View Full Version : A close shave in Japan.

Foreign Worker
17th Oct 2003, 20:09
My experience of Japanese the ATC'ers is a quite favorable one overlall.
Yesterday (Thursday) morning we were following a domestic Japanese 747 aircraft into Fukokoa and had them clearly identified on TCAS. To the absolute amazement of all of us on our flight deck, we observed this B74 descend to within 200 to 300 feet and within an estimated less than 5 miles of unidentified TCAS traffic, before the Approach Controller ordered the 747 to "Turn left, heading 220", in what can only be described as "urgent" intonations.
Why the crew of either aircraft allowed themselves to be compromised to this degree is not for me to querie. Nor that the controller might have been fully loaded up, but more, that just that this type of scenario can still be a part of our everyday life as a professional - whether we are Air Traffic Controllers, or pilots - we don't need to let complacency creep in.
Lotsa questions I know, and all credit to the Japanese controller who took the initiative to actively intervene!

19th Oct 2003, 02:37
Just a word of caution.
Please don't interpret TCAS as if it were Radar. You are not looking at relative positions of traffic you are observing a "time plot".
There have been a number of serious airmisses ( I think of France in particular and Area Crossing Tracks ) where crews have looked at the TCAS thought they were on a collison course ( which they weren't ) and then turned to make matters worse.

Let TCAS tell you when there's a problem.

Faire d'income
19th Oct 2003, 05:22
TCAS is completely unreliable in azimuth so your 5 mile estimate is irrelevant to the story.

Flap 5
19th Oct 2003, 15:02
Yes TCAS is not accurate in azimuth but we are talking about plus or minus a couple of miles here. When you are travelling at around 5 to 6 miles a minute on a descent a couple of miles is very little. Japan airspace is very busy and you are constantly being passed from controller to controller as you descend in to lower airspaces therefore the initial post is very relevant. There have been many TCAS incidents in Japanese airpspace and there are likely to be many more.

Faire d'income
20th Oct 2003, 01:13
When you are travelling at around 5 to 6 miles a minute on a descent a couple of miles is very little.

No kidding! I'm glad you pointed that out on a Proffessional Pilots discussion board.

The original post is sensationalism. The headline was
Close shave in Japan which is why most of us looked here.

Then we see it is based on an azimuth assessment of a TCAS display and a 'stressed' heading 220.

Try AB123 Right turn now 90 degrees climb to. ..break break...CD321 Left turn now 90 degrees descend to ....!!! That's is the sort of thing you hear for a serious conflict.

Also if the traffic was five miles or more away it is a complete non-event. The original post said within an estimated less than 5 miles . Even if the TCAS was accurate if it was say 4.6nm is it a close shave?

I take your point about busy airspace but the thread title should have been ' Traffic sometimes heavy in low level Japanese airspace'. :zzz:

Kaptin M
20th Oct 2003, 05:30
TCAS is completely unreliable in azimuth so your 5 mile estimate is irrelevant to the story. Completely unreliable in azimuth?
Contrary to your statement, Faire d'income, I have found TCAS quite reliable, and in agreement with ATC advisories as to position of other traffic, esp. within 20 nm.

Foreign Worker has indicated that the aircraft were on Approach Control, so normal separation standards were apparently in probable contravention of the regulations. It is only on the very rare occasion that controllers on Japan App. allow aircraft to come closer than 10 nm, in my experience.
Additionally, traffic into FUK often enters from the East and West ie. head on, and if this were the case in F.W.'s example, with 200'-300' vertical separation and a closure rate of say 400 kts, I can understand his/her concern.

Donald Dunbar
20th Oct 2003, 21:26
I would have to agree with Faire d'income. The rest of you clowns don't know what you're talking about.

Love Don

ATC Watcher
21st Oct 2003, 00:20
Right on dunbar, what a lot of bull :mad:
If we start getting such reports and such comments ......
Yes TCAS is very bad in azimuth ( within + and min 11 degrees )and you definitively cannot rely on it to assert positions. but that does not mean that in perfect conditions the display will always be wrong. And the " heading 220 " in a stressed voice ! Ah! personally I never met a unstressed japanese colleague did you ?
Advice : if you do not know what you are talking about, just read and keep away from the keyboard.
( maybe that should become the motto of this forum !)

Foreign Worker
22nd Oct 2003, 09:33
Do you respondents have any authoritative references to back up your claims that TCAS is unreliable?
Comments of the sort logged by Donald Dunbar indicate a distinct lack of knowledge and professionalism for his part.
My post was NOT made for the sake of sensationalism, merely a statement of fact - and devoid of any racist comments such as posted by ATC Watcher.

Akin to Kaptain M's comments, I too have found in over a decade of operation, that TCAS of today is an extremely reliable tool in providing the position - in both the vertical and lateral axes - of other traffic.

Another re-reading of my original post would have one realise that as the other traffic was unidentified, the ATC had no idea in which direction it was heading, nor any radio contact with it.
The instruction to turn onto a heading of 220, was a 50-60 degree course alteration for the 747.
My estimate of less than 5 miles was being generous - it appeared to us to have more likely been closer to 3nm.

ATC Watcher
22nd Oct 2003, 21:46
Forgein worker :my remark about stessed japanese controllers was supposed to be a joke, not a racist remark ! I have to watch my language in the future. my apologies.

I never said TCAS was unreliable but TCAS was never made to be a CDTI ( Cockpit display of traffic info ) is is a software logic aimed at issuing warnings to prevent a collision in last minutes close encounters. The display is an add on that was not even specified in the early certification process.
The poor azimut detection at large distances is a early, known deficiency but that does not alter the anti collision logic.
again TCAS should not be used to identify traffic.
As a controller I would add " never " and a pilot suing R/T sentences like " I see him on TCAS " will get him in trouble one day.
TCAS is a last minute anti collision device and should not be used for anything else.
The Collision avoidance logic (CAS) has also its flaws, especially in issuing reversals, (and Ueberlingen is a perfect example of this), but that is another story.

Ah, and a 50-60 degree turn can be odd, I grant you that ..

Flap 5
22nd Oct 2003, 22:04
Oh dear. What was a perfectly reasonable post has turned in to a slanging match.

I too have found TCAS to be reasonably accurate. I too have found that Japanese airspace is busy. If you put these two together the original post is of some concern.

The concern with some of the posts is that they show an amount of complacency which is worrying. If you get a TCAS warning you don't say 'Oh its okay, its inaccurate in azimuth', you react to it!

The comment by ATC Watcher was NOT racist it was a statement of fact. Some people will call any criticism racist! And unFaire D'Income my point about the miles per minute clearly needed to be pointed out as you are so keen to sweep it all away as irelevant!

22nd Oct 2003, 22:46
ATC Watcher: says "... The Collision avoidance logic (CAS) has also its flaws, especially in issuing reversals, (and Ueberlingen is a perfect example of this), but that is another story."

Ueberlingen appears to be a flaw about the Russian pilot belatedly following ATC instructions rather than his correct TCAS Resolution Advisory.

Upon receiving a "Traffic" advisory, my technique is to note the traffic quadrant where the target is, then switch the TCAS screen immediately to the 5 mile range, or as a last resort down to the 3 mile range, which as a last resort would afford me the option of a lateral maneuver away from the target, in addition to simultaneously following the vertical RA instructions.

I believe that the DHL crew at Ueberlingen could have sharply turned away from the target while complying with the vertical RA instructions. The radar plot did not record such an immediate turn, and as it was, the 757 tail had sliced into the TU-154.


24th Oct 2003, 07:33
I don't know what class of airspace we are talking about here, but we are in the throes of trading in our C airspace for E.

We will now have the (greater) possibility of unidentified traffic on the approach path of jets, and our training tells us to just to pass traffic (in those areas where we have radar anyhow). Do I understand from these posts that everybody is getting separated elsewhere???

Kaptin M
25th Oct 2003, 00:18
So as a matter of FACT, how UNreliable is TCAS on the (say) 20 mile scale?
I shouldn't imagine that it should be considered as untrustworthy as Faire d'income might lead us to believe, as the returns are transponder based from the identified traffic.
Perhaps on the 160nm and 320nm scale, the veracity is degraded due to time lag, but on closer in/nearer scales, one might assume that unless there is a radical heading change on the part of either party, the returns will ACCURATELY define the almost exact proximation in real time terms.
And it is upon THESE visual id's that TCAS will issue an alert.

It is therefore with more than a little difficulty, that I try to understand the abject criticisms of F.W's posting - except possibly to think that some ATC'ers might feel their positions threatened!?

Faire d'income
25th Oct 2003, 03:08
Don't ever suggest your reaction to a Tcas warning on a check. You will fail!

Any company's SOP in the event of an RA is to ( as the system will tell you ) Monitor Vertical Speed . Any further interpretation is at best misadventure. At worst your bank angle may prevent you from achieving the necessary climb rate.

Kaptin. TCAS II is unreliable in azimuth. TCAS III was being developed ( is it up and running yet? ) to overcome this flaw. It was envisaged that TCAS III would provide lateral as well as vertical commands during an RA.

I am not an ATC'er and if I feel my position threatened it is because there are fellow aviators who don't understand their systems or how to react to warnings from them!

Flap 5 your point was about as relevent as telling an F1 driver to brake for chicanes.