Board index FlightGear Support Tools OpenRadar

True/Magnetic discrepancy

OpenRadar is a standalone radar screen which connects to the FlightGear multiplayer servers. It is currently being developed.

True/Magnetic discrepancy

Postby timw » Fri Oct 09, 2015 10:50 pm

Hi,
I'm mapping out a STAR for an airport and am coming across a discrepancy which I don't understand, hopefully someone here can shed some light on it.

  • I have 2 navpoints defined by their LAT&LONG.

  • The Approach Plate for the airport shows the magnetic heading between the navpoints as 346°
    True Track | Magnetic Track : 342.6 | 346

  • In my xml file I draw a line from the 1st navpoint using an angle (docs say use magnetic heading) and length. I know instead I could just end at the 2nd navpoint, and that works just fine, this is just to illustrate the problem.
    Code: Select all
    <line start="VATRY" angle="346" length="37"/>
    but this misses the 2nd navpoint by miles.

  • If I apply variation which is about -3.4°, it pretty much lines up
    Code: Select all
    <line start="VATRY" angle="342.6" length="37"/>

This doesn't make sense to me, I have a magnetic heading, and a config that expects me to use a magnetic heading and they don't line up, Is there a step I've missed?
This also affects my loops and intercepts too, but I figured line was the simplest illustration.

cheers,

Tim
timw
 
Posts: 10
Joined: Fri Oct 09, 2015 10:00 pm

Re: True/Magnetic discrepancy

Postby Johan G » Sat Oct 10, 2015 1:31 am

How old is the approach plate? ;)

FlightGear uses the World Magnetic Model (WMM) (probably a bit older one of those models by now). This model will give varying variation depending on the date you have FlightGear use.

I once plotted out this perfect course, but would always end up drifting a lot. Finally I figured out why some additional work is needed if using a few decades old map. The magnetic variation varies over time. 2 degrees off does a lot when there is 20 NM between the checkpoints...

However that may not be it. Someone gave an excellent but slightly different answer a few years back, unfortunately I have a hard time finding it. :?
Low-level flying — It's all fun and games till someone looses an engine. (Paraphrased from a YouTube video)
Improving the Dassault Mirage F1 (Wiki, Forum, GitLab. Work in slow progress)
Johan G
Moderator
 
Posts: 5528
Joined: Fri Aug 06, 2010 5:33 pm
Location: Sweden
Callsign: SE-JG
IRC name: Johan_G
Version: 3.0.0
OS: Windows 7, 32 bit

Re: True/Magnetic discrepancy

Postby sanhozay » Sat Oct 10, 2015 9:02 am

Based on your example, it looks like OpenRadar is using true track in the line definitions instead of magnetic track. The documentation suggests it uses magnetic:

http://wiki.flightgear.org/OpenRadar:_E ... and_Angles

But if that's true, I'm surprised nobody has questioned it before, especially because declinations in some places are not just a little bit off, they are way off, e.g. Los Angeles has magnetic north 12*E of true.

At VATRY in the Irish Sea (I assume this is an Dublin STAR using VATRY at 52.5*N, 5.5*W), magnetic north is currently 2.91*W of true north, drifting 0.16*E per year. But in Dublin, it is about 3.4*W, which is what I assume is used on recent Dublin charts. My Flightgear 3.7 shows magnetic variation of 3.38*W at Dublin airport.

http://www.ngdc.noaa.gov/geomag-web/#declination

Given a true track of 342*, an attempt to fly to it using a magnetic compass bearing of 342* puts you west of your target because magnetic north is west of true north and therefore the needle points west of your waypoint. To compensate, you have to correct by tracking east of magnetic 342*, i.e. magnetic heading of 346*.

Side note: Flightgear was updated to WMM2015 in Jun 2015, so version 3.4 and 3.6RC/3.7 will be using a slightly different magnetic model. The differences should be insignificant but the latest versions of OpenRadar need to be using the WMM2015 model.
sanhozay
 
Posts: 1207
Joined: Thu Dec 26, 2013 11:57 am
Location: EGNM
Callsign: G-SHOZ
Version: Git
OS: Ubuntu 16.04

Re: True/Magnetic discrepancy

Postby wagnerw » Sat Oct 10, 2015 7:13 pm

Open Radar shows the magnetic declination that it gets out of the model as ToolTip text of the airport name at the right top corner... Simply hover with your mouse over it.
Please compare it to the declination on your chart! It is likely that the real magnetic declination varies from the result in FlightGears model. OpenRadar and Flightgear use the same mathematic model.
This difference must be included into your route definitions.
Your assumtion, that OpenRadar uses magnetic headings is right.

The heading indicator beside the mouse pointer is integrating the wind correction, additionally. So don't get confused: A slow flying aircraft in strong cross wind, will get another heading to fly to, as a fast aircaft in low winds. This calculation uses the wind speeds of your current metar. So it is not really correct, because wind speeds and headings change with heights and are interpolated along your flightpath. Unfortunally the flightgear internal wind speeds and wind headings are not exposed via multiplayer protocol, so OR has to estimate it.
In our experience this calculation is good enough.
wagnerw
 
Posts: 283
Joined: Tue Nov 06, 2012 8:35 pm
Callsign: D-W794

Re: True/Magnetic discrepancy

Postby timw » Sat Oct 10, 2015 10:45 pm

Thanks for all your replies, I'll try to cover the points raised and questions asked in 1 reply:

Open Radar shows the magnetic declination that it gets out of the model as ToolTip text of the airport name at the right top corner... Simply hover with your mouse over it.
Please compare it to the declination on your chart!

tooltip shows -3.6°, the chart shows -4°, different, but not sufficiently to account for the issue I'm seeing.

The heading indicator beside the mouse pointer is integrating the wind correction

I'm not even looking at aircraft headings and tracks yet, I'm just trying to get my STAR routes to line up correctly on the display, and bear a reasonable resemblance to the plate. I'll worry about whether aircraft line up to the same tracks later.

it looks like OpenRadar is using true track in the line definitions instead of magnetic track

This was my concern.

I assume this is an Dublin STAR using VATRY at 52.5*N, 5.5*W

correct.

magnetic north is currently 2.91*W of true north, drifting 0.16*E per year. But in Dublin, it is about 3.4*W, which is what I assume is used on recent Dublin charts. My Flightgear 3.7 shows magnetic variation of 3.38*W at Dublin airport.

You'll see the values I have towards the top of this post
The difference between VATRY and Dublin though is more significant than I'd considered, thanks for pointing to that. Maybe the issue is a combination of things and this is part of it.

How old is the approach plate? ;)

02 APR 15


So in summary, all I'm actually trying to understand is why the magnetic bearing on the plate disagrees with the magnetic bearing in OR and is the fact that it's off by approximately the amount of variation a coincidence?

I just tried another illustration, I plotted a line from the start of rwy 28 at EIDW (chart shows 279°M), by the end of the runway I'm 160m off and would be on the parallel taxiway. whereas 275.3° lines up pretty much spot on.

Perhaps I should try my test case in Los Angeles ans see how far off I end up there.
Last edited by timw on Sat Oct 10, 2015 11:13 pm, edited 1 time in total.
timw
 
Posts: 10
Joined: Fri Oct 09, 2015 10:00 pm

Re: True/Magnetic discrepancy

Postby sanhozay » Sat Oct 10, 2015 11:08 pm

timw wrote in Sat Oct 10, 2015 10:45 pm:So in summary, all I'm actually trying to understand is why the magnetic bearing on the plate disagrees with the magnetic bearing in OR and is the fact that it's off by approximately the amount of variation a coincidence? Perhaps I should try my test case in Los Angeles ans see how far off I end up there.

Does it disagree in OpenRadar, or is it that tracks entered in procedures files are true instead of magnetic? I think that is the crux.
sanhozay
 
Posts: 1207
Joined: Thu Dec 26, 2013 11:57 am
Location: EGNM
Callsign: G-SHOZ
Version: Git
OS: Ubuntu 16.04

Re: True/Magnetic discrepancy

Postby timw » Sat Oct 10, 2015 11:16 pm

I think it disagrees in OR. I just tried to draw a line of 279 down the centre line of EIDW rwy 28, and it's on the parallel taxiway by the end of the runway. No procedure files involved.
timw
 
Posts: 10
Joined: Fri Oct 09, 2015 10:00 pm

Re: True/Magnetic discrepancy

Postby sanhozay » Sat Oct 10, 2015 11:17 pm

What does the tooltip referred to by wagnerw say?
sanhozay
 
Posts: 1207
Joined: Thu Dec 26, 2013 11:57 am
Location: EGNM
Callsign: G-SHOZ
Version: Git
OS: Ubuntu 16.04

Re: True/Magnetic discrepancy

Postby timw » Sat Oct 10, 2015 11:19 pm

see my post above:
tooltip shows -3.6°, the chart shows -4°, different, but not sufficiently to account for the issue I'm seeing.
timw
 
Posts: 10
Joined: Fri Oct 09, 2015 10:00 pm

Re: True/Magnetic discrepancy

Postby Johan G » Sat Oct 10, 2015 11:22 pm

wagnerw wrote in Sat Oct 10, 2015 7:13 pm:OpenRadar and Flightgear use the same mathematic model.

I noted that simgear/simgear/magvar/coremag.cxx (permalink/commit 00a204) got updated from WMM 2005 to WMM 2015 September 2nd with commit 105438.

While the difference should be minute (at least for now), is perhaps OpenRadar currently using the older model?
Low-level flying — It's all fun and games till someone looses an engine. (Paraphrased from a YouTube video)
Improving the Dassault Mirage F1 (Wiki, Forum, GitLab. Work in slow progress)
Johan G
Moderator
 
Posts: 5528
Joined: Fri Aug 06, 2010 5:33 pm
Location: Sweden
Callsign: SE-JG
IRC name: Johan_G
Version: 3.0.0
OS: Windows 7, 32 bit

Re: True/Magnetic discrepancy

Postby timw » Sat Oct 10, 2015 11:31 pm

Wow!

KLAX, plot a line on the rwy 06L centreline bearing of 071° and I'm off by 1/3mile by the end of the rwy.

off, e.g. Los Angeles has magnetic north 12*E of true.

yep add 12° and I get perfect alignment.

Code: Select all
<line start="33.949174,-118.431079" angle="083" length="3"/>

This is looking quite a lot like a True/Mag issue.
timw
 
Posts: 10
Joined: Fri Oct 09, 2015 10:00 pm

Re: True/Magnetic discrepancy

Postby wagnerw » Sun Oct 11, 2015 6:52 am

I reproduce the issue... And I wonder, that nobody has found it yet. Thanks for telling me! I will try to find the reason...

Sorry guys, but, the more I look into it, I wonder how I could miss it. I wonder when it slipped in, or if it was as it is now, from the beginning...
wagnerw
 
Posts: 283
Joined: Tue Nov 06, 2012 8:35 pm
Callsign: D-W794

Re: True/Magnetic discrepancy

Postby timw » Sun Oct 11, 2015 9:19 am

Thanks for looking into it wagnerw and for such a quick response. Should I raise a bug-tracker ticket or is that not necessary?
timw
 
Posts: 10
Joined: Fri Oct 09, 2015 10:00 pm

Re: True/Magnetic discrepancy

Postby wagnerw » Sun Oct 11, 2015 8:35 pm

No not necessary... I only need an concentrated hour to change it... And time for tests... I come back
wagnerw
 
Posts: 283
Joined: Tue Nov 06, 2012 8:35 pm
Callsign: D-W794

Re: True/Magnetic discrepancy

Postby wagnerw » Mon Oct 19, 2015 6:51 am

Hi, a short update:
I have fixed this issue already, but working at this spot, I have also updated the magnetic model of the world to 2015 version already. Here we are a bit in front of flightgear. The model is already checked in into FlightGear (I actually took it from there), but it is not released yet (as much as I know). So for a short time, OR will be a bit more exact than FGFS.
Before I publish my fixes, I want to do another improvement... Please be patient for a couple of days...

Important: If you have implemented route.xml files for an airport and if you have used headings to define the route elements, you will have to double check, that the headings are defined in the correct magnetic headings. If you adapted them to correct my bug, you will have to fix it now. Sorry for this additional work!
Wolfram
wagnerw
 
Posts: 283
Joined: Tue Nov 06, 2012 8:35 pm
Callsign: D-W794

Next

Return to OpenRadar

Who is online

Users browsing this forum: No registered users and 1 guest