TCAS safe for ATC?

[Voel]

:confused:

Acft transmits blind on ATC freq that head-on traffic as same level. ATC descent the aircraft to a lower level. Opposite traffic calls thereafter that it already iniated TCAS desent. What know? Has any recommendation regarding the use of TCAS after the Ueberlingen accident been adressed already?

[Proceed As Cleared]

Has any recommendation regarding the use of TCAS after the Ueberlingen accident been adressed already?
Should've been adressed after the incident in Japan (January 2001) already...

http://aviation-safety.net/database/2001/010131-2.htm

From the report:

The commission recommended a.o. that the International Civil Aviation Organization (ICAO) put priority on computer orders over instructions from controllers to prevent similar incidents.

1.5 years later Ueberlingen happened...

[Scott Voigt]

In the US from the very onset of TCAS, ATC has in our rule books that we NOT intervene as soon as we know that a TCAS RA is taking place. We do issue traffic but we don't do anything else until the aircraft advises that they are finished with the TCAS event.

regards

Scott

[Del Prado]

we NOT intervene as soon as we know that a TCAS RA is taking place

Of course we only know of a TCAS RA by pilot reports. The controller will be ignorant of an RA for at least 5-10 seconds and possibly up to 20 or more seconds, that's plenty of time to issue an instruction which may be opposite to TCAS.

There seems to be more of a tendancy to give avoiding action in the horizontal plane only now, but will pilots follow this advice when a bank angle of more than 5 degrees can cause errors in TCAS ?

[Barry Cuda]

In the original question the "other" ac has taken the RA, so the pilot of "our" aircraft must be getting an RA and should therefore disregard ATC level instructions and follow the RA.

At least that's what I think!!!

Also, if a pilot reports an RA I will always give traffic info and, IF I THINK IT IS RELEVANT, avoiding action in the horizontal plane only...

[Phoenix_X]

Usually yes Barry. The other aircraft would indeed also receive an RA, and in most western companies that RA will take priority over any ATC instructions.

However, if the other aircraft has no TCAS (fortunately more and more rare nowadays), but only a mode-C transponder? 'Our' aircraft will create an RA based on the current vertical behaviour of the other aircraft. If ATC now issues an instruction to the other aircraft opposite to what it was doing it could at least cause a problem with the RA instruction. It may be revised to a more aggressive manouvre, or of course it could be worse...

[Lon More]

TCAS resolutions are supposedly confined to the vertical plane; for this reason it might be better to limit ATC instructions to turns - of course you may only make matters worse, depending on the radar resolution - or better still to only give traffic info.

[Max Angle]

but will pilots follow this advice when a bank angle of more than 5 degrees can cause errors in TCAS ? I can safely say that I would not follow any advice from anybody until the TCAS RA is over, you don't even do what your eyes tell you, the aircraft you see might not be one the generating the RA.

[Scott Voigt]

Actually, it is my experience that TCAS RA's are announced just about immediately over here...

regards

Scott

[Del Prado]

Scott, how can you know that ? You'd have to be on the aircraft and hear the RA to know for sure.

The reason I mentioned the delay in reporting an RA is the tragic Lake Constance example when there was a delay in the DHL RA report. I don't mean to lay any blame, just to show an example of the pilots' first priority to aviate and navigate before communicating an RA.
In an ideal world one pilot would fly the RA and the other announce it immediately. In the real world a delay of 5-20 seconds doesn't seem unreasonable.


Max Angle said

I can safely say that I would not follow any advice from anybody until the TCAS RA is over, you don't even do what your eyes tell you, the aircraft you see might not be one the generating the RA.

Is that what we might expect all pilots to do ? Are ATCO's training for this ? Is it worth giving any avoiding action during an RA incident ?

[atcea.com]

Here's the FAA 7110.65:

2-1-27. TCAS RESOLUTION ADVISORIES

a. When an aircraft under your control jurisdiction informs you that it is responding to a TCAS Resolution Advisory (RA), do not issue control instructions that are contrary to the RA procedure that a crew member has advised you that they are executing. Provide safety alerts regarding terrain or obstructions and traffic advisories for the aircraft responding to the RA and all other aircraft under your control jurisdiction, as appropriate.

b. Unless advised by other aircraft that they are also responding to a TCAS RA, do not assume that other aircraft in the proximity of the responding aircraft are involved in the RA maneuver or are aware of the responding aircraft's intended maneuvers. Continue to provide control instructions, safety alerts, and traffic advisories as appropriate to such aircraft.

c. Once the responding aircraft has begun a maneuver in response to an RA, the controller is not responsible for providing standard separation between the aircraft that is responding to an RA and any other aircraft, airspace, terrain or obstructions. Responsibility for standard separation resumes when one of the following conditions are met:

1. The responding aircraft has returned to its assigned altitude, or

2. A crew member informs you that the TCAS maneuver is completed and you observe that standard separation has been reestablished, or

3. The responding aircraft has executed an alternate clearance and you observe that standard separation has been reestablished.

NOTE-
1. AC 120-55A, Air Carrier Operational Approval and Use of TCAS II, suggests pilots use the following phraseology to notify controllers during TCAS events. When a TCAS RA may affect an ATC clearance, inform ATC when beginning the maneuver, or as soon as workload permits.

EXAMPLE-
1. "New York Center, United 321, TCAS climb."

NOTE-
2. When the RA has been resolved, the flight crew should advise ATC they are returning to their previously assigned clearance or subsequent amended clearance.

EXAMPLE-
2. "New York Center, United 321, clear of conflict, returning to assigned altitude."

----------------------------------

Once again, the danger lies in the lack of standardization around the world. For something a critical as TCAS/ACAS RA's, we very much need to be on the same page.


ATC 24/7 (http://atcea.com)

[Cartman's Twin]

Regarding how ATC respond to TCAS RA's it is 'good practice' not to give any vertical avoiding action once an RA is notified. (Having said that, in the proverbial heat of the situation I'm sure some controllers may try to climb/descend an a/c back to where they thought it should've been.

Traffic information is encouraged but we're encouraged only to give avoiding action in the horizontal plane and wait until you report 'back under my control'.

[West Coast]

"You'd have to be on the aircraft and hear the RA to know for sure"



We as pilots are of course trained to respond to RA's. We also are aware a response may cause us to deviate from altitude and the relative comfort that an assigned altitude provides. It only takes one guy to handfly the escape maneuver, the other is taught to immediately advise ATC so they are in the loop. Many times a RA is not a great surprise. Watching the TCAS we see someone heading towards us, it turn to a TA and finally a RA. This allows time to think about impending requirements such as ATC notification.

[Scott Voigt]

Del;

What West Coast said, also I have been both in the sims watching the training as well as in the jump seat (back when we could) and have watched it first hand (With VFR's who were 500 feet away, which by the way is perfect separation.). Also we can see that the aircraft hasn't even started the manuver yet when they report the RA... It's a training thing here that we started at the onset... One of the nice things of having ONE aviation authority for a LARGE bit of airspace <G>...

regards

Scott

[Right Way Up]

Pilots will normally advise ATC of RA immediately. That is the job of PNF. In the DHL accident from recollection, one of the pilots were out of the seat (had just gone to the toilet), so the SOP of PF performing the manoeuvre and PNF advising ATC of RA was flawed. The PF did what he was trained to do but there was nobody in the other seat to report the RA. It is a damning indictment of the aviation training setup. We practice standard "abnormals & emergencies" but never seem to think outside of the box.

[Blue heaven]

ATCEA.com - the need to standardise ATM procedures worldwide is possibly the most safety critical aspect we need to undertake at this time - urgently. ICAO set the standard which almost all countries have signed up to.

Eurocontrol (representing 41 European States) are working very hard on this aspect, in all areas of ATM. See eurocontrol.int/acas/ (http://eurocontrol.int/acas/) and go to the Bulletin pages for more info. Some good stuff there, based on the ICAO provisions.

One example: "One of the main messages for pilots to note from the ACAS training material is an emphasis on the need for pilots to ‘follow the RA’. Indeed, ICAO Doc.8168, ‘Procedures for Air Navigation Services – Aircraft Operations (PANS-OPS)’, Volume 1, Part VIII, Chapter 3, Paragraph 3.2 c) 1) of Amendment 12, states that: "in the event of an RA [Resolution Advisory], pilots shall respond immediately by following the RA as indicated, unless doing so would jeopardise the safety of the aeroplane".

Now wouldn't it be nice if the same could be said for the FAA, where they could standardise phraseology aspects on the other big issue, Runway Incursions, and move from "taxi into position and hold" to the ICAO parlance of "line up and wait". Everyone can be a winner.

Where theres hope.......:ok:

[atcea.com]

Blue, thanks for the references. Your reply is both thoughtful and informative.

Now wouldn't it be nice if the same could be said for the FAA

I agree the the US and the FAA are often the "culprits" in the efforts to standardize precedures. For many years, I believe our attitude has been, "We're the Big Boys - you wanna play, you play by our rules!"

Slowly but surely, that attitude is changing for the better. A good example is the FAA's acceptence of ICAO METAR and Airspace Classification procedures. I agree, however, we have a way to go yet.

ATC 24/7 (http://atcea.com)

[Scott Voigt]

Ok, to throw the bomb out there, just what is wrong with "position and hold." vrs Line up and wait???? Don't say just cause that is ICAO. That doesn't hold water with me. ICAO makes changes, just like that stupid one not so long about on changing to taxi to holding position.... Looks to me like they CAUSED more of a problem.

regards

Scott

[West Coast]

Prepare for thread liftoff...

[xe70]

I was working two days ago. Had two a/c, one climbing, one descending, nose to nose, with 1000 ft. vertical assigned ( the right way) and acknowledged. Passed traffic to both a/c. When a/c were approx. 5nm nose to nose, descending a/c, about 300 ft above cleared level, had an RA.
He announced TCAS climb, but only climbed about 100 ft, before announcing he was resuming ATC cleareance. The opposite direction climbing a/c did not get an RA. Climbing a/c was within 200 ft. of his cleared level.

I have been told by pilots, and assumed, when traffic is passed in similar cases, pilots decrease their vertical rates to avoid RA's.

[Phoenix_X]

Well, that's what most pilots have been trained to do, and it's what airmanship dictates. Unfortunately some obviously either forget or 'don't think it's important.' This seems to be the case here. Believe me, usually, passing the traffic info should prevent this.

It's odd that the other aircraft didn't have an RA though, if his TCAS system is switched on, the RA should be coordinated and both aircraft will get one.

[bookworm]

Ok, to throw the bomb out there, just what is wrong with "position and hold." vrs Line up and wait???? Don't say just cause that is ICAO. That doesn't hold water with me. ICAO makes changes, just like that stupid one not so long about on changing to taxi to holding position.... Looks to me like they CAUSED more of a problem.

I think you've just made the case for uniformity, Scott!

If everyone were to use the consistent phraseology of "line-up and wait", then the phraseology "taxi to holding position" wouldn't have been a problem. It's only because a significant proportion of the world chooses to use the non-ICAO "position and hold" that "taxi to holdin position" has a potential ambiguity to that subset of pilots. AFAIK, no one died uncovering the ambiguity.

[Jerricho]

Quick question for the drivers out there.

As we often run visual approaches here in the frozen north, I've lost count on the number of times I've passed the traffic on the one ahead, to recieve the reply "Ummm, we can't see them, got them on TCAS". Often makes me scratch my head as to what exactly is being implied.

[West Coast]

Nothing really. The TCAS will aid me in finding the traffic once you point it out. If as in the enroute environment we receive traffic calls well outside of viewing distance, I can keep myself updated to its position in relation to me. Given that, I don't really expect too many updates once I say I see it on the metal detector. Not required by SOP or practice, I guess you could make an argument it borders on excessive verbiage.

[Jerricho]

Thanks Westy. I know you and I have been here before about "situational awareness" and is the traffic on your TCAS actually the one I'm pointing out to you.

Excessive verbage is a good description, but I would hate to think a driver, having supposedly identified the pointed out traffic on the box, but never visually aquiring it then bases a "visual" call solely on the TCAS showing a target in the general direction (can anybody remind me of the azmiuth error please) correclating to traffic they can see there.

[West Coast]

Can't say as I have or have I seen anyone else call traffic in sight solely off the TCAS. Bad habit if they do. I would be afraid in some situations to do it. I could see doing it while not actually seeing the plane. This to be followed by something to the effect of maintain visual separation, the traffic will climb through your altitude. Likely cause the liar to fess up.

[bookworm]

Strikes me that visual acquisition is one of the least reliable of methods of keeping aircraft apart. It's all too easy to sight an aircraft, and then, because of the background clutter or change in motion geometry, lose it again. The most famous example (http://aviation-safety.net/database/1978/780925-0.htm) happened in West Coast's back yard. At least a contact, once on TCAS, tends to stay on TCAS.

Jerricho, what do you expect as a response to your traffic call? Are you using it for reducing standard separation? If so, do you accept "traffic in sight" as the criterion for reducing separation, or do you require a positive confirmation that the pilot of the following aircraft can maintain visual separation?

[Phoenix_X]

At least a contact, once on TCAS, tends to stay on TCAS.

Exactly right. Let me emphasise the 'a' .

Especially in the example given above, all that you know is that you've got some target on your TCAS. In Jerricho's example, that's not worth a thing. It could be anything, it could be totally different from the target Jerricho is trying to point out.

Jerricho, keep scratching your head. If the reply is 'we can't see him but he's on TCAS' I would advise you to ignore the last part. They can't see him. Period.

[Jerricho]

Bookworm, the response I am looking for one of two responses:

"Visual with the traffic", in which case the pilot will be cleared for the approach and instructed to follow the preceeding or;

"Not visual", in wich case I will vector them and be responsible for separation.

(Here in Canada, there isn't the same "reduced separation in the vicinity" as in the UK.)

As I mentioned, I am just curious as to the "got him on TCAS" call, as it means nothing to ATC, they're still going to separate you. Yet it's something I hear on a daily, if not hourly basis.

[Blue heaven]

Scott,

You didn’t read the full text of my message. US are a signatory to the ICAO convention. That to me, means that they agree to abide by the rules. If every signatory did the same, then we would not be in the silly mess we find ourselves in. Can you imagine a basketball or baseball game being played with differing rules being applied to either team? Mayhem!! And that’s what we’ve got.

So how best to fix it then?

Agreed, the ICAO change from “holding point” to “holding position” was a disaster. They however, have recognised this fact and have changed back. To me that shows an understanding of the issues and how best to resolve them. The “be reasonable – do it my way” approach is not appropriate here.

My hope is that such a lead can be followed by others, especially so in this case where on our side of the pond, the issue of runway incursions gives us so much grief, especially where the procedures are applied differently dependant on where the aircraft concerned comes from.

A touch of reasonableness I think should prevail, and the high horse attitude knocked over.

[West Coast]

"got him on TCAS"

I think some guys think they are doing you some sort of favor. The how and why of their thinking I can't explain.

[bookworm]

Especially in the example given above, all that you know is that you've got some target on your TCAS. In Jerricho's example, that's not worth a thing. It could be anything, it could be totally different from the target Jerricho is trying to point out.

But why is that different to a visual contact, Phoenix_X? It's just a shape in the distance -- it doesn't have an id-tag attached. At least on TCAS I'd know its level and probably have a better idea of its relative motion.

[Del Prado]

As Jerricho pointed out, there is an azimuth error inherent in TCAS (30 degrees iirc).

That's where the danger lies in using TCAS to 'identify' target aircraft.


But with the additional situational awareness afforded by TCAS, when the aircraft pops out of the clouds and traffic is seen, wouldn't the crew then be entitled to call visual ? provided they give a range and bearing to target aircraft that checked with the radar information ?

Does this example save RT loading by obviating the need for further traffic info ? after all, standard separation is still applied till the aircraft is visual.

Maybe 'we have him on TCAS' , in some cases at least, means we'll keep a look out and call you when we're visual.

[karrank]

"position and hold"

Maybe coz everybody else on earth will think about the instruction for a bit, than agree to hold where ever they are and report their position.

Back to the thread...

If the TCAS has gone off the controller has (probably) already farked it up (unless the other traffic is unknown to him/her/it). We have to be mindfull that if we notice traffic in unsafe proximity and attract the attention of one aircraft and take some sort of action the other may be reacting to the beeping on the dashboard.

This 5 degree bank thing is a worry, we currently take the above to mean vector like mad in such collision avoidance situations, and leave the vertical to the gizmo????

Our radar gear is designed to help us preserve separation standards, so it's ok to be about 6 seconds behind reality.

[Phoenix_X]

But why is that different to a visual contact, Phoenix_X? It's just a shape in the distance -- it doesn't have an id-tag attached. At least on TCAS I'd know its level and probably have a better idea of its relative motion

An excellent point :). Except for maybe seeing the a/c type and company, it's not much better than TCAS. I won't argue that!

However, the point remains that thinking you're clear of a TCAS target, may not
A) Mean you're actually clear due to the position error
B) Mean that you're clear of the aircraft that's conflicting with you

At least if you actually see an aircraft point (A) is solved. I agree (B) is not. Thanks for the heads-up! :ok:

[bookworm]

As Jerricho pointed out, there is an azimuth error inherent in TCAS (30 degrees iirc).

That's where the danger lies in using TCAS to 'identify' target aircraft.

The resolution of the clock code used for giving traffic info for visual acquisition is also 30 degrees. If you're really worried about the crew failing to tell the difference between 2 contacts that cannot be distinguished on TCAS, would you rely on the crew to get the right aircraft visually?

[Del Prado]

Bookworm, a valid point. I believe the problem arises when 11 o'clock traffic appears at 1 o'clock on TCAS. The clock code will always tell you whether traffic is left or right. (providing the controller doesn't get it wrong :oh: )

Can anyone else confirm the 5 degree bank limit when taking avoiding action ?

[West Coast]

"Can anyone else confirm the 5 degree bank limit when taking avoiding action ?"

Are you talking about action initiated by ATC or the TCAS? TCAS RA's provide no lateral instructions in the escape manuever, strictly in the vertical.

[Jerricho]

I have a similar question about banking/turning during an RA manoeuvre. Hammy will probably tell the story better, but was privy to a situation where a potential TCAS RA situation was about to unfold and the controller issued avoiding action involving a turn was given. Moments later "*** TCAS climb" came over the r/t and it looked as if the turn stopped. SOPs?

[West Coast]

Follow ATC instructions till overridden by the TCAS. Follow TCAS until overridden by the EGPWS.

[Miles Magister]

There is some excelent reading on the eurocontrol website.

Go to this link http://www.eurocontrol.int/acas/

Select Safety Information and scroll down to ACAS II Bulletins. There are 5 in all and well worth reading. I do not think you will doubt the wisdom of following TCAS once you have read them.

Miles

[Jerricho]

Follow ATC instructions till overridden by the TCAS

True, but what if in a turn and you receive a CLIMB RA?

[Giles Wembley-Hogg]

I don't know anything about a 5 degree bank angle limit for avoiding action, but on the 757/767 the TCAS assumes that the bank angle "will be less than 15 degrees for climb performance purposes".

Whilst we are talking about avoiding action (I admit that the following is mainly for the benefit of TMA controllers). When we are slowed to our minimum clean speed (close enough to holding speed as to be no different), a combination of a 30 deg banked turn and g loading and/or turbulence may result in a genuine stick shake warning. With this in mind, when you give an avoiding action turn, the actual turn itself will not be appreciably tighter than those you normally see.

The flap speed schedule only allows for "an inadvertent 15 degree overshoot beyond the normal 25 degree bank angle". We also get a "BANK ANGLE BANK ANGLE" warning when the bank angle exceeds 35 degrees. The up-shot of this is that to keep our heart rates to acceptable levels and to avoid bending the aeroplane we will not be ablt to turn much more quickly than normal in a TMA environment.

I actually had an RA not that long ago, and the concentration required to switch from calculating heights and ranges in order to follow a nice CDA to suddenly fly the aeroplane out of the (growing) red sector was quite stunning. All good fun, but I can categorically state that avoiding action in the horizontal plane is the only useful response from the ATCO and even then, with the flight director showing turn left, the controller saying turn right, the clearance requiring a descent and the TCAS requiring a level off - there was a lot going on!

G W-H

[ferris]

Giles- as an aside...

How hard was it to follow SOPs? Was the compulsion to follow the human (ATC) instructions strong? Did the controller get notified of what was happening in time to assist/stop assisting, or was he notified after it was all over? I'm just interested as the current system takes the controller out of the loop, without him knowing he is out (perhaps as it should be).

[45south]

Excuse my lateness in posting this but..........

If the TCAS units onboard the A/C "talk" to each other then they mus tx information can this info not be recived and displayed/intigrated with the radar display or is this to far fetched ????

This would enable the controller to be alerted to the TCAS RA and to " keep his mouth Closed??"

Comments its just an IDEA:confused: :)

[Jerricho]

I believe under new Mode S stuff that's being introduced, this will become eventually become available to controllers.

[Warped Factor]

With Mode S TCAS RAs can be downlinked and this is also a recommendation from the Uberlingen accident report.

It's debateable how useful this might be, given the possible timescales of a RA it is quite likely that by the time the downlinked info is on whatever display is put in front of the controller the actual event will be over and done with.

WF.

[45south]

From what I understand the information could be recived by an omnidirectional reciver and not subject to the delays of a rotating radar head.
How that info is intergrated is anyones idea it may only be a flahing light on the console!

so in real terms the delay in the controler being alerted in theory should only be 1-2 seconds after the A/c. :hmm:

If only pigs could fly????:uhoh:

[ATC Watcher]

45 South :
quote :-----------------------------------------------------------------------
If the TCAS units onboard the A/C "talk" to each other then they mus tx information can this info not be recived and displayed/intigrated with the radar display or is this to far fetched ????
This would enable the controller to be alerted to the TCAS RA and to " keep his mouth Closed??"
--------------------------------------------------------------------------------
Not that simple. Omni antennas do work, but with a range of 40 Nm max, many RAs are false RAs, and there are a few other problems.

Keeping controllers quiet is also not the miracle solution.

Remember that in Ueberlingen , and in all the other close calls resulting from a contradiction betwen RAs and ATC instructions, the ATC instructions were just BEFORE the RAs or at the same time....

Humans and automation do not mix well. The more I learn and known about TCAS, the more I think the only way out is to link it to the Auto pilot.
Until we come to that ( and we will one day) : Always Follow the RAs like your life depends on it , because it might.