Ground speed wrong in 750?

[RXP 2,338]

@fppilot this is a nice article, concise and to the point.

The RXP GTN takes values out of the simulator, shape them through embedded Shadin FADC simulation, and feed the GTN trainer with it. Therefore, the OAT and TAT displayed/used are those from the simulator simvar (datarefs in X-Plane). Ground speed is also directly taken from the simvar as-is and there is no further processing done within the GTN with it. There are subtleties though in the various data source paths feeding the GTN, and how it deals with these, and this is where our small adjustments will help a little in some cases! With this, the data displayed and computed is accurate to the sim-source data always now.

[fppilot 4,381]

Doing some more studying. Can get confusing but I worked through it.

From my flight yesterday, the Turbine Duke v2 was indicating TAS of 267 knots and -30°C for OAT/SAT. 
The RXP GTN750 was indicating -21°C for TAT.

----------------------------------------------------------------------
Most useful formula I found to determine TAT from SAT/OAT is this one. Example plugs in the values from my flight.

TAT= SAT + (TAS/87.1) ²

With my values:
TAT=  -30 + (267/87.1) ²
                    (267/87.1)  = 3.06544
                                        (3.06544) ² = 9.3969

TAT = -30°C + 9.3969°C  = -20.6031 (which the GTN obviously rounded to 21)
----------------------------------------------------------------------

So both the TDv2 and the GTN750 are correct. To each his own. TAT (GTN) or OAT/SAT (TDv2).  And proper use of either TAT or OAT/SAT, with an E6B that allows one of the other, returns identical ground speed GS.  Funny. I downloaded the Sporty's E6B iPad app this morning and it for the like calculation it provides entry option for only one temperature, True Temp. Based on my use True, SAT, and OAT appear to be interchangeable. 

There are other formulas out there but be aware that most require converting SAT °C to Kelvin then converting the result from Kelvin back to C°.
 

[fppilot 4,381]

Took another flight today and can conclude two things here.  First. Like my last reported flight in this topic, I started off with the GTN 750 in North Up orientation.  I took off and made numerous turns and like I reported earlier, the wind pointer never changed. Nada.  Then I switched the GTN orientation from North Up to Track Up.  The wind arrow changed position and then started following my turns.  I switched back to North Up and the GTN then continued to follow my turns.  Not sure if the pointer's initial static behavior is like that if starting in Track Up instead of North Up. Something to look at.  I have screen shots if you want them.

Second, I pretty much verified that when it is following correctly the pointer is off by approximately 40° clockwise.  I verified this with a real world forecast, FSGRW weather engine, ATIS to confirm FSGRW was placing surface wind and altimeter settings, observations of pointer direction and ground speed after turns, and finally a calculation using GTN readings of TAS, GS, Heading, and Track (course).  The wind pointer consistently showed wind direction at my altitude of 11,000 ft of from somewhere between 360° and 010°.  Applying the four readings mentioned above returned calculated winds of 60 knots from 324°.  Spot on to my FltPlan.com forecast.  That is in the neighborhood of 40° difference from the GTN pointer.

Heading 067°
Track 082°
TAS 204 knots
GS 225 knots.

Online E6B used is at: https://www.e6bx.com/e6b/

[RXP 2,338]

Hi Frank, I thank you for the additional tests, and I just wanted to let you know all these bugs are now found and fixed for the next update. It is just we are holding off the update a little longer because of our current work on the GTN for XP11.1 floating windows, and other work G500 related.

hint about ⅓ of the issue with the wind vector: I'm sure the 40º difference is the difference between the wind direction, and your speed in knots... (really, there was a typo in the code where we were reading the wind direction as 'knots' instead of 'degrees'....) This is ⅓ of the problem, the other ⅔ are GTN trainer related.

[fppilot 4,381]

26 minutes ago, RXP said:

I'm sure the 40º difference is the difference between the wind direction, and your speed in knots... (really, there was a typo in the code where we were reading the wind direction as 'knots' instead of 'degrees'....)

Then once I changed from North Up to Track Up and actually got the vector arrow to start turning with me, why did it turn 180° when I turned the plane 180°?  That would indicate it was coded to some element of direction.  It still had the same margin of error but it clearly turned as far as the plane turned. 

[RXP 2,338]

Yes in fact it depends on some other factors, but it is all fixed now for the next update.

[Gregg_Seipp 568]

On 12/12/2017 at 7:14 PM, RXP said:

Yes in fact it depends on some other factors, but it is all fixed now for the next update.

Any idea when that will come...roughly?

[RXP 2,338]

Roughly yes but you might not be pleased with the estimation! More seriously, the update is near ready but just held off the time we finalize working on the bug we've discovered in XP11.1 which didn't exist in XP11.05. It will be easier for us to release both simulator updates at once. It shouldn't take really long now, a day or two (roughly).

More of the XP11.1 bug here:

https://forums.x-plane.org/index.php?/forums/topic/48429-examplegauge-doesnt-call-mouseclick-callback/&page=3&tab=comments#comment-1304646

(additional contextual details 2 posts up in the linked discussion).

[Gregg_Seipp 568]

On 12/15/2017 at 2:07 PM, RXP said:

Roughly yes but you might not be pleased with the estimation!

Heh...not a worry.

[RXP 2,338]

We've now circumvented/mitigated the XP11.1 bug and we are conducting a few more tests prior shipping the updates for all simulators, with the GTN fixes discussed here!