Like it or not, that sounds like normalization of deviation. I mean, they wouldn't call it a "collision" alert if that wasnt the possible result of not doing something to avert it. The idea that "oh, they have them in sight" so it can just proceed w/CA flashing in red needs to be looked at. While an alarm is going off, your only assurance (as the controller) that there won't be a collision is because one pilot claims to have the other aircraft in sight. At the very least, it should be looked at in the context of people asking for "visual separation" and operating in extreme proximity to another aircraft.

 

That's one of the bigger issues. If you've ever been up in a tower, TRACON or ARTCC, you realize how much of this is the joint operation of the pilots and controllers together. Newer pilots sometimes think that ATC "is watching out for them", but it takes a significant amount of interaction and action on the pilot's side to have the desired level of safety and in many cases, no one is watching out for them, they are on their own. The system only works when both parts of the system work adequately.

 

I agree, but that's the system we have. CA is constantly going off during visuals into SFO. The entire airport wouldn't be able to exist if they had to treat those alerts seriously.

I think this is just a cultural difference between ATC and pilots. ALPA has pushed safety culture for pilots miles ahead of controllers. Multi-tasking in a cockpit had been pretty much illuminated, but good luck trying to tell controllers they shouldn't be multi-tasking.

 

If that is the case, that is a problem. They really need to redesign the software to minimize the nuisance alerts. Like exclude all aircraft on the ground, perhaps. Also, make it better at predicting conflicts in the air. I am guessing the designers of the system had a different idea of how the system would function and be used compared to how it is actually used.

 

Radar, FDR Disagree On DC Black Hawk Altitude

Investigators are trying to reconcile conflicting data.

https://www.avweb.com/aviation-news/...hawk-altitude/

 

Build a real set of parallel runways would be great start--we had controllers ignoring CA and pilots, under FAA approval, turning off a safety tool, TCAS. What that about normalizing deviance?

Yes I know know, the NIMBY enviros will fight it.

 

Very odd, mode C data is uncorrected on the airplane, so the pilots can't screw that up with a wrong altimeter setting.

The mode C is corrected for QNH on the ATC scope, so QNH correction is independent between controller and pilot... this has resulted in a few saves in the past.

*Assuming* the RJ was on path, since both mode C/ADS-B and it's FDR seem to agree on that.

That means the helo was high. ATC got bad mode C. Either pilots had bad altimter read out, or they just screwed up. But if mode C was bad, that might imply an issue with static system calibration, that affected both mode C raw data and altimeter indication?

If the pilots just set the wrong QNH, the controller should have seen the real altitude.

At night I wouldn't necessarily expect the helo pilots to be able to visually differentiate between 200' above water vs 325' above water. Especially if they were on NODS.

With all that said, if they called traffic they were responsible for not hitting it. Static system might be a contributing factor.

 

No kidding… spend some of that infrastructure money on a longer 01R over the swampy river… congress holds the purse strings and want DCA expansion right?

 

The ATC playback I saw showed the Blackhawk rising from 200' to 350' and also showed the CRJ decending and meeting the Blackhawk at 350'. So what was showing wrong?

 

I'm assuming you didn't see a phone video of the controller's actual scope?

What kind of data was it? ADS-B? I'm pretty darn sure that's GPS derived "altitude" not barometric. So independent from mode C and altimeter. That might actually check with the static system calibration hypothesis.

Don't know if the controller was referencing ADS-B or mode C. How do they deal with conflicting data there?