View Full Version : Boeing 777 FPA

5th May 2002, 13:51
Can anybody tell me whether FPA mode on the MCP is used to relieve workload during a VFR approach by letting LNAV fly the approach down to the threshold from a fixed point along the centreline?

I recall on the 757/767, building a RX (Runway Extension), performing an Intercept leg along the extended centreline and entering height/speed in the FMC allowed a pseudo 3 degree path to allow a VFR approach to the threshold flown by LNAV when an ILS is not available.

Don't fly glass, so if I don't make total sense - excuse me.

5th May 2002, 19:14
FPA is just that - a means of flying a fixed angle descent.

It's no different to VS in that if you are not on the correct profile for your, say 3deg, approach and select FPA 3degs then you will simply parallel the profile.

Also, if you ARE on the profile and, say, the machine ballons a bit as you select flap then it will NOT regain the profile but again will parallel it at your selected FPA.

I never used it, preferring VS which is less cumbersome to adjust because the increments are bigger.

What you are describing would be a VNAV profile where the AP would achieve specified points and angles between. On the 777s I flew this is not currently available due to software glitches.

Best way to fly a visual is by looking out the window and get your mate to call nominated check alts.

6th May 2002, 11:02
FPA is another method of manually selecting a descent in the aircraft. Instead of using V/S, FPA can be selected instead. Depending on approach reference speeds, 700 to 800 fpm on an approach equals out to roughly a 3 degree G/S. But, if it is easier to think in terms of a 3 degree slope...one can just select FPA..simple as that.

With regard to VNAV approaches, using the autopilot...Yes, it is done! and.....it works wonderfully. The only requirement is that the approach has to be properly 'coded' for use. We can determine this on the legs page and having a look at the approach. If the approach is 'coded', there will be an FMS notation depicting the G/S angle...i.e. 2.99 deg. Along with the GPA, there are 'hard' crossing altitudes for the autopilot to follow, or fix crossing altitudes.

If the approach is coded, then a VNAV approach can be used. So for instance on a localizer only approach, all that is required is to load the approach..say a normal ILS approach....verify that the approach is satisfactory for use, configure the aircraft properly for the approach, using LOC for lateral guidance and then selecting VNAV for vertical guidance.

When the aircraft FMA is annunciating VNAV Path, one only has to sit back and watch the airplane 'do-its-thing'. The FMC calculates the angle and descent point to begin down to MDA, the fix crossing restrictions are all taken into account, thus the airplane flies a continuous descent of MDA....if one is looking out the window at the PAPI or VASI, the aircraft is right on the slope all the way down.

The most disconcerting part of flying a VNAV approach this way is that at the point that we precalculate the descent point using our 'natural computer'..our brain....sometimes the airplane waits a bit longer to begin the descent and it is difficult to resist the temptation to select either FPA or VS to force the airplane to descend. If another method is selected and we do away with VNAV, then the whole exercise becomes an FPA or VS approach and manual altitude restrictions need to be adhered to with vigor.

The system works very very well....NDB, VOR, LOC are all being coded for VNAV. The accuracy of the approach is so good, it is expected that the VNAV approach MDAs to be down to 250 ft.

Hope that helps a bit.....Safe flying.

7th May 2002, 13:55
sorry,. my text should have said VNAV, not LNAV, of course we are constructing an artificial Vertical not lateral profile!

One thought though. With the advent of GPS, will we see the elimination of the need for ILS for suitably equiped transport category a/c?

After all a GPS-autopilot coupled approach could substitute for the ILS (and we could use the triple-INS and DME/DME as backups/additional voters).

It would also allow (along the lines of MLS) curved approaches.