AA 757 at Shannon, descended below MSA

phonypony

Crew was passed an incorrect QNH, 35hPa out, resulting in an altimeter over-read by about 1000ft. A very scary mistake, particularly given the visibility and cloud cover at the time.

From the AVherald http://www.avherald.com/h?article=43f6fe92&opt=0

Audio is on liveatc.net, EINN2 07/07/11 1030z, from about 09.10 to 10.40 in the MP3. Chilling.

sparrowcar

Disregard, found it

I can't find this on liveATC.net.... am I missing something?

highlydebased

This should be it;

http://archive-server.liveatc.net/einn/EINN2-Jul-07-2011-1000Z.mp3
http://archive-server.liveatc.net/einn/EINN2-Jul-07-2011-1030Z.mp3

cuterob

sparrowcar wrote: »

Disregard, found it

I can't find this on liveATC.net.... am I missing something?

go to liveatc.net .. click the atc audio archive link on the left list of links.. a calender should appear.. click 7th of july.. select the feed to einn shannon approach.. select the time to 1030-1100z and hit submit.. go to 9.10 in the mp3 and listen from there

pclancy

Wow.
I always write down the QNH of both airfields that i'm flying at on a pad during my preflight as well as some other basic notes about procedures and weather etc-doesnt everyone? Its easily available online and I would have thought on hand at major airlines. Does the pressure ever even get that high in Ireland? Looking out the window of my house right now I could average a guess that the pressure is low because of the rain clouds and wind, I don't get how the pilots wouldnt have questionned such a high setting compared to the last one they were using. A scary mistake by the ATC but I'm confused as to how the pilots just accepted it.

Question: Would they have got a radar altimeter callout of "1000" as they passed through that altitude regardless of the QNH setting? Scary stuff. Especially looking at some of the ground heights on that Shannon chart in the link.

hano

You would think Simon Hradecky would have got the flight number correct for his article, given the nature of the story.

highlydebased

hano wrote: »

You would think Simon Hradecky would have got the flight number correct for his article, given the nature of the story.

9636 as opposed to 9363 (what he reported)- easy error to make.

From a brief listen to the clip it seems atc initially reported 1030hpa, pilot mistook for 1003, corrected it to 1030, then it was properly corrected to its correct setting, 995hpa. Could have been a costly error.

Jammyc

Had the ATCO in question just started his shift or something? I wonder where they got 1030 from.

phonypony

pclancy wrote: »

Wow.
A scary mistake by the ATC but I'm confused as to how the pilots just accepted it.

Absolutely. There seems to be a tendency to take the ATCO instruction at face value and dial it in without thinking too much. On a VFR flight below the transition altitude you would have copped it straight away as you watched the altimeter rise and rise while twisting the knob... but changing from 1013hPa to another setting descending into another country obviously didn't make them bat an eyelid in this case. We're all only human and make mistakes, but I don't know what worries me more, the 2 crew not questioning it or the time it took for the ATCO to figure out the reason for the discrepancy and his insistence that it was 1030, even when they suggested the lower 1003

phonypony

Jammyc wrote: »

Had the ATCO in question just started his shift or something? I wonder where they got 1030 from.

The only thing i can think of, is the first time they were passed the QNH, the time was almost exactly 10.30 zulu. I have no knowledge of an approach controller's radar display, does anyone know are the two figures (time/QNH) close together on the screen, could he have confused them?

pclancy

Just listened to it properly now. In hindsight I don't think the pilots accepted it without question and it sounded like they would have never been let descend below a safe altitude. Otherwise EI-GCE being in the circuit at the same time might have got more training then they imagined

LLU

sorry to butt into the thread, just curious about something: from what I can gather, the gist of the problem here is that the planes barometric altimeter became incorrectly calibrated meaning that the pilots were working with an incorrect height reading. Worst case scenario: they could collide with the ground? My question is whether there is some other system which would detect them flying too low, such as radar on the ground or aboard, gps, etc? Thanks!

bombs away

highlydebased wrote: »

Could have been a costly error.

Not really considering modern aircraft of this type are fitted with gpws and radio altimeters which dont operate off barometric pressure settings

loldog

LLU wrote: »

..from what I can gather, the gist of the problem here is that the planes barometric altimeter became incorrectly calibrated meaning that the pilots were working with an incorrect height reading.

The gist of the problem is that Shannon gave them the wrong pressure reading. Why that happened is the problem, and why perhaps the crew didn't cross-check it with the ATIs.

.

loldog

bombs away wrote: »

Not really considering modern aircraft of this type are fitted with gpws and radio altimeters which dont operate off barometric pressure settings

Listening to the ATC, it seems the crew were unaware that they were too low. A gpws wouldn't be much help either if you're flying into a steep hillside.
.

phonypony

bombs away wrote: »

Not really considering modern aircraft of this type are fitted with gpws and radio altimeters which dont operate off barometric pressure settings

There have been many incidents of 'controlled flight into terrain' by aircraft, a handful of which had a GPWS installed. They may warn you of the danger, but it's up to the crew to interpret the danger and decide the appropriate action. It reminds me of the video on youtube of the Cessna 172RG making a gear-up landing, with the gear up warning blaring all through the approach...

highlydebased

bombs away wrote: »

Not really considering modern aircraft of this type are fitted with gpws and radio altimeters which dont operate off barometric pressure settings

Still does not make it acceptable to commit such an error though

radar0976

I presume there will be a preliminary report out on this over the next few weeks. Will certainly make for very interesting reading.

tippilot

Human error. It happens.(unfortunately) I feel for the guy. Nothing came of it thankfully. If it highlights a potential information conflict the way data is presented on ATC screens then at least something good comes out of it.

As much a damning indictment of American Airlines QNH procedures as anything else.

Wish the guy well.

lintdrummer

phonypony wrote: »

It reminds me of the video on youtube of the Cessna 172RG making a gear-up landing, with the gear up warning blaring all through the approach...

:eek: I had to find that and post a link:

https://www.youtube.com/watch?v=-K4QHpVXtxI

castie

emmettogara wrote: »

:eek: I had to find that and post a link:

https://www.youtube.com/watch?v=-K4QHpVXtxI

That looked quite on purpose.

Manic Moran

On a VFR flight below the transition altitude you would have copped it straight away as you watched the altimeter rise and rise while twisting the knob...

Not necessarily. On a jaunt I was on a few weeks ago, the pressure difference between KMEN and KTVL (13nm distant) was about a full inch of mercury (some 1,000ft altitude difference). 1030hpa seems to convert to 30.40", which is high, but not so high that it's unheard of.

NTM

xflyer

The thing that struck me about it was the fact that neither party thought it unusual. The controller most of all. In fact the day had exceptionally low pressure, 995 is quite low in this country. But the QNH passed was exceptionally high, 1030 would be quite rare. That in itself would be a reason to set off alarm bells. Perhaps he had a bit of blond moment.

On the other hand the crew really should have realised something was wrong. Any weather information would have told them they were flying into a low. How did they miss that? Maybe being American they are more used to inches of mercury?

Lucky the high ground around there isn't that high.

globemaster1986

xflyer wrote: »

The thing that struck me about it was the fact that neither party thought it unusual. The controller most of all. In fact the day had exceptionally low pressure, 995 is quite low in this country. But the QNH passed was exceptionally high, 1030 would be quite rare. That in itself would be a reason to set off alarm bells. Perhaps he had a bit of blond moment.

On the other hand the crew really should have realised something was wrong. Any weather information would have told them they were flying into a low. How did they miss that? Maybe being American they are more used to inches of mercury?

Lucky the high ground around there isn't that high.

Ya i think its fair to say the differences between using hPa and hg was a factor. I am more accustomed to using hg than hPa and if i am honest that makes more sense to me. I think tha JAA system is unnecessarily complicated with QFF, QNH, QNE etc, but thats just me:D

Mikael Pupina

The reason for the incorrect QNH value being issued during this incident is really quite simple. The METAR is displayed on the screen giving the current half-hourly observation. The time of the observation is displayed, in this case 1030UTC. The ATCO confused the time window with the QNH value window and issued the time that the Met observation was recorded. The repetitive transmission to AAL is indicitive of classic human error. The ATCO simply confused the display windows. How many times has this happened in ATC Centres and on Flight Decks in the Past?? Perhaps a HMI design re-think would help here.

A serious error indeed, but a simple error! Free Lessons learned and thankfully no one was hurt. Let's hope all other ATCO's have been made aware of this.

xflyer

Interesting explanation but both my points remain valid, plus note the layout of the Metar. This is the offending item:

EINN 071030Z 19011KT 9999 -SHRA FEW008 FEW018CB SCT020 BKN046 13/12 Q0995 TEMPO 4000 BKN008

The time is at the start, the QNH at the end. Hard to confuse to two in my opinion based on the fact that I read these on a daily basis, much as the controller does.

However and this interesting, when I pasted the actual into the box. This was the layout seen:

EINN 071030Z 19011KT 9999 -SHRA FEW008 FEW018CB SCT020 BKN046 13/12
Q0995 TEMPO 4000 BKN008

I wonder is the display in ATC goes to two lines as well? Definite scope for confusion.

I suppose it's just as well it wasn't the 1059 Metar.

rattrap

xflyer wrote: »

Interesting explanation but both my points remain valid, plus note the layout of the Metar. This is the offending item:

EINN 071030Z 19011KT 9999 -SHRA FEW008 FEW018CB SCT020 BKN046 13/12 Q0995 TEMPO 4000 BKN008

The time is at the start, the QNH at the end. Hard to confuse to two in my opinion based on the fact that I read these on a daily basis, much as the controller does.

However and this interesting, when I pasted the actual into the box. This was the layout seen:

EINN 071030Z 19011KT 9999 -SHRA FEW008 FEW018CB SCT020 BKN046 13/12
Q0995 TEMPO 4000 BKN008

I wonder is the display in ATC goes to two lines as well? Definite scope for confusion.

I suppose it's just as well it wasn't the 1059 Metar.

The metar is laid out very differently-all the information is laid out in boxes, but there's also an airfield data box that is used more often. This has the runway in use, time it was released, what ATIS information is current, the time and the QNH. I would say this is where the ATCO read it from, as they are in close proximity to each other in this box. The time can be turned off in this box-a notice has been released since which says this SHALL be standard to have the time turned off. The recording sounds pretty bad, especially because it was queried by the pilot. But, let he who has not sinned and all that.

the beerhunter

rattrap wrote: »

The metar is laid out very differently-all the information is laid out in boxes, but there's also an airfield data box that is used more often.

i'm not trying to throw stones here, but that airfield data box shouldn't be used in an approach position. it's only updated manually, when the atis is being updated. the metrep window is the only one which is current. i could see how the controller might have been just sitting in and given the info from the airfield data window, but it's unfortunate that he didn't check the other one when he got settled.

rattrap

the beerhunter wrote: »

rattrap wrote: »

The metar is laid out very differently-all the information is laid out in boxes, but there's also an airfield data box that is used more often.

i'm not trying to throw stones here, but that airfield data box shouldn't be used in an approach position. it's only updated manually, when the atis is being updated. the metrep window is the only one which is current. i could see how the controller might have been just sitting in and given the info from the airfield data window, but it's unfortunate that he didn't check the other one when he got settled.

Yeh, I dont use use it myself. It only takes half a second to throw your eyes over to the METREP, so that's where I take the QNH from. Just an explanation though-wasn't saying what a controller should and shouldn't do!

Bearcat

in airline world, a lot of outfits as SOP, on receiving the initial altimeter setting from ATC cross check it off the latest atis as a gross error check. Not with this crew/airline it appears................

Mikael Pupina

Yes I agree, double checking the current pressure setting is always good practice and so is having a conscious awareness of the general principles of Meteorology. The dogs on the street know that such high pressure (1030) is associated with really good weather here and that the weather here has been absolutely terrible with low pressure systems prevailing for weeks now.

I think that at a little awareness of the operating environmental envelope by any controller would not go astray and that it is incumbent on all ATCOs to possess a degree of knowledge about what is happening outside the window, especially those working in the Tower or Approach environment. Too often ATCOs display their lack of simple met knowledge. That said, this was human error and I'm not being critical in this case or "throwing stones" but I think the first rule should be is that ATCOs should make themselves aware of the operating environment. Would a pilot take off on a flight without checking the en-route weather? I don't think so. The environment is part of an ATCO's tool kit and as such he/she should be aware of it's performance /impact at any given time. The way in which the METAR is displayed IS a factor in this case but maybe alarm bells should have rang in the ATCo's head.

Bearcat

Mikael Pupina wrote: »

Yes I agree, double checking the current pressure setting is always good practice
.

I agree with the rest of your post but to use the term good practise I would read is just nice stuff to do, star on your report stuff etc.....not so.....the capt may end up in hot water over this and it is the commanders responsibility to ensure the correct altimeter setting was set irrespective of the atco's wrong transmission....hence i repeat most operators of airlines when initially given the altimeter setting, there is a cross check with the crew that the setting co -indes with the latest atis or latest metar recieved re gross error....this is what happened here it appears.

we may beg to differ but if lessons can be learned so be it. Do I recall when checkin into LHR on initial contact you must give type, atis, and qnh setting??

View Profile

Do I recall when checkin into LHR on initial contact you must give type, atis, and qnh setting??

That's standard practice in most airports when establishing contact with approach.

Bearcat

View Profile wrote: »

That's standard practice in most airports when establishing contact with approach.

verbalizing the QNH when checking in on initial contact is not mantadory at any european airports except LHR I am aware.......Likewise in the USA ....nope. Its certainly not bad practise and will error trap on both sides of the fence. I'd be all on for it.

tippilot

Bearcat wrote: »

View Profile wrote: »

That's standard practice in most airports when establishing contact with approach.

verbalizing the QNH when checking in on initial contact is not mantadory at any european airports except LHR I am aware.......Likewise in the USA ....nope. Its certainly not bad practise and will error trap on both sides of the fence. I'd be all on for it.

Yes it is. Verbalising both the QNH and the copied ATIS is standard practice. At UK airports they also require aircraft type.

phonypony

tippilot wrote: »

Yes it is. Verbalising both the QNH and the copied ATIS is standard practice. At UK airports they also require aircraft type.

Yup and yet it still doesn't always work as a x-check. Even today I heard a controller correct one aircraft who contacted him with information Sierra, 'ATIS is now Tango'. The next aircraft to check in also stated he had 'Sierra' and the the next 'Romeo'. Neither was corrected by the same controller....

Bearcat

tippilot wrote: »

Bearcat wrote: »

View Profile wrote: »

That's standard practice in most airports when establishing contact with approach.

verbalizing the QNH when checking in on initial contact is not mantadory at any european airports except LHR I am aware.......Likewise in the USA ....nope. Its certainly not bad practise and will error trap on both sides of the fence. I'd be all on for it.

Yes it is. Verbalising both the QNH and the copied ATIS is standard practice. At UK airports they also require aircraft type.

No it's not"...........I said European airports, Uk accepted. Think it's in their AIP if anyone can dig it out. Go to LIRF/KJFK etc not mandatory.

tippilot

I've no intention to get into a pissing match with you. It is standard practice. I do it daily. I hear it daily. If you fail to report it, you will be advised of both and must acknowledge. Legal requirement. You will find it under JAR Ops(somewhere).

May I also add that at first you stated LHR only and now have extended that to the entire UK. Also KJFK doesn't fall under the European proviso you earlier indicated.

As for the Fuimicino example: Italian ATC?? Bad example.

"Co-ordination in progress."

I'm also willing to bet that View Profile is in the business too. Seems to know what he's talking about. Easy to pick them out.

Oscar Bravo

Not really on topic but just a quick question. i set off from Knock VFR to Abbeyshrule on a QNH of 999,call into Shannon (127.5) at boundary outbound and they say Good afternnoon EI-??? Shannon QNH 1001 report 5 miles from Abbeyshrule.What QNH should i use?

tippilot

Oscar Bravo wrote: »

Not really on topic but just a quick question. i set off from Knock VFR to Abbeyshrule on a QNH of 999,call into Shannon (127.5) at boundary outbound and they say Good afternnoon EI-??? Shannon QNH 1001 report 5 miles from Abbeyshrule.What QNH should i use?

The Shannon QNH, ie the regional setting. That way all aircraft on frequency will be using the same pressure setting with obvious benefits regarding separation.