Wrong Runway Takeoff Alert!

[Hardy Heinlin]

I would use the distance reference omnidirectionally from the takeoff lat/lon, not just sideways from the centerline. 100 ft seems a little low when you're rolling forward a bit. That's why I'm interested in the actual fixed value which the real box uses; I mean there must be a fixed value, regardless of the amount of nuisance triggers this fixed value causes.

What do you mean by "% chance"? A "chance" in a random generator?


|-|ardy

[Will]

I think using a circle as a warning bubble would lead to false positives. Sometimes people intentionally taxi down the runway before applying takeoff thrust, or alternatively, they may have planned for an intersection departure (1000 feet from the end of a 12,000 foot runway, for example) but ATC routes them to use the full length. We don't want a 100% chance of a warning right at the point of initiating the takeoff roll, when you actually have more runway than you need, right?

Of course, I have no idea how the real thing works.

[Hardy Heinlin]

If you use a taxiway intersection, e.g. Alpha, you need to enter "A" on the FMC TAKEOFF page in the field where you normally enter the displaced threshold value in feet or meter. You see, this function has a very precise reference :-) -- not only for the alert but also for the IRS position update in case the GPS fails.


|-|ardy

[John H Watson]

Interesting. It's also listed in the Qantas books, dated May 3, 2004. No technical data, though, other than that Manual Bulletin.

[B747-400]

Hi,

Cpt. Obet responds ...

... the manual says 'specified parameters' and doesn't give exactly what they are. I think it's just important for aircrew to know they have been exceeded ...

So, as he is a trainings-Cpt., it should also be OK for us not to know everything about it 

Just to let you know.

Cheers
Hans

[Jeroen Hoppenbrouwers]

What do you mean by "% chance"? A "chance" in a random generator?

Yes. The point in time when the comparison is made is known: that is when takeoff power is set. At that moment, first calculate your precise ok/warning. And if warning, then roll the dice to see whether you lie and say ok anyway. We're generating false ok, not false warnings.


Hoppie

[Hardy Heinlin]

So, as he is a trainings-Cpt., it should also be OK for us not to know everything about it 

OK for the pilots -- but for us Sherlocks the fun starts with the research for the entire context ... :-)

[B747-400]

You're sooooo right   

[Hardy Heinlin]

Yes. The point in time when the comparison is made is known: that is when takeoff power is set. At that moment, first calculate your precise ok/warning. And if warning, then roll the dice to see whether you lie and say ok anyway. We're generating false ok, not false warnings.

I think, referring to the visual references in the outside view, such a random effect would be so unnoticeably subtle that it's not worth the programming efforts. When you are on a wrong runway in a "non-KSFO" layout you'll be typically more than 2000 ft off anyway. Maybe I'll try a rectangle, 500 wide, 1000 long, with a continuously good GPS ...

[Jeroen Hoppenbrouwers]

I do completely agree that it is not necessary to simulate exact behaviour of a system that is not exactly defined in the first place.

[Hardy Heinlin]

The new runway alert option is now available in PSX version 10.172:

https://aerowinx.com/board/index.php/topic,4191.0.html


Regards,

|-|ardy

[Will]

Hardy, what parameters did you decide on for this feature? I just took off of VHHH 07R, and had the caution message, even though my ND showed me on the runway. My heading was 074, so the caution must have been tripped by a difference in forward/aft, or else I was the unlucky recipient of a random false positive alert. Perhaps I was 400 feet prior to the start of the runway? The taxiway in MSFS delivers the aircraft to the pavement ahead of a displaced threshold, but its not too much (only about 50 feet, per the chart).

[Hardy Heinlin]

It's the same detection algorithm that the EGPWS terrain clearance floor alert uses when detecting the active runway. It involves course/heading comparisons, distance and other stuff. If you can do a straight-in approach to 07R without a runway related terrain warning, the system must be intact. Maybe you had the FMC's destination runway on the ND? The new runway alert just checks the FMC's origin runway.

[Will]

No, it was a flight from VHHH to NZAA, and the ND showed the proper runway and the SID, and it matched the visuals out the windshield. I verified the runway after I saw the alert.

I did taxi to the takeoff position using MSFS visuals as a reference, and that may have introduced some kind of error (see my note above about the displaced threshold).

From the discussion above, it seems there is some sort of randomness involved in whether or not an alert is generated. Is it possible to get an alert when the FMC runway matches the actual takeoff runway?

[B747-400]

Hi,

I tried a TO from VHHH / 07R, verified CDU RWY/SID entries and ND display (bird was exactly placed on TO position with PSXs move aircraft position window!), but also got the RWY disagree message short after my TO run began.

BR
Hans

[Hardy Heinlin]

I see, it's a bug. A pretty ridiculous bug :-)

Well, the alert system compares the runway ID entered in the FMC with the runway ID detected by the EGPWS algorithm. The latter one omits leading zeros. So the comparison of "07" and "7" will signal a disagreement. This happens world-wide with all runway numbers lower than 10.

As this bug was discovered only today, this alert option doesn't seem very popular, or you guys have been using just runways with higher numbers in the past six months :-)

I'll fix it in the next update.


Thanks,

|-|ardy

[B747-400]

I saw this message a while ago. But I assumed it as my mistake, because it was a 'quick & dirty' flight with my son, without proper preparation ...

Thx for your information. It's always interesting to learn a bit of 'behind the scenes' 

BR
Hans

[Will]

I saw it once before, but I also assumed it was user error... The second time I saw it was yesterday.

[andrej]

Sometimes I get it as well.

Last time when taking from VHHH (07R) and KBOS (RWY09). But as I am not using pure vanilla PSX (connection with MSFS), I thought that this was related to my setup.

Cheers,

[Hardy Heinlin]

That bug is now cleared in PSX update 10.174:

https://aerowinx.com/board/index.php/topic,4191.0.html


Regards,

|-|ardy