The reason track separation in RNP environment is still measured in miles and not inches is that all these rules assume a certain precision, so that even if aircraft deviate from the route for whatever reason, they are 95% or more likely to be within a certain tolerance. While that does not mean the other 5% are ok, though, protection is provided for a much wider tolerance.
RNP tolerance is half the spec, so 0.5NM either side of the centreline with 95% probability. And the protected airspace is twice the spec, so 2NM either side of the centreline. RNAV5 has less protection, IIRC.
The total error is in practice the “Flight Technical Error”, i.e., the total error introduced by pilot and autopilot performance, ranging from how precisely the needle is read to how accurate the aircraft control is.
for RNAV5, 2.5 either side would be within spec, and RNP1 0.5 eithter side, but controllers are so used to seeing precision that they will probably drop a hint well before
RNP tolerance is half the spec, so 0.5NM either side of the centreline with 95% probability.
It’s the whole spec. with 95% probability, isn’t it?, i.e. 1 NM either side of the nominal track for RNAV 1. At least that’s how I interpret the ICAO PBN Manual.
Timothy’s point (if I get why he asked)
I think (hope) I am being transparent.
One of the most common, if not the most common, flight situation over northern Europe at between FL80 and FL150 is that there is approximately 4/8 cover of bubbly clouds.
Sometimes those bubbly clouds are innocuous, but mostly they are bumpy, sometimes they are ice-bearing and at all times pax prefer to be between them rather than in them.
These clouds vary in size, but let’s say that they are generally 1½-2 nm diameter, with 1½-2 nm between them. That means that, under the circumstances I am defining, there is always clear sky within 1nm of the centreline of an RNAV5 route.
Generally, one can pick a route through these clouds, one cloud at a time and, after passing each one, looking for the best way past the next to get the CDI back in the middle, or even deviating slightly the other side. (That is exactly what one does VFR.). One can generally do this, ENR, without the CDI getting more than ½ FSD. One can maybe push that a bit on an SBAS box, where ENR is 2 nm FSD, not 5, so even ⅔ or ¾ is well within RNAV5 PBN tolerances, and “along the defined centre line of that route” is not precisely defined.
As the repeating pattern is of the order of 4nm, if one were to ask for every turn, one would be asking every 1-2 minutes, which occupies a lot of air time and mental space for the controller. Quite often, the controller will often say, in slight exasperation “you are free to manoeuvre for tactical avoidance” or words to that effect. As everyone sharing the airspace at those levels will be squawking C, it is reasonable to look at TAS and see that there is nobody anywhere near.
To me, it makes sense to quietly weave and not bother anyone, but it is precisely that that I am checking with my friends.
I am sorry if any of that wasn’t clear.
It’s the whole spec. with 95% probability, isn’t it?, i.e. 1 NM either side of the nominal track for RNAV 1. At least that’s how I interpret the ICAO PBN Manual
I agree. Here is a quote from the US AIM section 5−5−16. RNAV and RNP Operations, a. Pilot:
11. Definition of “established” for RNAV and RNP operations. An aircraft is considered to be established on-course during RNAV and RNP operations anytime it is within 1 times the required accuracy for the segment being flown. For example, while operating on a Q-Route (RNAV 2), the aircraft is considered to be established on-course when it is within 2 nm of the course centerline.
According to here there is no longer a need for the aircraft or pilot approval in Europe, so for N-reg owners no need for the FAA LoA.
The requirement for PBN training and signoff for the pilot remains.
To be precise, the requirement for PBN training and signoff for the pilot, together with the introduction of NCO.OP.116, replaced the need for an operational approval (which, for FAA aircraft, was previously documented as an LoA)
NCO.OP.116 Performance-based navigation — aeroplanes and helicopters
The pilot-in-command shall ensure that, when PBN is required for the route or procedure to be flown:
(a) the relevant PBN navigation specification is stated in the AFM or other document that has been approved by the certifying authority as part of an airworthiness assessment or is based on such approval; and
(b) the aircraft is operated in conformance with the relevant navigation specification and limitations in the AFM or other document mentioned above.
Note that the aircraft does need to be approved in the sense of NCO.OP.116(a).
The change (or something very similar) was adopted by ICAO a few months after it was introduced in Europe.