Near miss over East London
 

That piece of sky is extremely familiar to me. It is in cloud right now. I am looking at it as I type.

But for the 1000 feet, I assume that sometimes almost head-on conflict between LHR base approach and LCY departure is played out very frequently - that must sorely push the odds ?

I wonder whether London ATC's special new Mode S kit got a look in this morning, or whether it all happened too fast. (See http://www.pprune.org/forums/showthread.php?t=287326 post #15 )

Opposite track climbing and descending .... classic.

Sounds like a simple case of level bust then R.A. Someone bust their level climbing to 1000' below me last week. God bless TCAS. If the MCP Altitude was set correctly, I guess the mode S readout wouldn't show anything until the bust happened.

This was probably no more than a TCAS RA, certainly not a near miss.
TCAS works on closure rates and if the 146 had a high climb rate and the Air Canada had a high descent rate ( to maintain a CDA) the TCAS would alert. Both would have assigned altitudes that would avoid a collision.

146 and high climb rate in the same sentence :)

In my book RA = near miss!

Must've read a different book to you.

If I had designed TCAS I would've got it to provide an RA way before a near miss, but then again I didn't.

I have had RAs out of EGLC and seen the conflict aircraft. There was no way it was a near miss. TCAS works on closure rates as follows.

Traffic Advisory (TA) is generated when a predicted aircraft path penetrates the Caution area, 20-48 seconds before entering the Collision Area.

Resolution Advisory (RA) is generated when a predicated aircraft path penetrates the Warning area, 15-35 seconds before entering the Collision Area.

15-35 seconds is plentry of time to take action and has a margin built in to avoid a "near miss".

In a previous incarnation as a controller in this piece of airspace a very similar incident happened to me a few years ago. This also involved an airliner from across the pond arriving at Heathrow and a London City departure. The QNH was 992mb/hPa and when given this by the previous controller along with clearance to descend to 4000ft the pilot read back "2992 millibars" and descended with 29.92 inches set on the altimeters. Unfortunately this incorrect readback was missed. The resulting separation was around 500ft.

Not wishing to cast any aspersions, I offer this out of interest only.

Interesting view WaGA

As you will have gathered, I am not so confident of the routine nature of this one. Do remember that the LCY aircraft will have been airborne no more than about 30 seconds when this occurred. Also the LCY aircraft would have been in a right turn to the north and the Air Canada would almost be just completing a right turn south onto base.

Both we are told were changing level.

That's a hell of a lot of conflict resolution to expect from an automatic system which is trying to resolve closure rates which at first seem manageable and then both aircraft almost simultaneously actually turn towards each other!

Ever see the movie "Red October" ? What did Sean Connery remember about the closure rate resolution on the torpedo warhead after he turned at the last possible moment and faced it down head-on? Might be a bit cheesy, but you get my meaning.

Not sure how the accuracy of the original report is verified but if they really got "within 500 feet" per the report, then I am sure this one is being investigated.

LHR and LCY were using the same set up again this morning. It is probably the most common in daily use. Without the plates in front of me just now I am guessing, but that bit of sky just north of Canary Wharf where conflicts could occur in the overlapping STAR and SID might be as small as a cube with 2000 foot sides but usually when there's a LCY SID and LHR arrival converging, you see the Heathrow traffic either turned early nearer Tower Bridge or extended a bit downwind almost to overhead LCY 10 Threshold whilst the LCY traffic is obviously directed to keep its head down in the climb.

When the sky is free of LHR traffic, many of those RJs skyrocket out of City and in one sweeping turn exit to the north and/or then round to the east quite literally like bats out of hell. They often remind me of the Lightnings and Canberras skimming the horizons of my childhood!

Snapshots of the cubic mile centred on that overlap cube when both types of traffic are in view (as they frequently are at peak periods) would yield shots of a myriad of different dances indeed !

My point is that this tiny corner of London's airspace is becoming ever more concentrated, it is where the converging aircraft' crews have their hands full during take-off and approach, and it is the classic opposite track climb and descent scenario. Added to that is the wildcard for the LHR traffic is that LCY aircraft likely to launch off 28 out of LCY mightn't even be on the menu when you are about to turn towards Canary Wharf, but they very soon are up and at you by the time you are coming out of the turn unless the controllers and LCY pilots are right on the ball. The odds of a potential rapid onset conflict during normal operations are then surely as great as can be found anywhere ?

A possibly useful TCAS overview is at http://techaidproducts.com/PDFs/TCASUncovered.pdf

LeFreak : 146 and high climb rate in the same sentence...



Stop it now, its a lovely plane... especially with its 5 APU's! :)

SLIP AND TURN

Agreed. If indeed the separation was 500' then clearly something went wrong with one of the aircraft. As I am sure you are aware the TCAS doesn't take account of whether the aircraft are turning. Each calculation is based on rates of descent in protected areas from each aircraft. The collision threat is re-assessed irrespective of direction, So no more calculations would be made than if both were in a straight line.

Also both aircraft are on different frequencies which doesn't help. The only answer (short of moving one of the airports) is to ensure that rates of climb and descent are not excessive. This was part of my line training, even for my current airline that doesn't fly to City. In busy airspace we should reduce the climb rate to not more than 1500'/min when within 2000' of assigned level. Not written down, just good airmanshap.

Now RAs with military fast jet having a bit of fun in Pennine Radar, thats another thing.

slip and turn,

I don't wish to be rude but much of what you write here just indicates you don't really know what you're talking about.

LCY and LHR operate autonomously of each other. Simply put LHR arrivals descend to 4,000ft and cannot go below that until a set point that is clear of the LCY SID track. LCY departures climb to 3,000ft and cannot go above that until outside of the lateral boundary of the LHR RMA.

There's no interaction between the two units at all bar unusual circumstances i.e. LHR don't alter their vectoring if there's a LCY departure or not. The only thing that dictates where the LHR base leg is is how busy it is and how many are off the stacks on intermediate approach, there's a natural ebb and flow in and out and I stress again it's nothing to do with what LCY are up to.

Obviously if either an LHR arrival or LCY departure busts its level in the area you're talking about there's potential for an incident, but the same is true in many other parts of the TMA. This little corner of it is in no way unique.

Blimey a 146 at 3000' within 30 seconds of take off. Maybe the lightning analogy wasn't too far off...

Well I do think its a bit rude of you Roffa old chap, as even though I am not an overworked controller with not enough meal breaks, I have had quite a few hot dinners whilst watching that piece of sky in the last few years :)

I still haven't studied the plates but all I can tell you is that when they LHR & LCY are using these western runways (which of course is frequently) and when there is no cloud obscuring my view or theirs, more often than not, the LHR traffic turns south around the eastern edge of CW and then turns roughly west around my gaff.

Sometimes it extends downwind and might then turn 135 degrees back toward me, and less frequently it might turn earlier down near Tower Bridge.

I am little surprised but not overly so when you say that barring unusual circumstances the traffic from each airport is handled autonomously.

I guess you must be a controller.

I can equally well tell you that the City departures yield a more consistent pattern with usually at least 90 degrees of the right turn out completed by the time they reach abeam Canary Wharf. Not always however. An hour ago while I was having another hot dinner I noticed what appeared to be a merry dance of City traffic extending on the runway heading well past CW then turned north while something heavy joining the LHR approach from the north made a kind of S on the LCY side of the departing traffic.

I am sure what you say about the published procedures is correct but I am just reporting what I see. I am sure that if videos of the traffic were overlaid Red-Bull-Air-Race style then the overlaps would be quite apparent. Not too many level overlaps we would hope, and no simultaneous ones, but the tracks certainly would all the time and that just leaves our 1000 feet to do all the work.

If Red Bull had left their cameras in place the other week covering the sky just next door to this bit then maybe you could be more sure ...

The difference between this corner and any other in the London TMA is rather more obvious than you seem to be suggesting Roffa. As I am sure you appreciate, they are up to a rate of around 2.5M per annum passengers out of LCY I think and in small aircraft that's a lot of movements, especially if they are sometimes only half full!

Lets all use Standard?
 

If this is yet again a mis-set altimeter setting you would have to ask:

"Isn't it about time that we all started using QNH in Hectopascals/Millibars?"

Come on USA, if the rest of teh world can use it would it really hurt to conform?

Not a wind-up, just a question.

Brad

slip and turn,

I'm glad you've enjoyed a few hot dinners watching the traffic out of the window, I've enjoyed twenty odd years watching the same bit of sky from a radar display :)

Hopefully one of my colleagues will come along and back me up, but I assure you LHR traffic and LCY traffic is primarily separated vertically, not laterally, in that area. To do the former would simply not be practical. The respective controllers do sit only a matter of feet apart in the same ops room but can go a whole shift without having to speak to each other operationally, the procedures are set up such that in all configurations the two airports operate independently of each other (well when LHR are on the 27s and LCY on 10, though still independent ops the LHR folk take a close interest in how far downwind the Thames folk take their traffic, but that's another story).

If you see LCY departures doing slightly different things it's either because they're not following the SID track exactly or Thames, for their own reasons, have taken it off the SID early... but that is 99.9% certainly nothing to do with LHR traffic.

Below is an fairly old plot of inbound tracks to LHR, where you see the plots of traffic joining final approach at different ranges that's nothing to do with avoiding LCY traffic. It's just the natural way that the final moves in and out depending on the volume of traffic.

http://mysite.orange.co.uk/roffa/images/radarplot.jpg

Very much appreciate the trouble you've gone to Roffs but the fact remains that laterally the LHR inbound and LCY outbound tracks overlap all the time, and the adjustments I thought I was seeing weren't much to do with the various joins but all occurred around the final 180 degree turn sector.

Your plot doesn't show any of the LCY plots of course but does show the variation in LHR tracks in the area of the reported conflict i.e. around a point 70% up and 80% right on your plot (I am guessing).

Is there a Thames plot you could overlay?

As you say, its the vertical separation that keeps us safe.

Edit: Roffa it just dawned on me that you said you'd been looking at that sky on those screens for 20 years. I don't think I'd last 20 weeks. You deserve a medal if they haven't already given you one. Seriously. I'm sure they must have.

Any chance of any more of those "old" plots of inbound traffic to LHR or LGW?

Slip n Turn

I must step in and back up roffa here
London City departures do NOT affect Heathrow inbounds. City departures climb to 3000 then further climb on the SID when instructed by radar, Heathrow arrivals decend to 4000 = separated in my book.
In regards to where the Heathrow aircraft do their turns onto base and finals, this is purely down to conditions of the day (winds etc), amount of traffic and runways in use at Heathrow, and also who's controlling at the time I guess!! (I dont do radar!)
Hope that helps Slip n Turn
Roffa hope this backs you up!

Look sunshines, I didn't report the conflict, someone else did and THAT hasn't been denied :{

Actually I see taffy has toned down his post a bit so I am smiling more easily :)

Neither am I arguing about how separation is planned to be obtained.

I am just telling you what happens outside my window which has a bloody great 800 foot high glass steel and concrete pointer 4500 feet from my eye dividing the view almost East and West. If I say LCY departures routinely complete a turn to the north before they get west of Canary Wharf they do. If I say LHR arrivals usually turn onto base at the same place albeit higher, then that's what I've seen.

Why take potshots at the messenger ? I am sure that's not what Roffa intended when he asked for backup, or do you know better?

The only contentious opinion I have given (I think?) is that vertical separation alone in this uniquely busy corner seems to push up the chances of an error.

Take that last sentence apart if you like but don't question my ability to use my own eyes please to take bearings against known landmarks please.

Roffa has given us the luxury a nice plot to look at and thanks be to him. How about you guys showing us some LCY 28 departure tracks overlayed on the LHR 27 arrival ones and then you can tell us typically how many times a day LCY traffic climbs reaches or exceeds 3000 early in that right turn out. If it rarely if ever does, then my contention dissolves into oblivion, but if 3000 is frequently achieved that early in the climb out then please read or write on.