Posts about: "Pass Behind" [Posts: 111 Page: 5 of 6]¶

Following "5*why", keep asking...
Why did the did miss part of the message
Why was the incomplete read back missed
Why does missing a few words lead to this disaster within seconds.

Does this lead an answer like: helicopter in a very busy airspace, busy controllers (insufficient time to be fully focused on full read backs) being by default too close to (and even needing to cross) the glideslope.
​​​​​​

N123, join the downwind, your traffic is a 737 on final 2 miles out, turn base behind him, you are #2. I can do that at night unless there are other 737s lined up and then I have to figure out which one.
N123, do you see the closest plane lined up, pass right below and behind him and never mind all the other planes right behind. Ah......NO.
There is night visual and there is night nutty visual. The first example leaves a lot of room for error and time for ATC to see if it is going wrong.

No, it does not matter much as he was performing a visual separation ,and remember, he was instructed to pass behind so it may deviate from track. ( although they might have missed this instruction according the CVR )

So this just cracks me up. He's in the middle of the river where the route says it's up the East bank, and that's OK because the routes are not defined with no procedural separation from landing traffic. He's instructed to pass behind the CRJ, but that would involve him either holding short or deviating over the city at 200ft at night, but instead he chooses to plow right on. The helicopter is out of his standard altitude, and the jet is way above the glideslope, and ATC encourages them to sort it out themselves. And the helicopter crew are wearing NVGs. What could possibly go wrong.

Not quite funny once the body bags fill up.
Your litany of how the holes in the cheese lined up might be missing a detail or two, but any of those holes not lining up might have avoided this tragedy.
I think I agree with you.
The rad alt is right there.
At night over water at low level, the pilots I flew with did not ignore their rad alt.
It was a part of one's scan.
If I know that field elevation is 14', and my rad alt isn't at 200' or less on a route where max altitude is 200', a correction is needed now, before the error gets larger. (The separate issue of going behind, and that tower guidance apparently being stepped on, is another pair of holes in the cheese).
I am at a loss to understand the apparent magnitude of the altitude error (they were still too close laterally, yes), but as I've been out of the cockpit for a few years I am not aware of what's being taught these days.
On most airlines, they do have a barf bag, still, in the seat pocket in front of you. Suggest you vomit into that and avoid the choke hazard.

No, at this point foaming at the mouth might be a better descriptive.
While the intent was for them to "pass behind" (and thus not be on/under final when the RWY 33 traffic is on final) there was a missed opportunity to get the spacing (in time) needed.

When the NTSB has more to share it will be interesting (to me) to see what a balanced mitigation strategy is.

The diagram shows the route up the east ~half of the river, solidly over water. The inset seems to show it within perhaps the east quarter, but still 100% over water. Asking crews to follow the riverbank therefore seems to be a direct contradiction.

You could reasonably define the bank as the water's edge, and therefore expect crews to fly along an infinitesimally narrow path. Or as the space between the water's edge and the [edge of the flood plain | first flat area | something else], which would imply that the western boundary changes with the water level. Both imply the route is substantially above land.

Neither are useful for precise navigation, but the map and the description are probably 'close enough' if they are only needed for general route guidance and knowing that structures on the east bank need to be NOTAMed for helicopters, but probably not the west bank.

A good reminder that measurements/specifications without tolerance are often worse than useless.

If it quacks like a duck... this kind of "It can't be an X because we can't do it, so we'll call it a Y" leads to a culture that gets used to massaging the truth for convenience. Did we hear more on the Alaska door plug that was an 'opening' not a 'removal'?

If wake turbulence is to pose a problem then "go behind" is the worst place to be. OTOH, passing at right angles to the downwash means a very short exposure time and, with the wing-tip vortex one is experiencing up-and down-wash at nearly the same time at each vortex, unlike being parallel which generates roll in one direction for aircraft that are too small to intercept both at the same time. Entering one sideways sees an up-wash and then the immediate down-wash along that filament of flow with a smaller downward velocity until fully into the vortex sheet between the two vorticies; then the same on the other side when exiting. I would expect it to feel like dropping off a curb on one side and then popping up a curb on the other side with a small decrease in altitude.

I would also expect helicopter pilots would avoid doing it at all. I would not recommend testing.

From what I understand the delegation of separation used in the US is based on the "see and avoid ", It is basically delegating the positive control (ATC separation instructions ) from the controller to the pilot , who has to acquire the traffic visually and maintain visual contact and maneuvers to avoid it . = traffic info from ATC + See and avoid. . I do not think the ICAO forefathers designing the airspace classifications had this in mind when defining class B.
Listening to the NTSB , the only ATC instruction given : to " pass behind " was not received , and therefore not acknowledged by the crew , so we are here 100% in the good old "see and avoid" scenario I would say

IMO the fact that the helo crew didn't readback/acknowledge "pass behind" is irrelevant. Twice before, they told the LC that they had the CRJ in sight and requested visual sep, which was granted. Surely you don't specifically need to be told to not hit the CRJ after you've reported it in sight?

I think the LC saw what was unfolding and said that because he had concerns that the chopper didn't actually have the CRJ. He was right.

LC says, “PAT25, do you have the CRJ in sight?”

PAT25 says (for the second time), “Traffic visual, request visual separation”,
which is not true, evidently they were not visual at all because they crashed into the CRJ.

LC responds, “Visual separation approved, [pass behind the] CRJ,”
which clearance was issued on the false premise that PAT25 was visual. But PAT 25 only hears, “Visual separation approved…[pause]…CRJ,” because they briefly stepped on the LC’s transmission themselves. No read back is forthcoming; the LC should be expecting one but he fails to chase it up, because he’s too busy.

If PAT 25 realised he’s stepped on the middle part of the LC’s transmission to himself (which likely could have included an instruction) then he ought to be professional enough to say, “Stepped on, say again,” etc and make absolutely sure he knows what the clearance may have been. That was standard practice a long time ago, but no longer, it seems.

My original point was that in the Mover/Gonky video (post #1228) it was suggested that it’s perfectly OK to second guess what ATC might have said to you, reply to that, and then if no correction is forthcoming you can comply with your guess. As others have pointed out implicitly, that works if there’s only one error involved, but here there were three: an untrue statement, leading to a wrongly issued clearance, and a missing read back.

System broken.

Good points, especially the first, Keep in mind that no readback was required, and that type of instruction doesn't require a readback and according to the AIM is not even a type that "should" be read back, so the controller wouldn't be surprised if he didn't get one. Agree with Capn Bloggs;even if PAT25 didn't get the particular "pass behind" transmission, visual separation, which came up twice, carries the implicit instruction "Don't hit the other aircraft" ( no sarcasm intended), regardless of the method employed.

In this scenario, there is zero information in the call to "pass behind" that indicates any imminent danger. All it does is further legitimise the previous two approvals for visual separation.
Now if the controller had said "you look to be tracking very close to the CRJ are you sure you can pass behind?" or similar, then maybe the helo crew would have got excited. A call like that might have even triggered a "holy sh1t" moment about the TCAS "Traffic". But as far as they were concerned, they knew they had the traffic in sight and could do the visual sep thing and even if they had heard "pass behind" they would have said/thought "well, obviously". Except they had the wrong aircraft. ATC had an idea they had the wrong aircraft but didn't get the message across.

As for the reference, same thing. The helo crew could have read-back "pass behind" but it wouldn't have achieved anything.

If they would get the "pass behind", they would have waited for some illuminated object to pass from left to right before crossing that runway extension line. But they happily entered the final approach 33 sector as if they didnt expect any landing traffic at all on 33. In fact they decided to cross BEFORE, since "their traffic" seemed to be still very early in the circling procedure and in visual contact 3-4 miles on the nose.

Yes, the RT added no more traffic information than before and completely failed to get the imminent danger across to the IP on the helicopter. However someone may have got a hint because there was a subsequent conversation between crew members about whether they should move further east. But we don’t know yet if that conversation was actually about traffic deconfliction. It could just be about accurate tracking of the left bank of the Potomac or something else entirely.

Assuming they correctly received/understood that the object they were to pass behind was landing runway 33 , not runway 1. That seems to be in some doubt.

Because without that information, they could IMHO quite happily look at the A319 approaching runway 1, intend to pass behind it to head south down-river until the A319 was no longer over the river, and loiter around the runway 33 approach until that happens.


Shift the times by ten seconds and the same accident could have still occurred.

Visual simply doesn't work at the required level of safety if there are multiple aircraft to be visual with.

Even a non-pilot, non-engineer (and worse, an SLF/attorney) is able to understand the difficulties created by relying on visual separation at night, and more specifically, relying on it given the facts of runways in use on the night of the accident (i.e., that the clearance to the helicopter did not make it clear that the traffic advisory was meant to refer to an aircraft "circling" for approach and landing to Runway 33).

Reading this thread since the night of the accident, many have noted the "wrong-thinking" (for lack of a better term) underpinning the way traffic was managed by FAA and ATC.

I'm not enthused about the litigation that almost certainly will be intense once it commences. But reading the thread I've started to wonder.....

What would a chronological reconstruction of each incremental decision by FAA about the operation of DCA look like, a chronology that would (of course) include each Congressional enactment requiring or allowing further intensity of operations? The airport did not go from a nominal operational environment, with typically safe procedures and airspace usage rules very similar to or the same as any other major urban airport in the country, to the situation which obtained on January 29, overnight - or so it would seem.

(Yes, reconstructing the facts to build such a chronology would involve quite a lot of discovery activities in the presumably forthcoming lawsuits, but I'm not veering off into any further comment about why that would matter or which party or parties it would help or hurt..... other than to say, very often, cases are won and lost in discovery.)

Some years ago, on a trip to Capitol Hill hoping to find gainful employment on a Congressional staff somewhere, on the return flight from what then was Washington National, the aircraft's cockpit door was open as the boarding process was being completed (it was 1987). I recall being seated close enough to the flight deck - leaning a little into the aisle (a Midway Airlines DC-9 iirc) - to be able to see the pilot in the LHS and part of the D.C. skyline through the cockpit's front windows. Maybe that is why my mind somehow can't quite comprehend the recollections from kidhood of reading about the midair collision over New York City which is featured prominently in the book Collision Course , together with the events of the night of January 29.

Added: Wall Street Journal, March 6: "Air-Traffic Staff Rules Tightened After Crash." Also reports previous "close call" incident. Highly recommended reading (and I didn't see anything to contradict the above post, WR 6-3).

That was made pretty clear in the NTSB update over a week ago, yes. If your mental map is that the active runway is 01 (which it was) then unless you have the information that someone is landing on 33 and that's your traffic, you'll be looking for traffic approaching 01... particularly if you are familiar with flying in this area due to your mission requirements. If you know that your traffic is approaching RWY 33, then you look for it in a different direction, particularly if you are familiar with the local flying area.

(Not sure why this little hamster wheel is still spinning, but this is PPRuNe).

Hiya WR 6-3 , firstly, please do accept my apologies if I came over a bit harshly re ASIAS. I always read you input as, while the detailed legal nitty-gritty is generally way beyond me, you always provide a refreshing viewpoint which is good to read! The aim of my Post, based on my interpretation of the above quote, was just to emphasise the existence of ASIAS is just as a tool or, rather, a set of tools but which has no real "proactive function" in itself. Yes, it represents a very small part of the Flight Safety chain - but as a resource in the main. As you say, there were loads of examples of what happened at DCA and my post was an explanation as to who, I believe, should have been looking at it. Maybe they did - and just ignored it? No idea!

In the UK, the airport operator is totally responsible for the safety of operations on or above their field out to 4000 ft on approach and departure - about 12 miles. The CAA who regulates aviation here, along with the Dept of Transport, but in conjunction with other stakeholders such as NATS (our en-route ATC organsiation), airlines and airfield operators etc, designs the airspace and sets the generic rules for it's use. As you said, the main rule is "Don't crash anything into anything else!". However, at the airfield, it is the airport operator who needs to ensure that the generic rules work in their environment. If a particular rule is legal but, if applied in their location, is unsafe, they can't apply that rule! That's what the Safety Case should say.

A relevant example? I used to regularly (for work) transit London S > N and N > S along the red dotted line below - at 2400ft as it kept me just out of the London TMA but as high as possible so, if it had all gone quiet on the engine front, I'd have the best chance to get the plane onto the ground with me alive and no-one else hurt. As part of the transit, I had to cross the final approach into London City (green line). So, say heading N, I'd chat to Thames Radar while over South London and would ask for a London City crossing. They'd chat to City ATC to co-ordinate and take me to the Thames before handing me to City Tower. My route over the Approach has aircraft at 1200 ft descending on the ILS and so I would have crossed anything landing at 1200 ft above that. So, while on first glance, it was fine I never, ever got that clearance with an aircraft landing - and for very good reasons. If the aircraft on the ILS had had to do a go-around, life could have become interesting very quickly. I was always held until the landing aircraft was well clear (ie ATC could see it was almost at the airfield) before they cleared me across. Indeed, legally, they could have given me a "You see the landing? Pass behind that"! But, no - always positive clearances. Got my vote ....... + gave me chance to enjoy the views of Greenwich and the City of London and "delayed" me a couple of minutes. OK, a bit unfair as LCY is "sleepy hollow" compared to DCA but, despite that, they were very friendly ..... but very strict!!!!!



So, when you are discussing (I think!!!) whether ATC elected to give PAT25 a "You visual? Pass behind" or whether it was "ordered" from on-high as it was deemed safe to do, ultimately, it is the ATCOs call on the day. Now, before I get flamed, do I blame the ATCO ? No! Well, why not? After all, he gave PAT permission! And yes, it was a "legal" clearance. But was it a wise clearance? When unpacking that little lot you have to look at a raft of Human Factors which influenced that ATCO on that night. He was on his own so, probably, the only way he could cope with all he had on his plate was to try and shift some responsibility onto PAT25 - one less thing for him to juggle. But even then, he needed to be monitoring which he clearly was - but while very busy with other approaches and departures so he just picked up a concern too late as the audio shows - "Are you sure you see the jet?". Another factor - was the strategy to use Route 4 while 33 was active something ATCers at DCA, over time, started in an effort to cut down radio traffic and speed things up? If so, had it been assessed and then monitored for adverse safety? While anecdotally, it seemed people were aware of "close calls", had any analysis taken place looking at the Databases? In the UK certainly, all the Airport Operator responsibility.

To the final "accountability" part re payments to families, the thing we have here is (and, folks, correct me if I've got this wrong) the FAA set the rules and the operate the Rules. Here, the CAA sets the rules and the Airfield Operator operates the rules. That is important - for eg, huge argument here in the UK about just how independent the UK's Military Flight Safety organisation is ..... as it is part of the Military. IMHO, what we have here are valid generic rules, some of which were simply not suitable in this specific location - just like crossing London City with an aircraft on approach - legal ...... but very unwise. As to who carries the can, well, as you said WR 6-3 , the legal beagles will get to the bottom of that ..... but it will take some time. I'll leave that side to you!!! My main question out of all of this is, why were not these issues flagged up by airport management for further investigation given it was something of an open "secret"? And where else (around the world) is this happening?

And, on that cheerful note...........! H 'n' H

Hiya Capn , that is precisely what I've been saying in my last few posts (while waiting for various builders and plumbers to get back to me!) - have a look at my last few posts. It's also worth reading Post #1365 by layman54 re aircrew to which I was responding - hence the limited reply I gave concentrating on the topic their post had raised.

My view - they were specifically asked to visually identify a/c A. Now, given the stream of a/c, the distance the a/c were away, the angle of the stream, etc, etc how could they reliably pick out a/c A? Almost impossible I'd say. But they said they had, which then formed the basis of the clearance. But, as you say, they were actually looking at a different a/c, a/c B, not deliberately, but mistakenly thinking it was the one ATC had asked them to look out for and pass behind, a/c A.

You are spot on in that there was no way for ATC or the helo crew to verify that the one the helo crew were watching (a/c B) actually wasn't the one that ATC thought they were watching (a/c A). That, as you say, was a very significant weakness. So they were on course to avoid the one they were watching. But that wasn't the one ATC had asked them to watch.

You are dead right in how difficult "see and avoid" is. I'm lucky as I only flew between small UK Regionals and, particularly when in my home patch, I often got from Twr "Visual with a/c A on Final? Join Final as #2 to a/c A!" type stuff. Worked OK when it was only me and a/c A. Any more and usually Approach sequenced me if I'd arrived SVFR by placing me under Radar Control for a short while before then handing me off to Twr to join the cct visually once faster traffic was well ahead of me. That's if they didn't forget me completely .... which happened once ...... but that's another story - they were most apologetic when I reminded them I was still on the last vector they'd given me!!!! "Sorry about that, I'd completely forgotten about you!"! Charming!