7 last miles ... according to flight data recorders, T2CAS several times warned the crew about the dangers. Yablontsev personally tested this system in the North Caucasus and preparing Flight Operation Manual for SSJ-100, he wrote that reaction for the alarm should be "immediately begin to climb." However, in Indonesia for some reason the pilot did not believe the system: he was convinced that they flies over a flat terrain toward the airport and even ignored the recommendations of its own, when descenting to 800 feet.

And they never saw the sides of the canyon which were maximum 3-500m away?. It would mean they accidentally and unknowingly always flew pretty much in the middle of that canyon for almost 5 miles without ever getting close enough to one side to see it and ask themselves wtf???
Plus a shouting TCAS where there should be flat terrain underneath when returning to Jakarta ?
Sorry, This sounds a bit strange for my taste.
I rather prefer to wait for the official report hoping it will be published and be objective without cover up.

Henra - some of the earlier posts and pictures show how easy it is to have mist & cloud around the crash site - he could probably see the mountain out of one window and thought he was in over flat land to the north - couldn't see the other ridge coming up fast on the other side..........

Why not to put it simple?

IFR?
Check your plates for MSA.

VFR?
Have a look out, stay clear of clouds/traffic and obstacles, but don't blame any other person e.g. ATC.:=

But there shouldn't have been any mountain to any side of the aircraft if they had just made the mentioned reciprocating error.
They would have been North of the mountain range all the time. In that scenario any green solid mass of terra firma reaching up to their flight level, be it to the left or to the right, should have initiated instant concern - and action.
As they were released to FL60 and hit the mountain at ~FL62 I see no indication for a vigorous attempt to climb until maybe the last 5 or 10 seconds.

Can we review for a moment? Looking at PJ's post #288 the crash appeared to be south to north. How does that fit with a 'reciprocal error' heading south?

If I remember correctly that was one of two possible scenarios proposed by him based on the originally reported coordinates which were not quite exact.
The other was North to South. The latter one seems to be the more likely and generally accepted one. That second one is also the basis for the speculation about this reciprocal error.

If this map is correct:
Файл:SSJ100 crash on Salak ru.png — Википеди�
(red point = crash site, 2211 - Mt Salak)
it is unlikely that plane flew from South to Nord. The crash site lies behind the peak in this direction. The crash site lies in the end of canyon which heading is ca 200, opposite to the direction to the airport.

henra:

There was discussion much earlier that the TCAS(TAWS) may have had a regional terrain database, which did not include the area in question.

Agreed! As single Issue I would agree it may not have been very reliable and there could be an explanation why to diregard it.
It is just the combination with other Issues that makes it -in my Eyes- not the most likely overall scenario.
Looking back at most accidents Occam's Razor has an excellent hit rate. Maybe this is one óf the exceptions.
But I only tend to believe it if there is substantial evidence pointing in the direction of a scenario that needs (too) many assumptions. So far I don't.

What scenario prefers Occam's Razor in this case?

CFIT - They lost situational awareness and flew into a mountain they weren't expecting to be there.

Yes, of course. Obviously they didn't expect the mountain to be there. But how did they come to lose situational awareness?

Henra, you are quite correct - my original scenario, south-to-north, is incorrect.

The flight path was north to south, striking the other side of the ridge where I thought the original site was - everything fits. This knowledge and mapping actually came from a German site and it is worth re-reading this thread to see their work, which I think is correct.

Edit - here's as good a map and theory as any, posted by mcgyvr81

Thanks for that link, PJ. We sum this up, as I and many others said earlier as
'LOST'. Why? I sure reams of PDFs can be written on the psychology of this, but I still think my post#591 says it all and we should close the folder and go 'WHAT?!!!?

BOAC;
Yes, fully agree.

I haven't read it yet, (haven't covered all recent posts here) if the TAWS database included this area. Anyone?

I really hate to say this but I honestly believe it is something along the lines: Sightseeing gone wrong.
Lost situational awareness and maybe thought they were in a different valley. One they saw before where you could fly through. Or thinking they were on the outside of the mountain range and flying alongside it (while the outer ridge being obscured).

This would explain why they did not initiate a climb once entering the canyon, without requiring total obscuration of both ridgelines while accidentally following exactly the direction of the canyon without seeing it.

Happy to stand corrected by the official report, though

The Indonesian NTSC has just released the Preliminary Report in to the investigation of this crash.

http://www.dephub.go.id/knkt/ntsc_av...4_Released.pdf

Re the report, ATC never delegates Responsibility for Terrain Clearance to the pilot.

The keyword "visual" is found wanting when the pilot & atc agree on 6,000'.

Also ATC doesn't give a 'clearance' to 6,000'. He says '6,000 copied'. Doesn't sound like a clearance to me. Have to say they approved the right orbit after that.

you wonder if the pilot misread the MSA as 6000 ft??

had it been 7000ft instead of 6900ft he might have noticed it ..........

A few confirmed facts but still a few questions :

Fact : Flight rules/PLN was IFR. Req alt was 10.000
MSA was 6.9 until 25 NM. Radius HLM impactwas 28 NM .
Question 1 : what was the MSA at that point ?
fact :a/c was not under radar vector : reponsibility terrain clearly PIC.
Fact : discrepancy in the report on the ATC R/T phraseology used for the a/c to descend to 6000 ( below MSA)
Question2 :was the PLN vertical request made to descend below MSA followed by an ATC instruction ( XX descent to 6000 ) or was the request made in such a way that it was understood to be going down visually ? then the " 6000 copy" mentioned at the begining of the report could be understood.

It would be nice to see the R/T transcript.

I suspect that you're asking a great deal for perfect dialogue from Jakarta Radar Control

What I'd like to see is the transcript for the 07:28 UTC conversation when the pilot asked for approval to make a right turn "the Controller approved the flight to make an orbit to the right at 6,000ft"

If he did say 6000ft then the crew probably thought (wrongly) they were OK

yes but I fly as well, and when in IFR ( or VFR in controlled airsace ) if I request a descend and I hear back : " copied " I am never going to start decending , and would ask again if we are cleared to descend. Anywhere on the globe. Therefore I think the full exact R/T exchange would be helping us understand why ( and if) they did take this " copy" as an ATC instruction .

Normally not the kind of assumption a 57 years old test pilot with over 10.000 h would do. In this situation (outside radar vectoring) The ATC clearance / responsibilty is only with regard to separation with other traffic, not with terrain.
A guy like this would know that for sure.

The clue(s) are somewhere else I would say.

if you mean orbit=2x Radius then here it is:

(Turn Radius) NM = TAS↑2 / 68652 / tg φ

( φ = Bank Angle)

Couple of points regarding the report...

Below is a summary of the main crash site coordinates, with sources, (copy/paste into your browser examine original sources).

Interestingly, the Indonesian NTSC has two different coordinates in their "Immediate Recommendations" document and their "Preliminary Report" document.

The estimate from PPRuNe member sledge1984 coincides very closely with Sergei Dolya's estimated position and the NTSC Preliminary Report position.

The estimate from mcgyvr61 which had the aircraft flying up the valley and striking the almost-vertical face is, to me, the most viable crash site estimate. I believe this because the terrain on the east side of the valley, is not an "85deg" vertical face as seen in crash-site photographs, and Google Earth images which seem to fit the site photographs.

The discussion on TAWS by A Van (with some contributions by others) is a must-read if only to further understand the workings and limitations of TAWS/EGPWS and T2CAS.

There are a number of theories which have been discussed in the thread concerning the TAWS - that it was intentionally ignored/turned off, that it may not have had database information for the area, or, according to A Van, may possibly not have had sufficient "granularity" to sense the mountain peak, (you have to read the post and take a look at the links to understand this interesting and important point).

The preliminary report makes no mention of the TAWS so we will have to wait for the final report.

First, a contour map supplied earlier in the thread:

http://www.smugmug.com/photos/i-kz6N...-kz6NNwg-L.png

Here's an image summarizing a few of the crash site estimated coordinates, with sources:

NTSC Immediate Recommendations document crash site coords

NTSC Preliminary Report crash site coords

Dolya/Map crash site coords

PPRuNe mcgyvr61 probable crash site coords

PPRuNe Sledge1984 est crash site coords

http://www.smugmug.com/photos/i-pHpb...-pHpbSFx-L.jpg

Unfortunately you are using the map (which is pretty small scale for this sort of thing - looks like 1:100,000) as evidence for the detailed local terrain

You'll find that even the Ordnance Survey get it wrong in areas of steep terrain - it's very difficult to render accurately - if you ever read Wainwrights Lakeland Guides he points this out all the time - especially on places such as Scafell Pike or Great Gable

You'd probably have to map at 1:10000 or even 1:5000 to accurately show the terrain on Salak - note the map shows no "sharp ridges" whereas all the photographs of the mountain ( not just at the crash site) show steep ridges all over the place- essentially they are below the resolution of the mapping

which is another reason to stay clear of rapidly rising terrain :(:(

More detailed terrain on OpenStreetMap:
OpenStreetMap

thanks - looks a horrible place to be flying around

Not necessarily related but following in Aviation News today:

I read that Armavia has financial problems and economy issues are true reason for effort to return the a/c. As far as I know, there is the only one safety issue at first several SSJ - some sensor in climatisation system - if not absolute accurate mounting an cabling it gives false alarm of dehermetisation of cabine.

There are several non technical problems for foreign air companies - for example, when flying from Armenia to Russia for technical check they reportedly wait 2 days for Russian customs formalities. Russia also have not a scheme of financing the supply of a/c to a foreign companies by providing government guarantees for lower rates for loans.

It could be only teething.

This is slightly off topic, but relevant. Sky Aviation of Indonesia who had the highest number of employees on board (the majority as flight attendants) has just acquired their first jet aircraft which is a B733. I wonder what the status of their Sukhoi orders are now? Will they cancel all orders and now pursue with the 737? As much as the Sukhoi aircraft itself was most probably not to be blamed for the crash, it will cause a stigma for Sky flight crew and potential passengers as well. Indonesians are quite superstitious.

For 'quite' read 'very'. ;)

When I worked in Jakarta, one of the lifts in my office building was thought to be haunted and the local staff would not use it. As a consequence, it was the fastest and most convenient lift in the building.

if you used it I can see why the the locals gave it a wide berth.............. ;);)

Russian delegation will fly to Indonesia next Wednesday to discuss the final report. Group will be led by Deputy Minister of Industry and Trade and the visit will spend several days. Final report is supposed about half of October.

Was just on here to look for an update on this.

Thanks for the info Karel.

Russian group of specialists finished negotiations in Jakarta and together with Indonesian side signed the protocol that fixed "agreed positions about results of investigations and determines the content of the future conclusions".

Indonesian side works on the report and will negotiate it with many instances, incl. NTSB (USA) and BEA (France).

Final report is not supposed before end of October.

Given that international relations between Indonesia and Russia were fraught for many years I doubt they'll be too accomodating to any "pressure"

on the other hand they absolutely HATE a fuss so there may be some changes made

Pressure by whom, on whom ?

Why would you think that the parties involved aren't going to agree on probable cause on this occasion ?

I believe that Russian point of view could be useful for Indonesian investigators. They know very well the aircraft, they very carefully analysed FDR and CVR records, they understand Russian mentality, they know differences in air traffic customs and rules in Russia. Let as remember that captain was not much experienced in foreign flights.

By my opinion Russians would like to read clearly in final report that there is no reason to suppose that there was any problems with SSJ. They understand that it was caused by human factor, mostly by pilots, maybe they suggest that captain was a little confused by non standard reactions by ATC. There is no conflict with known facts.

I don't think that "pressure" is the best word for discussions with Russian or with NTSB and BEA.

"Pressure by whom, on whom ?"

Dave, Dave

EVERY time there is an accident report released this forum explodes with people alleging "pressure" by one side or the other or by the Masons or the Templars or little green men from New Mexico

I just thought I'd get in first ;)