PMDG aircraft and 3 reds PAPI autoland

[Turlad 0]

I wonder if anyone else is experiencing a similar issue...

On all of the 3 latest PMDG A/C - NGX, 777 and 747 I've got an issue where the autopilot / autoland will routinely fly into 3 reds / 1 white on the PAPIs. It's fine with 2 whites 2 reds until the last 300 to 400 feet then you see the transition to 3 reds. 

In my real life experience the autoland should not do this. It flies the glideslope and maintains 2 whites 2 reds until after crossing the threshold. PMDG hasn't always had this error - I've updated my fsx nav data from the https://www.aero.sors.fr/navaids3.html website to match current data. The autoland worked better before granted, and if I revert back to FSX default nav data I get 2 reds 2 whites for an appropriate amount of time. But surely the PMDG should just take the ILS from fsx and fly it. Why am I seeing a duck below the ideal 3 degrees?

Now I know the ILS antenna and PAPI aiming point do not exactly match in real life, and the geometry is different for bigger and smaller aircraft, but that's not the issue. 

Is there something I need to do to get the PMDG aircraft and fsx ILS / PAPI data to match up?

Thanks.

 

Adam Turley

[JRBarrett 3,070]

I believe that the updated ILS NAV data from the Hervé Sors site will reposition the glideslope transmitters closer to their r/w coordinates - but does not adjust the locations of VASI/PAPI lights? Perhaps that could account for some of the problems, (if the original approach path lighting was linked to the original glideslope location in the default scenery).

There are SOME airports where there actually is a disparity between the electronic glideslope and the visual glideslope. The ILS charts for such airports will usually have a note on the vertical path depiction to the effect "VGSI is not coincident with glideslope".

I don't know if the problem is exacerbated by the fact that the 747 and 777 are physically large aircraft, or if it is something specific to PMDG coding of the ILS system in their aircraft.

[scandinavian13 5,192]

1 hour ago, Turlad said:

I wonder if anyone else is experiencing a similar issue...

It's not an issue.

1 hour ago, Turlad said:

In my real life experience the autoland should not do this. It flies the glideslope and maintains 2 whites 2 reds until after crossing the threshold. PMDG hasn't always had this error - I've updated my fsx nav data from the https://www.aero.sors.fr/navaids3.html website to match current data. The autoland worked better before granted, and if I revert back to FSX default nav data I get 2 reds 2 whites for an appropriate amount of time. But surely the PMDG should just take the ILS from fsx and fly it. Why am I seeing a duck below the ideal 3 degrees?

It should follow the guidance of the glide slope signal. Period. End. Autoland pays no attention to the visual indications. Discrepancies will exist, and are noted where they exist. This is all stuff you would know from real world experience...so, what real world experience are you referring to here?

1 hour ago, Turlad said:

Is there something I need to do to get the PMDG aircraft and fsx ILS / PAPI data to match up?

Read the charts.

On just about any chart you look at, you'll find this curious line: "VGSI and ILS Glidepath not coincident."

In other words, the ILS GS WILL NOT put you on the glidepath that the visual glide slope indicator. Curiously, you indicate that you know this, but for some reason, persist on the point that this is somehow odd. I'm not following that logic. How are you determining that's not the cause of what you're seeing?

[threexgreen 3,139]

Recently did an autoland at O'Hare and had three reds as well. Didn't think it was an issue though. As the others said, noted on the chart http://155.178.201.160/d-tpp/1704/00166IL10L.PDF

[Bryn 72]

I see this a bit. If there is a note in the chart, then it shouldn't be a surprise (especially as it would be briefed?) and adds to the realism. But sometimes it's not expected.  In this case it's probably a small issue in the AFD file, where the glideslope location is not the same distance from the threshold as the PAPI. The aero.sors data might be exacerbating this problem if it's using an arbitrary location for all glideslopes, but I don't know what that service does.

Usually after each flight, I have a 'simulator write-up' of little things I need to fix, and I include this if I do an approach where I see 4 reds or 4 whites, but centered GS (and there's no note in the chart). I then go and edit the AFD so it doesn't happen again (either using ADE or AFX). It only takes a minute, and then the airport is good for the next time I pass through.

Bryn.

[Turlad 0]

55 minutes ago, scandinavian13 said:

It's not an issue.

It should follow the guidance of the glide slope signal. Period. End. Autoland pays no attention to the visual indications. Discrepancies will exist, and are noted where they exist. This is all stuff you would know from real world experience...so, what real world experience are you referring to here?

Read the charts.

On just about any chart you look at, you'll find this curious line: "VGSI and ILS Glidepath not coincident."

In other words, the ILS GS WILL NOT put you on the glidepath that the visual glide slope indicator. Curiously, you indicate that you know this, but for some reason, persist on the point that this is somehow odd. I'm not following that logic. How are you determining that's not the cause of what you're seeing?

 

I'm sorry but it is an issue. It's an issue that's only recently cropped up, presumably due an FSX navdata update. I've been flying PMDG in FSX since 2008, and something recently has changed. I'm trying to work it out.

I'm well aware that autoland doesn't pay attention to the PAPIs however, for my day job I'm an FO on the 737-800 with a few thousand hours and I've done many dozens of autoland /  CAT3A approaches to the same RW's / airports I'm testing in FSX. You don't consistently get 3 reds on the PAPI from 400ft onwards at every airport. Yes, there are some airports where the PAPI doesn't exactly match the slope angle or intersect at the same touchdown point. These are usually the places that aren't certified for CATII/III... What I'm seeing is the PMDG aircraft consistently flying below the visual indication and you can just tell that the picture / slope is too shallow. It hasn't always done this - something recently has changed - probably with my install, and I was looking for suggestions.

Incidentally, this only happens with PMDG aircraft - the default aircraft and other add ons show the ILS and PAPI much more matched up, even with the new nav data. 

As I say, I'm a big PMDG fan, not trying to be critical. There's simply a discrepancy somewhere and I'm trying to work out what it is. 

 

Adam Turley

[scandinavian13 5,192]

5 minutes ago, Turlad said:

As I say, I'm a big PMDG fan, not trying to be critical. There's simply a discrepancy somewhere and I'm trying to work out what it is. 

Come with examples, and people can look into it (either people in the forum, or myself). Vaguely stating "this is an issue" without being specific about where you're pointing is naturally going to draw some ire, because of a number of things:

1. FSX VGSIs are occasionally missing, in the wrong spot, or not set properly
2. FSX data is over a decade old, unless you've updated it
3. VGSIs and ILS GS data rarely match

These are all variables. In order to point categorically at something (which was vaguely the aircraft, based on your initial post), those must be controlled, or at least a scenario stated so that someone can research.

Either way, pointing from the outset usually isn't the best approach unless you're absolutely sure, and have the data to back that up.

[cannonplug 126]

4 hours ago, Turlad said:

I've updated my fsx nav data from the https://www.aero.sors.fr/navaids3.html website to match current data. The autoland worked better before granted, and if I revert back to FSX default nav data I get 2 reds 2 whites for an appropriate amount of time

Based on what you said here and Jim's post below yours, Hervé's update is what seems to be causing the issue. I had a cursory look in his forum (http://aerosors.freeforums.net/board/4/fs-navaids-fixes-updates)  but didn't see any topics in this regard. You may want to pose your question there, however, to see if he has any insights.

By the way, I've seen the same behavior.

Cheers!

[JRBarrett 3,070]

2 hours ago, Bryn said:

I see this a bit. If there is a note in the chart, then it shouldn't be a surprise (especially as it would be briefed?) and adds to the realism. But sometimes it's not expected.  In this case it's probably a small issue in the AFD file, where the glideslope location is not the same distance from the threshold as the PAPI. The aero.sors data might be exacerbating this problem if it's using an arbitrary location for all glideslopes, but I don't know what that service does.

As far as I know, Hervés updates definitely do not put ILS localizers or glideslopes in arbitrary locations - to the contrary, he uses official data sources to locate the ILS transmitters at their exact current lat/lon coordinates, and corrects frequencies that have changed since the release of FSX (P3D as well, since it also uses the old FSX default data).

The updates only affect airports which use purely default data. Any payware or freeware add-on airports with updated AFCAD files will take priority, and will not have their own updated ILS info (if present) changed by Hervé's updates.

[rototom 198]

10 minutes ago, JRBarrett said:

to the contrary, he uses official data sources to locate the ILS transmitters at their exact current lat/lon coordinates

Probably shouldn't move anything unless you're updating the entire airport. If the default runway was not in the exact lat/lon real world location?, moving the ILS equipment without moving the runway and papi would create a dependency.

[BlueStar 323]

9 minutes ago, JRBarrett said:

As far as I know, Hervés updates definitely do not put ILS localizers or glideslopes in arbitrary locations - to the contrary, he uses official data sources to locate the ILS transmitters at their exact current lat/lon coordinates, and corrects frequencies that have changed since the release of FSX (P3D as well, since it also uses the old FSX default data).

The updates only affect airports which use purely default data. Any payware or freeware add-on airports with updated AFCAD files will take priority, and will not have their own updated ILS info (if present) changed by Hervé's updates.

Jim,

The problem is the default FSX airports are not in the correct location.  By this I mean the airport location maybe 50 to 100 feet off.  When I build a new AFCAD using actual current lat/lon coordinates, this mean that the airport must be moved slightly to get the runway/LOC/GS/PAPI into the correct relationship.  Google Earth is excellent for helping out.  Even then it normally requires a couple of flight tests to get the correct look.  I highly recommend Airport Design Editor for making these changes. 

blaustern

[Turlad 0]

2 hours ago, scandinavian13 said:

Either way, pointing from the outset usually isn't the best approach unless you're absolutely sure, and have the data to back that up.

Well my initial post was intended as a feeler, to see if others had experienced the same problem, and for all I know there could be an easy fix... From the other replies on this post it appears some at least have noticed it as well. 

I can collect data / screenshots etc but obviously to do it meaningfully will take a little time. At the moment it's just a general observation / feeling and a little bit of knowledge and experience makes it fairly obvious.

If the nav data is suspect, it would be a little tedious to have to manually correct every RW in the sim. It still doesn't explain why the default aircraft seem to have no issues.

[downscc 1,327]

I have never found a PAPI or VASI located on an airport in the simulation world where the FAA data says it should be.  Most the time even pricey scenery developers get it wrong because they accept the default locations, and they almost always miss that there is no back course and that the DME is not exactly located where the localizer is.... the trivial list is long and one you would not notice were you not a anal as I can be with such things.

[JRBarrett 3,070]

5 hours ago, Bluestar said:

Jim,

The problem is the default FSX airports are not in the correct location.  By this I mean the airport location maybe 50 to 100 feet off.  When I build a new AFCAD using actual current lat/lon coordinates, this mean that the airport must be moved slightly to get the runway/LOC/GS/PAPI into the correct relationship.  Google Earth is excellent for helping out.  Even then it normally requires a couple of flight tests to get the correct look.  I highly recommend Airport Design Editor for making these changes. 

blaustern

Some airports are off, others are correct. It's kind of hit or miss. In addition to using P3D as a standalone sim, I also use it as a scenery generator, driven by Aerowinx PSX. In that application, PSX controls the position of the P3D aircraft, using its own world model and runway database - which is very accurate. In many cases, the two correspond perfectly, though some are pretty far off.

The problem with FSX default data is that there were/are a significant number of ILS approaches where the placement of localizers and glideslopes were wrong from the very beginning: Localizers offset next to the runway, instead off aligned with the centerline - (NOT LDA installations which might be that way in r/w - but airports where the LOC should be on center) - or, glideslopes  co-located with the localizer, instead of at the proper position etc. Such problems are to be expected, considering that MS was trying to emulate 20,000+ airports and related facilities, in many cases using automated tools, and probably with a variety of data sources of various vintages.

I have found that Hervé's updates fixed many of those problems (though not all) - especially where ILS approaches have changed frequencies since FSX was released. Could definitely see where it might break the default ILS/PAPI correspondence though...

In any case, before departing on any flight in P3D, I always spend a few minutes taking a close look at the destination airport facilities using Flightsim Commander - checking runway numbers, ILS LOC alignments and frequencies etc - all with the aim of having no "last minute surprises" in the approach and landing phase. FSC is an excellent tool for this purpose.