ATCC Controllers' Read Binder...

NOTAMS, FAQs and other info for users of ATCC

October, 1998

#1: Speech Enhancement Add-On Coming This Month

A Speech Enhancement add-on will be posted to this web site toward the end of this month (free download, around 8MB), adding essentially two more voices to ATCC ("variations on a theme"). Included will also be a beta version of the speech recognition driver to interface with the IBM series of speech recognition products (such as IBM ViaVoice). The speech recognition part will probably be more of a curiosity than practical (it seems to make the most errors at the worst times) (and don't buy ViaVoice just for this beta version), but the two new voice modifications do help enhance things. It's free, anyway!

The rumored New Sectors, more from New York, Chicago and L.A. Centers, are still due out by the end of the year, and no new news on the next big, perfect version of ATCC except it's still being planned.

#2: Working the Same #^&@ Sectors

The most experienced ATCC user has no more than two years experience with the Sectors (beta testing started around October of 1996), so if ATCC sectors ever seem "routine," imagine the situation of most controllers stuck with the same sectors for (on average, probably) 10-15 years. Ten years of working Sector 82! Full-time! What situations could you possibly see after even five years (or two years? One year?) that you haven't seen before? Detroit arrival...hmmm...descend to 240. ORD departure. Nobody running parallel, climb all the way up. MKE arrival, look over here, up there, nobody there, descend to 280. Ho-hum, tap your fingers, same routes, same airlines, same callsigns. That's the truth for most of it, well over 99 percent, that you've seen it before, it requires no thought, and the clearance is automatic. Somewhat like driving a car, you have to be alert, but are you really putting thought into it?

THAT is how the system is designed! You need to be able to save your "thought" energy for when you really need it, when the deviations start, or when a huge mess gets handed off to you from somebody sleeping at the next scope. If you had to thoroughly analyze everything at every moment (such as if there were no preset routes, just random flights), you would quickly exhaust yourself and burn out before you even got a year on the job. This is why the system is designed to (hopefully) be as predictable and routine as possible, as much as possible--to save you for when the inevitable unplanned chaos does occur. You'll be ready for it, then.

#3: When the Routine Fails

"Failure" means a Deal, of course. One category of "stupid" errors (and most of them are, in retrospect) are situations you have handled hundreds, or thousands of times before, and all handled the same way with the same, deal-less results. This one may have been just a tiny bit different, but as part of the Routine you lump it together with the standard scenario it most similarly matches--except it is just different enough not to work, which you don't realize until too late.

One example, for sector 66, is an oceanic arrival, and a JFK jet departure. Oceanic arrival and JFK departure??!! Correct, you just climb one and descend the other, and they miss each other by 20,000 feet. They've never been a factor in the history of sector 66 and never will be! As part of the Routine, they should never even be taken into consideration. Don't even waste your time checking the two for possible conflict, because you can go 20 years and never have them come together.

Until...the routine is slightly altered, slight enough that while you still see it is different, it doesn't by itself change things much. Then another part of the routine is slightly altered, and another, until they cascade exponentially and suddenly you have a situation. Maybe sector 22 calls up and asks if the arrival can enter your sector descending to FL180 instead of at 180, because he got stuck up high for traffic. Sure, you say, 18, descending to 18, what's the difference? He's still off your scope, anyway, and you never have traffic down at that part of the sector.

Five minutes later, NY Departure calls up and says reference that departure he's flashing at you, he's going to be stuck at 8,000 until he passes that Cessna, or maybe could you let him take him 5 miles south for the climb, then back on course? Sure, you say, why not? Ho-hum, three aircraft on your scope, cross CAMRN at 11,000, blah blah blah.

The departure checks on, and by rote you look up and down the airway, nope, no traffic, so climb all the way up. Ho-hum... then the oceanic arrival checks on, you look for who he's tied with, nope, no other JFK arrivals; look for props cutting across on the airway at 14 or 16,000, nope, no props, so descend to 12,000. Roger. Same old routine, year 14 of working the same procedures at the same sector!

A large proportion of deals have that moment of initial recognition that something is just a little wrong--just a little concern, but not much, with maybe a polite request to get back on course, or "could you expedite your climb a little," and in this scenario it might be asking the JFK departure to "tighten up your turn a little" back on course when they first appear south of where the last 2,000 JFK departures usually go. But no big problem, in and of itself.

Then the oceanic arrival reaches OWENZ, and as a foreign carrier not familiar with the area, asks, "Confirm, direct Canarsie at this time?"... Uhhhh, "Negative, sir, I need you on course, and cross CAMRN at and maintain 11,000."

"Ok, CAMRN at 11,000, uhhhhhhh...." and the backslash hesitates as you figure the crew is scrambling to figure out where CAMRN is. Blip, blip, the backslash is still moving direct CRI.

"Egypt Air 880, verify you are direct CAMRN, that's C-A-M-R-N."

"Ahhhhh, that is affirmative, turning direct CAMRN now, thank you."

Yawn. This is a tiny little break to the routine, but happens in every sector, every day, actually. But then combine it with another, the JFK departure just a little south, then noticing they are making a nice wide turn, too, and now it is rapidly becoming a situation.

"Uhhhh, Delta 462, tighten up your turn please....in fact, turn left heading 090."

"....OK....We're a little heavy today but we'll do our best....heading 090, DAL 462."

Another radar hit...hmmmm....something's not...

"Make it heading 070 now...no, make it 040..can you tighten it any more?!"

"We'll try."

Too late, BLINK-blink-BLINK-blink-BLINK and your nice 14 year Routine is broken!

This author's deal at real-life L.A. Sector 4 (though later classified as a "non-event") fell into this category, "a few minor changes to the Routine that multiplied suddenly."

#4: Non-Routine Deals

Another category of easily avoidable deals occurs when there is no Routine, and the controller doesn't realize he has to snap out of his monotony and think up a solution until the pilot mentions something about his TCAS going off (anti-collision "radar"), or of course the BLINK-blink-BLINKing of the snitch.

One example of this might be a Lear Jet climbing out of ORD in sector 82, toward ELX and along the north half to CRL and somewhere east. In a perfectly plausible scenario, while climbing out of FL240 and still west of ELX, he might say he has a "request." Things might be slow, with only a couple other aircraft on frequency, so the controller would say "go ahead."

"Dispatch has asked us to change our destination to Indianapolis, so we'd like to go ELX, then GIJ, OXI, direct IND."

ORD to ELX, to GIJ? Hmmmmm, that's certainly out of the norm, but sure, no traffic. "Cleared to IND via direct ELX, direct GIJ, direct OXI, direct, and, uh, you'll probably need to descend soon anyway so level it off at your discretion, then pilot's discretion descend and maintain FL240."

Non-standard phraseology for a somewhat non-standard routing, but after working sector 82 for fifteen years, controllers will get that way, and it would certainly be understood by any Lear pilot out there.

Crossing ELX at FL250, the Lear begins its 320 degree loop around to GIJ, as another ORD checks on, climbing toward GIJ and J146. Any certified sector 82 controller would certainly note the non-standard situation going on, and realize the conflict, but gee, how many times do they work jets in 320 degree turns climbing then descending toward GIJ? This is where errors can be made, when there is no routine to base decisions on. The 737 is climbing really well, and the Lear is descending already below FL250, and can easily be descended further with a call to the sector below...hmmmm...it should work, but then again...hmmm...

And all the "make a decision and make it work" training comes back at this point, so the controller makes the decision to go for it, but it just doesn't work. Conflict was recognized, but action taken was inadequate to ensure separation is one of the clichés on FAA Operational Error reports, for situations such as this. With no routine to fall back on, it is sometimes hard to get into "have to think about this" mode when you've been coasting along in Routine for hours or days.

Controllers are aware of all of these pitfalls (they're some of the first things mentioned in training), but still blunder into them!

#5: The Media and Deals

As has been mentioned in a previous newsletter (and still true), deals happen "all the time" (about 3-5 a month per Center), and while it is open public information, they aren't publicized. Sometimes, though, the media finds out, and must somehow report the facts and describe all of ATC (and probably a complex situation) in a few paragraphs or a 15-30 second blurb on the evening news. The Associated Press ran a story on a recent deal at L.A. Center (south of sector 19), which the L.A. Times headlined "Pilots Swerve to Avoid Collision." The reporter (smugly?) computed they were "8 seconds" away from collision, based on "1,000 MPH closure rate"...yea, if they were head-on, but....

...they weren't head-on, and the pilots didn't swerve to avoid collision, it was a 3-mile "backshot" deal due to a weather deviation (see #4 non-routine deals above!)...so yes, it was less-than-minimum separation, but how do you explain it to the media (and the non-ATC world?)

Also complicating that story was the controller union (knowing how the media reports such things) mentioning understaffing, and that there was "only one controller" where there "should have been two." Should there have been two? There was no D-side, sure, but would a D-side have prevented it? Maybe, or maybe having D-sides all the time makes things less interesting for those controllers and they tune out sooner, watching the clock for their next break. As usual, it's a more complex situation than can be explained in a few paragraphs of a newspaper, so it sounds much more sensationalistic than it is.

Another example was a recent newspaper article that mentioned how controllers "reserve altitudes" for the President when Air Force one comes through a sector. Wow, make way for the President! Center controllers have to laugh at this (they do shut down airports for A1, but not altitudes!). Evidence in the article actually pointed to some controller explaining that if Air Force 1 is flying along at 16,000 feet, "these departures" would only be climbed to 15,000 feet...(ummmmmm, routine positive separation?!)...so yes, it is kind of like "reserving" 16,000 feet just for the President...but not really.

ATCC users should be able to tell what is really going on when the media reports on ATC, too... it's usually (always) much more mundane (and safer) than they describe.

#6: D-Side Fun

Speaking of mundane, D-sides ("handoff" controllers) have to struggle to keep their duties interesting. When there were large numbers of controller trainees in the pipeline (prior to about 1992), D-sides were staffed by eager newbies who actually did find it a challenging job, because of course it was all new and mysterious. But once those trainees got signed off, D-sides had to be staffed by regular, fully certified controllers. Why work somebody's D-side when you can just work the whole sector yourself? Very boring. There is also a conventional wisdom that having two controllers at the sector is "better" (despite now having somebody to talk to and get distracted by), so all controllers need to know how to make the best of things when inevitably assigned to a boring D-side.

Duties vary by Center, but the main responsibility is handling coordination with other sectors. Except some sectors don't have to coordinate much at all, so D-sides may or may not have much to do. The big duty usually ends up being to keep strips in the strip rack sorted by time (0243 comes before 0242? Or after? hmmmmm), which isn't much fun at all, so some Centers allow other things like handling incoming calls as well as outgoing (thus having the power to approve requests, like "can I give MSR880 to you descending to FL180 instead of at FL180?") which takes some of the load off the R-side, but means the D-side has to pay attention to what is going on. And/or they will allow D-sides to take as well as make handoffs, which gives the D-side even something more to do but can be a little surprise if it is busy and the R-side has no idea who is checking on because he never took the handoff and hasn't gotten to that part of the scope in his scan.

Keeping things interesting, despite the need for routine and sameness, is somewhat a concern of the System, but mostly it falls to individual controllers. Just as some people don't mind driving long distances, some controllers don't mind working the same sectors for 20 years, while others get stir-crazy after just a couple. This is why the FAA put in a clause that a controller may go to the front of the transfer-request list to move to a different facility--only after ten years at that facility! So to be faithful, ATCC users should work the same ATCC sectors for at least eight more years, to get the proper "feel" for what goes on.

(We hope to come out with the new sectors before then, though!)

END

The Read Binder is updated at the beginning of the month. All information is for use with Xavius Software's Air Traffic Control Center^TM only, is the opinion of the author(s), and does not necessarily reflect the policies or practices of the U.S. Federal Aviation Administration or Federal Aviation Service. Send your questions or comments to support@xavius.com and we'll be glad to help!