PDA

View Full Version : Heathrow ATC approach and handover questions.


AdamFrisch
1st Jun 2010, 00:37
I was just listening to an approach and landing at Heathrow found on YouTube, and couldn't help but wonder why certain procedures are in place and why things are done in certain ways. I would greatly appreciate if someone would explain it to me.

As the approach took place the aircraft (in this case a Cathay Pacific heavy) went through and Approach freq for the initial hold and was brought down to 4000ft. After that it was handed over to the Director freq. Once speed had been reduced further and it was established on the ILS, it was handed over to Tower and given a clearance to land. After touchdown it was further handed off to Ground.

That's four handovers to get the aircraft down. Four instances where information has to be passed from one controller to another internally so everyone's on the same page. Four instances where potential screws up could take place. To an outsider with zero clue about ATC procedures, that seems rather inefficient. I'm sure there's a reason for it.

But to me it would seem a lot more time efficient for one controller to handle the whole thing from App to the Ground, as that would save four briefings and handovers and four potential screw ups. I understand that the frequency would get busier this way, but perhaps there's a way to separate approach and departure or something? Why is it done this way?

I also noticed how much time gets wasted on repeating callsigns over and over again. Has there ever been any talk about assigning each approaching aircraft a number in the sequence instead, and having that as the callsign for the busy landing seq, like:

"Heathrow Approach, Cathay 257, FL100"
"Cathay 257, you're number 10 for landing"
...
"10, descend to FL90"
"10 to FL90"

Etc, etc. Daft idea?

When you listen to ATC at Whitman Field during Oshkosh (when it becomes the busiest tower in the world), it's insane how much traffic they manage to wrangle. And it's all pretty much down to the fact that the approaching airplanes RT is kept to a minimum.

gone_fishing
1st Jun 2010, 00:51
When all units are open at Heathrow, aircraft will be initially handled by one of two Intermediate (INT) controllers, whose job it is to take aircraft from the holding stacks (BNN, LAM, OCK and BIG) and combine them into one stream - one handling the north (BNN and LAM) and the other the south (OCK and BIG). They will then hand the aircraft off to Final (FIN). The aircraft from the north (all assuming westerly operations) will normally be handed off on a heading of between 120-130 degrees and descending to 4000ft at 220 knots. The south will be the same, except heading something like 070-080 degrees. When aircraft are being vectored in this standard way, very little if any co-ordination is made between the appropriate INT controller and FIN controller, as it is all standing agreement and a silent transfer can take place. Normally, aircraft will be instructed to "callsign only" with FIN. FIN's job is to merge these two streams (one from INT N and one from INT S) into a single stream on the ILS and hand-off to AIR (Tower) at, normally, 160 knots to 4d. All this reduces RT and co-ordination and is the reason for having several frequencies - the job just couldn't be done by one controller.

I honestly hope you can see the issue with using sequence numbers instead of flight numbers for RT callsigns: "Number 11, oh, you're number 10 now - turn left..." and all the other impracticalities.

They do a fantastic job at Heathrow, like the rest of UK controllers and work some very crowded airspace. I'd say the system is pretty efficient - complying with all the regulation that many US controllers don't have so much of around their necks.

lemony
1st Jun 2010, 01:28
one controller to handle the whole thing from App to the Ground


Do you mean take it turns to control aircraft, ie four controllers each doing an aircraft from app to ground or one to try and fit everything in there on their own which I think would be an impossibility.

AdamFrisch
1st Jun 2010, 02:05
I honestly hope you can see the issue with using sequence numbers instead of flight numbers for RT callsigns: "Number 11, oh, you're number 10 now - turn left..." and all the other impracticalities.

No, I was thinking more of it as a queue number - you would keep that number until you landed, no matter if you were number 1 for landing or 100. It was just a way for the ATCO to assign a number at the time of arrival.

HEATHROW DIRECTOR
1st Jun 2010, 06:38
With respect, Adam, you don't appear to have much understanding of aircraft operations at a major airport so i hope the following will help.

Allocating a landing number in the way you suggest is out of the question. With modern radar each aircraft can be easily allocated a transponder code and this is processed by the radar to show the aircraft callsign on radar together with information on height, speed, etc. The additional work involved in changing this to a sequence number would be enormous and would benefit nobody.

With your suggestion, at the end of the day ATC might be saying: "You're number 600 to land". What would that achieve? Controllers involved with a particular flight have all the necessary information in advance of talking to it so direct liaison is often not necessary. Many years ago the final director at Heathrow would telephone the Air controller in the tower and tell him the landing order - maybe a dozen phone calls an hour. Now, the tower has radar so they can see the landing order and phone calls are not required except in unusual circumstances.

The system employed at Heathrow is pretty very similar to that at any major airport around the world. You can't possibly compare such operations with Oshkosh, where everything is VFR. Commercial aircraft fly IFR in Controlled Airspace which requires them to be properly separated for safety reasons. The landing and take-off rates achieved at major airports are about the maximum possible given the size of aircraft involved. If you want to see the slickest handling of traffic on a single runway take a look at Gatwick..

Gonzo
1st Jun 2010, 08:12
Adam, in many cases there is no transfer of information during a handover of an aircraft from controller to controller......We have what we call standing agreements where the next controller will 'expect' aircraft to be given to him/her in a certain place and a certain situation, and if this is the case no information needs to be given.

timelapse
1st Jun 2010, 08:45
I think to answer your question properly we need to go back to basics a little bit.

If, taking a very wild view, there could be one controller doing the approach tower and ground phase there probably would be, but because of various reasons the tower/ground and approach jobs are split. They require different training and skills and different equipment. Also the workload required to combine them would be unworkable if there was more than a few aircraft per hour wanting to do anything. Also bear in mind that a lot of airfields require air traffic control to vehicles wandering around, anything wanting to cross runways, etc.

So, you have more than a few aircraft in an hour (so it's too much work for one person), and just say that your controller is qualified and able to do the ground, tower and approach job. How do you go about solving the problem that his workload is too high? You have one choice really.. you get another controller on board to help with the work who is equally qualified.. but how do you split the workload?

In your world that you suggest, you assign each controller one half of the planes and the other controller the other half. This reduces the workload by half for each controller (supposedly).

If you consider the realistic consequences of this, you will see that the only thing it does is hugely increase the workload for both controllers. If you think about it, every single aircraft movement would need to be co-ordinated with the other controller to ensure that there is no conflicting traffic. Every runway crossing, landing or take off, climb, descent or route clearance would need to be checked with the other controller to make sure they didn't have any vehicles on the runway or conflicting aircraft anywhere near. The situation would be unworkable.

The other way to split the workload is to give each controller their own area of responsibility and draw a line between them. The approach controller is given a certain amount of airspace, the tower/ground controller is given some too, and the runways and taxiways. You then set up agreements between them (as Gonzo said) so that you can present the other controller with aircraft in a certain position doing a certain thing, and they will expect it to be there (so therefore won't have anything else that would conflict). This is exactly what happens at Heathrow (except it's split 2 more times because of the huge volume of the traffic).

I can see why you wouldn't really see the best way to do it unless you saw how it worked from the other side - so maybe it's worth a visit!

AdamFrisch
1st Jun 2010, 09:21
If you consider the realistic consequences of this, you will see that the only thing it does is hugely increase the workload for both controllers. If you think about it, every single aircraft movement would need to be co-ordinated with the other controller to ensure that there is no conflicting traffic. Every runway crossing, landing or take off, climb, descent or route clearance would need to be checked with the other controller to make sure they didn't have any vehicles on the runway or conflicting aircraft anywhere near. The situation would be unworkable.

Thanks for your responses. It makes sense.