There are some things that non-airline pilots shouldn't be making definitve statements about. Google RNP.

Couldn't agree more.

JR an ILS on 34 is like an Ipad in the cockpit- is it a need or a want?

All this discussion about the "dangers" of an NPA onto 34 and nothing said about the visual approach over sheed. There's your dangerous approach. 400' high with 6 miles to run and a howling quartering tailwind on base.

I disagree. Flying a ML 34 VOR/DME approach (or any NPA really) is relatively simple IF the crew understands the FMC, what it's got programmed and how it achieves what is programmed.

Am quite surprised the amount of interest this subject has sparked. It's good really as it sparks debate & keeps these potential dangerous situations in everyone's mind.
Humans learn form their mistakes, us Aussies included we are not immune here'
I look at it like this as the big question here now is do we need an ILS on this particular rwy. The answer is neither yes or no as it's much like a railway crossing where boom gates are installed. Do we leave the boom gates as they do the job if they are obeyed or do we install a bridge over the tracks to make it safer?
The whole idea of any inst App is to get the plane in the best position to land, end of story. The ILS improves that position doesn't guarantee you will get in but improves it much like the bridge over the tracks, improves yr chances of getting across the tracks but doesn't guarantee it (bridge might be blocked due car crash!)
ILS means it's safer, we don't need to be a rocket scientist to know that.
The 'boom-gates' (VOR) do the job, safety comes under the heading 'it's enuf 'till the horse has bolted'
As usual like all 'crossings' it's okay 'till someone dies!
The answer really is subjective.

Wmk2

I'd say the question comes down to this:

We have identified a tendency of at least one foreign airline is to select an idle thrust descent mode to minima when conducting an NPA. That is downright dangerous.

Since we as a county do not want an airliner splattered on the ground, we have two choices - either restrict those airlines' operations to Australia until they can demonstrate suitable training for NPAs, or install ILS approaches on all runways those airlines will utilise.

For the relatively small cost of 1 ILS installation compared to banning airlines, I would go for the ILS.

I think a review of altitude infringements on NPA's to Melbourne reveals its more than one airline and includes Australian & US airlines.

Hi Everybody,
Interesting thread to say the least.
In my time an NDB approach (No such thing as an ADF Approach, an ADF is the piece of equipment in the aircraft, the NDB is the radio beacon on the ground!) in a F27 or a Viscount properly executed down to near minimums was an extremely satisfying thing to do accurately. And everyone knew how to do one, ie Wynyard, Tassie. Time and motion study.
I'm sure JL will agree with me.
I'm sorry if I'm showing my age but to give an NDB/VOR approach to students in the simulator without any aircraft emergencies is an extremely good handling exercise for all pilots, with, or without, RNAV, PRNAV or GNAV or whatever you want to call it and a good lesson in crew management.
In my 30 years on B747/B744 I think I only ever did the real thing at JFK 13L, Tabriz, and on RWY06 VOR, and sometimes RWY24, at Manila at some very interesting times! There was no ILS on RWY06 and always a few CB's! (LAVS on the B744).
In my opinion its perhaps not a bad idea to keep these approaches (VOR/RNAV) in the Sim. RNAV approach into Kansei to ILS/GS at 1200 feet was also good effort.
Boeing and Airbus are continually designing the flying pilot out of the flightdeck through automation but these procedures must be taught!
I know, it's very hard to teach students the loop these days, ie stick/control column, throttle/s, rudder/s and scan! (Winjeel and Vampire days, eh! No, not that one!).
They just want to play with the autopilot!
Oh well, back to the Port and Stilton...
Cheers.
JO.

Firstly, I didn't make a definitive statement, I expressed an opinion. You can tell because it started with " I think.." In this instance I may be wrong. Although, my reading so far indicates that RNP is a navigation tolerance standard which is technology agnostic - it can be achieved with VOR, INS, DME and possibly other systems. I understand RNP originated in Europe and uses angular clearance tolerances rather then height or vice versa. Either way it requires full procedure redesign.

Secondly, I think the Queenstown GPSS RNAV RNP approach (which is the poster girl for RNP approaches) has been around for maybe 10 years? Or more? I think it was originally developed using an Apollo CNS-80. It doesn't really seem to be catching on. Which is my point, that (especially with airline capital expenditure programmes) these things are long linked.

As I understand it, a precision RNP approach requires an additional altitude input. This can be from either WAAS (back to the WAAS debate) or a precision barometric input, which I assume has to be part of the Flight Management System. This barometric input can also turn a VOR NPA into a VOR RNP precision approach.

If I read the draft CASA CAAP correctly (AC-91U-II-C-5), it says that Australia will only develop RNP VNAV approaches (ie precision approaches) if we "acquire a GNSS augmentation system". In other words, we'll only get precision RNP approaches if we install WAAS beacons, which I understand is not on the agenda.

Which brings us full circle back to ILS.

How did I do for a non-airline guy?

There are now RNP-AR approaches for BNE, CB, ML, CNS (I thinK). They're catching on in a huge way. RNAV approaches into CTAFs and procedural control aerodromes are the norm for QF 737s.

QF 767s can do RNP-AR into BNE. Small issues with the design (both procedural and FMC programming) of RNP-AR into CB and MEL but they should be up and running for the 767 within 6 months.

This is on top of the RNAV approaches that QF GE powered 767s CAN do in virtually every major aerodromes around Australia.

There is no reason why any 777 or 744 operator can't go through the appropriate motions and avail themselves of the RNAV straight in to 34 at MEL.

Proved my point.

Jetstar uses RNP down to .3 at OOL and it gets us down to 439'. .11 will get us to 367'. We also use it in BN.QF use RNP approaches in preference to ILS on most occasions but a QF pilot could give you more detail. GBAS ILS like approaches are already being trialled in Sydney. Reading stuff on RNP and using it on a regular basis are not the same thing. I'm not trying to be superior but I do get annoyed when people make statements about airline operations and procedures that they really do not know much about. The long term future for ILS will be to go the way of the NDB.

I don't disagree with the argument for having an ILS installed but someone has to agree to pay for it and I can't see too many hands going up.

Perth had only a VOR on 03 for years. Much whinging, moaning and complaining by all and sundry made no difference, despite the relative difficulty of the VOR approach, especially in bad weather (yes, we do get bad weather from the north occasionally, like Melbourne, by the sound of it). Then an ILS was installed. All the problems with landing on 03 went away, instantly.

Old Akro, no disrespect intended, but you are a bit off-track. Even RNP AR approaches (the whizzbang ones like ILSs/Queenstown) do not need a WAAS (airlines are doing them all over Australia now, see the AIP SUPPs), but do need a lot of organisation, training (initial and ongoing) and equipment standards. Too hard/impossible for many operators, including, I suspect, Thai. Could they do even an RNAV-Z to 34 (higher MDA though)? The capabilities of the 777 may be the issue.

As at the Gold Coast, on balance, we should bite the bullet and just chuck in an ILS on 34. The safety improvement is marked, the cost amortised quickly. I agree with the government paying; it's national infrastructure. If that crew is also buzzing around domestic Thai with VORs then they should have done better. However, if they are international types, then it is reasonable that they expect an ILS at one of the major gateways to Australia.

Agree. Never done it, don't want to.

ML RWY 34 has both GNSS & RNP approaches. They have the same MDA (330 ft) as the VOR approach. They have the same 3 deg approach path (naturally) and RNP approach seems to approximate the DME arc and transition of the VOR approach.

The RNP approach seems to remove the circle-to-land minimum altitude, which makes the Australian NVMC visual approach clearance requirement interesting.

As I understand it, the RNP requires CASA approval which is primarily about flight crew training and equipment certification. The RNP requires RF which is some sort of flight management system distance measuring function - I guessing DME based. I think this is an example of an RNP approach that could be flown with DME & VOR and doesn't require GPS

Correct me if I'm wrong, but I believe this is still a non-precision approach (which would require an additional height reference - typically a WAAS beacon).

The Thai incident was NVMC, so visibility & MDA were not an issue. The issue was the early & mid procedure altitude reference, which for the VOR approach is basically read the chart, read the DME and check the altimeter. For whatever the reason, the crew on the night (and a number of other crews by reference to ATSB reports) didn't do a very good job of this.

So, unless the RNP hardware feeds the flight director, then the Australian non-precision RNP approach isn't much of an improvement over either the VOR or the GNSS NPA. The main difference (I think) is that the FMS of a RNP aircraft will fly the curved transition.

As I understand it, the benefit of the ILS for all RWYS at ML is that the crew falls into a more practiced routine and gets vertical guidance. I don't think the Australian implementation of RNP achieves this.

I found a US FAA reference that says an ILS costs USD$3m to install and a US AOPA reference that suggests they cost USD$18,000 pa to maintain. If that is all it is - then why the fuss? Mildura is spending more than that on a new terminal building. If it costs significantly more in Australia, then we should hold the blowtorch to the belly of AsA and find out why they are so uncompetitive. - Or subcontract it to the FAA.

If the visual star onto 34 is dangerous why fly it? Emirates refuse it & good on them. If you all refused to do it maybe you'd get an ILS? Which international airlines are flying RNP approaches in Australia?

Old Akro, sorry mate, but you're rapidly losing the plot.

Not they aren't. VOR 430 AGL, RNAV-Z 450 AGL, RNAV-P 381 AGL.

Wrong. RF is short for Radius to Fix, meaning the FMS is capable of flying a constant radius turn to arrive over a fix/waypoint on a particular track (ie not simply a join-the-dots box). This is needed for "curved" RNP AR approaches. It's got nothing to do with DME and is all done in the box itself.

No! I said before and I'll say it again, RNP AR does not require WAAS. As for being a precision approach, you could say it is (exact definition irrelevant) because vertical guidance is provided and, I suspect, MUST be followed by either the autopilot or manually via the FD.

Nothing Australian about it; RNP ARs are the same world over as far as I am aware and they do provide vertical guidance (just as any half-decent FMS will provide vertical guidance for the VOR or RNAV-Z/GPS NPA).

With that, I agree. :ok:

Jeppesen don't have this procedure. It looks like a CASA special for approved operators only. So far, I can only find Jepp RNP charts for YMML & YBBN.

Maybe, but not completely, I'm just trying to learn & getting some bits wrong.

I screwed the MDA's - you are correct.

I pretty sure I'm correct about DME feeding the FMS - how else does it know the track miles to run? The INS needs updating. It can come from GNSS or DME. I'm guessing the FMS uses DME as default.

By definition, a precision approach requires an additional altitude reference. ILS does this with the radio based glideslope. GPSS does not have sufficient altitude accuracy, but it does when augmented with ground based reference (WAAS). The other alternative is a sensitive barometric input which I presume is something sophisticated that exists in FMS world only.

Without an additional altitude reference, I think we strictly speaking have RNP - LNAV approaches. This allows curved approaches and the fuel saving of shorter " Green" track approaches that AsA boasts about as well as reduced lateral obstacle clearances, but does not have the vertical accuracy of a precision ILS approach unless it has altitude augmentation.

My reading (mainly of FAA material) is that not all RNP approaches are alike. The CASA CAAP identifies 4 flavours of RNP approach. Only 2 of the four has ILS equivalent MDA's and accuracy - one using WAAS and one which requires localiser augmentation. See AC 91U-II-C-5. Its categorisation of sensitive barometric input differs from the design notes that I read on the Queenstown RNP design - it might just that the NZ stuff is old.

If you look at the YMML RWY 27 ILS & RWY 27 RNAV -M (RNP), the RNP approach has a DA(H) - being careful to get it right - for Cat D is 610ft & 931ft respectively. This reflects the greater altitude uncertainty of the (non WAAS) RNP NPA approach.

I am presuming (although I may be wrong) liability issues will mean that aircraft systems will not display vertical guidance ( a la ILS) for approaches with only C129a GNSS input. I contend it is the absence of this vertical guidance that contributed (or at least exacerbated) to the Thai incident.

But, this is off-topic, although its been interesting.

My fundamental thesis is that (whether through ability or not) the Thai crew joined a list of others who messed up an NPA approach to RWY 34. Rather than saying they are just not good enough and we are better, I'm saying there have been enough Australian, US & Thai crews to make similar mistakes that we should get on board with the rest of the world and put in more ILS approaches. If we can't do it for less than the cost of the carpark modifications at Melbourne, then we should be asking why we can't install them at International benchmark costs. If anyone mentions us being a small market, I'll scream. This level of equipment is an international market. A bunch of airfares & some airfreight from the US might add $100k, not $millions. The reference I found from the FAA was no less than Randy Babbitt about a year ago.

Visual approaches
 

Just mulling over this, checked Jepps and I see under Departure, Approach and Landing Procedures:


Still curious as to why it's necessary to issue a visual approach clearance?

UTR

Mate, they had a perfectly serviceable VNAV path to follow. They just chose not to use it. It's in the report. They selected FLCH when the aircraft pitched up to intercept the correct approach path. Poor situational awareness or knowledge of systems or training or something.

I was kind of wondering if we'd get a point of view like angryrat's :D

Taking that & Keg's opinion, I've concluded an ILS is not required on R34 :ok:

I've also concluded that you lot are like the People's Front of Judea, or is that the Judean People's Front?...............

It is an easy approach, especially with an FMS, and if it were Ozzies involved, an incident like this would embarrass the hell out of the operator and it would lift it's game pronto. But the reality is that we have no control over international operators and if one of their aircraft pranged in such a situation we'd be a laughing stock. Catering for the lowest denominator? I suppose. But that's what we are doing ourselves, introducing easier/safer approaches eg GPS NPA and RNP. It's all relative.

I resent that, but standby while I draft a resolution......

Old Akro you really don't know what you're talking about in this instance.

A typical FMC in a current model wide, or narrow body, jet CAN get input from VOR or DME but they also get input from dual GPS and VOR/DME input is typically inhibited. With RNP AR they must be inhibited to ensure the required FMS accuracy to fly the approach.

RNP AR does not require WAAS.

Not all FMCs can fly RF (Radius to Fix) legs which is a requirement for RNP AR approaches. The AR btw stands for Authorisation Required...the plates are not published for public consumption.

I very much doubt Thai is RNP AR approved...they MAY be GNSS conversant which is the GPS approach available to GA aircraft. GNSS is a VERY different animal to RNP AR.



An FMC gives you along track distance not slant distance direct to the station like a DME.

That a particular FMC can't do RF legs doesn't mean they can't fly curved tracks - they can.

ANY 777 FMC is perfectly capable of providing vertical guidance, that's what VNAV is, and the aircraft will follow it very well if the auto thrust is working. The LNAV (lateral guidance) and the VNAV is fed to the Computers and presented to the crew and auto pilots via the Flight Director. IT LOOKS JUST LIKE YOU'RE FLYING AN ILS.

When you programme the FMC to fly an approach a bunch of height/speed way points are coded into the legs starting at the RWXX waypoint (50' over the threshold) and back through the FF waypoint, CF waypoint and on back around the DME arc (in this case) and through the IAL transition all the way to TOPD.

In the cae of the 777 if you leave the damn thing alone in LNAV VNAV it will accurately fly the correct profile and decelerate to the required speeds at each constraining waypoint. LITERALLY all you have to do is configure the aircraft and set the MCP ALT (alt window) to each successive altitude constraint until the FAF then to the minima (once you're at least 300' below the missed approach altitude you set that and the aircraft will continue descending to the minima)....extend the flaps when the command bug tells you to and put the wheels down before it gets expensive.

It's EXACTLY like flying an ILS from TOPD...for all intents and purposes. Yes to a higher minima but from an flight instrument interpretation and manipulative stand point an NPA in a modern jet is functionally identical to an ILS.

Now if you insist on not using VNAV and instead use VS (Vertical Speed) or FLCH (Flight Level Change) all bets are off...You better know how to do that pilot sh!t...and you damn well better understand the differences and what happens when the Auto Flight Guidance System goes to 'On Approach' mode.

As an aside RNP AR is a thing of beauty...although not perfect. I'd suggest within 5 years, 10 tops, ILSs will be as rare as NDBs.

Yep. Agreed. But guys said that when we moved to nosewheels, went from range finding to ADF, went from ADF to VOR, etc. I was reminiscing recently about the old days of DED reckoning, drawing tracks and deviation lines on maps and 2 minute accuracy for full reporting. Going to MET briefing to speak face to face with a forecaster. Working out winds with a computer. There was real pride in navigation. Now I fly the GPS. Something is lost, something gained.

I lament the death of the slide rule and loss of skill with 7 figure log tables. I miss drafting tables with cane cored scales, an assortment of sharpened clutch pencils. I am dismayed by ABS, stability control and even synchromesh in cars - all of which degraded the skill and enjoyment of driving.

But there are 2 choices and I vote to not be left behind.

Maybe. But I understand about the first half of your points and thought I had incorporated that. Maybe I didn't express it properly.

The thing that I don't understand (and I thought admitted to) is whether the FMC gives vertical guidance for non-precision RNP approaches. You've explained that. Thanks.

VNAV gives accurate vertical guidance for NDB, VOR, GNSS, RNP AR. The tolerances vary.

Doesn't matter whether you're flying an NDB, VOR, GNSS, ILS or RNP AR in a modern jet...keep the needles centred and you arrive safely at the minima.

That may be the problem Chimbu. You can fly into the ground with the needles centred.

You just have to know what the needles are telling you!

All this going around in circles by way of explanation seems to ignore one key point.
A crew of a supposedly premium international carrier were unable to interpret information presented to them such that they screwed the pooch in a fairly unsafe fashion.
I am utterly astounded after reading the report for the 4th time that they managed to get themselves into the position they did.
It's not as if ATC at MEL is hard to understand, neither is the coding for the approach (I looked at it last night in our FMC), there are no traps for the unwary.
Thinking even more deeply into it, if I was feeling more of a Luddite than usual (no comments please CC) and I'd decided I was going to use FLCH to get to platform altitude...then a basic understanding of ALT followed by VNAV being armed would have me flying the prescribed vertical profile...even then...V/S at about 700-800fpm on that approach would keep me easily within limits.
Is it really that hard?
:ugh:

The important question is not whether or not a competent pilot should be able to fly the approach without an ILS, the question is whether an ILS would reduce the chances of someone crashing in Footscray etc. one day... and if it would, whether reducing that chance is worth the cost of an ILS.

Bloggs we do have control of international airlines. The US & the EU ban unsafe operators from their airspace. Australia is too busy kowtowing & PC. They pick on easy targets that can't defend themselves.

I agree with andrew. A stitch in time saves nine.

ANZ has RNP-AR capability, someone was asking about international airlines prior.

RNP-AR is available to DA's around 300 ft AGL to multiple destinations and those charts wont show up in the Jepps because they are operator specific, only Brissy and Melbourne are available on the general subscription. For QF we can get these approaches to CNS, TSV, ISA, OOL, HBA, ASP, AYE, DRW, ZNE, KTA, BME, PHE, KGI, PHE, ADL, name an airport and a procedure can be built to get to ILS plus about 50-100ft DA's depending on terrain, with no ground based aid augmentation. I don't fully believe ILS will be gone in 10 years because of the risk with a GPS shutdown, but it is the way of the future.

Just one more clarification Arko, since RNAV approaches there ain't no "precision" or "non-precision" approaches any more. RNAV approaches vary in their precision so that terminology is gone. In Sydney QF have been operating commercial RNAV approaches to autoland (called GLS which is GBAS assisted GNSS updated RNAV). RNP-AR approaches vary in their precision depending on the RNP figure used (a tighter RNP gives a lower minima). VOR and NDB approaches flown in LNAV/VNAV are just as precise as RNP-AR approaches (if they are coded to the RWY threshold). In fact, in QF the approach aid doesn't even have to be working.

About the incident- anyone could have done what the Thai crew did. Australian jet crews over the years have nearly landed on roads, botched go-rounds, taken off at night without lights, nearly ran wing tanks dry, nearly hit hills in the hold, descended below altitude restrictions. Anything else? I prefer to look at this incident as a learning exercise.
With respect to pilot standards, we are not taught these days to excel, we are taught to be average. Very few colleagues I fly with practice hands on flying, let alone manual thrust. Very few will also continue the approaches if visual early just to see where the minima really is and what you would be looking at ( always fun at Cooly and Maroochy.........sorry Goldy and Sunny). This is part of our culture nowadays. The sim is treated as checking tool. Not a training device. Schedules are crammed. Rosters make you want to do the minimum amount of work required. Sops do not contain words like "Captain's discretion"or "common sense". And if anything goes astray, JUST see what a JUST culture will do for you.

About the approach type debate- RNP will quite happily fly you into the ground even with VNAV engaged if you miss set the QNH.

Fathom an ILS will 'fly you into the ground with the needles centred' too -it's called auto land;)

Honestly, between the FMC, FD, FMA and RADALT it's really not rocket science.

Clark y agree completely - that doesn't make it any less a training and standards issue.

What in that report gives anyone the impression that THAT crew on THAT day would have done better self positioning for the ILS?

Now that's a term rarely heard nowadays - but as an exercise in practicing basic use of radio aids to find the airport it smartens the scan rate if hand flown.

While Thai bashing gets people thinking more about commonsense DME v height as a double check on profile management, you would be unpleasantly surprised at the astonished looks of dismay sometimes seen when in the simulator both CDU's are failed and the crew are asked to find their way procedurely to the airport ILS from 50 miles out, hand flying raw data and no radar vectors.

But it is just this sort of handling practice that keeps the pilot sharp rather than lazy in his old age. But then again, does he need to be sharp when the automatics are so reliable nowadays?

The counter argument of course being that that sort of combination of events would never happen in real life. Like the double engine failure at low level that Sully had with his A320 ditching in the river.:E Or the Garuda double engine failure in a 63,000 thunderstorm coupled with a total electrical failure when his APU failed to start because of a long stuffed battery and of course the all flaps up ditching that was the result. Of course that could never happen, either.:E But it did.

I did not see anything in the ATSB report about the efficacy or otherwise of the ground based warning system that Airservices must surely have in place when aircraft get too low and provides a warning to the Approach controller.

Surely the flight was already low when the pilot contacted the TWR?

I am sure Dick Smith and others in the past have mentioned that the ATC system has warnings like this.

I will have to enquire with my controller acquaintances??

You are correct except that the first thing you do when conducting a RNAV approach is turn off the DME update.

The FMC builds a profile from the end of the runway with fixes at certain points based on a 3 degree slope. It's all done in the computer with no external reference.

I always do a 3x check from the end of the runway anyway to ensure we don't have a Air NZ scenario again. (Type 'A Free Lesson' into YouTube if you don't know what I'm talking about)

To all those who state that an ILS on 34 would fix all the problems regarding foreign carriers and NPAs, 16R in Sydney has an ILS but it's out for maintenance for nearly a month. Guess what approach had to be flown to get into Syd? As an aside did Emirates end up landing after curfew with a screaming southerly?

Stick and rudder skills are obviously important, as is spatial awareness.

The issue with this incident however, isn't the lack of stick and rudder skills, but the lack of awareness of engaged autoflight modes and the inability to use and monitor them correctly.

Hi Lookleft,

There is quite a difference between an outage of an ILS for one month vs the lack of the facility at all when it comes to risk.

It doesn't matter whether its out for an hour what matters is that professional flight crew can operate the aircraft in the environment that they are presented with. Having an ILS on 34 in Melbourne (and that would also have to include 05 AD 11DN 30HB 14R LT) might make it easier but it won't make it safer if the person at the controls does not have a good knowledge of how his/her aircraft works. I flew an approach into Sydney yesterday that gave me all the appearances of an ILS with an LDEV and a VDEV indication and FD's that were centered. I would have been in for a nasty shock however if I thought that it was going to autoland as well.

Modern FMC's, autopilots and glass cockpits were also supposed to take away the risk of an NPA but all we have is a different set of problems. If you want to eliminate the risk then 16R shouldn't be used at all while the ILS is U/S. That would also imply that if an ILS was installed on 34 then the airport would be unavailable if it required maintenance and the only runway that could be used was 34.