OpenMarine

Full Version: How to get True Wind Direction instead of apparent using Signal K.
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3 4
Hi folks,

Please bear with me here since I am a complete novice when it comes to boating and navigational equipment.

I have a project to setup a weather station at home using a decommissioned Gill ultrasonic windsensor I got from where I work.
My plan was to just get environmental data and log it to a database. 

After extensive googling I found that using Signal K on a Raspberry Pi was a good way to decode the iiMWV nmea sentence coming from the sensor. 
Especially since there also was a plugin to write to a influx database.

I thought I had it all figured out until I see that the wind angle is "apparent" and not true direction. 
I see the sensor nmea sentence is sending true direction 0-360 degrees but that is changed to environment.wind.angleApparent in the data browser in Signal K.

Could anyone please point me in the right direction on how to just pass through the "True ground wind direction" that is sent from the sensor??

Andreas
True wind angle has to be calculated, as on a moving vessel the speed of the vessel and the current the vessel is traveling in contribute to the wind speed and direction that is detected by your sensor.

If the sensor is stationary, apparent wind speed will be the same as true wind speed. Apparent wind will be the angle referenced to the forward position on the sensor (+/- 180 degrees). I assume that you will be pointing this reference to north. On my ultrasonic sensor, it's a little mark that I have to align with the centerline of the boat. There are ways in OpenPlotter to calculate true wind direction and speed, but they require a vessel heading and speed to create the numbers, which I'm assuming you don't have.

It's a little work, but in your situation you can calculate wind direction using Node-Red, by taking in the NMEA0183 string you described and calculating wind direction based on apparent wind and a reference to north. I would be a simple calculation, from +/- 180 to 360. You would then output the appropriate SignalK keys for True wind speed and direction.

If you are just viewing the information in Grafana, you can make the calculation part of the query that creates your graph.
Thanks for your quick reply.

The reference that is 0 degrees is pointed north yes.
My sensor range is already 0-359 degrees and when I cat /dev/ttyUSB0 in terminal I see the correct wind angle, $IIMWV,090,R,010.11,M,A*18.
But in Signal K the environment.wind.angleApparent vary around 3.0 -> -3.0 when browsing the data from Influx in Grafana,

Is a re-calculation necessary when the value I want is already being sent to Signal K?
SignalK outputs in radians, so you'll need to convert (back) to degrees. You can do that in Grafana.
Ah, I see. I googled and found the formula for conversion but no clue how to do that in the query in Grafana.
Is there no setting or tweak in Signal K to output degrees?
No, unfortunately not. SignalK outputs in radians.
Thanks for taking the time to answer.
You mention I could do that in Grafana but math is not my strong side. I see 1 radian is 57.29 degrees and I could use the "math" function in the Grafana query, but not sure about the formula when negative goes from 180-360 degrees and positive from 0-180.
if you download the Signal K plugin "Derived Data" this will calculate the True Wind Speed and Angle based on Apparent wind speed and Angle, Speed over ground and Heading
generally you want to know the true wind over water rather than ground for sailing performance
(2022-04-10, 04:50 PM)dreeas Wrote: [ -> ]Ah, I see. I googled and found the formula for conversion but no clue how to do that in the query in Grafana.
Is there no setting or tweak in Signal K to output degrees?

Signalk works with SI units, it's down to whatever is displaying the data to do any unit conversions. 
In Grafana there's a math function in Select >>

[Image: I84xuH3.png]

Also useful for wind is adding a peak gust plot >>
[Image: Q70i8Ix.png]
Pages: 1 2 3 4