While looking at the various instrument approaches for Napa Valley (KAPC), I noticed a discrepancy between the RNAV (GPS) Y/Z RWY36L missed approaches and the ILS/LOC RWY36L approach:

annotated KAPC ILS36L plate

annotated KAPC RY36L plate

annotated KAPC RZ36L plate

Why is it that the ILS approach is able to use a turning missed approach to keep pilot and plane clear of the high ground to the north, but the RNAV approaches to the same runway force you to either have visual contact with the high ground early, or outclimb it? To me, it'd make more sense to have the RNAV procedures use a similar missed approach to the ILS, as then you could have an all-aircraft RNAV procedure with low minima, vs. one with high minima that works for everyone and one with low minima that requires a significant (more than twice the norm) missed approach climb rate.


4 Answers 4


As best I can understand from the TERPS, it's because there are certain criteria for RNAV missed approaches that wouldn't be met by copying the ILS missed approach.

Specifically, I found the following (see chapters 7 and 15):

  • In RNAV missed approaches "turns shall not exceed 120°", but the ILS missed approach requires about a 180° turn. There's some related information here about the FAA testing turns greater than 120° in RNAV instrument procedures, it looks like some avionics can't handle them properly.
  • RNAV missed approaches are made up of legs between waypoints, and the first leg after the MAP has a required minimum length "to allow the aircraft's stabilization on course immediately after the MAP". Unfortunately I couldn't identify what that length is, because the FAA's PDF quality is abysmal and some information is unreadable, but the ILS missed approach requires a turn after a climb of less than 300' (from 214' to 500'), so the distance traveled in that time would presumably be quite small. If the RNAV minimum leg length is significantly greater than that distance, then it might be necessary to continue straight ahead, as both RNAV approaches do.

I'm certainly not a TERPS expert and there may well be other reasons for not using the ILS missed approach, but those two points seem to show that the criteria for designing missed approaches is different enough between procedure types that you can't just 're-use' them freely.

  • $\begingroup$ I'd think that you could emulate the ILS missed in a RNAV procedure with two appropriately placed waypoints...but that doesn't deal with the leg length issue you pointed out, though! $\endgroup$ Dec 24, 2015 at 6:36
  • $\begingroup$ @UnrecognizedFallingObject Yes, I think that's the point: RNAV routes are constructed out of straight legs that have a minimum length and can't be joined at more than 120°, which means a genuine curve isn't possible and joining together a series of legs might result in a 'circle' with a large radius. At least, that's my understanding of the TERPS. $\endgroup$
    – Pondlife
    Dec 24, 2015 at 18:17
  • $\begingroup$ This is definitely the right answer. The GPS is fix-to-fix, so the Missed Approach Fix has to be able to be sequenced as the next waypoint in the RNAV equipment, given the turning constraints. $\endgroup$
    – rbp
    Dec 24, 2015 at 18:49

Look at where the missed approach points are on each approach. The ILS has you very low, and possibly unable to clear the terrain to the northeast, hence the turning procedure. The RNAV has you going missed almost 1000 feet above the ILS, so terrain clearance would not be as much of a factor. For either ATC or convenience reasons they probably prefer you to go missed to the north east, hence they have the RNAV go that way.

  • $\begingroup$ That's not true for the RNAV Z procedure -- the minima on that are much closer to ILS minima, yet it uses the same missed as the RNAV Y procedure, just with a minimum climb rate requirement. $\endgroup$ Dec 23, 2015 at 3:12
  • $\begingroup$ The inclusion of a minimum climb gradient on the RNAV Y is due to the lower minimums for that approach. If it were an LPV only approach with only a 250ft DA, there is a good chance there would be a missed approach procedure similar to the ILS. However with the inclusion of the less restrictive (climb gradient-wise) MDA minimums, the use of a straight out missed is possible. The minimum climb gradient is no doubt based on the 250ft LPV DA, since including multiple MAP that are based on the minimums used for the approach are not desirable. $\endgroup$
    – Scott W.
    Dec 23, 2015 at 3:26
  • 1
    $\begingroup$ the NPAs could use the same missed approach procedure as the ILS currently does -- your argument is nonsensical (just consider for a moment what the missed approach procedure for the LOC approach to 36L is) $\endgroup$ Dec 23, 2015 at 4:26

GPS approaches and missed approach routing are designed point to point to point, so "climb to XXX then direct ABC" doesn't work -- there is no defined track to sequence to after the straight-ahead leg. So the GPS approaches can't use the same procudure as the ILS does.

There does seem to be a terrain or obstacle issue somewhere on the missed approach routing of the GPS approaches. So you can either fly the "Y" approach with a high MDA (so you start the missed approach higher, with less altitude needed to clear the obstacle), or you can fly the "Z" approach that takes you lower but requires a higher than standard missed approach climb gradient.

There is probably a reason that the approach designer didn't use a missed approach track that would look more like the ILS missed approach track (but done point to point to point), avoiding the obstacle even with a normal climb gradient, but we can't tell what that reason is. A minimum length of each segment and similar TERPS criteria would be my guess, but that is only a guess.


Could it be because you are using two different navigation systems?

Yes, you might be on the RNAV/GPS for the approach and have the ILS on a second display. But, why wouldn't you just use the lower DA on the ILS anyway?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.