Narrative:

Events took place in the middle of the night; and we were on the ground in mexico city for over six hours; so fatigue was probably a factor. The toluca 6A departure out of MMMX requires a loop around the city for noise abatement; but terrain is also a significant factor if getting too wide outside of the loop. There is a right turn at 6 DME to a heading of 174; and then another right turn at a second fix to a course of 263 (the tlc 083 radial inbound); creating a triangle of sorts. Both of these fixes are depicted on the departure as 'fly over' and not 'fly by' points. We were flying to airfields that are not in the aircraft's limited database. Therefore; approaches and departures had to be hand-built. The two pilots spent about 45 minutes trying to correctly build the departure in the box (they had plenty of time due to loading delays). The copilot was concerned that he could not input the points as 'fly-over' points. In the air; when you go direct to a fix; you can also insert an intercept heading. However; none of us knew how to work that into the flight plan waypoints on the ground (if it is even possible in the old box). The displayed path on the nd treated the waypoints as 'fly-by'; meaning it cut the corners of the triangle and was inside of our required course. There were therefore two loops: an outer loop; which was our intended course and the one depicted on the SID; and an inner loop that cut the corners of the waypoints; which is what was depicted on the nd as we flew the departure.as the copilot was very concerned with being inside of course; he briefed that he would takeoff in heading select mode. He would overfly the turn point; and then manually turn to the next heading. For the first turn point; it worked reasonable well. The PF may have started turning just a hair early; or maybe turned with too much bank; because we rolled out slightly inside of course. After completing the right turn at 6 DME; the PF initially but the heading bug to direct to the next fix; and the no-wind heading direct to the next fix was around 165 instead of 174. This is how I know that we were slightly inside of our 'outer loop' intended course; although we were still outside of the 'inner loop' purple course depicted on the nd. I expected the PF to correct back to the left until the heading to the next fix was 174; but he instead rolled the heading bug to 174; paralleling the course slightly inside of where we should have been (I didn't speak up because the deviation was minor and the next turn point would come quickly).although we had briefed 'fly over' vs 'fly by'; and the copilot was concerned about being inside of course; apparently the PF was focused solely on the first turn point; because for some reason he turned to the second heading way too early. Instead of overflying the second waypoint and then turning; he rolled in a heading of 263 at least two miles prior to the turn point. I don't know if the plane even got to a heading of 263 before he realized he had turned early; because he then put in an intercept heading to correct back left; but it was not nearly aggressive enough (maybe a 230). I said 'you are cutting inside of course' as I saw the plane was set to cross almost perpendicular to the purple line on the nd (the 'inner loop'...the entire reason we were flying in heading mode instead of LNAV was that the depicted loop was 'too far inside of course;' yet we were about to cut well inside of even the inner loop). The PF asked the pm to put direct to the third waypoint with an inbound course of 263; and was distracted with this request as we were about to cross inside of the loop. I said 'at this point you're better off just going LNAV' but got no response. The corrective action at this point was a hard 70-90 degree turn to the left to at least intercept the 'inner loop;' followed by an immediate right turn to 263. Right as we were crossing inside of the inner loop; the pm said 'I'll just put you direct' when the PF wanted the 263 inbound course. As he expressed his concern with that and they hadn't yet modified the route; I said more direct 'just go LNAV' (which would have turned the plane left back to 174; we were crossing the depicted line at this point). The pilot ended up putting direct to the third waypoint; which gave a course of 250 instead of 263 (and the plane was pretty much on that heading already). Had he put 263 like the PF requested; the plane would have gradually corrected back to course; but not modifying the route at all and just engaging LNAV with what was originally built would have been a smaller overall deviation. Both pilots seemed to think that our overall deviation was minor and no more than a mile; not realizing how badly we cut that second corner (we were around three miles inside of course). After discussing the direct 250 vs direct with a 263; the captain made a comment 'well no one has yelled at us yet;' and as if on cue the controller told us 'aircraft X; you are not on the procedure.' I got the ATIS and read it off; and noticed that only the captain rolled in the correct altimeter setting. On the descent/arrival checklist; the PF seemed momentarily confused that the pm had a different altimeter setting; and I spoke up again with no response. That's when I noticed that the copilot had his flight volume punched off. He hadn't heard me read ATIS; and that's why he didn't respond to any of my inputs on the departure. I reached over and punched his intercom on. Once we landed and shut down and had a chance to discuss; I got a pen and paper and drew out what I had watched happen on the departure. The PF stated 'I disagree' and remained convinced that the only problem was the captain not putting in the 263 intercept when he went direct to the third waypoint (and while that would have helped; the plane was already headed west and well inside of course when they were having the discussion about whether or not to put in the intercept heading). While part of this course deviation was pilot induced and could have been flown better; it would have been nearly impossible to fly the departure perfectly; as we could not depict our desired course correctly. Things snowballed; but really began with not being able to pull the departure from a database or manually build one with 'fly over' points.this aircraft should be limited to destinations that are in its database; until the ng box can be upgraded. Our course deviation was mainly a noise abatement/ATC issue. However; there is significant terrain nearby (i.e.; had the deviation been outside of course instead of inside). The weather was severe VFR; but had we been IMC the event would have been much more harrowing; and possibly a safety issue.had this been an rnp-1 or RNAV-1 departure (which I believe all RNAV departures now are in the us; and is becoming the international standard); we would not have been legal to fly it (as all procedures must be pulled by name from a database without pilot modification). Although the departures out of guadalajara and mexico city do not state 'RNAV1' and it was technically legal; and there were contributing factors of pilot situational awareness and CRM; the single biggest factor that led to our course deviation was the limitation of the old ng box. The lack of a database also caused problems on the way. We executed not exactly a go-around; but we had to break off the approach and get vectors for a second try. The approach was changed last-minute from the one we were expecting and had hand-built at altitude; and by the time we had it set up again; because it couldn't quickly be pulled from the database; we were too high to make the runway.

Related reports:

Google
 

Original NASA ASRS Text

Title: Air carrier Relief Pilot reported that after takeoff they were unable to maintain departure procedures due to the limited database in the aircraft.

Narrative: Events took place in the middle of the night; and we were on the ground in Mexico City for over six hours; so fatigue was probably a factor. The TOLUCA 6A departure out of MMMX requires a loop around the city for noise abatement; but terrain is also a significant factor if getting too wide outside of the loop. There is a right turn at 6 DME to a HDG of 174; and then another right turn at a second fix to a course of 263 (the TLC 083 radial inbound); creating a triangle of sorts. Both of these fixes are depicted on the departure as 'fly over' and not 'fly by' points. We were flying to airfields that are not in the aircraft's limited database. Therefore; approaches and departures had to be hand-built. The two pilots spent about 45 minutes trying to correctly build the departure in the box (they had plenty of time due to loading delays). The copilot was concerned that he could not input the points as 'fly-over' points. In the air; when you go direct to a fix; you can also insert an intercept heading. However; none of us knew how to work that into the flight plan waypoints on the ground (if it is even possible in the old box). The displayed path on the ND treated the waypoints as 'fly-by'; meaning it cut the corners of the triangle and was inside of our required course. There were therefore two loops: An outer loop; which was our intended course and the one depicted on the SID; and an inner loop that cut the corners of the waypoints; which is what was depicted on the ND as we flew the departure.As the copilot was very concerned with being inside of course; he briefed that he would takeoff in Heading Select mode. He would overfly the turn point; and then manually turn to the next heading. For the first turn point; it worked reasonable well. The PF may have started turning just a hair early; or maybe turned with too much bank; because we rolled out slightly inside of course. After completing the right turn at 6 DME; the PF initially but the heading bug to direct to the next fix; and the no-wind heading direct to the next fix was around 165 instead of 174. This is how I know that we were slightly inside of our 'outer loop' intended course; although we were still outside of the 'inner loop' purple course depicted on the ND. I expected the PF to correct back to the left until the heading to the next fix was 174; but he instead rolled the heading bug to 174; paralleling the course slightly inside of where we should have been (I didn't speak up because the deviation was minor and the next turn point would come quickly).Although we had briefed 'fly over' vs 'fly by'; and the copilot was concerned about being inside of course; apparently the PF was focused solely on the first turn point; because for some reason he turned to the second heading way too early. Instead of overflying the second waypoint and then turning; he rolled in a heading of 263 at least two miles prior to the turn point. I don't know if the plane even got to a heading of 263 before he realized he had turned early; because he then put in an intercept heading to correct back left; but it was not nearly aggressive enough (maybe a 230). I said 'You are cutting inside of course' as I saw the plane was set to cross almost perpendicular to the purple line on the ND (the 'inner loop'...the entire reason we were flying in heading mode instead of LNAV was that the depicted loop was 'too far inside of course;' yet we were about to cut well inside of even the inner loop). The PF asked the PM to put direct to the third waypoint with an inbound course of 263; and was distracted with this request as we were about to cross inside of the loop. I said 'At this point you're better off just going LNAV' but got no response. The corrective action at this point was a hard 70-90 degree turn to the left to at least intercept the 'inner loop;' followed by an immediate right turn to 263. Right as we were crossing inside of the inner loop; the PM said 'I'll just put you direct' when the PF wanted the 263 inbound course. As he expressed his concern with that and they hadn't yet modified the route; I said more direct 'just go LNAV' (which would have turned the plane left back to 174; we were crossing the depicted line at this point). The pilot ended up putting direct to the third waypoint; which gave a course of 250 instead of 263 (and the plane was pretty much on that heading already). Had he put 263 like the PF requested; the plane would have gradually corrected back to course; but not modifying the route at all and just engaging LNAV with what was originally built would have been a smaller overall deviation. Both pilots seemed to think that our overall deviation was minor and no more than a mile; not realizing how badly we cut that second corner (we were around three miles inside of course). After discussing the direct 250 vs direct with a 263; the captain made a comment 'Well no one has yelled at us yet;' and as if on cue the controller told us 'Aircraft X; you are not on the procedure.' I got the ATIS and read it off; and noticed that only the Captain rolled in the correct altimeter setting. On the descent/arrival checklist; the PF seemed momentarily confused that the PM had a different altimeter setting; and I spoke up again with no response. That's when I noticed that the copilot had his FLT volume punched off. He hadn't heard me read ATIS; and that's why he didn't respond to any of my inputs on the departure. I reached over and punched his intercom on. Once we landed and shut down and had a chance to discuss; I got a pen and paper and drew out what I had watched happen on the departure. The PF stated 'I disagree' and remained convinced that the only problem was the Captain not putting in the 263 intercept when he went direct to the third waypoint (and while that would have helped; the plane was already headed west and well inside of course when they were having the discussion about whether or not to put in the intercept heading). While part of this course deviation was pilot induced and could have been flown better; it would have been nearly impossible to fly the departure perfectly; as we could not depict our desired course correctly. Things snowballed; but really began with not being able to pull the departure from a database or manually build one with 'fly over' points.This aircraft should be limited to destinations that are in its database; until the NG box can be upgraded. Our course deviation was mainly a noise abatement/ATC issue. However; there is significant terrain nearby (i.e.; had the deviation been outside of course instead of inside). The weather was severe VFR; but had we been IMC the event would have been much more harrowing; and possibly a safety issue.Had this been an RNP-1 or RNAV-1 Departure (which I believe all RNAV departures now are in the US; and is becoming the international standard); we would not have been legal to fly it (as all procedures must be pulled by name from a database without pilot modification). Although the departures out of Guadalajara and Mexico City do not state 'RNAV1' and it was technically legal; and there were contributing factors of pilot situational awareness and CRM; the single biggest factor that led to our course deviation was the limitation of the old NG box. The lack of a database also caused problems on the way. We executed not exactly a go-around; but we had to break off the approach and get vectors for a second try. The approach was changed last-minute from the one we were expecting and had hand-built at altitude; and by the time we had it set up again; because it couldn't quickly be pulled from the database; we were too high to make the runway.

Data retrieved from NASA's ASRS site and automatically converted to unabbreviated mixed upper/lowercase text. This report is for informational purposes with no guarantee of accuracy. See NASA's ASRS site for official report.