Hello folks, just entered a holding at best holding speed of 212 knots. Flew towards to fix using LNAV and directly opposed to the inbound leg. Plane flew straight ahead for about a minute after passing the fix, perfect so far. Then it made a 180° to the left and ended up on the holding outbound leg pretty precisely. Interestingly enough it flew almost that whole outbound leg (remember, still opposite direction) before finally making a sharp turn inbound the fix (see below), but now obviously at a very bad angle to capture the actual holding afterwards. Why doesn't it fly like a ~210° degree turn at maximum bank angle directly to the fix (or even turn farther, and intercept the inbound leg right away like a true parallel/opposite entry), and then enter the hold? Behold, MS Paint has been used in this illustration (it is supposed to be a right hand holding). The left case is how the plane flew it. The right case would seem a little more convenient I think. Any thoughts? Oh and BTW, anyone managed to get the EFC filled with data? Won't accept anything here, just does nothing.

I think you have addressed two issues that I have also experienced. First off the holding scenario you illustrated seems to be consistantly the way the NGX is going to enter a Parallel hold. I experienced an even more disturbing scenario, however: when entering a hold after miss at LEPA the direction would specify a teardrop but FMC entered it as a parallel and then uncomfortably performed the same maneuver you experienced but upon passing the hold point proceeded to enter the non-protected area and fly the hold backwards! Tried this twice and each time same result! It was displayed correctly on the screen but flew it opposite direction and on the wrong side! The second issue you may have alluded to has plagued me since the release: very often the approach or miss procedures in the NGX datebase are lacking either altitude or speed entries. Without them the VNAV will automatically accelerate to 240K. After FX ATC announces the runway in use, I try to go into the CDU to select and then "clean-up" the procedure. Very often, however, any attempt to enter speed or altitude displays "invalid entry" msg. I have not experienced these anomalies in the B744 product. I am sure these are "bugs" in the release but PMDG is way behind in even acknowledging the receipt of trouble tickets so we will have to wait until the dust settles a bit, I guess. Craig

Hey Craig, thanks for getting back on the issue. First of all, I'm sure you know the holding is not equal to the protected area. That goes actually quite a bit beyond (can't remember the exact values, would have to dig up the docs), and it's perfectly normal to leave the depicted pattern for a parallel entry (again, don't know where it was exactly but I think it's in the FCOM). Still, it's not clear to me why it doesn't just fly the entry as any usual IR rated pilot was taught to do. That way you'd certainly intercept the holding way better than shooting thru the fix and to find yourself on the non-holding side at a great angle. As to your speed issue, were there any constraints on your leg page that would yield 240 knots? What did your HOLD page say? Because for me I indeed was flying at 240 (descent speed restricition I believe), LNAV/VNAV were flying, and right before the hold there was a DECEL point where it would correctly slow to 212, as indicated on the HOLD page. No manual input from my side. I will try out a few other scenarios asap and see what it does. EDIT Yeah the buffer area extends 5 NM around the actual hold and has initially ~1000' obstacle clearance, decreasing to about 200' at its outer boundaries.

Edited August 15, 201114 yr by badderjet

You misunderstood my point. I understand that a parallel procedure brings the aircraft initially into the non-holding side. My point was that the NGX correctly diagrammed the approach as a standard (right turns) which from the direction of entry would have indicated a teardrop approach. The FMC instead flew the approach as a non-standard (left turns) entirely on the wrong side of the inbound and continued even when established to fly the second circuit the same. I don't know how to draw a diagram on this computer so I hope my verbal explanation suffices. As to the second point, I have no problem when the approach or miss procedure already has speed or altitude restrictions but if they are blank and I try to add them it will not allow my entry. The FMC's on PMDG other products do and my understanding from real world is that FMC can and often needs to be ammended due to changes dictated by ATC so this is a glich! Craig

The left case is how the plane flew it. The right case would seem a little more convenient I think.

According to the standard IFR holding entry rules - at the angle you drew it, how the aircraft was approaching the fix, either entry is fine, both teardrop and parallel entry will work. The left is equally correct as the right one and only pilot's preference would determine which one is more 'convenient'.

No no I misunderstood nothing, I'm aware of what you mean (I think). I haven't had time so far but I will try various other entry scenarios as soon as I get the chance (including teardrops, maybe I can repro the same behavior here). Then I will also check about the speed restrictions, this sounds a little weird. Also, as I indicated above, FCOM suggests one were able to fill the EFC field but so far it wouldn't do a thing for me here.

According to the standard IFR holding entry rules - at the angle you drew it, how the aircraft was approaching the fix, either entry is fine, both teardrop and parallel entry will work. The left is equally legal as the right and only pilot's preference would determine which one is more 'convenient'.

I don't argue with that Michael, totally agree. In this case however it opted for parallel (I didn't decide anything, LNAV did) and flew it in a fashion I would not describe as beneficial. In my picture the left and right are both the same, i. e. parallel entries, just flown a tad differently.

This is very interesting, as I've been practicing with the NGX the missed approach procedure for the VOR rwy15 approach at KSYR( http://tiles.skyvector.com/sky/files/tpp/1108/pdf/00411V15.PDF ). So far in my experience, the NGX has flown the missed approach inconsistently, and it usually behaves as the OP described in this thread. However, it did one time fly a parallel entry and the hold just right...but I can't get it to do that predictably! I may well be at fault and am interested in any guidance as to how to best achieve this. Thanks for bringing this up, Etienne.

No prob Wayne, glad you're in our boat here. I'll make sure to shoot that approach and give the MAP a try thereafter. Oh and thanks for the chart link, is certainly gonna be helpful. EDIT Here we go - well, at took some time to try out that approach. On the first attempt I was surprised by another "automatic" go around. AP was active in LNAV/VNAV and beautifully flying towards SYR VOR when around 3000' it decided to go around without my knowledge or intention. Plus frame rate dropped to 1, so I had to reload. Second attempt went a bit better, although I got another huge stutter once I pushed TOGA (but at least I got to do it myself this time). LNAV opted for a teardrop entry. BTW The FMS figured 213 knots would be appropriate for the hold and maintained that speed. The beginning of the entry looked perfect. Then it flew WAY past the depicted holding (God knows why) and turned back to a perpendicular course to the inbound. It would really not have had to as speed was really moderate. Then turned inbound and tracked quite nicely. Still, I don't know why it went so far out in the first place. Maybe it's normal though, so see the attached.

Edited August 15, 201114 yr by badderjet

Thanks, Etienne, for trying that approach...sounds like the NGX remains unpredictable in terms of its ability to fly these approaches automatically. I tried again last night but had success only by employing HDG SEL and flying the parallel entry and hold in that way. It's odd that I did get the NGX to fly the entry and hold via LNAV/VNAV once...but I haven't been able to reproduce this.