Say Again? #9:
Maiden and Me
Every so often, a major radar outage makes the headlines. Flights are delayed, safety is compromised and some news reporter theorizes that it's time for a major upgrade to our national system. Just how bad is it, anyway? For an insider's view, read what Don Brown has to say in this month's
As I was rummaging around in my brain for some interesting stories to amuse pilots, I kept coming across the same thing again and again: Maiden. It dawned on me that much of my career has revolved around Maiden. From the day I walked in the door at Atlanta Center, to the present day, Maiden has influenced my life far more than I would have thought possible for a machine.
Maiden is an Air Route Surveillance Radar (ARSR). It's an ARSR Model 1 to be exact. That means it was made in the late 1950s. What a coincidence, so was I. Maiden sits right smack in the middle of the eastern portion of Atlanta Center (ZTL). It is the main ARSR site covering the area around Charlotte, N.C., with overlapping coverage provided by several sites along the borders of ZTL.
The first thing I remember about Maiden was my instructor showing me what it looked like on the old broadband display. I remember how primitive the display looked when compared to the computer-generated display we call narrowband or RDP (Radar Data Processing). The targets on broadband were all different sizes, depending on how far away from the antenna they were. I bet some were ten miles wide. The borders and airways looked like they were drawn with an Etch-a-Sketch.
One time, when the computer failed, the supes went from sector to sector handing out "shrimp boats." If you don't know what shrimp boats are, don't feel bad; I didn't either. I looked at the supe and said, "What do you want me to do with these?" He gave me a grease pencil and said to write the call sign of each airplane I had a strip for on a shrimp boat. I said, "You're kidding me, right?" He wasn't. Imagine trying to write with a grease pencil on something about twice the size of your thumbnail. Shrimp boats were the manual version of datablocks. They were clear plastic and you'd lay them down on the scope (back when the scopes would lay down flat) and push them across the scope, keeping pace with the radar target. Fortunately, the computer came back up and that was the last I saw of shrimp boats. Well, except for the card games where we used them as poker chips.
The only other time I ever saw broadband used was for weather. Controllers found the "raw" radar was a better display for depicting thunderstorms. During the thunderstorm season controllers would constantly switch from narrowband back to broadband to look at the weather. It's one of the reasons older controllers at the Center are more comfortable vectoring aircraft around thunderstorms than the younger controllers. We learned to compare the weather display on narrowband with the raw data from broadband, figuring out what the limitations were in the process.
My fall from grace into the abyss of being a safety rep started with Maiden. At one point in the mid-80s Maiden had a terrible ring around problem. No, not the kind Wisk would get out of your collar. "Ring around" is when a radar displays the same target over and over at the same distance from the antenna but on a different bearing. It forms a "ring" of targets around the radar site. Besides being annoying, it confuses the computer that is reading the data from the radar and displaying the datablock. It plays havoc with the ground speed readout also. So if the datablock wasn't jumping from target to target, the speed would change by 100 knots or more. Of course, this had to happen right where two major routes crossed or I wouldn't have a story to tell — or at least one less story to tell.
Anyway, there I was, a grunt controller with about five years' experience. The crisis after the strike of 1981 was over and the FAA had gone back to "managing" its controllers. The old guys recognized the signs, but I was blissfully ignorant. I started asking questions. "How do I go about getting this fixed?" I'd ask my supervisors. Time after time I was told they'd take care of it. The problem persisted. I started asking other people. The older controllers told me nothing ever got fixed — I was wasting my time. I kept asking anyway.
Then one day another controller overheard me talking about the problem and she asked, "Are you filling out the bullseye forms?" I didn't know what a bullseye form was. She explained that whenever they had a radar problem they'd document it on a bullseye form, turn it into their supe and AF (Airways Facilities) would work on the problem. Sounds simple enough. That was my first introduction into the concept of "two FAAs." What was common knowledge on one side of the control room was unknown on the other side of the control room. It was also my introduction to my best friend.
She gave me one of the forms and I filled it out the next time I saw the ring around. I gave it to my supervisor and asked where I could get another form. He reluctantly took it and said he'd find me another one. He never did, so I went over his head and got a stack of forms from his boss. That was my first adventure with the "chain of command." I filled them out whenever I could and turned them in to my supervisor.
Plans and Good Intentions
During this same period I was also helping form the National Air Traffic Controllers Association. At the same time, the FAA was redesigning the airspace on the East Coast. Many of you probably remember this project. It was called the Expanded East Coast Plan. Controllers had a lot of concerns about this plan, mostly in regard to training. Entire Centers were going to change, literally overnight, with completely new airways and procedures. Imagine waking up one morning and discovering every road in your town was different — and you were the delieveryman.
One thing led to another and I wound up being volunteered to represent NATCA for Atlanta Center in a meeting with the Secretary of Transportation and the the FAA Administrator. As you can imagine, that was quite an education. I'll spare you the details. Suffice it to say the EECPlan went into effect without our concerns being addressed to our satisfaction.
On the very first day of the Expanded East Coast Plan, about 40 miles from the Maiden ARSR, COA458 and COA703 passed "about 0.6 miles" from each other at Flight Level 350. I loved that "about 0.6 miles" phrase. My sources said it was "about" zero miles and zero feet on the altitude. This was pre-TCAS mind you. You can read the short version on the NTSB website.
In that I had an interest in the event, I took the trouble to order the long version of the NTSB's report. As any of you that study anything about aviation safety know, there is rarely just one cause in an incident or accident. That's mostly due to those redundancies that safety people are always droning on about. COA458 was being maneuvered to meet an unfamiliar restriction for the EECP. The aircraft it was being maneuvered around had gone NORDO for a few minutes forcing the controllers to go to "Plan B." The coordination for Plan B got fouled up. The icing on the cake occurred when the datablock on COA458 displaced by about 15 miles south immediately prior to this NMAC (Near Mid-Air Collision). The computer was chasing one of the "ring around" targets. Because the computer was confused about the whereabouts of the target, the Conflict Alert feature never went off until the controller reinitiated the track on COA458, telling the computer where the aircraft really was. By that time, it was too late.
Remember the bullseye forms I'd been filing? They'd been finding their way into the "round file." Unfortunately for my supervisor, I'd been making copies before turning them in. Whoops. My fate was sealed. When I made the copies available to the NTSB, I graduated from nuisance to troublemaker. Oh well. The NTSB still ruled it controller error and the FAA solved its problem. It doubled the length of the Radar Trouble Report (bullseye) form. Funny thing, a controller working a pile of airplanes doesn't really have time to fill out a lengthy form. So the form doesn't get filled out very often. Never underestimate the cunning of a bureaucrat.
The Blind Leading the Blind
Perhaps the most significant effect Maiden has had on my life is the fact that it doesn't work on a regular basis. I would assume that this has something to do with Maiden being as old as I am. I require more maintenance than ever too. Much of maintenance used to take place during slow periods, i.e., the midnight shift. It wasn't as noticeable then. Now, due to those good old government cutbacks that politicians are always ranting about, there's no money to pay the night differential and overtime. That's assuming that AF even has enough technicians left that they actually could assign someone to a midnight shift.
Before I wander too far off course, what all this means to you is: "Radar service terminated." The best response I ever got to that was, "Center you can't terminate me I'm IFR." Now that was funny. Anyway, what all this means for me is that I deal with non-radar issues more than most controllers east of the Rockies. Atlanta Center just happens to own a big chunk of airspace in the highest section of the Appalachian Mountains. Those mountains block off the other ARSR sites that would provide some overlapping radar coverage. When Maiden is taken out of service for repairs, one of the sectors in my Area goes non-radar.
For me, it's both a blessing and a curse. Working non-radar on a regular basis has given me a lot of insight into various issues in ATC that I would not have otherwise. Working in a high-profile facility like Atlanta Center means I can get those issues in front of a lot of people that controllers in Salt Lake Center can't. The cursed part takes place when I say something like, "Non-radar is the foundation that the house of Air Traffic Control was built upon." It seems that hardly anyone — pilots, controllers, managers or politicians — understands what that really means. I've uttered many a curse (in frustration) while trying to explain it. It's like trying to explain grits to a Yankee.
Let's Make a Deal
The worst Operational Error I ever had was during a Maiden radar outage. The weather wasn't that good, just good enough for us to get into trouble. There was an overcast layer at about 2,000 feet AGL. Pilots could depart VFR but arrivals needed an instrument approach to get under the clouds and land. As you can imagine, shooting full approaches (with procedure turns) quickly backed things up.
The worst problem though was with the VFR departures of aircraft requesting an IFR clearance. I think I had about five of them inside of 15 minutes. The one I remember the best was a Mooney off of MRN, heading to the northwest. I'll spare you from having to look at the charts. The Minimum Enroute Altitude at MRN is 3,600 and rises quickly to 7,500 northwest of the airport. There's only one way to get out of MRN, IFR, in a non-radar environment: follow the standard instrument departure to the NDB and then to either SUG or BZM VOR. Neither one of those VORs is to the northwest of the airport.
Let's pause right here and let me ask you a question. Have you been paying attention to everything I've written so far? Not just in this article, but in all the others I've been writing for you? Do you remember this from "Say Again? #2: The GPS Mess"?
d. Area Navigation (RNAV).
1. Random RNAV routes can only be approved in a radar environment.
I told you I'd bring it up again. Can you see the problem?
Here's a guy who departed VFR expecting to pick up his IFR just like he always does. But in this case, he's only got 2,000 feet of VFR airspace and that is shrinking as he heads into rising terrain — and I don't have any radar. Now then, what could make this situation worse? Do you remember this from "Say Again? #1: A Wing and a Prayer"?
In my opinion, frequency congestion is the number one bottleneck for en route ATC.
Hindsight being 20/20, the best thing this guy could have done would have been to return to the airport, land and then call FSS for an IFR clearance right off the ground. But he didn't know that. And the only way to explain it to him is to clog up the frequency even more while we play "20 Questions." Do you remember this from "Say Again? #3: ATC 101"?
And despite what you may hear, you can't tell if a controller is busy by listening to a frequency.
Non-radar just doesn't sound very busy on the frequency. You just can't work that many airplanes non-radar, so it never will. I don't know if you can imagine how much coordination is involved in non-radar, but I know you can't imagine how busy a controller's brain is — not unless you've ever done it before.
So what does this pilot do? He starts to explain to me what a fix he's in. I realize that of course. Even if I didn't know it to start with, I've already heard the story from four other pilots that did the same thing this guy did. That's how I got into this mess. As I'm dealing with the situation, two or three other aircraft call at the same time. This is where I made the (nearly) fatal mistake, "All aircraft this frequency standby, King Air 23456 are you on the frequency?" He was. And so was somebody else I wasn't aware of.
I really wanted to talk to that King Air bad. I had to figure out a way to get him up and out of the way of the approaches. I knew he was going to get hung up under the holding pattern at the VOR (the same VOR serves the inbounds and the outbounds during non-radar) but if he could cross 10 west of the VOR at 6,000 I could run another arrival at 5,000. I got what I wanted. I got him up to 6,000. Of course he let me know he was "looking for higher" and guys in the holding pattern wanted to know "what number are we?" and the Mooney continued to let me know what a sticky wicket he was in. So when I heard, "Cherokee 34567 is over BZM at 8,000" I had to ask, "Who was that?" Quickly followed by, "What'd he say?" and "Oh &%#@ !"
You probably already guessed it was one of the guys I told to "standby" earlier. I didn't know it was him of course. I didn't even know he was in the sector. You also probably guessed that I had someone in the holding pattern over BZM at 8,000. I did. There's nothing like issuing a couple of "immediate" clearances non-radar to ruin everybody's day. Of course, the frequency gets amazingly quiet once you do. I'm not trying to make excuses by the way. It was controller error. The controller (me) failed to maintain control of the frequency and therefore lost control of the sector. It's something that you might want to keep in mind the next time I threaten to bite your head off if you key that mike with one more request.
Let's see...where were we? Oh yeah, Maiden. The next way that Maiden started affecting my life was with a variation of the old ring around problem. It started generating split beacons — not split side by side mind you, but really split. The site would paint a target on the southeast side of the site, say a 150 bearing at 30 miles, and then it'd generate another target 180 degrees opposite at 30 miles, on a 330 bearing. They were far enough apart that you wouldn't associate one with the other. The false target looked just like any other target. You would see a dozen a day at random times. They'd last for three-five turns of the radar and then disappear as suddenly as they'd appeared.
Being the safety rep, I started filling out those new and improved Radar Trouble Report forms again — all 25 items. I went to my counterpart in the FAA and complained about the split beacons. I filled out NASA Aviation Safety Reporting System forms on the site. I even tracked down the radar engineer and spoke to him about it. At every turn I was reminded that the wheels of government turn slowly. The problem was being worked on. Give it time.
I may have mentioned it before but it's worth repeating, controllers are an impatient lot. We're expected to get it done — right now — and to do it right, every time. We, in turn, transfer that attitude onto the people we deal with. I'm sure pilots have noticed this attitude on occasion. The FAA's managers have noticed it too. It's one thing to notice it on the radio. It's quite another when it's in person, toe-to-toe. I step on a lot of toes as a safety rep.
Seeing a Ghost
As this problem dragged on for weeks I go about my business as usual. One day, I was sitting at a sector, just minding my own business, doing whatever it is controllers do when they get bored. I figured it was time to switch a particular airplane to Approach. I look up and "N12345 traffic twelve o'clock your altitude." Even as I keyed my mike I knew I was too late. They merged before I'd finished speaking. Same altitude, center of target to center of target. They were going to hit.
I just knew it. That's the third time in my career I've seen two IFRs get that close. (I could tell the other was IFR by the beacon code.) It's the first time I was ever working one of them. It takes about 10 seconds for the targets to come apart. It's the longest 10 seconds of a controller's career. I'm sure I scared the pilot just by the tone of voice I used. He never saw the airplane by the way. It's hard to see what's not there. It was just Maiden putting up another false target. You can read about a similar incident in NASA's ASRS database. I filed that report about one week before this incident.
You've Won a Free Vacation
After a visit to the doctor and a three-day weekend at Zombie Land (what do they put in those pills anyway?) I showed up at work ready to get back on the horse that threw me. The Flight Surgeon wasn't so sure though. He wanted to talk to me, so I got him on the phone. After making sure I'd been off the drugs long enough to be legal, he started asking questions about the Maiden radar site. Flight Surgeon — Radar. Hmm. Maybe I wasn't all there. I couldn't figure out why a doctor was asking about a radar. Whoever clued him in about the radar site forgot to clue him into the fact I was the NATCA Safety Rep. Once I informed him of that, he decided that my concerns about a radar site really weren't a sign that I was mental and therefore weren't relevant to restoring my medical certification.
One day, I was discussing my latest tactic concerning Maiden with a pilot and he asked the question, "Just how long has this been going on?" I told him 15 years. Then he asked, "Just how long do you think it will take you to get it fixed?" Ouch. Point made. I don't have another 15 years left in the FAA to work on it. But maybe I won't have to.
The world changed in many ways on September 11th. One of those changes is that people better understand the significance of having an ATC surveillance system that doesn't depend on participation by the pilot. Hopefully that will translate into improving the reliability of our radars. The best way to do that is to replace the ones that are obsolete and make sure we have the technicians to maintain them.
It will be expensive — very expensive. The reason I'm telling you about it is that it's your money. A privatized ATC system could care less about primary targets. They don't generate revenue. An up-to-date radar system that includes primary/surveillance capabilities will have to be funded by your tax dollars. I can hear the "Gulp" from here.
There's a couple of bright spots to look at. First, as expensive as a nationwide radar system is, it's cheaper to do it on the ground than it is in the air. In other words, it's cheaper than flying AWACs all around the country 24/7. You're already paying for that. Secondly, the primary portion of our radar is where we derive our weather (precipitation) data. They've made a few improvements in that technology since the 1950s. It'd be nice to incorporate them into our radar sites.
Now, you are paying attention, right? It was just last month in "Say Again? #8: Air Traffic Chaos" I was telling you that our weather radar wasn't very reliable. Now you know another reason why it isn't very good. Here's yet another reason: Some of our sites don't have primary (some people call it search radar) capabilities. That's right; we don't get any weather data at all off those sites. I'd tell you where those sites are, but the FAA would probably try to fire me for violating national security. It's like the TFR issue. You have to avoid them but we don't want to tell anyone where they are.
Controllers and Comedians
I hope some of you are really thinking and wondering where I'm headed with all this. I know I've said I try to keep my readers amused (and I do) but I'm a safety rep., not a comedian. I have a purpose. Trust me, I have enough to do without writing this column and AVweb doesn't pay me enough to endanger my day job by flirting around national security issues.
Some of the "higher skills set" in the FAA had a plan to save the taxpayers millions of dollars. They wanted to decommission primary radar at all the long-range radar sites. Everybody (all the alphabet groups, AOPA, ALPA, NATCA, etc.) told them it was a bad idea. This argument went on for years. It ended on September 11, 2001.
... or so we thought. The rumor mill is cranking up again that we may not get our long-anticipated upgrades for the radars. They may try to drop primary radar again. No, they're not that stupid. It's a budget ploy. Who's going to pay for it? DOD or FAA?
Why am I telling you? Because it's your money. That's why. And just in case you happen to talk to your Congress person anytime soon, you might like to have a say about how your money is spent. It's still your money, whether DOD or FAA spends it. Upgraded radars with primary capabilities can provide you with better weather data, the ability to paint the bad guys and the ability to paint the good guys when their transponder quits because they're on fire and need a vector to the nearest airport. Is it worth the investment? You decide. It's your money. It's your safety.
Have a safe flight!
Facility Safety Representative
National Air Traffic Controllers Association