. . . thread drift
 

roulette, others;

As one retired but still very much involved, the sidebar discussion between knowledgeable contributors on RNP AR/SAAAR, VEB / GQS has been invaluable; - in particular, the disagreements have been educational and helpful.

For those many pilots and instructors who may have only a barebones knowledge of this aspect of RNP work this and previous parts of the thread which dealt with NPAs has been well worth reading and thinking about.

There are a number of publications both in Canada (TC) and the U.S. (FAA) which put flesh on these ideas and I for one have downloaded them to further educate myself as we are doing these approaches and I want to understand what is at stake.

My thanks to all who took the short diversion so the rest of us could benefit.

roulette:

My airline (the late TWA) first got approval for CAT II in the 707 with baro alone. It didn't work out very well. So, they added radar altimeters to the entire 707 fleet.

My view was (and is) that CAT II in the 707 era was a dicey proposition. The later fail-operational auto-land birds made CAT II safe (L1011 and 757/767 in our case).

As to the VEB, we get a lot more out of Baro VNAV than with LNAV/VNAV. I think we've hit the stops on that one.

LNAV/VNAV, as an SBAS level of service, is far better than the Baro VEB.

LPV is better yet. An RNP AR FMS suite that would switch from RNP AR/VEB to LPV the last 1.5 miles, or so, would yield far better results than even RNP 0.10 with VEB Baro.

The GQS is not going away, though, for any approach with vertical guidance (excluding advisory vertical guidance, which is sorely misunderstood by many operators and may have been a factor in the KBHM accident).

For the discussion, both criteria considers RNP to be non-precision.

VEB and GQS are good concepts, but need to be developed further.

With either criteria, with RNP, VEB, you have a minimum of a 200' obstacle clearance with a 50 foot minimum momentary descent, hence it will never get below the 250' min.

The procedure design by third parties is usually very platform specific. The FMC combinations can act differently, even between same aircraft with different airlines, depending on the configuration. Obstacles must be independently mapped and verified.
The FAA and regulators require significant amount of money be spent to certify and validate the procedure for use, and there is ongoing maintenance of the procedures for every cycle.

Third party developers include also EO, weight limit, and DEP procedures, which are not defined in the criteria.

All of this costs a lot of money...

underfire:

The FAA considers RNP AR, LPV, and LNAV/VNAV to be neither precision, nor NPAs. The current party line is that they are approaches with vertical guidance (APVs).

Quite separate concepts. In the US the GQS applies to all APVs, as well as ILS and GLS.

The FAA has a different take on it. Because it is barometric VNAV, it has non-precision required obstacle clearance (ROC), which is the same for NPAs and Baro VNAV. The maximum possible height loss at DA in FAA RNP AR is accounted for in the transition to missed approach criteria.


That can be true for specials but third party developers in the U.S., of which there are two, must develop to public RNP AR criteria. They cannot account for OEI, unlike in some other countries.

So much so most of business aviation avoids getting involved, even with fully compliant airframes.

Again, not so for public RNP AR procedures in the U.S.

No doubt about that!

I think you know better than that. Alaska Airlines, and Naverus procedures in AK and PSP are designed with EO missed.

Do you think that PSP, and all the procedures in AK, and CA are designed without EO? There is the weight limited RNP DEP for Burbank.

There are RNP DEP in the US....what public criteria details DEP?

Alaska Airlines, Naverus, and Jepp all are certified as 3rd party designers, and all have significant exemptions to the criteria to do this.

There are what, about 3 third party developed PUBLIC RNP procedures in the US, and those were all required by the FAA to cert the developer, and I dont think any of those are in use.

Baro Cat II evolution to RA
 

@aterpster - Sounds like you may well have known my old pals W.R. and Larry DeC ! If so, you're well aware of the reasons for the eventual transition to RA (even at TWA), and the fact that the RA transition had nothing whatsoever to do with any Baro shortfall. In fact the transition was driven largely by needs of AFDS, the PB-20 (and PALS) on the B707, and the soon to come GPWS (bless Don B's heart) [remember TWA 514?]. We both loved the L1011 and its pioneering Cat III approvable autoland, but it is too bad it never fully incorporated either the MONA AMBAC, or AINS-70, or we otherwise could have probably started with [real] RNP two decades sooner. As to LPV, it is nearly useless and obsolete, and contributes nothing now, or will ever, to [real] RNP, being both unnecessarily angular and airspace wasting straight-in, and no better (and probably a lot worse) than [real] RNP [filtered inertial] core performance in big air transport jets. [Real] RNP is already vastly better than any LPV or APV will ever be for a broad range of reasons, and as a result it is much safer. Just for a few examples, it addresses all relevant risk factors, isn't exclusively based on (flawed Gaussian) extrapolations, and it fully protects the core down to the TDZ and back out. With both the present scientific VNAV VEB and RA floor available, it is the best offering out there, for any ops which do not have an ILS or GLS reference. Down the road, especially when we add another 30 to 60 SVs (e.g., Galileo et al), and with networks of GLS/GBAS on the way, it gets even better. That's why we already (for a decade now) have provision for RNP .003 in AC120-28D (see Table 4.5.1-1 Page 17). Further, RNP ALWAYS addresses both rare-normal and suitable non-normal conditions as well as normal performance (TERPS and PANS-Ops don't, can't, and shouldn't). That is just one more reason why hardly any real RNP airlines even use FAA's pseudo RNP (lite) at this point.

FWIW I took off on rwy 6 yesterday. They have scraped the crash site down to bare dirt. 18/36 was closed.

Tom,

There are some RNP transition to GLS final procedures out there, with a 15'+/- vertical accuracy, and even to CAT III Autoland, but the RNP 0.003/0.006nm horizontal containment I have not seen allowed. (that is a total width far less than the wingspan, which does not make sense)


Do you know where that is promulgated?

I would certainly like to understand your definition of REAL RNP. It appears that you are speaking of 3rd party custom RNP design, which includes EO and Departures, not the FAA/ICAO all engine approach RNP, 18 degree bank angle, and uncompensated baro, etc (and no DEP).
Is this what you are saying?

As a quick respite from the discussion of the alphabet-soup of approach procedures, it occurred to me that the LOC approach plate specified that it wasn't available at night if the PAPIs were out. Does that mean it is available if the PAPIs are out during daytime? In which case, how are you supposed to determine the vertical profile in a way that you can't do at night? And what about daytime IMC? Can it be used with the PAPIs out then? As it stands, it seems a bit of a strange limitation.

You can see the obstacles and terrain during the day, at night, not so much.

I just got a new Jepp Chart for the LOC 18 approach into BHM, and the night section in the minima has been removed. Only the note about VGSI remains, but there is now only one set of minima.

LOC, RNAV, VNAV, VPATH, RNP 0.3, SAAR, approach construction, TERPS, PANSOPS, JAR, blah blah blah.

The NTSB was quick to release some snippets of information from the CVR, such as "sink rate" and "runway in sight" when the aircraft may have already been below the MDA.

I would be very interested in hearing the PNF (pilot not flying) call outs as the aircraft reached minimums at 1,200ft and the response of the captain.

We can speculate about the how and why this unfortunate accident occurred, but I would bet the farm the NTSB report of Probable Cause will read simply, as it did in the March 29, 2001 crash of the Gulfstream-III, N303GA at Aspen, CO:


PROBABLE CAUSE The National Transportation Safety Board determines that the probable cause of this accident was the flight crew's operation of the airplane below the minimum descent altitude without an appropriate visual reference for the runway.

They may have difficulty in only doing that with the published call out of 'runway in sight'. Some explanation will need to be put forward for that call.



"What Do you Need?
When you are on an instrument approach, the regulations (14 CFR 91.175(c)(3)) state that you may descend below MDA or DH only if:
The aircraft is continuously in a position allowing for normal rate of descent to a landing on the intended runway;
The flight visibility is at or above the visibility required to complete the approach; and
At least one of the following visual references for the intended runway is distinctly visible and identifiable to the pilot:

reason for call: runway in sight

1. pilot thought runway was in sight

2. pilot heard sink rate alarm and said what he said a thousand times before in a pavlovian response.

3. pilots lied so that recorder would befuddle ntsb


dozens of pages about the details...time to go straight to the problem...if you are doing a non precision approach you have to BE CAREFUL as you may get DECEIVED and make a big mistake.

flarepilot:

In this case, had they followed the PAPI from MDA to the runway, we wouldn't have this thread.

1. Was the PAPI on?

2. Why no MSAW alert in the tower?

Those are two big questions.

I think there were posts about the PAPI's and MSAW.

If I remember, the PAPI's were on and the tower had disabled the MSAW due to so many false alerts.

I'll go back through and see what I can find.

Of the options only number 1 makes any sense. The crew were not trying to ace a check ride they were on a real night non-precision approach with terrain close beneath them.

So - the question is were the runway lights in sight but distorted in some way causing the crew to wish to descend more rapidly, or was some other set of lights thought to be the runway lights?

I've recalled a number of airports with warnings on approach plates...something like: CAUTION, do not mistake freeway lights for runway lights.

sure you have seen something like this...

Old Boeing Driver:

The only information I have are from the three Robert Sumwalt video tapes, still on the NTSB site.

My recollections:

He said there was no MSAW warning. He didn't say it was either working or not working.

His only comment about the PAPI was that it was flight inspected the next day and the lights were properly aligned. He made no comment about whether the PAPIs were on or off at the time of the accident.

flarepilot:

Yep.

But, KBHM Runway 18 is a black hole approach.

My understanding is that the a/c had a high rate of descent with engines at idle when it impacted? With terrain on the approach and a high RoD I cannot see how any MSAW system could have helped, on or off, in a practical time scale.

Should not the big question be why the rate of descent when 'visual'?

LOWEST common DENOMINATOR
 

(Bold is mine)
Since years :p I am collecting for fun tens of quotes in media from very high educated people refering to "LOWEST common denominator"...
They are not lapsus but loss of focusing on what is behind a group of four words(four in French, three in English) . It is a good exemple of how something used by (wrong) rote may bring trouble in a logical thought and brain.

Concepts of "HIGHEST common DENOMINATOR" and "LOWEST common MULTIPLE=LCM" are taught to everybody before Highschool and not difficult to understand, and I am sure, roulette, you know them very well, and you are in excellent Company! It warns us seriously about the limit of "logical" thought that we may wait from a pilot at ground, in flight, and in the middle of an approach a little tired with low visibility.

The only way to overcome such failure is replay and replay and replay again any sequence which is developping in parallel brain processes : that is "TRAINING". It needs airline money.;)

Dumb question probably, but what is "VEB"? :confused: :{

Dumb answer - 'Vertical Error Budget', but any more I know not, but I will remember it at the next party to impress folk:rolleyes:

Thanks BOAC. I'll add that to my CVE. :}

Oh - go on................

Aha - you are talking about "A band of musical amigos hell-bent on sharing songs and making light, with a seemingly unquenchable thirst for quality wines and German shots." - yeah!:hmm:

Captain's Vast Experience. :D:ok::8

Capn Bloggs:

The Vertical Error Budget was developed so the designers could use less obstacle clearance in the final clearance of an RNP final approach segment than that used in LNAV/VNAV approaches for the unbathed masses.

The OEM has to perform additional tests that demonstrates a particular airframe type's Baro VNAV system performs better than the standard Baro VNAV system.

The "budget" comes in when the designer calculates the RNP AR final approach segment required obstacle clearance, which is proportionally less as the RNP value used decreases from RNP 0.30 to as low as RNP 0.10.

I'm not an engineer nor a designer but it seems to me some of it is based on smoke and mirrors.

Thanks Terps. :ok:

ok465:

Photo taken of the crash by a news helicopter that morning after the crash shows the PAPI on.

aterpster
 

Thanks for your reply. I am still sifting through all the postings, but I believe you have it correct.

I have a really good memory, it's just short sometimes.

Were landing lights found on?
 

wondering if landing lights were on during this apch? and to clarify I mean the airplane's landing lights not runway lights (for those not in the USA).

anyone know the position the switches in the cockpit were found? does the 'bus have retractable landing lights?

I would also like to know the status of all lights on the plane including nav, rotating beacon, strobes etc

and cockpit lighting

flare:

Quote:

---

Were landing lights found on?

---

It usually takes about 6-8 months for the factual docket to be posted by the NTSB.

@underfire & aterps ref RNP criteria, VEB, et al
 

Except - and I know you know this - that the FMC fitted to the A320 was no more advanced than those fitted to the B757 and B767 which had been plying their trade for 6 years at that point. If pilot representatives had pushed that point aggressively at the time then is it not possible that an understanding could have been reached more swiftly?

I do not think he is emphasizing the fmcs:hmm:

Well, he is when he uses the word "automation", because the FBW systems are not automation - nor do they preclude handflying.

But his main point is about maintaining situational awareness...

DW why are you picking on us so much today:}

"Understanding" will only occur when the body count gets unpalatable. Why do you think nothing has happened up until now, let alone in the 6 years (your number) between 75/76 and the Deathstar? With the fundamentalist magenta agenda still in full force, reason will not be listened-to.

Besides, the pilots of those early magenta days were seasoned hand-flyers/brain-users. It takes years before the skills of the old hands atrophy, whereas the Children of the Magenta never had them.