A strong possibility that the somatogravic illusion was a factor IMHO.

"erbuscap" what on earth are you talking about?
really?:E.
You obviously haven't studied enough your basic PPL stuff, ehh?
They talk there about Altitude and Heights. Go back to the books, come back and rephrase your statement a bit.

My take on the accident.

The captain had been in to this airport many times before.

Given the weather he knew he was in for a very rough ride on approach.

He thought he slip in over the river to get visual and it all went pear shaped.

For the guys who fly in the western developed countries you have to understand turning back to base will cause you big problems not least if there is a politician/high ranking policeman etc on board:=

I believe this is standard after an accident in some places

that is the current chart still i think

Looks a terrible error even if it didn't cause the accident

Jack Harper,

Yup, I agree with you - well spotted; I think I would have missed this in my briefing.

Approach Chart for PAKSE

The part in question appears to be the "NON·PRECISION RECOMMENDED ALTITUDE DESCENT TAB" as described by Jeppesen glossary-legends at:

http://ww1.jeppesen.com/documents/av...ry-legends.pdf

It is designed to be an aid for non precision approaches giving a distance vs altitude table for a steady descent approach to a runway. According to Jepp:

"General Description: The Recommended Altitude Descent table, shown to facilitate the CDFA technique *, contains "check" altitudes that correlate directly to the Vertical Descent Angle (VDA) used in conjunction with the final approach segment of the procedure."

* Continuous Descent Final Approach (CDFA): see Continuous Descent Final Approach

In this case it would place the aircraft at an altitude of 645 ft (300 AGL) at the middle marker which appears about right. Remember its non precision so would not be used in low ceiling situations.

Its amazing how much information is on these charts but you must do your homework to use them properly. I hung my wings up years ago so would appreciate any further opinion or clarification.

jfill, you are not quite right here. Let me put in my five cents as retired military ATCO and PANSOPS procedure designer (+MEPL IFR).

The whole procedure looks to me that it was constructed pre-PANSOPS with the VOR/DME as FAF at 3.8 NM from TD and the MAPt most probably at the threshold for RWY 15. Note the obstacle situation in the departure sector (terrain 3006 ft MSL). The charming aspect is that you could have flown the procedure even with DME inop or the aircraft not equipped accordingly by using a time table to determine when passing the MAPt

With the introduction of PANSOPS and its excessive requirements (just ask any PANSOPS guy to fit a CAT C holding at 5000 ft MSL to an NDB surrounded by a 10 x 15 NM controlled airspace) there was apparently no chance to keep the MAPt at passing threshold at 400 ft AGL. Using the required climb gradients for the missed approach they had to relocate it starting 1.9 NM outward from the THR RWY15 and change the FAF to DME 2.3 to allow for the required max descent gradient. The designer had to find a compromise between the required minimum visibility and minimum ceiling for the approach. Putting the MAPt closer to the TD would have resulted in a higher MDA, and putting it further out would have resulted in higher visibility requirement.

It is safe to assume that the pre-PANSOPS minima for Pakse were a lot lower, but PANSOPS insists on creating procedures based on the "worst case" scenario - sloppy flying, inaccurate on-board instruments, shoddy nav aids and hideous wind speeds (omnidirectional) not compensated by the pilot.

Your interpretation of the required altitudes based on this misleading table shows the inherent danger of sloppy charts. Breaking out of clouds at 300 ft AGL on a 500 ft/min descent rate in a non-precision approach and still being almost 2 NM from the RWY at the "middle marker" (actually it is the MAPt based on DME as there are no marker beacons for this type of approach) does not meet any safety standards.

The correct straight-in minima are:
- Ground Visibility 3200 meters
- MDA (MSL) Cat. A-C: 949 ft MSL
- HAT (AGL) Cat. A-C: 645 ft AGL
- Ceiling: Cat. A-C: 700 ft AGL
- No credit for lights (despite SALS)

There are a lot of those in South East Asia. Thailand, Malaysia, Cambodia Vietnam abound with these 'hybrid' approaches. The aviation authorities need to amend them - it's just laziness.

There isn't such a thing as a PANSOPS pilot. There are those who are more used to PANSOPS and those more used to TERPS, but pilots aren't certified on each standard. I think you will find many of us are used to both.

This approach is sloppy, but shouldn't have been too difficult on the ATR with it's flight management system. We don't know the exact reason and the circumstances as to why they went around. It wasn't the approach that killed them It was impacting the ground on go around and we don't know why that happened either.

Strong possibility? I doubt that. Hadn't the captain been flying for 30 years? Somatogravic illusion is a catch phrase used all too often to explain a so far unknown reason for a crash.

There would be hundreds of crashes every month all over the world that could be blamed on somatogravic illusion especially for aircraft catapulted from aircraft carriers where the acceleration on take off is mind-boggling on the senses.

Dear Dan,

it seems that I created a bit of misunderstanding with my last post.

The current procedure at Pakse is according to PANSOPS. At least this is what the Jeppesen chart states. It just appears to me that a VOR placed smack in the middle of the approach, being neither IAF, IF nor FAF shows that the original location was calculated pre PANSOPS.

With the "PANSOPS guy" I of course meant a certified procedure designer, not a pilot.

You call the approach sloppy. As a retired PANSOPS procedure designer I cannot understand this statement. With the adverse affects of location, terrain, available navaids and the PANSOPS restrictions it looks pretty decent to me. I have flown some rather hair raising approaches in California single-handed IFR night w/o autopilot, this one - not taking into account adverse weather - is pretty straightforward and easygoing to me.

Like any spatial disorientation, the somatogravic illusion doesn't respect experience. There are many case involving 10,000 hrs plus pilots. Unfortunately, the somatogravic illusion is hard to identify - largely because so few pilots who succumb live to tell the tale. Very few accident reports identify it - the true figure is probably far higher than the statistics reflect.

In this case, a crash soon after a GA in poor weather/at night bears all the hallmarks of the somatogravic illusion.

There aren't hundreds of crashes per month anywhere, but somatogravic illusion accidents do feature heavily in carrier launches. During one period during the Vietnam war, the USN were losing a pilot a month to it.

Aterpster - no comments on the chart error?

MOE - from where did you get the MDA you posted?

Looking at the IAP published in the Lao AIP , a mistake was actually made by Jeppesen . The table in the AIP plate shows distance to/from PAK versus Height and NOT altitude . It is an oversight because in the Vientiane (VOR/DME 13) charts , the same height info is published on the AIP ( no altitude reference , just height ) but Jeppesen converted it to altitudes on their plate . Interestingly enough , on the more recent plates designed for Luang Phabang , the AIP gives you both Altitude and height in brackets instead of height only .
Conclusion : Oversight from the designers of the original Vientiane/Pakse plates who should have given both references instead of just the height AND oversight(actually more of a wrong info ) from Jeppesen on the Pakse plate .
Probably unrelated to this accident but interesting nevertheless .
MDA for the straight in approach is 990 ft AMSL , MDH-645 ft ( Cat A,B,C) . Info published in the AIP as well .

I have to admit it - from the Jeppesen chart. But I did also check whether it made sense to me. The giveaway are the horrendous minima for the circle to land. Without a detailed map depicting terrain and other obstructions it is a "best guess" but I assume that the required obstacle clearance in the missed approach area can only be met by adjusting the MDA from a possible 400ft above TDZEL to 645ft above TDZEL and additionally by relocating the starting point of the missed approach (MAPt) to 1.9 NM from the THR.

Another point is whether a height should be added to any given altitudes in an approach procedure apart from the MDA/DA. To me adding a height above TDZEL to an altitude restriction 5 NM from the runway is irrelevant at best and dangerously misleading at worst. Depending on the underlying terrain these heights based on the TDZEL can deviate considerably from actual values. The Smolensk crash showed this clearly.

BOAC:

That is the current chart. Before I would comment on that strip I would need to see state source.

I'm out in Hawaii at present so my resources are limited.

I do know that strip is usually the result of state source.

Dan:

"somatogravic illusion" covers a bit of ground. On a Missed, the addition of power, if you remain on your primary attitude reference (call it Artificial Horizon/Attitude Gyro/Attitude Indicator, or whatever the term is in your aircraft) should not be a "sudden acceleration." Smooth addition of power and increase in pitch is how it's supposed to be done. The illusion in any case is mitigated if your instrument departures in crap weather are flown on the instruments. Pilots do this every day in bad weather.

Are you suggesting that the PF got the leans or that his scan broke down?
If so, on what basis, from the info available, other than "aircraft hit the ground?"
Agreed. Nor does gravity.
Are you suggesting that pilots are neither trained nor educated regarding this phenomenon?
The year is currently 2013, not 1968. Would you do us the favor of not mixing past and present tense? The USN no longer loses an aircraft a year to this feature of flying, no less one per month. (There are ample other causes, thanks very much. :suspect:)

Why do you think that is?

I'll tell you what I think the reason is: the phenomenon is known, and a thing called aviation physiology training works to mitigate that risk. I went through that training back in the early 80's for the first, and not the last, time when it was already well established.

Question for you: Do professional pilots in commercial aircraft not get training and education on this fundamental physiology problem?

You've all missed the big picture.

The guy was riding a bucking bronco in that weather.

If you've ever experienced those conditions you'll know what I mean.

Trying to get in to a difficult strip marginal VFR with no aids and being thrown around the sky.

Basically. Lonewolf, having been through both 'schools' like you, the answer is essentially 'no'. I recall (A USAF film, I think, from the 60's?) called 'Get on the Gauges' which was pushed hard in the British military. I suspect that unless the trainers in civilian aviation have had the same sort of strict indoctrination, it is not properly 'taught'/stressed to the same extent - but then of course, the 'manoeuvres' and potential exposure to it are far less.

In any case, like others, I suspect this has nothing to do with this crash.

YOP - what do you mean 'no aids'? Were both the NDB and VOR/DME off the air?

When you get that low below the weather all the aids let you down.

Private pilots often operate below minima and fly a combination of VFR/IFR.
That's the killer trying to mix both and all the kit in the world won't help you when it comes to basic panel flying in bad weather low level.

I've landed at lot's of airstrips in bad weather at night with nothing more than car headlights on the threshold. Severe turbulence is another issue.

Basic panel flying is a skill lacking in a lot of commercial pilots these days hence the Air France accident.

However in this case I think trying to get in on a visual approach in severe turbulence beat a very experienced pilot.

Ye Olde

Indeed. Very experienced flying with a Russian ADI and add to that no METAR.

It will be interesting to see what QNH he was working with.

God alone knows what he was doing anywhere near Pakse on a day like that.