IFR Approach Debate

One significant difference is the term MSA. In the US, it is the the acronym for the "Minimum Safe Altitude" and may only be used in an emergency. It is not an operational altitude on an approach and an approach clearance does not authorize descent to the MSA altitude. In Canada and other countries, MSA is the acronym for "Minimum Sector Altitude" and it may be used when cleared for the approach. This is from the current Canadian AIM:



In the US, the TAA (Terminal Arrival Area) is considered as a large RNAV random feeder route from the enroute environment with an altitude specified and the pilot, when cleared for an approach, is permitted to descend to the charted TAA altitude, once inside the TAA on receipt of an approach clearance, unless otherwise restricted by ATC.
Thank you for that clear explanation. So if I'm IFR in the US, I'm already within a TAA, and I hear "cleared for the approach" with no other instructions attached, then I can immediately descend below my previously-assigned altitude to the appropriate TAA altitude?

(In real life, of course, I'd still verify with ATC, but I'm just curious about the rules.)
 
Yes, even if you were cleared for the approach when outside the TAA, unless you were restricted by ATC, on entering the TAA, you may descend to the charted altitude. Normally ATC will give you an altitude to maintain until inside the TAA in this case, but without this, you remain at your last assigned altitude until inside the TAA, then you may descend as charted.
 
Yes, even if you were cleared for the approach when outside the TAA, unless you were restricted by ATC, on entering the TAA, you may descend to the charted altitude. Normally ATC will give you an altitude to maintain until inside the TAA in this case, but without this, you remain at your last assigned altitude until inside the TAA, then you may descend as charted.

John, agreed on if you're within the TAA, because the pilot would be on a published route or segment of the approach (the latter in this case), however, if outside the TAA on a random route, then the controller would be in error if issuing the clearance without an altitude to maintain until one of those two cases were true. The 7110.65 is crystal clear about that requirement. A pilot should push back if it's missing, most notably if their currently altitude is not the same as the minimum altitude associated with the segment they'll eventually be joining (ie, you're at 5k and the segment you'll be joining calls for 2500).
 
John, agreed on if you're within the TAA, because the pilot would be on a published route or segment of the approach (the latter in this case), however, if outside the TAA on a random route, then the controller would be in error if issuing the clearance without an altitude to maintain until one of those two cases were true. The 7110.65 is crystal clear about that requirement. A pilot should push back if it's missing, most notably if their currently altitude is not the same as the minimum altitude associated with the segment they'll eventually be joining (ie, you're at 5k and the segment you'll be joining calls for 2500).

I very much agree, that is what I would expect from ATC, so my answer was more academic.
 
And that radial was his cleared route, which was my point. He wasn't off wandering around like the OP thinks he could do.
His problem was leaving the assigned altitude after being cleared for the approach before being on some segment of the approach.
And, among other rule changes, that accident caused the rule to maintain your last assigned altitude until on a published route or segment.
 
Looks like the OP hasn't returned, too bad, would've been nice to get a little clarification.
 
There are lots of small details that can cause confusion between countries. For example, the first time I flew to the U.S. (VFR) 18 years ago, I was shocked to hear that tower was clearing more than one airplane to land on the runway at the same time, and actually asked for clarification because I thought they'd made a mistake.

In Canada, only one plane is cleared to land at once, so if you're #3, you won't hear "cleared to land runway XX" until #1 and #2 are already down; in the U.S., "cleared to land" seems to mean "you're precleared for when your turn comes," and they'll give it to all 3 at the same time. (I'm used to that now, of course.)

I learned to fly here in the US many years ago and I never got used to it. It just does not add up with any logic my brain can devise.... It just seems to me not logical or safe that I can own the runway and be cleared to land when clearly there's someone on the runway or about to take it in front of me....& I sure don't like the feeling when someone behind me is cleared to land on top of me while I'm still slowing!
 
"cleared to land" really means "when you cross the threshold, I promise all the separation requirements will be met."

If you really want some excitement, watch new pilots' faces when they see that the separation requirements don't necessarily mean that the runway will be CLEAR as you're landing. Under the right circumstances, there could be plane rolling out 3001ft ahead of you.

Then, of course, there's Oshkosh where they keep the tower warm by setting fire to the FAA order on Air Traffic Control :)
 
I don't understand the comments saying the controller was wrong. Exactly what is wrong with a 90 degree turn at an IF? It's the most common turn at an IF on a TAA. You are 2.4 from the next waypoint and almost 7 from the FAF. I'd be flying direct to AGNSS per the instruction and turning onto the FAC.

Although I think the controller could have been more clear, I don’t see anything wrong with a 90 degree turn either. If using a navigator, it’s even less of a problem since it gives you a precise queue for when to start the standard rate turn for it to work out well.
 
"cleared to land" really means "when you cross the threshold, I promise all the separation requirements will be met."

If you really want some excitement, watch new pilots' faces when they see that the separation requirements don't necessarily mean that the runway will be CLEAR as you're landing. Under the right circumstances, there could be plane rolling out 3001ft ahead of you.

Then, of course, there's Oshkosh where they keep the tower warm by setting fire to the FAA order on Air Traffic Control :)

And in the right circumstances (helicopter) the plane ahead could be a lot closer than 3000 ft. I was maybe 1,000 ft following a Cessna into KRYY the other day.
 
I learned to fly here in the US many years ago and I never got used to it. It just does not add up with any logic my brain can devise.... It just seems to me not logical or safe that I can own the runway and be cleared to land when clearly there's someone on the runway or about to take it in front of me....& I sure don't like the feeling when someone behind me is cleared to land on top of me while I'm still slowing!
There was a time in the US it wasn’t that way. The rule was the Controller couldn’t give the Landing Clearance until the preceding aircraft crossed the Landing Threshold
 
any idea roughly when that changed? My memory is a bit fuzzy if it was ever that way in my time or not
 
There was a time in the US it wasn’t that way. The rule was the Controller couldn’t give the Landing Clearance until the preceding aircraft crossed the Landing Threshold

I wasn't aware. I started my flight training in Dec '96, so it must've been before that. During that time, I've seen transponder usage change, position and hold change, taxi phraseology has been ravaged several times, climb/descend via debacle, but I don't recall that one.
 
There was a time in the US it wasn’t that way. The rule was the Controller couldn’t give the Landing Clearance until the preceding aircraft crossed the Landing Threshold

My last real go around, because of traffic, in the Citation was because a C150 had not crossed the line at KHEF around 2010.
 
I wasn't aware. I started my flight training in Dec '96, so it must've been before that. During that time, I've seen transponder usage change, position and hold change, taxi phraseology has been ravaged several times, climb/descend via debacle, but I don't recall that one.
I knew it was sometime between 81 and 02. Now I know it was 81 and 96
 
My last real go around, because of traffic, in the Citation was because a C150 had not crossed the line at KHEF around 2010.
by "crossed the line," I'm guessing you are referring to clearing the runway? If so, that would make sense because same runway separation isn't applicable for a turbine powered aircraft, the runway has to be clear (not past the hold short lines, though, but the plane has to be off the runway and still moving).
 
There was a time in the US it wasn’t that way. The rule was the Controller couldn’t give the Landing Clearance until the preceding aircraft crossed the Landing Threshold
I wonder why the U.S. changed that when other countries didn't.
 
I've been flying for thirty years. During that period, I remember multiple landing clearances being permitted, and then there was some accident that caused them to institute the one-at-a-time rule. I didn't know they had changed it back to allowing multiple clearances until I noticed them doing it at my home field recently.
 
Yes, even if you were cleared for the approach when outside the TAA, unless you were restricted by ATC, on entering the TAA, you may descend to the charted altitude. Normally ATC will give you an altitude to maintain until inside the TAA in this case, but without this, you remain at your last assigned altitude until inside the TAA, then you may descend as charted.
Then, there was PADL.
 

Attachments

  • inattention-and-slopp.pdf
    469.1 KB · Views: 17
What an absolutely ugly NTSB determination. The pilots were given told to "maintain at or above" an altitude that was 3400 feet below the safe altitude and it was somehow the pilots fault for failure to realize the controller screwed up and that somehow their trivial readback error was somehow causal?

Given that at the point they were given that clearance, they were not even in the TAA area and on a off-airway route, the thing that should have driven the controller is the MVA which is not disclosed to pilots in any form, and the pilots would have to intuit what the minimum IFR altitude would be based on OROCA and inquire if the controller really meant that (because you can legally be within the grid block OROCAs).

Oh, and couple that with the fact that the MSAW warnings were going off and the scope dope didn't bother think that anything might be wrong.
 
What an absolutely ugly NTSB determination. The pilots were given told to "maintain at or above" an altitude that was 3400 feet below the safe altitude and it was somehow the pilots fault for failure to realize the controller screwed up and that somehow their trivial readback error was somehow causal?

Given that at the point they were given that clearance, they were not even in the TAA area and on a off-airway route, the thing that should have driven the controller is the MVA which is not disclosed to pilots in any form, and the pilots would have to intuit what the minimum IFR altitude would be based on OROCA and inquire if the controller really meant that (because you can legally be within the grid block OROCAs).

Oh, and couple that with the fact that the MSAW warnings were going off and the scope dope didn't bother think that anything might be wrong.
Yep, I thought the NTSB had its head way up the dark place.
 
PADL was a case where the pilot was cleared to an altitude below the charted TAA altitude and cleared for the approach. One should never accept a clearance for the approach below the charted segment altitude, even if it is above the MVA. You can be cleared to a fix, maintain an altitude until crossing the fix, and then cleared for the approach, but once one is established on a charted segment of the approach and cleared for the approach, the chart determines the minimum altitudes. The clearance was non standard and the pilot read it back incorrectly, but once inside the TAA, the charted segment altitude is the minimum. At the time the pilot received the clearance, they were approximately at 30 NM from ZEDAG and at 5900 feet so on entering the TAA, the segment is a published segment with a minimum of 5400 to ZEDAG, so the descent should not have been continued to 2000. Here is a link to the NTSB report https://reports.aviation-safety.net/2013/20130308-0_B190_N116AX.pdf and here is an excerpt from the ATC communication transcript:

0803:33 AER51: Anchorage Center Ace Air fifty one current weather down into Dillingham requesting RNAV one nine approach any chance we can get direct ZEDAG?
ARTCC: Ace Air fifty one cleared direct to the Dillingham Airport via direct ZEDAG ZEDAG transition. Maintain ah maintain at or above two thousand until established on a published segment of the approach. Cleared RNAV runway one
niner approach to Dillingham Airport. Remain this frequency.
AER51: We'll stay with you. Cleared to ZEDAG transition for RNAV one nine approach into Dillingham. Maintain [ARTCC controller dialing the DLG FSS] two thousand until a published segment of the approach Ace Air fifty one.
 
What an absolutely ugly NTSB determination. The pilots were given told to "maintain at or above" an altitude that was 3400 feet below the safe altitude and it was somehow the pilots fault for failure to realize the controller screwed up and that somehow their trivial readback error was somehow causal?

Given that at the point they were given that clearance, they were not even in the TAA area and on a off-airway route, the thing that should have driven the controller is the MVA which is not disclosed to pilots in any form, and the pilots would have to intuit what the minimum IFR altitude would be based on OROCA and inquire if the controller really meant that (because you can legally be within the grid block OROCAs).

Oh, and couple that with the fact that the MSAW warnings were going off and the scope dope didn't bother think that anything might be wrong.
Whell, sure ATC didn't live up to the expectations that they should act like a third crewmember and save the crew from their mistakes. That expectation comes from TWA 514, but really, even before then when pilots were supposed to know where they were all the time and operate accordingly the onus is on the crew. TWA 514 had one officer who seemed to know, but got overruled by the two with their hands on the flight controls. Admittedly, the chart was misleading for a pilot who wasn't a "student of the game"—hence all the fallout afterward. In this case, though, the chart has everything you could hope for as a pilot. So, what I wonder is, "Did the captain have a copy in front of him?" The first officer was offline and not apparently in the loop confirming altitude clearances with the captain and may have clipped the only chart to his control yoke. Maybe. IDK. The bottom line is ATC's function is to prevent pilots from going where they want, nice as they are as regular people. But if they make a bad enough mistake they will KILL you. Fly like it.
 
That expectation comes from TWA 514, but really, even before then when pilots were supposed to know where they were all the time and operate accordingly the onus is on the crew. TWA 514 had one officer who seemed to know, but got overruled by the two with their hands on the flight controls. Admittedly, the chart was misleading for a pilot who wasn't a "student of the game"—hence all the fallout afterward.
All other inattentions, bad training, lack of "fleck" aside, had Round Hill intersection been in the profile view as mandated by inter-agency charting specifications at the time, the accident would not have happened. Period.
 
Oh, and couple that with the fact that the MSAW warnings were going off and the scope dope didn't bother think that anything might be wrong.
As in the Piper they planted on the mountain a few hundred feet from JLI VOR.
 
All other inattentions, bad training, lack of "fleck" aside, had Round Hill intersection been in the profile view as mandated by inter-agency charting specifications at the time, the accident would not have happened. Period.
I added it here, but don't see how it changes anything. What am I missing?

Round Hill.jpg
 
Last edited:
Yep, I thought the NTSB had its head way up the dark place.
This one was kinda like the Medford one. In order to comply with the rules for doing the Approach, the pilots would have to climb once getting established on the published procedure. In this case when entering the TAA. At Medford as I recall, it was joining a DME Arc. That don’t follow the logic check. Pilot gotta go whoa, this ain’t makin NO sense. I’m not going to Monday morning quarterback what should have been probable cause and what should have been contributing.
 
If it had been on the profile it certainly wouldn’t have been 1800. Probably 3700. I’ll entertain arguments for maybe 3000 or 3400, but my money is on 3700.
Why? The MEA on the plan view was 1800'? :dunno:
 
They weren't cleared for the approach at the time they crashed. Their clearance limit was the hold with an EFC.
 
Why? The MEA on the plan view was 1800'? :dunno:
It’s not an MEA, but generically speaking, yeah, call it MEA. The segment altitude is 1800. But what you depicted showed the altitude AT Round Hill as 1800. You can look at thousands of Approach’s where you descend to a lower altitude after crossing the Fix at it’s Minimum Altitude. To illustrate further, look at Procedure Turns. There is the ‘high station’ altitude. The Minimum Altitude until established inbound. And then the altitude you can descend to after that. I’d agree that if Round Hill had been charted on the Profile View and depicted correctly, it likely would have eliminated the confusion in the cockpit.
 
Why? The MEA on the plan view was 1800'? :dunno:

The way you depicted it is as an altitude AT Round Hill, not the segment altitude AFTER Round Hill. The altitude after Round Hill is already depicted as 1800, the altitude at 6 DME.

In the Profile View, the beginning altitude, at Round Hill, would currently be charted to the left of the fix, and would be 3700 (maybe*).

* Maybe, because current standards require that any segments converging on the IF have a common altitude. In this case, assuming the published 3000, 3400 and 3700 are all the minimum allowable for obstacles, today they would all be raised to 3700.
 
Why? The MEA on the plan view was 1800'? :dunno:
The way you depicted it is as an altitude AT Round Hill, not the segment altitude AFTER Round Hill. The altitude after Round Hill is already depicted as 1800, the altitude at 6 DME.

In the Profile View, the beginning altitude, at Round Hill, would currently be charted to the left of the fix, and would be 3700 (maybe*).

* Maybe, because current standards require that any segments converging on the IF have a common altitude. In this case, assuming the published 3000, 3400 and 3700 are all the minimum allowable for obstacles, today they would all be raised to 3700.
That wasn't required in 1974. The three initial segments could have remained different and the lowest of the three shown at Round Hill in the profile. If 1,800 wasn't good after Round Hill then an intermediate step-down fix would have been required. But the crash occurred a few miles prior to Round Hill, and 1,800 was good after Round Hill, at least as to DME. I don't know about the fix displacement area.
 
That wasn't required in 1974. The three initial segments could have remained different and the lowest of the three shown at Round Hill in the profile. If 1,800 wasn't good after Round Hill then an intermediate step-down fix would have been required. But the crash occurred a few miles prior to Round Hill, and 1,800 was good after Round Hill, at least as to DME. I don't know about the fix displacement area.
So I think you're saying Round Hill should have been an IF (which I don't remember being charted in those days), correct? That's what you mean by saying it was missing, since it is along the route from the IAFs. All that aside, once radar released them for the approach they ought to have seen the MDAs for the two initial approaches segments they were within on the plan view. I don't think they even looked.
 
So I think you're saying Round Hill should have been an IF (which I don't remember being charted in those days), correct? That's what you mean by saying it was missing, since it is along the route from the IAFs. All that aside, once radar released them for the approach they ought to have seen the MDAs for the two initial approaches segments they were within on the plan view. I don't think they even looked.
I was involved in the investigation and saw the FAA’s TERPS maps, Round Hill was the IF and you’re correct, IFs weren’t charted with “IF” until many years later. One of the crew members saw the initial segments and mentioned it. But, he was overruled. All of them, and me, had received lousy training by TWA’s reg instructor about when you can descend. I knew he was wrong and cited the ILS at KONT to him two years earlier. My concerns fell on deaf ears. I had to testify about that after the crash.
 
I was involved in the investigation and saw the FAA’s TERPS maps, Round Hill was the IF and you’re correct, IFs weren’t charted with “IF” until many years later. One of the crew members saw the initial segments and mentioned it. But, he was overruled. All of them, and me, had received lousy training by TWA’s reg instructor about when you can descend. I knew he was wrong and cited the ILS at KONT to him two years earlier. My concerns fell on deaf ears. I had to testify about that after the crash.
My only connection was my trip to Key West was scrubbed at the last minute due to the weather concerns. The trip back from the airport usually took 55 minutes, but was about three hours that day. The next morning I read about the crash. Followed the investigation closely afterward. There were a lot of discussions around the airfield between professional pilots during that time. Not everybody had the same opinion as to who to blame.
 
I came up with the same solution slightly differently: If he's going 3 miles per minute and if he's about three miles from AGNSS (to be clear, we don't know where he started his turn—we're all just guessing) he's one minute out at the most. In order to rollout on final it'll take a one-half minute standard rate turn if started at exactly the right distance. That distance in terms of seconds is 120sec/pi/2, or roughly 20 seconds, which is one mile @ 3mi/min.

However, maybe he doesn't like the way his automation snatches the plane into a sudden bank, so for passenger comfort he elected to steer right earlier than the FMS would, because he's smarter than the dumb machine. Maybe he knows the wind is going to temper the 30° bite he chose, too, making it more like 20° in reality, or less even. So the 10 seconds he used up in the 30° turn is almost all deducted from the (worst case) 40 seconds remaining until the FMS wakes up and jerks the plane onto final. After turning, at the most, he has 30 seconds until one mile from the centerline, and then 20 seconds will be used to make the remaining 60° turn. Seems well within a captain's prerogative to me, if MY math is correct (feel free).
We'll have to agree to disagree. I don't know what FMSs you've flown behind, but none of the ones I use "jerks" the plane anywhere. If anything, they are annoyingly smooth (at least laterally). The know where you are, where you want to go, what the winds are and when to start the turn to exactly roll out on the next course.

Either way, I don't think the correct answer when cleared to a fix is to put the plane in heading mode and "roll your own" vector to final. It may have worked in this case, but it shows a poor understanding of how to follow clearances and use the automation.

The fact that the OP started a thread that created this much debate and hasn't checked back in speaks volumes to me. Perhaps he did go and listen to the LiveATC feed and maybe what happened wasn't exactly what was presented here. I guess we won't know until the @drgwentzel can come back and clarify some stuff for us.
 
Last edited:
I added it here, but don't see how it changes anything. What am I missing?

At TWA there was no training, ground school, or otherwise, on U.S. MSAs, since they aren't operational altitudes. Perhaps in international ground school, I don't recall.
 
We'll have to agree to disagree. I don't know what FMSs you've flown behind, but none of the ones I use "jerks" the plane anywhere. If anything, they are annoyingly smooth (at least laterally). The know where you are, where you want to go, what the winds are and when to start the turn to exactly roll out on the next course.
On a Cardinal? I checked his profile (I should have done it before): drgwentzel | Pilots of America My thinking now is he wasn't flying a jet @ 3 miles/minute less than one minute from station passage (AGNSS is over TEB VOR, btw).

Either way, I don't think the correct answer when cleared to a fix is to put the plane in heading mode and "roll your own" vector to final. It may have worked in this case, but it shows a poor understanding of how to follow clearances and use the automation.
Yes, that's where I agree to disagree. The pilot may prefer not to use automation for no reason at all, imo. When to begin the turn is a matter of technique and we don't know when he actually did. To do it in one continuous standard rate turn manually would be more luck than skill, so taking a reasonable first cut seems like a conservative attempt to avoid overshooting to me. Now, if he started the turn at 5 miles where he was cleared for the approach at Cardinal speeds I would be forced to agree to agree with you. :)

I'm now wondering about this "pilot monitoring". If this is a Cardinal—who the heck is he?
 
On a Cardinal? I checked his profile (I should have done it before): drgwentzel | Pilots of America My thinking now is he wasn't flying a jet @ 3 miles/minute less than one minute from station passage (AGNSS is over TEB VOR, btw).

Since in his narrative, he said NY said "UE4388 cleared for..." I was assuming that was United Express flight 4388 and he was probably in an RJ of some sort, not a Cardinal


Yes, that's where I agree to disagree. The pilot may prefer not to use automation for no reason at all, imo. When to begin the turn is a matter of technique and we don't know when he actually did. To do it in one continuous standard rate turn manually would be more luck than skill, so taking a reasonable first cut seems like a conservative attempt to avoid overshooting to me. Now, if he started the turn at 5 miles where he was cleared for the approach at Cardinal speeds I would be forced to agree to agree with you. :)
And it may be all well and good in this instance. Obviously nobody have him a number to call (or maybe they did and that's why he's not posting), so it worked out. What concerned me the most was this statement to the pilot he was flying with:
He simply stated, '...cleared for the approach.' Therefore, we can fly any heading or altitude which will intercept the localizer and glide slope and fly the approach effectively and safely."
I don't agree with that.

Also, the breakdown of CRM was alarming to me.
 
Back
Top