FS2004 Strange Occur: AI Plane can be "slew followed" but not waited for at arriv airport

[b3burner]

I'll try to be as brief & to the point as possible, though that's sometimes hard for me-- I tend to ramble on as much as the day is long:

I made a flight plan for a plane to leave on a Mon AM from Bakersfield, CA to South Lake Tahoe, CA (KBFL to KTVL) 07:00hrs local dept to 07:57 local arriv scheduled.

If I slew my user plane all the way from take off to landing, I can follow the plane fine and it does what it's supposed to do (minus flying through a mtn at 8000' because it's in the way upon final approach). But if I sit at Lake Tahoe starting at 7am and wait 57 minutes for my plane to show up from Bakersfield to arrive it never shows up. Why is this? I know it most likely got off the ground and in the air-- per my earlier slew follow test, so it doesn't make sense that something would have happened to it, simply because I was waiting at the arrival end instead of following it.

I read the part in the manual about the 37 min rule, as it pertains to midroute flights, but I'm not sure I understood it right. I think it said that planes that it deems to be flying too slow get erased and never make their destination as arrivals, but later spawn at the next airport for their next departure. Well, that appears to be happening, as the plane does have to set up for a 09:00 departure from Lake Tahoe back down to Fresno, CA (KFAT)... so this would seem to make sense. But upon checking the cruise speed of the plane in question in the aircraft.cfg for the WOAI set, it is at 477 KTAS. That would seem plenty fast. It is not set at the 200 or 250kts that the manual would seem to indicate that WOAI steps down the speed, thereby causing a problem. So if the plane is at a proper 477, why might it be disappearing while it's in midair between Bakersfield and Tahoe (and I'm waiting in Tahoe for it)?

Or is there more going on here? Case in point, I have a another flight plan using the same WOAI aircraft with the same tail number assignment within 100 nm radius as where that first flight is flying. Do shared tail numbers of the same aircraft in the same vicinity cause clashes.. or is that immaterial?

Thanks, John

 

[gadgets]

Without the traffic file, John, its hard to say. I doubt it has anything to do with the 37-minute rule, given flight time is only 42 minutes (+15 for takeoff and landing and taxi) - if you used AIFP to compile the file. Tail number is incidental in the performance of AI aircraft.

What does the Traffic Database explorer tell you?

Don

 

[tgibson]

Hi,

Two of the possibilities: the plane is disappearing inside the mountain (happens if it is inside long enough), or it is an IFR flight and is being routed further from the Lake Tahoe airport than you expect. Wait for a longer time (at least another 15-30 minutes) and see if it shows up.

 

[b3burner]

Without the traffic file, John, its hard to say. I doubt it has anything to do with the 37-minute rule, given flight time is only 42 minutes (+15 for takeoff and landing and taxi) - if you used AIFP to compile the file. Tail number is incidental in the performance of AI aircraft.

What does the Traffic Database explorer tell you?

Don

Okay glad the 37 min rule can be eliminated as a possibility. I did use AIFP to compile, and it was a WOAI aircraft I "pilfered" for my own private use. And glad to hear the tail number assignment is incidental has no bearing (though that would have been a cheap, easy fix... so in a way, sorry to hear it's not that).

But what is "Traffic Database Explorer"? Is that Traffic Look? I was using Traffic Explorer 2005 and Traffic Map 2005-- older module programs given to me by somebody else in the community, when I got started in this about 5 yrs ago. What might "TDE" tell me, if I manage to find it?

 

[b3burner]

Hi,

Two of the possibilities: the plane is disappearing inside the mountain (happens if it is inside long enough)

Thanks for the suggestion as a possibility, but I really don't think it's in there long enough. It's a small sliver of a mountain that it passes through in the last 40 seconds before landing, when approaching from the south and prepping for rwy 36 landing. It's in and out of the mountainside in less than 5 seconds. The only reason it even hits it is because the AI aircraft insists on lining up perfectly with the runway at that point, and the mountain side just so happens to be in the way. I think in real life a pilot would forego the perfect line-up and steer around the darn thing, and then worry about lining up again later! LOL

or it is an IFR flight and is being routed further from the Lake Tahoe airport than you expect. Wait for a longer time (at least another 15-30 minutes) and see if it shows up.

Hmm.. this could be a real possibility. Though it was due at 07:57 local and I think I waited until 08:35 first time around before finally giving up. I'm curious just how long I ought to wait? At some point, the plane does have to be parked there in order to be ready to depart from Lake Tahoe back south to Fresno by 09:00 local. Of course the really frustrating thing is if I move my plane to another airport far away, and THEN come back to Lake Tahoe-- well it will be sitting there waiting for me... (as if) it were sitting there all along, as if to say, "Hahh! Where have you been? I was here waiting for you. You just were blind as a bat and never saw me!" (Sneaky little bugger).

The other odd thing, was when I was sitting at Tahoe waiting, and I knew at about 07:25 it was probably about half way between Bakersfield and Tahoe, and still hadn't come into my view (of listed AI planes), I thought I'd quick slew SSE and try to meet it midway in mid air. At some point (presumably south of Yosemite) I did manage to enter its air-space, as I saw its description appear in the AI roster window. So at that point I thought to myself, "OK, let me back-up-slew with my joystick, just enough to make it disappear out of sight again" ... as I assumed it was headed in my direction. My presumption was that if I made it disappear out of sight by slewing backwards by about a mile or two (and the plane was averaging a speed of about 7 miles / minute, then it really ought to take less than 20 or 30 seconds for it to re-appear again in the AI roster list... since it in essence should have caught back up into my airspace again. ---> Yet it didn't. I never saw it again. Go figure.

 

[b3burner]

The only other thing I think I might try is to sit at the end of the next leg in Fresno, and see if when the same plane is leaving Lake Tahoe at 09:00 local, whether or not it arrives in Fresno where I can watch it land. Maybe the issue is mountains near the arrival end, but mountains at the departure end have no effect. Total guess (and probably an uneducated one at that), but just a thought.

 

[gadgets]

Or, you could send us the traffic file so that we could attempt to duplicate the issue.

Incidentally, if you left the cruise speed as WoAI has set it, AIFP would halve that. So, how long would it take to make the flight by Cessna?
Traffic Database Explorer is the integral AI monitoring tool accessible from the Flightsim Tools menu under the Traffic Toolbox item.

Don

 

[b3burner]

Okay, I'll try sending the file. This is a WOAI Southwest Airlines B737-300W (winglets) jet with a cruise speed of 477.0 ktas-- as listed in the aircraft.cfg file for the plane. I didn't want to mess with that number since doing that would probably mess with every other aircraft in the WOAI collection, and I don't want to do that.

I have this one plane with the California State Bear livery making numerous flights all around the state that criss-cross each other at various times throughout the week in various locations. A large airport circuit, a medium airport circuit, a regional airport circuit, and a muni airport circuit-- north/south back and forth. It was originally the areas where these flights which all share the same plane same livery same flight number (and are in each other's 100nm AI radius) that I thought there might be a conflict. But you already said that shouldn't be an issue. Of particular interest is the FP (I think fourth one down from the top)-- monday morning Flight 2 KBFL-KTVL and Flight 3 KTVL-KFAT; which is of particular interest.

How long would it take to make the flight by Cessna? I have no idea. I didn't try plugging in a Cessna into the AIFP edit window. should I do that? what would that prove, or help me to discover? I'll try it later.

Okay, I think I found the Traffic Database Explorer. It was one of my other monitors I found by clicking on the modules tab in my open simulator. Regardless of which one I opened, and I think I tried all 3 of them I have last night, I never saw my make-believe (self-made) Southwest Airlines Flight #2 from KBFL to KTVL ever arrive, or move into Lake Tahoe airspace (within reasonable range of my user aircraft) at anytime between 07:30 and 08:30 local time-- despite its scheduled arrival of 07:57 local (15:57 UTC).

 

[jvile]

I'm sure Don will fiqure out the problem but as a side note you said

The other odd thing, was when I was sitting at Tahoe waiting, and I knew at about 07:25 it was probably about half way between Bakersfield and Tahoe, and still hadn't come into my view (of listed AI planes), I thought I'd quick slew SSE and try to meet it midway in mid air. At some point (presumably south of Yosemite) I did manage to enter its air-space, as I saw its description appear in the AI roster window. So at that point I thought to myself, "OK, let me back-up-slew with my joystick, just enough to make it disappear out of sight again" ... as I assumed it was headed in my direction. My presumption was that if I made it disappear out of sight by slewing backwards by about a mile or two (and the plane was averaging a speed of about 7 miles / minute, then it really ought to take less than 20 or 30 seconds for it to re-appear again in the AI roster list... since it in essence should have caught back up into my airspace again. ---> Yet it didn't. I never saw it again. Go figure.

You cannot do that. If a plane enters your visual zone and then leaves your visual zone it will not reappear. The plane can only enter the visual zone once. When you backed away from the plane and it left your visual zone there is no coming back. The plane will eventully popup at the gate or could reappear at a later time and land at its next airport.

 

[b3burner]

I'm sure Don will fiqure out the problem

Well I hate to take up too much of his time with it, but if he does manage to find an answer that would be great. Maybe I've been overlooking something obvious in the past year I've been making my own AI flight plans. We shall see.

but as a side note you said... <SNIP>

You cannot do that. If a plane enters your visual zone and then leaves your visual zone it will not reappear. The plane can only enter the visual zone once. When you backed away from the plane and it left your visual zone there is no coming back. The plane will eventully popup at the gate or could reappear at a later time and land at its next airport.**

Wow! Really? I had no idea. Because I could have sworn I've done that before with success. Maybe slewing back into it, then it will reappear, but waiting for it, it won't? OK... well regardless I've learned something new. I guess attempting to witness scheduled AI planes in mid flight is really more hit-and-miss than I thought it was. Thanks for the information.

 

[b3burner]

I wanted to add, that I tried sitting at the Fresno, CA airport (KFAT) beginning at 09:00, when the same SWA Cali Bear aircraft was due to depart South Lake Tahoe (KTVL) on its next leg in the flight plan. It was due to arrive in Fresno at about 09:45 local as flight #3, and this time it completed the journey, showed up, and I saw it taxi to its gate as it was supposed to.

So at least I know the aircraft is capable of showing up at its destination airport and I can "bear" (pardon the pun, given the livery on the side of the plane!) witness to the fact that it did indeed show up, without my having to slew anywhere in order to meet it half way. It's just for some reason it does a better job of this departing from Lake Tahoe than it does arriving at Lake Tahoe. So now I am beginning to wonder if the local terrain around the Lake Tahoe airport is causing an issue?

1. Maybe if I manually raise its cruise altitude, it will approach higher and miss mountains?

2. Maybe if I write it as a VFR flight and force it's cruise altitude at FL179 or lower it will find a path that's safer?

3. Maybe if I set the weather with a south wind heading 180, it will force a northern approach over the Lake onto rwy 18, thereby avoiding mountains on approach, and giving it plenty of nice flat lake below that poses no obstructions?

I don't know. I need to experiment more with this. I'll get back to it on another night. Thank you to all you who have offered suggestions and help thus far.

 

[tgibson]

I have seen some flights that simply do not work properly, even though the syntax is just fine. I don't know what happens to the planes, but they never show up at the arrival airport, no matter what I try. I just give up.

 

[gadgets]

I have just run the flight plan and, with my user aircraft parked to the side of the runway, watched it land at KVTL. So, it would seem Jim's explanation "fits". (Thanks, Jim, I too was unaware of this "single appearance" limitation.)

Don

 

[jvile]

Using a flight plan I duplicated your FP.

I sat at KBFL and at 7 am local time my SWA pushed back. I waited for it to takeoff and turn north toward KTVL then I used the GO TO and went to KTVL. I sat there and watched the clock on the Traffic Toolbox Explorer window. The plane spawned in the visual zone at 7:25 and it was in the South/EAST sector of the airport descending. It aligned with RWY 36 and landed at 7:35. At no time did it go below any mountain tops or fly through any mountains.

In this picture the plane spawned at 7:25:02. My plane is in the middle of the black traffic map with all white letters. I am sitting at rwy 36, The SWA plane is to my lower right descending out of 24,966ft at 354KTAS.

In this picture the plane continues on a WEST heading and descending. You can look at some of the ATC data which says the current alt is 23,769 and the desired is 16,000 ft which in order to cross the next ATC target the plane must decent at 2158 FPM

This picture I backed out in slew mode to see the plane turning final. Looking at the data the plane must be at 8300 ft. when it reaches the Approach Start Decent (SD) target. it will then dsecend to the Touch Down (TD) target box 6264 ft which is airport runway elevation. If the plane overshoots the runway a certain percentage and cannot stop at the Landing Stop (LS) target with wheels on runway then you will see the plane take back off and declare a missed approach.

What I can suggest is for you to change the plane to another model. Runway 36 does not have any type approach code and if the planes flight dynamics are not the best in the world it might have problems trying to fly a hard coded approach.

 

[jvile]

(Thanks, Jim, I too was unaware of this "single appearance" limitation.)

Don

I think I spoke too soon. I was working with FS9 when I kept seeing the problem. FSX does allow multiple appearances.

 

[gadgets]

I was using FSX and a known-good 737. No issue.

Don

 

[b3burner]

I have just run the flight plan and, with my user aircraft parked to the side of the runway, watched it land at KTVL. So, it would seem Jim's explanation "fits". (Thanks, Jim, I too was unaware of this "single appearance" limitation.) [Later to be determined just an FS9 thing].

I'm pleased you got it to work, though I'm racking my brain trying to figure out why it wouldn't work for me? I'll reply to jvile separately, though I suspect I know why yours worked and mine didn't. See my next post.

 

[jvile]

I also used the default B737 with no issues.

I did go back and find my previous problem when planes do not allow multiple appearances. It happens in small countries that have close in borders. Some of my curved approaches come out of Itlay and into Austria which is the visual zone based on the airport I am sitting at. However to align with the approach at a high altitude the plane will fly back out of Austria and into Germany when making a 360 turn. Once it leaves my visual zone in Austria it never comes back but will evetully popup at the gate.

It seems to be more related to courtry codes and the AI plane loses its slot number and cannot reestablist with ATC.

 

[jvile]

I just noticed b3burner is using FS2004. All my test was with FSX.

I will go back and retest in FS2004.

 

[b3burner]

Using a flight plan I duplicated your FP.

I sat at KBFL and at 7 am local time my SWA pushed back. I waited for it to takeoff and turn north toward KTVL then I used the GO TO and went to KTVL. I sat there and watched the clock on the Traffic Toolbox Explorer window. The plane spawned in the visual zone at 7:25 and it was in the South/EAST sector of the airport descending. It aligned with RWY 36 and landed at 7:35. At no time did it go below any mountain tops or fly through any mountains.

My first thought was "incredible!" how did you get it to appear when I couldn't? I am also amazed the plane landed at 07:35 for you, because in my flight plan I let it go by the ETA projection without entering an earlier time, and it was due at 07:57. Maybe your AI plane had a faster cruise speed in its aircraft.cfg file? That's all I can figure on that one.

Then I got to thinking about it, and I accidentally overlooked the most obvious thing of all: Just because I sent you an attached copy of my flight plan didn't mean the WOAI SWA California Bear B737-300W came with it! I forgot the traffic files don't include the aircraft! Silly me! So we're in essence comparing apples to oranges, because you duplicated the flight using your own aircraft... not the one I was having problems with. So would it help if I send you a copy of the SWA California Bear livery plane... or at this point, are we beating a dead horse, and have we sufficiently determined that the issue is the plane I'm using, and there's probably little need to prolong this any further? But if you want me to send the plane, I will. Would be interesting to see if you have the same problem as well.

My other thought is I am doing all this in FS2004, so if you were doing your tests in FSX, I don't know if that was a good comparison either. Might it be advantageous to see if you could replicate the issue I'm having using the plane I was using in FS9? Or again, is this getting to be purely academic and a tad on the pedantic side? If this is starting to get old, I'm content to thank you for your time and effort at this point, and let it go.

In this picture the plane spawned at 7:25:02. My plane is in the middle of the black traffic map with all white letters. I am sitting at rwy 36, The SWA plane is to my lower right descending out of 24,966ft at 354KTAS.

Probably little need to re show the pictures in my reply, I saw what they were, but I continue to be amazed that you were able to get it to work. Yes indeed I see you sitting at KTVL, and yes that is the SWA AI plane (of your collection) coming in from the SE sector, as it should be.

In this picture the plane continues on a WEST heading and descending. You can look at some of the ATC data which says the current alt is 23,769 and the desired is 16,000 ft which in order to cross the next ATC target the plane must decent at 2158 FPM

Amazing all the data you are deriving from those pull down maps. I think my traffic maps don't show nearly as much as those. I think I have that program, but I probably got used to using the other ones in my collection, and I typically don't go to that one as much.

This picture I backed out in slew mode to see the plane turning final. Looking at the data the plane must be at 8300 ft. when it reaches the Approach Start Decent (SD) target. it will then descend to the Touch Down (TD) target box 6264 ft which is airport runway elevation. If the plane overshoots the runway a certain percentage and cannot stop at the Landing Stop (LS) target with wheels on runway then you will see the plane take back off and declare a missed approach.

I have actually seen the missed approach phase before, but I've usually seen it much more exaggerated than that. Like if I try to watch a plane land and spawn my user aircraft near the airport 5 minutes before the AI is due to land, it will still be at 27,000 feet and practically right over the airport. It will try to dive bomb in at a steep angle because it knows it's right there, and then level off around 10,000 feet, and wheels-up in its missed approach and continue on and start climbing again. They do do strange things. That's why if I'm going to spot an AI landing at the arrival airport, I usually try to time my appearance at least 20 minutes before it's scheduled arrival, so as to allow it time to set up a normal approach. At least that's been my experience with FS9. I haven't really messed with it in FSX yet.

What I can suggest is for you to change the plane to another model. Runway 36 does not have any type approach code and if the planes flight dynamics are not the best in the world it might have problems trying to fly a hard coded approach.

That's a good point, and one I wasn't aware of. I'm not sure I'm familiar with the term "approach code", but do you mean to say whether it has VASI lights, ILS, and/or other landing features? Or do you mean something else? And as far as a "hard coded approach", I take it you mean programming departure/arrival on a specific flight leg? Or do you again mean something else? But I can see what you mean that if I had somehow programmed in some hard-coded approach (whether I had intentionally done it or not), and KTVL Rwy 36 can't handle that, then it would present a problem. Also, I'm thinking maybe length of runway could be a factor as well if the plane lands and slows like a rock, but then again, I think I've programmed AI planes to land at airports with runways shorter than that.

At any rate I certainly do thank you for your time and input. I think I'm beginning to realize I just happened to get unlucky by picking the wrong plane for that specific leg. I bet if I ran the same route with a Beech or a King Air, I'd be just fine.

-- John