind. The performance of the boat is related only to the wind the boat 'sees' which is apparent wind. 'True' wind as calculated by many instruments only uses the log speed as the other input so is not true speed at all - it is a good approximation in strong winds with minimal tides but in light winds with alot of tide it is a long way out.
with regard to the origonal query I'm not actually sure you need a wind instrument for that - the one time I don't look at mine is when close hauled when I just look at the tell tales.
 
Yes I asked Raymarine about this and they replied that it only took SOG & COG into account when boatspeed was greater than somthing like 3 knots. That's bollix too as I I have never seen any step change in their "True Windspeed" display with an increase in boatspeed. I guess they just didn't want to admit it wasn't true windspeed, it would be a "trade descriptions" lapse in any other product!
 
I did an experiment on heading data display. I compared readings from my Simrad autopilot display with those on my radar display (direct NMEA feed) and a Seatalk ST60 Multi display (NMEA/Seatalk bridge feed). I "pulled" the compass off heading 45 degrees using a magnet which was then quickly withdrawn. The radar display took 6 seconds longer to return to correct heading than the autopilot display and the Seatalk dispaly took a whole 23 seconds longer. However, I didn't look at the response setting on the Multi but 23 seconds is a pretty useless result! This was done when I only had a 1Hz compass NMEA output. Perhaps I should repeat this now I have a 10Hz heading data rate input.
 
Actually if you read the original post the word "direction" is not mentioned.

In the context of most boat instruments True Wind usually refers to TWA, YWS
 
Just ran your theory through our software using:

AWS 30 kts
AWD 045 deg
COG 000 deg
SOG 7 kts

On a heading of 315 degs, TWS is 23.0 kts and TWD 000 deg.

Trying to do it your way we get a result of TWS=25.5 kts and TWD=011 degs. Your method simply doesn't work, sorry.
 
Isn't it nice when everyone can all agree.
This is when everyone is sat at the computer. God knows what results people are getting out on the water.
It's all a bit esoteric isn't it? A previous thread couldn't even get a concensus on the Starboard Tack rule.
 
Going to get a bit controversial here, but it would help if people actually knew what they were talking about before they posted. Too many posts are 'I think' or 'I believe' when in fact they haven't a clue

In some of the more esoteric subjects, that's fine, but when people are posting inaccurate info on quite basic subjects........
 
I believe I think I know that our ST50 wind has stopped calculating what it thought to be true wind (if the manual is to be believed) since we changed the failed ST50 tridata for an ST60 tridata (I'm sure I believe the decription on the box and invoice that it is an ST60 tridata and I think I have installed it correctly)

Certain enough? Or perhaps I don't have a clew ... oh no - thats a mobo!
 
Why bother with true at all, just tack and you know near enough.... Tide will of course also have an effect on apparent...
Back in the 70's we spent days on "Gunfleet", sailing different wind angles with different sails to build up polar diagrams fir the boat and wind speeds..
What it did was tell us when to change headsails for max speed...But we still didn't make the cup team...
Surely your windspeed has to be coupled to a GPS signal to arrive at the true wind direction... Or should you have two true winds one relative to the ground and one to the water ? The mind boggles... Just sail the bloody things... Close your eyes and you'll learn more in 5 mins than you did on your last cruise..
Mind you shouldn't do this solo!!
Of course I have GPS coupled to the laptop for longer trips ( Don't want me eyelids sunburnt) /forums/images/graemlins/cool.gif /forums/images/graemlins/blush.gif
Have fun Cheers Bob E....
 
From my son, a computer programmer, who in a past job wrote software for a well known brand of marine instruments...


Here's how to do it, well some C++ code anyway. The inputs are boat speed through the water, apparent wind speed and apparent wind angle. This method takes no account of tide set and drift, and boat leeway. The 'true' wind data calculated is really wind speed/angle over the sea rather than over the land, because no account of tide is taken.

The code...
<font color="red">
TReal result;
TInt error;

TReal windAngleApparentRadians=iWindAngleApparent * KDegToRad;
error=Math::Sin(result, windAngleApparentRadians);
if(error!=KErrNone)
return;
TReal l=result * iBoatSpeed;
error=Math::Cos(result, windAngleApparentRadians);
if(error!=KErrNone)
return;
TReal q=result * iBoatSpeed;
TReal p = iWindSpeedApparent - q;
error=Math::Sqrt(iWindSpeedTrue, (p*p + l*l));
if(error!=KErrNone)
return;
if(p>0.0)
{
error=Math::ATan(result, (l/p));
if(error!=KErrNone)
return;
}
else
return;
TReal windAngleTrueRadians=result+windAngleApparentRadians;
iWindAngleTrue=windAngleTrueRadians * KRadToDeg;
</font>

In the code above the calculations are done in radians rather than degrees, but the formulae used boil down to those below. A special check has to be made when using arctan to avoid a division by zero error, so without the check, the formula on the last line may not always work. <font color="blue">

l = sin(wind_angle_apparent) * boat_speed
q = cos(wind_angle_apparent) * boat_speed
p = wind_speed_apparent - q
wind_speed_true = square_root(p*p + l*l)
wind_angle_true = arctan(l/p) + wind_angle_apparent

</font>
 
I have just plotted the folowing:
COG 000
SOG 7kts
AWS 30kts
AWD 045 (thus wind from green 045)

results true wind from G 055 speed 26 kts

Thus on a heading of 315, the TWD is 010 TWS 26 - which confirms your plotting
and immediately reveals a fault in your software.

if the AWS is 30, and the TWS is 23, and your own speed is 7, the only course that satisfies this is directly into wind! /forums/images/graemlins/blush.gif

BTW I am using a system that has been in use in the military for at least 30 years for calculating courses within helicopter operating limits from ships - always worked for me.
 
Angus

Thanks, this is very interesting. When I ran some sample data through your algorithm the results were close to what my instruments were generating for 'true wind'. However, this method is totally incorrect as true wind is what would be experienced by a fixed object anchored to the seabed. Therefore heading and both SOG and COG have to be taken into account on a moving platform such as a yacht.

As you say, it takes no account of tides or leeway, nor heading. Since many of us spend 100% of our time in tidal waters this illustrates why 'true wind' from most intruments is not worth having.
 
Talbot

I deliberately chose a COG direct to windward to illustrate the error in your method, although the apparent wind is 45 degrees off the bow. If your course is direct to windward then it follows that the apparent and true wind direction cannot differ. Surely you cannot argue with this?

I'm no stranger to landing helicopters on ships, and since the ship would generally be turned direct to windward for the landing the error in your method may not have been apparent or even significant.
 
Tome,

I really dont understand where you are going here.

Your last post states that you chose a COG direct to windward, but the earlier post talks about a apparent wind of 45 degrees off the bow. I suspect that we are talking at cross purposes here somewhere and I am getting very confused.

Your work with helicopters may desire the ship turning into wind, but my experience was of making full use of the helicopter landing limits to minimise interference with other tasking, thus cross deck was the norm, and even a wind abaft the beam was quite usual. Thus considerable expertise was required in working apparent to true and back to apparent.
By preference military pilots prefer the wind from abt 30 degrees on the bow from the same side of the aircraft as the pilot.

I am not trying to be difficult here, just trying to understand what the @@@@ is going on /forums/images/graemlins/confused.gif
 
Having gone to the thread, rather than just answering the individual threads, I see where the errors are occuring. The plotting method I have used so far is assuming no significant impact from extrenal influences such as current and drift. The simple plotting system is what it says on the box - a simple system. if you want to add in the difference between COG and SOG and ships head and speed through the water, they can be plotted just as easily, but just takes another stage in the plotting cycle, and shows the same answers as your first post.

I accept for a yacht or slow moving ship that the extra stage is necessary, but your example was using a combined drift and current effect of 5 knots towards 072 (assuming speed through the water was the same 7 knots as that over the ground). yet you didnt plot that (or comment on it in your post) I can only assume that it was so obvious to you that you felt it didnt need mentioning, - but if it was so obvious, why did you not include it in the plotting - it also gets away from the original discussion, where it was claiming that you need to have an accurate electronic compass input - now I may be getting senile, but that will stilll only give course through the water, and if that and speed through the water is the basis for the true wind calculation rather than speed over the ground and course over the ground, simple plotting based on my method is just as accurate. If you want to go down the ultimate accuracy where you tie COG, SOG from your GPS (or plotting) into the equation, then I still contend that plotting will still work out, but you will need to do a more complex plot.
 
[ QUOTE ]
Your last post states that you chose a COG direct to windward, but the earlier post talks about a apparent wind of 45 degrees off the bow. I suspect that we are talking at cross purposes here somewhere and I am getting very confused.


[/ QUOTE ]

Believe me, I'm not trying to be obtuse. It's a subject which is clearly misunderstood and it happens that we've done a lot of work and trials here. Nothing to do with sailing or helos- it's to do with predicting movement of towed surface arrays which have significant windage. When out sailing, I couldn't really give a monkeys about true wind!

I chose an example where both COG to windward and apparent wind 045 are true to simplify matters. The boat is sailing on a starboard tack at an apparent wind of 45 degs, which is a feasible point of sail. However, there is clearly a significant cross-tide running across her bow which is pushing her COG directly into the wind. Since she has a good angle of attack, she will still sail happily on this course.

Since the boat is making way direct to windward, the apparent and true wind are now coming from the same direction and it follows that the true wind speed is equal to apparent minus SOG in this unique instance. This makes the true wind speed 23 knots. I think you'll agree with this?

So, an instrument which correctly calculated true wind would show it as 23 knots and Green 45 degrees.

What I'm trying to illustrate is that with an incomplete vector summation you will get significant errors. I think if you study what I've described you'll probably agree that true wind 26 knots and Green 55 degrees is incorrect?

The vector summation is fully described in the paper I referenced. You do need SOG,COG and heading to calculate true. Without heading, you will get errors both in the direction and speed of true.

Hope this makes sense?
 
Aaah, the light dawns!

If you are assuming no tide or leeway then your method works. As already pointed out, I sail 100% in tidal waters so tide and leeway are always significant and 'true wind' without taking this into consideration is useless.

The reason you need a fast compass is that SOG and COG are referenced to a heading (true or mag north). Therefore, it is necessary to also reference the platform-relative apparent wind to the same heading reference in order to generate the vectors needed for the summation, and then to convert the resulting true wind back to heading relative for display on the instrument.
 
You must log in or register to reply here.

Share this page

Top