PPRuNe Forums - View Single Post - LHR mixed mode proposal
View Single Post
Old 8th Dec 2007, 08:56
  #9 (permalink)  
120.4
 
Join Date: Mar 2002
Location: LONDON
Posts: 314
Likes: 0
Received 0 Likes on 0 Posts
Mornin' all... somebody call?

To answer the initial question: The longer turn-on to finals must become the norm for reasons I will try to explain. Please bear with me.

1. In mixed-mode the traffic must turn on vertically separated. (The TWASS concept mentioned in the Consultation document in whcih vertical separation is not maintained is a fail dangerous operation and there is ample evidence to prove that it is only a theoretically possible concept. It is not ICAO and the regulator is unlikely to be happy with it. I would refuse to actually do it.)

2. Understandably, the government wants as much CDA as possible and this means that the joining altitudes have to be as high as possible but for every 1000' up you put them you have to go an extra 3nm out. The current TMA design would only permit joining at 4000' and 5000'. Future changes to use 6000' and 7000' (and thereby provide full CDA are in the pipeline, and they have to happen for Runway 3 anyway.)

3. Problem. Traffic is only deemded separated when BOTH sides are established on the Loc. To be absolutely independent and therefore fully efficient, traffic must be able to descend regardless of the other side. Imagine a case where the high-side is on the Loc, approaching the glide but the low-side is not yet established. The high-side would not be able to descend and may require repositioning. The two Final Directors would spend too much time talking to each other instead of vectoring the traffic.

4. Solution. Have the traffic join at a MINIMUM range that is equal to the high-side glide path plus 3nm. In this case that means 15nm (5000' @3 degrees) plus 3 equals 18nm as an absolute minimum. Then, the low-side will always establish 3nm outside the range of the high-side glide path and therefore the high side will always be able to descend. Simulator trials demonstrated that in fact you need plus 5nm to allow for joining errors. 15 +5 =20nm, minimum.

Does that help?

.4

Last edited by 120.4; 8th Dec 2007 at 10:09. Reason: Morning .7, love you too!
120.4 is offline