Article: 13127 of rec.aviation.misc Newsgroups: rec.aviation.misc Path: newshost.ncd.com!ncd.com!olivea!spool.mu.edu!agate!library.ucla.edu!csulb.edu!csus.edu!netcom.com!jclark From: jclark@netcom.com (Jeffrey A. Clark) Subject: Re: NBC-TV report on TCAS to air July 14th (long) Message-ID: Organization: NETCOM On-line Communication Services (408 261-4700 guest) References: <1994Jul9.052132.5553@peck.com> Date: Sun, 17 Jul 1994 01:06:16 GMT Lines: 115 cprice@chuckles.broadvision.com (Chuck Price) writes: Ok, First, let me start out by saying I'm a controller in Southern California, and second, let me get the "opinions expressed are not the FAAs" out of the way... > The upshot of the segment was that near-misses are occurring as a result of > TCAS (3 this year if I recall), and there are people out there who say it > is just a matter of time before two airliners fly into one another while > the pilots obey the TCAS warnings and "ignore" the controller. This is a real concern of controllers. What remains to be seen is if in fact TCAS really WOULD put two planes together of course, perhaps not, but it makes us nervous. Now I'll be a rare breed of controller and concede that controllers DO make mistakes, including nasty mid-air-collision- types of mistakes, so TCAS can't be all bad if it catches a controller's mistakes. I would presume that in fact it has saved lives by catching controller errors. What would be interesting to know is what will the final equation be, in terms of lives saved/lives lost due to TCAS. The Seattle incident that was mentioned is called a "crossing manuever" by the guys and gals who programmed the TCAS logic. (to quickly explain, TCAS takes two planes that are about to level off 1,000 feet apart, and thinking that they are going to collide, decides to separate them. Sometimes it TELLS them to level off as the controller had already instructed, but sometimes due to the rates of climb/descent of the aircraft involved, decides that if the climbing aircraft climbs even FASTER, and the descending aircraft descends even FASTER, then they would miss by a greater distance, than by using the "level-off" method. Supposedly, this algorithm is going to be modified and will be programmed OUT of TCAS by December. The "crossing manuever" glitch is not the largest concern that most controllers I know have. The two largest concerns that I've heard center around terrain avoidance, and non-transponder equipped aircraft. TCAS has no way of knowing when it is instructing an aircraft to descend into a hill, and I would warrant that few airline pilots know the height of EVERY mountain in the area (like I am required to know....) An advisory to descend like the one in the show could be disastrous. (I suppose though that Radar Altitimters might be able to give terrain proximitity warnings, but I'm not sure if they're required in all larger aircraft like TCAS is, nor am I sure how well they could prevent a terrain incident). The other concern for us is that TCAS has no knowledge of non-Mode-C equipped aircraft. True, MODE-C is required near all TCAS, and inside Class C airspace, but a large portion of the approach is outside these areas. Furthermore, you would be shocked as to how many aircraft operating inside Los Angeles' Mode-C Veil do not have valid or operating Mode-C. I have already witnessed a TCAS "false alarm" where a turbojet climbed off an approach into an aircraft 1000 feet above him with no transponder. >Also, I am suspicious about how eager the ATC union was to bash >TCAS. While it may not be true, I couldn't help wondering if the ATC >union views the introduction of systems like TCAS, which infringe >on ATC's job, as an economic threat. Indeed, if the Rutan ATC system >(computer based, distributed processing, with little if any human >ATC involvement) were successfully installed, would they be for >or against it? I really don't know, but since the segment was so >unbalanced, with about 99% sensationalism and 1% data, I couldn't >come to a conclusion. Well, I'm not in the union, but I don't believe that even they perceive it as an economic threat. It is unlikely that even with a Rutan type system that controllers would become extinct as a species. At the present, there are too many variables that can't be anticipated or programmed for. And in any case such replacements would probably not be for a long time. What you will find true of controllers is that every single one of them that I have met, regardless of whether they are great controllers, bad controllers, good ones or grouchy ones, all of them care intensely about the separation of aircraft, and nothing scares us more than seeing two planes that we had separated go towards each other. It's more than normal fright, for a controller, it's a bone- chilling highly emotional soul-wrenching sight, and a lot of our reactions to TCAS may in fact come from the gut - it will have to earn our trust over time before we come to like it. Controllers also have a custom to protect the "herd" and when a fellow controller makes a critical mistake, or has lost the ability to do their job correctly, they are ostracized and in many cases, forced out. For a controller to see TCAS make a mistake may generate the same response - a judgement is made that 'the box can't control traffic and can't be trusted, which may be premature. Finally, you may have noticed that each sector you fly through isn't controlled by a committee. There is only ONE controller for that sector, and he is ultimately responsible for what happens there. When we train new controllers, the first thing that must be done is teaching the apprentice how to think almost exactly like the instructor. Only when the apprentice has been "brainwashed" into the habits and thinking of his mentor will he be permitted to work extremely busy and complex traffic because there must be an _overall strategy_ involved to make a situation work. The Airspace system is so incredily complex, and what is demanded of controllers is so site-specific (you need to give a particular airport so much space between aircraft, only certain types of planes on this runway please, noone can fly near this military base, etc etc.), and the streams of aircraft flying through each other can be so tangled that TCAS may never be able to comprehend it. It may resolve the first "conflict" it sees, (and it may even be a legitimate conflict), but it's resolution could conceivably put the jet in even greater peril in a blink of an eye. Another controller will not share my strategy, just like no two players play chess the same way. If a second controller takes over and makes a bad move, the whole game could fall apart, and TCAS is like a second controller. I LOVE the fact that it issues traffic to aircraft, and I'm not particular peeved when pilots tell me about the calls they're getting... I am not so proud that I can't admit that I forgot to call traffic to you. But when I tell you that a plane is not a conflict for you, and the box overrules me, and you believe the box and not me, you may end up in even worse trouble than you are currently in. Honestly, it's such a complex problem, and such a mixed bag, that if I were the FAA I'd make the same decision - keep it, and use it like it's being used. But it's scary nevertheless. -- jclark@netcom.com "Welcome to the game that moves as it plays..."