CROSSAIR CRASH -

[5 APU's captain]

At least I am agree with NorthernSky.
Descent below minimum - is it professional???
AND: in Aeroflot it is not possible to put into the cockpit two guys without at least one year experience on type for any one...

[scrubba]

harry & soddit,

my understanding of both the 737 and rj dfgs is that, if you set mda or any intermediate altitude below your present altitude and enter a vertical mode, a subsequent selection of ga will result in the airplane transitioning to pitch ga mode and toga thrust and climbing. you will be without a capture altitude until you set altsel above your present level (watching out for a capture as you wind through) so you are in a form of open climb. i do not believe that selecting ga will allow the pitch mode and a/t mode to continue as before.

we set up for a level segment at mda -if visual we just pickle off the a/p, if not we set ga altsel after mda capture.

are you both sure that the rj does not always climb when ga selected (if it has not already captured the altsel altitude)?

[Shaggy Sheep Driver]

Slopwith - re your comments about 24 departures at EGCC. There wasn't really room to build a second runway on the Manch site. But (with the promise to the local populace of 1000s of jobs - which surprise surprise - have not materialised) they shoehorned in a 24L/06R, on the far side of the exisiting runway to the terminals, with a massive offset to the SW because of its proximity to the existing runway. There isn't even room for a taxyway. Because of the offset, a centreline-tracking t/o from 24L would have traffic climbing out *very* low over Mobberley (not Wilmslow, which is south of the field). So they had to introduce 'silly' SIDs.

It's a mess of the airport's own making, not the fault of the folks who live in Mobberley and Knutsford.

SSD

[NorthernSky]

My post had nothing to do with individual events or crew member's actions, and everything to do with statistical risk.

Statistical risk is what we base commercial flying on. If, and it's still 'If', the Crossiar crew made an error, they did so whilst operating in a statistically risky environment/regime over which they had no control.

The noise lobby don't perceive NPAs to be in the same category as unlit pedestrian crossings or unroadworthy school busses. That is the problem. The reason they don't have that perception, is that we (the aviation community) have kept it from them in our pursuit of public confidence.

[scrubba]

soddit,

i was talking about full autoflight npa.

harry said:
[quote]By selecting MDA on the ALT SEL as you go down the approach IF you select GO AROUND using the GA buttons on the throttle face,with A/P & A/T engaged, then the throttles will close and the aircraft will descend.<hr></blockquote>

my understanding (based on bae tsg and honeywell efis/dfgs publications) was quite different. i don't have access to the sim and my pax would not enjoy me practising full autoflight ga just to prove a point. but i figured you guys might have better access to the sim.

so, once again: i do not believe that selection of ga while descending will result in entering the flight condition you describe, ie level change to mda,unless there has been a significant mismanagement of the airplane's energy state. i believe that, if you are in vs with the appropriate speed set in the vsel, the at will maintain speed with power. if you then decide to quit and hit a toga button, the ap will transition to a pitch ga mode and the at will go to toga max or the trp power and maintain the vsel speed. i further believe that it will look for an altitude logic - failing to find one will result in an "open" climb (for want of a better description) rather than some mystical recognition that the target altitude is below the present level and therefore the airplane must be dived back to the mda!

one of us is very wrong - if it is me, then i need to have a solid basis for understanding my error, such as a manufacturer's advice or something. it wouldn't be the first time i got it wrong, but in discussing this i found that i am not alone in my belief of what the airplane will do.

have you actualy seen this result that you report or are you going on what you were taught? nothing personal, just seems real important to me.

[H-M]

scrubba:

You are wrong...
Believe me I am flying the AVRO RJ every day!!!

[Capn Bloggs]

All,. .. .Firstly, I do not intend to prejudge the reasons for the Crossair prang: it was very sad that it happened, and very close to our hearts, being a Non-Precision Approach and therefore more difficult and dangerous (the previous callous comments by some contributors being noted!), especially when you don’t do them often.. .. .I’m sorry Soddit, H-M and Harry, I’ve got some bad news for all of you. Scrubba’s musings are correct. If the TOGA buttons are pressed when coupled in V/S, with the A/T engaged, descending towards an altitude set on the Alt Sel ie below the aircraft, the aircraft does a Go-round. It does not continue descending to the Altitude set on the Alt Sel!! I was on a test flight a few days ago and we did just that (to test the Pprune rumour): a GPS approach, coupled, in V/S, and the machine went for the heavens when the TOGA buttons were pressed at 200ft above the MDA, which was set on the Alt Sel.. .. .I might add that this scenario is easily simulated in line flying (take note H-M!): disengage the AT and the AP, but leave the FD on with a lower altitude set on the Alt Sel. Hit the TOGAs below 2000ft AGL and see what is annunciated. The result is the same!. .. .Soddit, while your comment about the implications of Vmin may be correct (in typical Pommie fashion, the Mom Vol 1 is very hazy about what happens, but it does indicate that LVL CHG will activate, the A/T will control to Vmin and therefore the aircraft will keep descending), it is completely irrelevant to this discussion, which is about what would happen if G/A was selected. On that point, the Vol 1 IS reasonably clear: the thing does a Go-Round regardless of what's selected at the time!. .. .One of our drivers in the crewroom said “who, in their right mind, would design a system where, when you pressed the TOGA buttons, the jet kept on going down?” Not even the Poms? The concept is totally ridiculous.. .. .Soddit, your comment about setting the Alt Sel in manual flight: “it would not matter what you had dialled into the Altitude window” is a tad naive too. As I hope you are well aware, the Alt Sel provides a Departure Warning when it is “busted” by 300ft: given that most MDAs are around 400-600ft AGL, that would provide a very handy last-second warning that you’d just busted something! I can tell you, if I heard the hoot and subsequent flashing Alt light, I’d be doing something pretty quick! Therefore, it seems to me to be a bloody good idea to have the MDA set if flying completely raw-data (given also that the PNF should have his FD on so that it will “capture” the MDA, providing a “fly up” cue that something is very wrong).. .. .I agree with scanscanscan and “Better be a Boeing otherwise I’m not going” that a catchbasket approach is the best. Let’s face it, Non Precision Approaches are the exception rather than the rule, so why not have everything going FOR you? Our company has an SOP that states that the A/P must not be operated in open descent mode ie without an altitude set to capture (BELOW, of course), whether in VMC or IMC. This would hopefully prevent an unintended descent below the MDA.. .. .So, WHY IS the Alt Sel set to the Missed Approach Final Altitude by some operators (and manufacturers?). Could it be that they consider it more important to have the Level Off Altitude set so the Go Around is not completely messed up by spearing thru the Alt while trying to set it? Perhaps. But, if a policy of doing a short level run at the MDA, to allow enough time for the Alt Sel to be set to the Missed Approach Altitude is employed (especially in something as manoeuvrable/powerful as the 146/RJ), then this paranoia about not setting the MDA is unfounded. Of course, the 300ft/nm descent profile can be easily adjusted to cater for a short level run before the Missed Approach point. And most importantly, it PREVENTS a crew flying through the MDA inadvertently. Not that there is normally any rush to do a missed approach anyway, on most Non-Precision Approaches that I’ve seen the MAPt is well past the point where the MDA coincides with the 3deg slope. . .. .Another trick we use is to set 300ft on the Rad Alt for Non Precision Approaches: if “gringo” says “Minimums” (ie you’re below the MDA) and we’re still in IMC, a Go Round is mandatory.. .. .All the best!

[White Knight]

Capn Bloggs. I do hope you're not going below MDA while IMC. (last paragraph of your post) . .. .On our fleet (BA Avro RJ100) the number in the alt sel window is the altitude to fly to in the event of a missed approach. On a non precision approach we level the aircraft at platform altitude, when ALT captured the alt sel is wound up to the missed approach alt and at the FAF VS mode is used to descend towards MDA - never below. Also we never fly the aircraft level at MDA hoping to see the lights. Nothing seen is an immediate go around.... .. .And yes pressing TOGA at any time gets the GA TRK mode with TOGA power.

[4dogs]

Folks,. .. .I must have missed this discussion first time around. However, alerted by this new post, I confess to being somewhat alarmed by the postings of some folks who claim to be driving the RJ. It seems to me that they either do not understand the equipment they are driving or they have deliberately embarked on a campaign of misinformation.. .. .PPRuNe is a rumour network that is widely read worldwide. It is disappointing when folks get on board and spread crap about operating issues when they must know there are always impressionable young folks around who actually may treat the information with more rigour than it deserves. Rumours should be sensibly constrained to the information flows about people, companies and what might term social items. Otherwise we might have to revert to APRuNe - the amateur network.. .. .For the folks that Bloggs has mentioned - the BAe MOM Vol 4 Instructors Guide has stated (since 1993!!) that MDA should be set in the Alt Sel for non-precision approaches. If you are doing something different, then clearly the managers authorising the different procedures have analysed the risk and cheerfully accepted the transfer of unlimited liability from the OEM to themselves.

[sabenapilot]

I have done jumpseat rides on both Sabena (Now SN Brussels Airlines) and Lufthansa RJs to Florence, (where untill recently there was only an NPA available and I can confirm you 100% that both company SOPs apparently call for the Missed Approach Altitude to be set on the MCP rather then the MDA.. .. .I don't know about the reason for this as I am not type rated on this plane, but if highly reputated carriers like SN, LH and LX all have this company procedure overriding the Bae MOM, then there must be a very good reason for that!. .. .BTW, I have flown both Boeing and Airbus at Sabena (I am now flying for Air Lib) and it is not that uncommon for a bigger carrier to have SOPs in place which differ from the MOM procedures.

[tetraNG]

When you use full automatics in an RJ no matter what mode you are using Vs or LVL CHG,the autopilot will revert automatically to LVL CHG if for any reason the airspeed will drop below Vmin for the current configuration,thus if the ALT SELECT is selected above your current height,the RJ will use max thrust and will CLIMB otherwise if it is selected below,the thrust levers will move to idle and apparently the aircraft will stall.That is the only reason we select the missed approach altitude when executing a NPA.That is the only way we have been trained from BA training Captains at Woodford.And for sure anytime you hit the TOGA buttons the aircraft will execute a go around no matter what altitude you have selected. Regards to all.

[rover2701]

As a humble engineer who worked in BAe customer support on the BAe146/RJ for many years, we were taught in ground school and also found through experience when testing the FGS iaw the relevent MM, that whenever the toga switches were operated with the autothrottle engaged the power levers would automatically move to the target N1 for toga and if a/p coupled would pitch the aircraft nose up irrespective what other mode on the Flight Guidance System was selected. In all the time I was involved with the type and checked the system out I never found that otherwise than that stated happened. . .Please I bow to greater knowledge if I am wrong. I cant believe that Honeywell or indeed any other manufacturer would make and programme the software in a dfgs to do other than that. The consequences would be to horrific to contemplate.. . . . <small>[ 13 March 2002, 19:15: Message edited by: rover2701 ]</small>

[tetraNG]

100 feet above minimums PNF calls "100 above" and PF selects ALT on the MCP and the aircraft levels off at MDA.But in order to do this you have to have a PF who has control of the aircraft and a PNF who watches and calls.If there are two PF in the cockpit then you are missing one crew and there is no automatic protection against that.Hope it wont happen again.

[Capn Bloggs]

White Knight,. .. .I can confirm that I/we don't intentionally go below the MDA in IMC (unless I really need to get in!...just kidding).. .. .IF we do, accidentally, then the Rad Alt would bleat (as well as the departure warning from the Alt Sel, of course, being set at the MDA!) and our SOPs require the PF to immediately respond appropriately. In other words, Go Round!. .. .Cheers, Bloggs

[5 APU's captain]

There was just one problem there: the PIC and SIC where BOTH unexperienced on AVRO....

[Capn Bloggs]

5 APUs,. .. .All the more reason to have a failsafe system ie setting the MDA until captured, in place...

[IONA]

Rubbish.. .The ALT SEL system fitted to the RJ and the 146 is not capable of accurately acquiring an altitude within the tolerances required of an MDA ( +50ft to -0ft ).. .BAe do not condone the use of the ALT SEL for this purpose, and in fact teach the setting of G/A ALT on passing the FAF.. .There are plenty of other 'failsafes' available to crew, including the use of SOP's or ROP's, setting of a Rad Alt if company permit, and good instrument scanning.. .These aircraft are not in the same league as Boeing / Airbus, and crews should be aware of the limitations and correct use of systems such as the ALT SEL.. .It scares me to thing that there are operators still utilising this dangerous practice, and publicly purporting it to be safe.. .In any case, there is no evidence to suggest that had MDA been set in the ALT SEL of the crash aircraft circumstances would have been any different. It would appear that the Capt was aware of his altitude, both from the PNF call of Minimums, and from the Rad Alt call. He attempted, for reasons unknown to me to fly level at this altitude to the MAP. This is the real lesson from this tragedy. The flying of non-precision approaches using 'step' altitudes, steep dives followed by level-offs needs not only to be discouraged by operators, but totally banned. It is simply unsafe to attempt to operate large commercial aircraft in this way.. .When a non-precision approach must be flown, a glideslope should be mathematically planned by the crew to give a constant angle of descent, and the aircraft should be flown fully configured from the OM / equivalent position at the very latest down to MDA, whereupon either a visual landing, or an immediate G/A should be performed.. .It's very sad to see that CFIT accidents such as this still occur with frightening regularity, despite the efforts of the flight safety foundation, aircraft manufactures, regulatory bodies and the industry in general. Despite quantum leaps in technology such as GPS, EGPWS, and even the good old ILS - we still see modern aircraft colliding with the ground.( think Air Inter - Strasbourg ) I guess it all comes down to cost ? Tombstone technology is something the NTSB often refer to - the cost in real lives before the industry takes firm action.. .. .As for the exact cause of this particular accident, should we at least not wait for the final report before critizing Crossair ( Swiss ) SOP's ?

[scanscanscan]

IONA you state BAE do not teach the catch basket set to MDA method and you sound very sure of your statement.. .However could I refer you to the post of 4Dogs a few pages previous to yours... .4 Dogs states the BAE MOM Vol 4 Instructors Guide today currently does recommend this and has done since 1993.. .Can you explain how both of you can be right and hold opposite opinions?

[4dogs]

soddit,. .. .I stand corrected. The Instructors Manual I was shown is maintained by an excellent technical library and that library is now following up on the amendment status - it appears that the level of support from BAe may have had a hiccup somewhere.. .. .Given that there are many NPAs that are not landing approaches but circling only, can you just fill us in on the technique employed at the bottom - do you select ALT Hold at MDA or do you fly manually as is suggested elsewhere to ensure IRT accuracy?. .. .Without prolonging the debate, how do you get back to Vmin with all automatics hooked up and two crew?

[Capn Bloggs]

All,. .The RJ Vol 1 that I have (updated to Sep01) has, wonder wonder, the MDA set in the Alt Sel on a VOR approach: section 8.1.3 page 69 (correctly dated Jun93) refers. So IF the Instructors Manual says otherwise, obviously BAe don’t think it’s a big-enough issue to change the Vol 1.. .. .For you bush-lawyers worried about non-compliance with OEM SOPs, if I was the lawyer of a family of a victim, I reckon I could mount a pretty good case for severely criticising any SOPs that DON’T use the catchbasket approach simply because of a fear of Vmin.. .. .TetraNG,. .On what basis do you claim that the aircraft will stall if it gets to Vmin with the ALT SEL set below the aircraft? The Vol 1 indicates that the FGS will revert to LVL CHG and therefore continue descent to the ALT SEL. When it goes into ACQ, the A/T then reverts to maintaining Vmin (probably with a surprisingly large amount of grunt!), so I cannot see how the jet would stall. Surely that wouldn’t get thru the certification process? “If Vmin is reached while descending to a ALT SEL level below the current altitude the jet will stall”. I don’t think/hope so.. .. .Soddit,. .Don’t be too harsh on us colonialists. Remember that it was I who blew apart the thread’s “theory” about what happened if you pressed TOGA. Further, we have much experience doing NPAs of all kinds, GPS, NDB, VOR and consider ourselves qualified to comment (if I may be so bold, I’ve probably done as many NPAs as you’ve done ILSs). Your insinuation that “bigger/more means that you must be right” is a tad imperialistic. It’s not what or how much you’ve got, it’s how you use it.... .. .4Dogs,. .How do you get to Vmin? Firstly, you’d have to wind the Vsel back, then secondly you’d have to forget to put the flap out, then thirdly you’d have to sit there and not notice the yellow line and red squares coming up at you, or the nose 10deg in the air, ...”what was your latest sim score again, Bloggs?”.. .. .IONA,. .I’m afraid I’m going to take you to task over some of your comments:. .. .Given that the BAe Instructor Manual appears to recommend setting the Missed Approach Altitude (I suppose 4dogs’ amendment is still on the ship?), is there any other basis for your classifying the practice of setting the MDA as being “dangerous”?. .. .On what basis do you state “the ALT SEL is not capable of accurately acquiring an altitude within the tolerances required of an MDA...”? Subjectively, you don’t know your jet very well if you think it doesn’t do a great job at levelling off at ANY altitude, let alone at an MDA. My experience is that it does it perfectly, every time (better than you, I suspect). Not that that is particularly relevant in normal circumstances because the crew would be monitoring it and would prevent it from going below. It’s only when the crew would have descended/let the jet descend below the MDA inadvertently that it would help. To my knowledge, there has never been an accident caused by an autopilot that aquired an MDA and went 50ft below in the process, something I cannot say for handflown approaches.... .. .The last sentence of your post quite correctly says that we should wait for the crash report, but earlier on you virtually conduct the investigation yourself, surmising that the Capt tried to fly level at the MDA! On what grounds do you say that? Give us all a break and stick to the facts. The articles in Flight certainly don’t indicate the captain was trying to do what you’re hypothesising. We’re learning things with each post (well some of us are!) so please keep your comments civil and professional. I can assure you that I’m not here in the position of enthusiastic amateur, and have raised the issue of the ALT SEL/MDA from a background of years of experience at these approaches. And I don’t see any direct criticism of Crossair SOPs in this thread.. .. .I agree with you re the constant-descent procedure; in fact it would probably surprise soddit that some of the jet operators downunder have been using this technique for more than 25 years for the very reason that it is counter-productive to getting to the MDA in a controlled fashion by diving, levelling and diving. Out of a network of around 18 ports, only 5 have ILS, so my base does many NPAs, both Straight-In and Circling. While they certainly are more difficult than an ILS, they are not hard provided you stay current and keep your hand in. On that note, a couple of my colonial colleagues have said to me that at least one major European 146 operator does NOT use the constant descent technique, Florence being the place which regularly brought their pilots unstuck for that reason. A subject of much mirth over a beer was one guy recalling watching A330s and the like diving at the dirt at Paris when the ILS was off for maintenance because the crews were so unfamiliar with NPA/Visual/System management. Funny? Obviously not, and this raises the main point of the thread currently and why I keep harping on the MDA in the Alt Sel: IF NPAs in Europe are so infrequently done and it seems, feared (or at least flown poorly by some), then why would you not use a perfectly capable saviour? The only thing bad that’s going to happen is that the jet will level at the MDA! Big deal!. .. .However, I refute your comments about level segments at the MDA: in the context of 146/RJ, I can assure you that, at ISA+25 (yes, 40deg C) at 2000ft AMSL an RJ100/146-300 with -507s can easily hack flying level on 3 engines, let alone 4, at the MDA in the landing configuration. Just add 15% N1, my son. Now while I don’t necessarily love the idea of flying level at the MDA, you are spreading furphies by your seeming paranoia on the topic, suggesting that it should be banned. I suggest you look at your Landing WAT charts: particularly at the temperatures you Europeans operate in normally; you’d be able to do loops at the MDA (figuratively speaking!). If you know what you’re doing, it’s a “piece of *iss”.. .. .Now, back to the topic in hand, and perhaps BAe can chip in here:. .. .APART FROM THE Vmin ISSUE, WHY IS SETTING THE MISSED APPROACH FINAL ALTITUDE SO MUCH SAFER THAN SETTING THE MDA??. . . . <small>[ 24 March 2002, 14:42: Message edited by: Capn Bloggs ]</small>