737 FMC - can you force fly-over?
Thread Starter
Join Date: Jun 2006
Location: Europe
Posts: 716
Likes: 0
Received 0 Likes
on
0 Posts
737 FMC - can you force fly-over?
Hello all,
I was trying to code the circling approach into Dubrovnik (LDDU) runway 30 into the FMC on a 738. It's the last page of this document: http://ldzo-fir.klaeuserich.net/LDDU.pdf
(With the risk of sounding like a magenta line warrior, I specify that I was not planning to use it, but explore the various functions, as I'm new to the type.)
Doing this, I tried to force a few of the waypoints to be fly-over (the base turn point and GR), as per the charted procedure. However the FMC uses it's turn anticipation and creates fly-by waypoints, effectively tightening the procedure.
The captain didn't know, and I didn't find anything about it the FCOM. On my previous type I could lineselect the waypoint and enter a /0 (like an uproute or downroute waypoint), to force it to fly-over instead of fly-by. Are there any similar functions avaliable in the 737?
(And the disclaimer again: it's for practice and system knowledge, not because I'm unable to find the cockpit windows.)
I was trying to code the circling approach into Dubrovnik (LDDU) runway 30 into the FMC on a 738. It's the last page of this document: http://ldzo-fir.klaeuserich.net/LDDU.pdf
(With the risk of sounding like a magenta line warrior, I specify that I was not planning to use it, but explore the various functions, as I'm new to the type.)
Doing this, I tried to force a few of the waypoints to be fly-over (the base turn point and GR), as per the charted procedure. However the FMC uses it's turn anticipation and creates fly-by waypoints, effectively tightening the procedure.
The captain didn't know, and I didn't find anything about it the FCOM. On my previous type I could lineselect the waypoint and enter a /0 (like an uproute or downroute waypoint), to force it to fly-over instead of fly-by. Are there any similar functions avaliable in the 737?
(And the disclaimer again: it's for practice and system knowledge, not because I'm unable to find the cockpit windows.)
Most Airlines only allow crews to fly LNAV or NAV approaches from "data base approaches". If you modify the approach or construct your own then you cannot fly it below MSA using LNAV or fully managed modes.
So, if you select a particular approach or departure in your 737 FMC that contain fly over waypoints then they will already be correct and you shouldn't modify the waypoints or the altitudes.
So, if you select a particular approach or departure in your 737 FMC that contain fly over waypoints then they will already be correct and you shouldn't modify the waypoints or the altitudes.
Last edited by nitpicker330; 9th Jul 2012 at 11:39.
Join Date: Jul 2008
Location: Euroland
Posts: 147
Likes: 0
Received 0 Likes
on
0 Posts
To force a flyover point, line select the point where you want to turn base and use it to create a point on extended downwind at a distance of 0,1nm.
For example type FI001116/.1 should do it. (FI001 being the name of the point where your original fly by point was of course)
For example type FI001116/.1 should do it. (FI001 being the name of the point where your original fly by point was of course)
Last edited by Tom!; 9th Jul 2012 at 12:03.
Join Date: Mar 2011
Location: engineer at large
Posts: 1,409
Likes: 0
Received 0 Likes
on
0 Posts
Coding it yourself is fraught with disaster, and actually, may be against the law. ICAO and FAA have very specific rules on this. If you over-right a point in your database, it may affect another procedure somewhere else that it was used.
You may be able to create TF legs, but never a specific coded turn radius as you dont have access to the CNF waypoint database. You also dont know what to code, or code 'at or above'.
with turn anticipation, all points must be flyby, or the ac will cross a waypoint, then start to turn, and intercept the TF leg as it can....
one would also have to understand turn radius, turn speeds, and bank angle restrictions, or the database will just disco it on load anyways...with a coded turn, the ac stays a certain distance from the CNF point, beginning tangent from the TF, maintain distance from CNF to tan to the next TF...if any of the waypoints do not work this out mathematically, it will disco on load as well...
I really cannot see how one could code this on the FMS....
You may be able to create TF legs, but never a specific coded turn radius as you dont have access to the CNF waypoint database. You also dont know what to code, or code 'at or above'.
with turn anticipation, all points must be flyby, or the ac will cross a waypoint, then start to turn, and intercept the TF leg as it can....
one would also have to understand turn radius, turn speeds, and bank angle restrictions, or the database will just disco it on load anyways...with a coded turn, the ac stays a certain distance from the CNF point, beginning tangent from the TF, maintain distance from CNF to tan to the next TF...if any of the waypoints do not work this out mathematically, it will disco on load as well...
I really cannot see how one could code this on the FMS....
Last edited by FlightPathOBN; 9th Jul 2012 at 14:23.
Join Date: Jan 2006
Location: Between a rock and a hard place
Posts: 1,267
Likes: 0
Received 0 Likes
on
0 Posts
bfisk,
Think you need to spell it out again 
Not sure if you can change a fly-by to fly-over in the FMC.
What I've done sometimes to use LNAV instead of HDG SEL is I've created a PBD point along the same axis (i.e. downwind leg in this case) 0.X nm further ahead. This of course requires good "guesstimation" of what the 'X' should be to create a perfect fly-over point.
(And the disclaimer again: it's for practice and system knowledge, not because I'm unable to find the cockpit windows.)
Not sure if you can change a fly-by to fly-over in the FMC.
What I've done sometimes to use LNAV instead of HDG SEL is I've created a PBD point along the same axis (i.e. downwind leg in this case) 0.X nm further ahead. This of course requires good "guesstimation" of what the 'X' should be to create a perfect fly-over point.
Tom is correct and (along with Fred) is also the only one who can actually read a question , answer it and not pontificate!
Expect more of the 'finger pointers' but you have your answer.
I specify that I was not planning to use it, but explore the various functions,
Join Date: Mar 2011
Location: engineer at large
Posts: 1,409
Likes: 0
Received 0 Likes
on
0 Posts
So if you dont know the syntax of a coded waypoint, CNF point, or other coded ponts that just the FMS sees, how do you know you wont over-write one in the database? FI001 could easily be in a database...
Depending on how the FMS is set up, you can get many different values for the turn anticipation, so you would have to experiment on how to get that to work.
This is an example of the same coded procedure, and the associated tracks for the different aircraft types and FMS manufacturers....
Specifically, the green track, and others, cut all the corners, and magenta even misses a turn...
Depending on how the FMS is set up, you can get many different values for the turn anticipation, so you would have to experiment on how to get that to work.
This is an example of the same coded procedure, and the associated tracks for the different aircraft types and FMS manufacturers....
Specifically, the green track, and others, cut all the corners, and magenta even misses a turn...
Last edited by FlightPathOBN; 9th Jul 2012 at 16:15.
Per Ardua ad Astraeus
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes
on
0 Posts
bfisk - as above - cannot be done. Not worth the effort (unless you are intending to submit a pic of the screen for an art exhibition) but extend past the point (on inbound track) for half your g/s in miles per min. and you'll be close to a fly-over.
Keep exploring
) but extend past the point (on inbound track) for half your g/s in miles per min. and you'll be close to a fly-over.Keep exploring
Thread Starter
Join Date: Jun 2006
Location: Europe
Posts: 716
Likes: 0
Received 0 Likes
on
0 Posts
Thank you all for your answers, both related to my question as well as to the inferred consequences. All views are welcome.
Again, letme point out that the whole thing was never used and the approach loaded "clean" from the database well ahead of ToD.
As I'm new to the type, I find these things easier to learn from practice than reading about it in the FCOM. Perhaps not critical knowledge, but the more you know...
Again, letme point out that the whole thing was never used and the approach loaded "clean" from the database well ahead of ToD.
As I'm new to the type, I find these things easier to learn from practice than reading about it in the FCOM. Perhaps not critical knowledge, but the more you know...
Join Date: Mar 2011
Location: engineer at large
Posts: 1,409
Likes: 0
Received 0 Likes
on
0 Posts
If you really want to look into this, here is how DTA is supposed to be calc'd.
Fly-over waypoints make you set many parameters, so you can fly over them, fly-by waypoints give the FMC the idea of what to do, and it handles the turn...
you can see why included angle TF legs have such unusual results...
there are several early versions of FMS, which would do a flyover anyways, then turn to get back to the TF leg, this is one of the reasons why the containment's are so odd looking...
this is an example of what the procedure design might look like for that turn you were looking at...
fly-over
vs fly-by...
Fly-over waypoints make you set many parameters, so you can fly over them, fly-by waypoints give the FMC the idea of what to do, and it handles the turn...
you can see why included angle TF legs have such unusual results...
there are several early versions of FMS, which would do a flyover anyways, then turn to get back to the TF leg, this is one of the reasons why the containment's are so odd looking...
this is an example of what the procedure design might look like for that turn you were looking at...
fly-over
vs fly-by...
Last edited by FlightPathOBN; 9th Jul 2012 at 19:58.
Join Date: Mar 2011
Location: engineer at large
Posts: 1,409
Likes: 0
Received 0 Likes
on
0 Posts
If you want to play with coding, at least be aware the syntax of the information in the database.
Public procedures, as you know, use 5 letters for waypoints that are pronouncable.
CNF waypoints used for FMS calcs, as 5 letters, all consonants, that cannot be pronounced. (although I have to wonder how some of the waypoints can be pronounced.
With coded procedures, such as RNP, there is now some reasoning to the naming convention. It is typical now to use the last 2 letters of the ICAO airport code as the first 2 letters of the waypoint, unless that is already used, then its the first 2 of the code or negotiated.
The IF will be a standard 5 letter pronounceable waypoint, but since the procedure waypoints are not announced over the radio, they do not need to be pronounced, so you will see XX123.
Now, the numbering, needs to be descending on the approach and increasing with the missed.
As an example YMML, would be ML600(VIP), ML598, ML596, ML594(FAF), ML300, ML302, ML304(HOLD)
Other than that, the numbering is up to the designer, and the navdatabase supplier to sort out....
I know, much to do about nothing, but that is how coding works....
Public procedures, as you know, use 5 letters for waypoints that are pronouncable.
CNF waypoints used for FMS calcs, as 5 letters, all consonants, that cannot be pronounced. (although I have to wonder how some of the waypoints can be pronounced.
With coded procedures, such as RNP, there is now some reasoning to the naming convention. It is typical now to use the last 2 letters of the ICAO airport code as the first 2 letters of the waypoint, unless that is already used, then its the first 2 of the code or negotiated.
The IF will be a standard 5 letter pronounceable waypoint, but since the procedure waypoints are not announced over the radio, they do not need to be pronounced, so you will see XX123.
Now, the numbering, needs to be descending on the approach and increasing with the missed.
As an example YMML, would be ML600(VIP), ML598, ML596, ML594(FAF), ML300, ML302, ML304(HOLD)
Other than that, the numbering is up to the designer, and the navdatabase supplier to sort out....
I know, much to do about nothing, but that is how coding works....
Last edited by FlightPathOBN; 9th Jul 2012 at 20:46.
