Airbus A320 crashed in Southern France
Join Date: Dec 2007
Location: Brighton
Age: 43
Posts: 53
Likes: 0
Received 0 Likes
on
0 Posts
Beyond reasonable doubt
You seem to have conveniently left out a few details in presenting your case - important details. Such as the (independently confirmed) act of deliberately selecting 100ft on the MCP and commencing a descent.
What we have are a series of confirmed (to beyond reasonable doubt) events, for which there is only one plausible explanation - Which is why the conclusions already reached hold up, and have been made public. I appreciate this is slightly different to concrete proof, but air accident investigations quite frequently fall short of the 'concrete proof' standard and settle in the realms of 'no other plausible explanation' or 'most likely explanation'.
There persists a lot of unreasonable (IMHO) doubts being raised here.
Hypoxia has been ruled out (beyond reasonable doubt).
Incapacitation has been ruled out (beyond reasonable doubt).
Who was were and who took which actions have been established (beyond reasonable doubt).
...all from combinations of CVR and Mode-S ES data.
I don't believe the FDR will add much, if anything, to these conclusions.
What we have are a series of confirmed (to beyond reasonable doubt) events, for which there is only one plausible explanation - Which is why the conclusions already reached hold up, and have been made public. I appreciate this is slightly different to concrete proof, but air accident investigations quite frequently fall short of the 'concrete proof' standard and settle in the realms of 'no other plausible explanation' or 'most likely explanation'.
There persists a lot of unreasonable (IMHO) doubts being raised here.
Hypoxia has been ruled out (beyond reasonable doubt).
Incapacitation has been ruled out (beyond reasonable doubt).
Who was were and who took which actions have been established (beyond reasonable doubt).
...all from combinations of CVR and Mode-S ES data.
I don't believe the FDR will add much, if anything, to these conclusions.
Before writing my post I carefully scoured NYT and other sources to find out what hard evidence is out there. Unless I missed something, no other assertions are possible exept what the three witnesses have said.
Before you can add anything, you've got to say what your source is, just like I have done. So...
How, precisely, has hypoxia been ruled out?
How, precisely, has incapacitation been ruled out?
Precisely which actions have been established?
Who took them?
FDR may or may not add anything. We must wait and see.
Last edited by eezeegeebee; 29th Mar 2015 at 20:47. Reason: quote mark bracket
Join Date: Jul 2013
Location: entre ici et là
Posts: 65
Likes: 0
Received 0 Likes
on
0 Posts
M. Brice Robin conclusion
First wings folded makes a nice summation Here
The two cents I would like to add is that the remit of the public prosecutor is to conduct a criminal investigation concerning an aircraft accident, period. He is to determine if a criminal act has been committed that caused the accident, he is not tasked with determining any other possible causes such as may be related to spontaneous equipment / software failures or deficiencies which may have caused the accident -that falls under the provenance of the BEA. Thus M. Robin is only looking for evidence of a crime. If he discovers enough the evidence to reasonably conclude that a crime has likely been committed then he will report such; he tells he has and he did. While there may be other possible reasons for this accident such as an unrecoverable equipment or software failure either due to the nature of that failure or due to pilot error in response due to inexperience it is not M. Robin's issue to address, that is BEA's issue. To sum up M. Robin was tasked to determine if enough evidence existed to indicate that a crime had likely been committed, he believes he has done so.
The two cents I would like to add is that the remit of the public prosecutor is to conduct a criminal investigation concerning an aircraft accident, period. He is to determine if a criminal act has been committed that caused the accident, he is not tasked with determining any other possible causes such as may be related to spontaneous equipment / software failures or deficiencies which may have caused the accident -that falls under the provenance of the BEA. Thus M. Robin is only looking for evidence of a crime. If he discovers enough the evidence to reasonably conclude that a crime has likely been committed then he will report such; he tells he has and he did. While there may be other possible reasons for this accident such as an unrecoverable equipment or software failure either due to the nature of that failure or due to pilot error in response due to inexperience it is not M. Robin's issue to address, that is BEA's issue. To sum up M. Robin was tasked to determine if enough evidence existed to indicate that a crime had likely been committed, he believes he has done so.
Apologies for not mentioning those.
Join Date: Dec 2004
Location: In Hyperspace...
Posts: 395
Likes: 0
Received 0 Likes
on
0 Posts
How, precisely, has hypoxia been ruled out?
How, precisely, has incapacitation been ruled out?
Precisely which actions have been established?
Who took them?
Who took them?
You are right that little has been precisely determined - but little rarely is. Many facts have simply been determined to be the only plausible conclusions.
Many are understandably unwilling to accept this at face value. Many are also crying foul that info has been released this early in an accident investigation - a very unusual step.
But remember - this accident clearly seems not to have been an accident at all, and those conclusions have been able to be drawn pretty quickly and conclusively - which to my mind means different rules apply.
As for sources - all of this comes from what the French authorities have reported to be on the CVR, and I have no reason to doubt that.
Join Date: Jan 2001
Location: UK
Posts: 2,044
Likes: 0
Received 0 Likes
on
0 Posts
NigelOnDraft said:
Quote:
Was me. With >500 Capts and >500 FOs on the fleet, and doing, say 100 "trips" a year (trip varies from 1 to 5 days), you might see the issue
Wondering if any airline ever came up with the idea of forming small teams within their crew pools who fly regularly together. Human beings work best in groups of up to 30 people - part of our evolutionary heritage.
Not knowing your colleagues in the cockpit may be seen as advantageous or at least convenient by the airlines, but I don't think it leads to a high level of well-being among the crews. You'll probably find that your conversations with your colleagues don't go much beyond the basic introductions, as does your familiarity with each other.
Quote:
Was me. With >500 Capts and >500 FOs on the fleet, and doing, say 100 "trips" a year (trip varies from 1 to 5 days), you might see the issue
Wondering if any airline ever came up with the idea of forming small teams within their crew pools who fly regularly together. Human beings work best in groups of up to 30 people - part of our evolutionary heritage.
Not knowing your colleagues in the cockpit may be seen as advantageous or at least convenient by the airlines, but I don't think it leads to a high level of well-being among the crews. You'll probably find that your conversations with your colleagues don't go much beyond the basic introductions, as does your familiarity with each other.
Leave aside the technical merits, would prove unworkable with any sort of bidding system e.g. our Bidline
Mix in Part Time options, Union work, differing leave etc. and it would not work IMO. Some of our Cabin Crew contracts try it from time to time - not for safety but "customer service", but I think it tends to fail for the same rostering reasons?Personally, I do not see it as an issue - we are all trained to the same SOPs, and if anything, at the end of a long tour, things might "relax" a little more than the author of the SOPs intended
I refer to the 'swiss cheese effect' as its a good example of how many individual events or factors unfortunately tie up to this one tragic event. Obviously I can only go by the media reports but to me Andreas major issues were:
1)His personality to be very detailed and controlling
2) Strong love for flying throughout most of his life. Was into glider flying (what real pilots do) and was even cabin crew so flying a major part of his life.
3) Depression/ untaken medication?
4)Relationship issues
5) Big possiblity of losing Class 1 due to eye sight and his whole career.
6) Reports of not getting on or gelling with the Captain
7) Reports of being bullied.
8) Now this last one is beyond my knowledge of the human brain but ability to act and do something morally horrific.
To me tragically that day these holes all lined up and he was presented with an opportunity and took it. Maybe he was taking medication which dulled his senses and sense of morality or conversely not taken his medication and he acted erratically. Maybe the chances of all these lining up was that chance in a billion.
A lot of pilots (sorry should really say system programmers) will have point no 2, and definitely some captains I used to fly with had 1 as well ( can be a very good thing). I am sure a lot of us have had or are having 4 and definitely in a flying career will have 6. So each individual item except number 8 is something that can happen or is a perfectly normal trait or event.
Its this number 8 which we need to figure out if its in our button pushing pilot society and how to find it which is the next big issue.
1)His personality to be very detailed and controlling
2) Strong love for flying throughout most of his life. Was into glider flying (what real pilots do) and was even cabin crew so flying a major part of his life.
3) Depression/ untaken medication?
4)Relationship issues
5) Big possiblity of losing Class 1 due to eye sight and his whole career.
6) Reports of not getting on or gelling with the Captain
7) Reports of being bullied.
8) Now this last one is beyond my knowledge of the human brain but ability to act and do something morally horrific.
To me tragically that day these holes all lined up and he was presented with an opportunity and took it. Maybe he was taking medication which dulled his senses and sense of morality or conversely not taken his medication and he acted erratically. Maybe the chances of all these lining up was that chance in a billion.
A lot of pilots (sorry should really say system programmers) will have point no 2, and definitely some captains I used to fly with had 1 as well ( can be a very good thing). I am sure a lot of us have had or are having 4 and definitely in a flying career will have 6. So each individual item except number 8 is something that can happen or is a perfectly normal trait or event.
Its this number 8 which we need to figure out if its in our button pushing pilot society and how to find it which is the next big issue.
Join Date: Jun 2005
Location: London, UK
Posts: 197
Likes: 0
Received 0 Likes
on
0 Posts
Well it's a ridiculous system then as in effect the French prosecutor is passing a subjective judgment as to whether a crime has been committed not even a week after the crash, and whether a crime has or not been committed cannot be determined until the full "proper" aircraft investigation has been completed. So his subjective "gut feeling" will either be right or wrong and so many hares may have been set running unnecessarily. How can such a daft system be justified when in many cases the prosecutor will be wrong?
Only half a speed-brake
... deliberate actions required by the F/O to keep the FD door locked out after the 5 mins are up ...
Join Date: Mar 2014
Location: UK
Age: 74
Posts: 54
Likes: 0
Received 0 Likes
on
0 Posts
'Being remembered and changing the industry'
After the 3 other recent crashes he is not really going to be remembered much. Even today after the Air Canada plane crash landed last night, the news has already moved on, as it has from the gigantic search in the Indian Ocean. And we moved on from that to Ukraine. Newspapers are fickle.
I think the general public are a little saturated with air disasters and other disasters these days.
And changing the industry ? Did he want to be 'remembered' by a new key code on the door mechanism ?!!!
Seems unlikely.
This was a guy that ran half-marathons and seemed all around successful. He is not going to kill himself because he had an argument with his gf. And if his eyes really were going then LH would have found him something. At least he can still run, he was good at several things.
I can imagine the newspapers have been able to dig up far less dirt on him then they would have on many other staff. Most of his acquaintances had only good things to say about him. Ask yourselves what your acquaintances would say about you.
I see no motive.
Last edited by rideforever; 29th Mar 2015 at 21:50. Reason: missing bit
Join Date: Dec 2004
Location: In Hyperspace...
Posts: 395
Likes: 0
Received 0 Likes
on
0 Posts
I was looking whether this is an established fact or just media derivative of whad had actually been released. No luck so far, can anyone help?
To explain it any other way would require a series of largely implausible scenarios, ie:
1. Neither the captain, nor anybody else, tries to enter the code (implausible)
2. Everybody on that aircraft who is supposed to know the code has forgotten it (implausible)
3. Given 1 and 2 above, the keypad must have been overridden from inside the flight deck, using a guarded switch in a position unlikely to have been activated accidentally (only plausible explanation)
4. Override functions for 5 mins (with LH, confimed by LH CEO), then entry can be attempted again. Continued denial requires further deliberate action from the FD (not possible if incapacitated).
5. Door failure - certainly possible, but doesn't address at all the primary issue of the aircraft being commanded into a descent. Or why the FO didn't respond to multiple attempts to get his attention, from both inside & outside the ac.
The only plausible explanation for what transpired is the deliberate action of the FO - when viewed in the context of the other deliberate action taken (commanding descent).
Personally, I don't believe any rash or premature conclusions have been reached - no matter how much we would naturally wish to defend 'one of our own'.
Join Date: Jun 2005
Location: London, UK
Posts: 197
Likes: 0
Received 0 Likes
on
0 Posts
Quote:
How, precisely, has hypoxia been ruled out?
Noises of shouting, banging on the door, screaming from pax, clearly audible captain's voice from outside the FD - multiple sources show, IMHO, that Hypoxia can be ruled out BRD.
Doesn't mean the aircraft wasn't depressurised or slowly depressurising or filling with smoke - those banging on the door could have had either toilet masks or portable oxygen masks on between shouts and they could have been wearing PBE.
Quote:
How, precisely, has incapacitation been ruled out?
Deliberate selection of an ALT below MSA and deliberate actions required by the F/O to keep the FD door locked out after the 5 mins are up - rules out his incapacitation BRD.
So an Airbus has never done something that isn't in the manual? Yeah right. So it's completely impossible that an FMA glitch changed the altitude? Or that the remaining pilot wound it down for some other reason?
Incidentally who that flies the A320 knows the time periods that your airline has programmed into the door system. What's the time delay after you enter the emergency code? I bet most of you have no idea. If you enter it again before the time delay has completed, does the time start again ... ? How familiar was the Captain/crew with the way the door system works?
It's perfectly possible that the aircraft had a different code programmed to the one the crew were trying. How often is this tested in your airline?
How, precisely, has hypoxia been ruled out?
Noises of shouting, banging on the door, screaming from pax, clearly audible captain's voice from outside the FD - multiple sources show, IMHO, that Hypoxia can be ruled out BRD.
Doesn't mean the aircraft wasn't depressurised or slowly depressurising or filling with smoke - those banging on the door could have had either toilet masks or portable oxygen masks on between shouts and they could have been wearing PBE.
Quote:
How, precisely, has incapacitation been ruled out?
Deliberate selection of an ALT below MSA and deliberate actions required by the F/O to keep the FD door locked out after the 5 mins are up - rules out his incapacitation BRD.
So an Airbus has never done something that isn't in the manual? Yeah right. So it's completely impossible that an FMA glitch changed the altitude? Or that the remaining pilot wound it down for some other reason?
Incidentally who that flies the A320 knows the time periods that your airline has programmed into the door system. What's the time delay after you enter the emergency code? I bet most of you have no idea. If you enter it again before the time delay has completed, does the time start again ... ? How familiar was the Captain/crew with the way the door system works?
It's perfectly possible that the aircraft had a different code programmed to the one the crew were trying. How often is this tested in your airline?
Join Date: Mar 2015
Location: Southampton
Posts: 1
Likes: 0
Received 0 Likes
on
0 Posts
If things were getting as hectic as we've heard, I can imagine them stopping trying that and just trying to beat the door down. It *could be* that the door would have unlocked if only they'd tried the code towards the end.
We'll only know for sure when the investigation is complete. Claiming "facts" now is almost certainly premature.
Join Date: Mar 2015
Location: NYC
Posts: 1
Likes: 0
Received 0 Likes
on
0 Posts
Doesn't mean the aircraft wasn't depressurised or slowly depressurising or filling with smoke - those banging on the door could have had either toilet masks or portable oxygen masks on between shouts and they could have been wearing PBE.
Join Date: Jan 2001
Location: In the boot of my car!
Posts: 5,982
Likes: 0
Received 0 Likes
on
0 Posts
Rushed
I am sure we would all like this to be a mistake as murder of 150 innocent people by a mad pilot is the worst possible scenario for all of us
I like most started this protecting the pilots from those who said murder and insisted it was a pressurisation problem!
You say either the prosecutor is right or wrong implying it's a 50/50 thing it's more likely to be 98 % he is right 2 % he is wrong (( all the revealed evidence points overwhelmingly that way (( and I am sure non of us want that
Don't put to much on the AAiB reports about being definitive as that is not always the case
I am sure we would all like this to be a mistake as murder of 150 innocent people by a mad pilot is the worst possible scenario for all of us
I like most started this protecting the pilots from those who said murder and insisted it was a pressurisation problem!
You say either the prosecutor is right or wrong implying it's a 50/50 thing it's more likely to be 98 % he is right 2 % he is wrong (( all the revealed evidence points overwhelmingly that way (( and I am sure non of us want that
Don't put to much on the AAiB reports about being definitive as that is not always the case
Join Date: Oct 2000
Location: UK
Posts: 36
Likes: 0
Received 0 Likes
on
0 Posts
What worries me most about this whole sorry affair is the knee-jerk reaction that seems to be arising, both from the airline industry itself, and some of the folks on this thread. I speak only as one of your lowly revenue-creating customers, so of course my opinion counts for nowt, I guess. But, I'll press on ...
I know that there are probably thousands things more likely to go wrong with a flight than having a suicidal pilot finding the perfect opportunity to drive into the ground; but us pax kinda rely on the fact that aviation is highly regulated and has a risk management regime that is probably second to none as far as transport safety and security is concerned.
So knee-jerk reactions are not what is to be expected; measured and appropriate responses are. Suddenly requiring two people on the flight deck at all times as a response to an incident that hasn't even had its proper investigation completed - what sort of signal does this send out to the general public? I'll tell you: it says that the industry considers this to be a problem that is potentially so serious that it needs to be urgently addressed. And then we have people here on the forum saying that Flight Attendants should not be allowed on the flight deck. Are we to believe that FA's are not subject to the same screening and security measures as those in control of the flight? Are we saying that potentially all crew are not to be trusted with our collective safety?
Perspective is needed here: despite all of the screening and safeguards, this guy got through the net, although it remains to be seen as to what exactly happened; nevertheless this was an extremely rare incident and from a risk perspective I'd have more of an issue with there only being two motors keeping me aloft instead of a comfortable four.
I would much rather see the airline industry say: we don't need to mess about with the way that the flights are operated, it's fine as it is. And let's face it, it probably is. Probably being the operative word.
Over and out.
I know that there are probably thousands things more likely to go wrong with a flight than having a suicidal pilot finding the perfect opportunity to drive into the ground; but us pax kinda rely on the fact that aviation is highly regulated and has a risk management regime that is probably second to none as far as transport safety and security is concerned.
So knee-jerk reactions are not what is to be expected; measured and appropriate responses are. Suddenly requiring two people on the flight deck at all times as a response to an incident that hasn't even had its proper investigation completed - what sort of signal does this send out to the general public? I'll tell you: it says that the industry considers this to be a problem that is potentially so serious that it needs to be urgently addressed. And then we have people here on the forum saying that Flight Attendants should not be allowed on the flight deck. Are we to believe that FA's are not subject to the same screening and security measures as those in control of the flight? Are we saying that potentially all crew are not to be trusted with our collective safety?
Perspective is needed here: despite all of the screening and safeguards, this guy got through the net, although it remains to be seen as to what exactly happened; nevertheless this was an extremely rare incident and from a risk perspective I'd have more of an issue with there only being two motors keeping me aloft instead of a comfortable four.
I would much rather see the airline industry say: we don't need to mess about with the way that the flights are operated, it's fine as it is. And let's face it, it probably is. Probably being the operative word.
Over and out.
Join Date: Oct 2013
Location: ALandDownUnder
Posts: 115
Likes: 0
Received 0 Likes
on
0 Posts
The changes which come out of this crash must go well beyond recommendations such as always having 2 people in the cockpit. The industry as a whole MUST look at ways in which crews health/mental health can be better monitored.
Join Date: Mar 2014
Location: N. California
Age: 80
Posts: 184
Likes: 0
Received 0 Likes
on
0 Posts
If FD gets depressurised the door gets unlocked automatically.
Join Date: Dec 2014
Location: Schiphol
Posts: 475
Likes: 0
Received 0 Likes
on
0 Posts
Facts, Lessons, Damages, Punishment - Questions
One of the lessons of this and previous accident investigations has been, that it is not clear to the international public how BEA and French public "Procureur" investigations are operating side by side during aerospace accident investigations. Where goals and work run in parallel, where they overlap, how independence is assured - while working with a substantial set of common evidence.
This understanding has at least two sides. First, part of the public not taking the time to do (at least some of) the required homework. Second, the way in which French authorities make clear to especially the international audience, how they operate side by side.
It appears that both public and authorities have to improve their act. So we should appreciate the work of some pprune members to help us with our homework. And I hope more is forthcoming.
The officials on the French side appear to be doing their job, but not with respect to improving understanding about what they do in general and how that works out in this specific case. Not an easy case because the focus changed from an accident to, as it seems, an act with intent.
As a consequence there is a lot of confusion with the general public, and in for instance some pprune posts we can see posters judging the procureur investigation by accident investigation standards.
A partially separate and partially overlapping issue is leaking information. In this case there have been a number of 'serious' leaks. The strict control of CVR and FDR information, including legal and regulatory back up of this, suggests that the probability of leaks originating from the accident investigation side was low. The fact that a high French military (in some countries part of the police is military - I wonder which side it was) person or even official is said to be the source, is in line with that probability. This means that information control on the procureurs side can clearly conflict with the accident investigation side. And also conflicts with the legal requirements on the accident investigation side.
One question that I have had for many many years now is how 'serious' serious is. We do not have the transcripts or other prove available, but in spite of that there are judgements all around, and a lot of damage to the feelings of friends and relatives of the victims may have already been done. Lack of information from the official side and confusion appear to be extremely painful. I wonder if preliminary information that may have to be changed later would reduce or increase the suffering of victims and relatives, and also on the "improvement of safety" efforts on the aerospace professionals side.
So how sure are we, that the present framework is better than one with more and earlier official transparancy and openness. More transparancy by earlier and more complete official presentation and publication of information that includes how that information should be judged at the time of publication. Transparancy that reduces the pressure on all parties - investigators, procureurs, prosecutor, victims and relatives, authorities, relatives, professionals, and the general audience.
Heads of state, government ministers, high ranking police and military - we only need to look over the last two years - have made statements that were out of order, factually and technically incorrect, etc. I have great admiration for the way in which investigators have responded to this. But it should not be necessary for them to do this.
The context today is much different than it was years ago. Modern public requirements and technology are quickly eroding the foundation on which the present framework is built.
This post basically contains questions. In this case the lead is French. But the questions are of course the same or similar for other recent cases in other countries. However, we learn by accident investigations, so lets keep the scope confined to this case of GermanWings9525. I hope some people can help me with improving my 'homework' by sharing their insights and views.
This understanding has at least two sides. First, part of the public not taking the time to do (at least some of) the required homework. Second, the way in which French authorities make clear to especially the international audience, how they operate side by side.
It appears that both public and authorities have to improve their act. So we should appreciate the work of some pprune members to help us with our homework. And I hope more is forthcoming.
The officials on the French side appear to be doing their job, but not with respect to improving understanding about what they do in general and how that works out in this specific case. Not an easy case because the focus changed from an accident to, as it seems, an act with intent.
As a consequence there is a lot of confusion with the general public, and in for instance some pprune posts we can see posters judging the procureur investigation by accident investigation standards.
A partially separate and partially overlapping issue is leaking information. In this case there have been a number of 'serious' leaks. The strict control of CVR and FDR information, including legal and regulatory back up of this, suggests that the probability of leaks originating from the accident investigation side was low. The fact that a high French military (in some countries part of the police is military - I wonder which side it was) person or even official is said to be the source, is in line with that probability. This means that information control on the procureurs side can clearly conflict with the accident investigation side. And also conflicts with the legal requirements on the accident investigation side.
One question that I have had for many many years now is how 'serious' serious is. We do not have the transcripts or other prove available, but in spite of that there are judgements all around, and a lot of damage to the feelings of friends and relatives of the victims may have already been done. Lack of information from the official side and confusion appear to be extremely painful. I wonder if preliminary information that may have to be changed later would reduce or increase the suffering of victims and relatives, and also on the "improvement of safety" efforts on the aerospace professionals side.
So how sure are we, that the present framework is better than one with more and earlier official transparancy and openness. More transparancy by earlier and more complete official presentation and publication of information that includes how that information should be judged at the time of publication. Transparancy that reduces the pressure on all parties - investigators, procureurs, prosecutor, victims and relatives, authorities, relatives, professionals, and the general audience.
Heads of state, government ministers, high ranking police and military - we only need to look over the last two years - have made statements that were out of order, factually and technically incorrect, etc. I have great admiration for the way in which investigators have responded to this. But it should not be necessary for them to do this.
The context today is much different than it was years ago. Modern public requirements and technology are quickly eroding the foundation on which the present framework is built.
This post basically contains questions. In this case the lead is French. But the questions are of course the same or similar for other recent cases in other countries. However, we learn by accident investigations, so lets keep the scope confined to this case of GermanWings9525. I hope some people can help me with improving my 'homework' by sharing their insights and views.
Last edited by A0283; 30th Mar 2015 at 17:46. Reason: Insert "Procureur" instead of 'public prosecutor', because too much is lost in translation.
Join Date: Mar 2014
Location: wales
Age: 81
Posts: 316
Likes: 0
Received 0 Likes
on
0 Posts
This happens when there is a pressure differential between FD and cabin. We (or at least I) don't know what that differential is; we don't know how tight the air seal is between the two spaces. If airflow between the two spaces is not completely constricted, a slow leak in either space to the outside would not open the FD door on a pressure differential. Below about 10,000 feet, there probably wouldn't be enough of a differential to unlock the door either.
Unlocking the door, in case of cockpit decompression (the door then opens towards the cockpit under differential pressure).
http://nicmosis.as.arizona.edu:8000/...329_DSC_25.pdf