Poorly Flown Approaches

[pelmet]

On 11 January 2023, the Sunwest Aviation Ltd. Fairchild SA227-DC (registration C-GSHY, serial
DC-897B) was operating as flight CNK214 on a cargo flight from Calgary Intl. (CYYC), AB to
Edmonton Intl. (CYEG), AB. Onboard were 2 pilots. The flight crew was conducting a hand-flown
instrument landing system (ILS) approach to Runway 02 at CYEG. The weather was at minimums
for the ILS approach and there was a 5 knot tailwind. At approximately 1.0 nautical miles from the
threshold of Runway 02, the flight crew initiated a missed approach. During the hand-flown missed
approach, the tower controller observed the aircraft turning 35 degrees to the left towards the flight
path of an aircraft (Air Canada flight AC234, Boeing 737 Max 8, registration C-FSIP, serial 61215)
conducting an ILS approach to Runway 12. Radar data shows that CNK214 continued to descend
while deviating left of the localizer. ATC instructed CNK214 to fly heading 360 and climb to 5000
feet above sea level. At 1111:11, radar data shows CNK214 at 0.9 NM from the threshold of
Runway 02, between 100 to 200 feet above ground level, 800 feet west of the localizer. During the
missed approach, there was a loss of separation with AC234, established on the ILS for Runway
12. Separation reduced to (2.1 nautical miles lateral and 300 feet vertical spacing where 3 nm or
1000 feet is required. Neither aircrafts' flight crew had the other aircraft visual due to instrument
meteorological conditions; AC234 did not receive a TCAS resolution advisory. After CNK214
levelled off at 5000 feet above sea level, CNK214 received radar vectors for the ILS to Runway 12.

At approximately 1 NM from the threshold of Runway 12, CNK214 was approximately 475 feet
south of the localizer and full-scale deflection below the glideslope. At 0.64 NM from the threshold,
CNK214 had descended to approximately 100 to 200 feet AGL and remained at that height until
0.4 NM from the threshold of Runway 12, at which point CNK214 began to climb in the missed
approach. Following the second missed approach, CNK214 returned to CYYC without further
incident.

[TG]

It almost looks like some sort of instruments failure.

[Eric Janson]

Looks more like a complete lack of manual instrument flying skills - very lucky not to crash or hit something imho.

Not sure why you'd continue an approach with a full scale glideslope deflection and probably a full scale localiser deflection.

Really have to wonder what was going on in that cockpit.

[co-joe]

...At 1111:11, radar data shows CNK214 at 0.9 NM from the threshold of
Runway 02, between 100 to 200 feet above ground level, 800 feet west of the localizer. ...

100' agl, 800' west of the loc? Wow that's close to a cfit. There's basically just farmer's field there but most runways that would have been a much different outcome.

[fish4life]

If it’s not an instrument failure they should lose their IFR rating

[YC87DRVR]

Well the clues point to lack of experience. With the constant recurring hiring add and *BONUS* being offered to fly the SW4 for Sunwest and the info provided in that cadors.
Let’s hope that’s not the case, but with how desperate some operators are to fill seats I wouldn’t be surprised. Can’t be many people lining up to fly metros in this market.

[MUSKEG]

Really have to wonder what was going on in that cockpit.

Most likely a total of 1000 hrs between the 2 pilots.
Loss of situational awareness I’m thinking.

[scdriver]

Here’s a similar one I came across:

A Morningstar Air Express Inc. Boeing 757-2B7 (C-FMHA/MAL7080) from Calgary, AB (CYYC) to Vancouver, BC (CYVR) was cleared for Runway 08L on the area navigation (RNAV) Z approach. MAL7080 descended to 1300' below the glide path. The last assigned altitude was 2000'.

Not quite as bad, but 1300’ below the glide path could be a really bad day if you weren’t out over the strait…

[CpnCrunch]

Here’s a similar one I came across:

A Morningstar Air Express Inc. Boeing 757-2B7 (C-FMHA/MAL7080) from Calgary, AB (CYYC) to Vancouver, BC (CYVR) was cleared for Runway 08L on the area navigation (RNAV) Z approach. MAL7080 descended to 1300' below the glide path. The last assigned altitude was 2000'.

Not quite as bad, but 1300’ below the glide path could be a really bad day if you weren’t out over the strait…

Here is the track log for that flight:

https://flightaware.com/live/flight/MAL ... R/tracklog

It looks like they descended to 1000ft at 6.8 miles, and they should have been around 2200ft at that point on the glideslope (or 2000ft as assigned). It's an LPV approach, so it's strange that they could have gotten so badly below the glideslope. Maybe CDI was on VLOC and they didn't notice the flag?

[Scuderia]

Here’s a similar one I came across:

A Morningstar Air Express Inc. Boeing 757-2B7 (C-FMHA/MAL7080) from Calgary, AB (CYYC) to Vancouver, BC (CYVR) was cleared for Runway 08L on the area navigation (RNAV) Z approach. MAL7080 descended to 1300' below the glide path. The last assigned altitude was 2000'.

Not quite as bad, but 1300’ below the glide path could be a really bad day if you weren’t out over the strait…

Doesn't make it much better, but I believe the author meant "1300' ASL, below the glide path" vs 700' AGL there.

1,300' is a charted minimum on that approach but ATC will often instruct not lower than 2000' until X DME to provide separation from VFR traffic below.

At any rate, all those altitudes are below the LPV glidepath or SCDA.

[rookiepilot]

It's an LPV approach, so it's strange that they could have gotten so badly below the glideslope. Maybe CDI was on VLOC and they didn't notice the flag?

That is a reasonable guess, and has killed more than once.

I recall my instructor beating on me to always remember that, and it stuck with me. It's an easy thing to overlook and especially with a change in approach.

[CpnCrunch]

That is a reasonable guess, and has killed more than once.

I recall my instructor beating on me to always remember that, and it stuck with me. It's an easy thing to overlook and especially with a change in approach.

It's surprisingly easy to do. When I got my plane and was practicing instrument approaches in VMC, there were a few times that the LPV approach didn't get activated in the GPS, and both glideslope and localizer needles in the CDI were perfectly centred. I just didn't realise that the little checkered box was the flag. That is the Garmin GI-106, which I feel isn't the best UX, but once you get used to it it's fine. Also there are various reasons why a GPS might not go into approach mode (even if you've activated it). All just a matter of knowing your equipment and keeping an eye on it.

[pelmet]

It's surprisingly easy to do. When I got my plane and was practicing instrument approaches in VMC, there were a few times that the LPV approach didn't get activated in the GPS, and both glideslope and localizer needles in the CDI were perfectly centred. I just didn't realise that the little checkered box was the flag.

Whenever you are doing an approach and you think you are a great pilot because the needles are staying perfectly centered....be suspicious.

Alitalia lost a DC-9 in Zurich when a faulty glideslope resulted in a centered needle without a flag.

https://www.facebook.com/watch/?v=1929238810609593

Air New Zealand almost lost a 767. A possible hint.....needles suddenly moving from a deflected position to centered and then staying there.

https://www.youtube.com/watch?v=GelRBhJ4gmI

[CpnCrunch]

Whenever you are doing an approach and you think you are a great pilot because the needles are staying perfectly centered....be suspicious.

Alitalia lost a DC-9 in Zurich when a faulty glideslope resulted in a centered needle without a flag.

https://www.facebook.com/watch/?v=1929238810609593

Air New Zealand almost lost a 767. A possible hint.....needles suddenly moving from a deflected position to centered and then staying there.

https://www.youtube.com/watch?v=GelRBhJ4gmI

Generally with an ILS approach you are coming from underneath and with lateral deviation, which helps to avoid this issue. With RNAV it tends to be straight in a lot of the time, although still from underneath.

One quirk of LNAV approaches is that the advisory glideslope doesn't seem to start until the FAWP. Before FAWP you'll just get lateral deviation, and the glidepath needle will be perfectly centred no matter what your height is. I can't remember if the glideslope is flagged in that case or not.

I just realised the issue I had with the GI-106b is that it doesn't in fact have an obvious flag like a checked pattern, or red GS. What it has are little orange arrows, which is confusing at first, because there is no obvious reason why little red arrows would mean "this thing isn't working". Obviously if you read the manual you'll know, but who reads their CDI manual?

[altiplano]

Really have to wonder what was going on in that cockpit.

Most likely a total of 1000 hrs between the 2 pilots.
Loss of situational awareness I’m thinking.

Need an ATPL to be a CA on an SA227... so more than 1000... but from where? It ain't deep I'm thinking.

Metros are hot little airplanes... if these guys survive they are going to learn a lot in the time they are in there. More than the college kids riding around in airliners today as a first job. I look forward to flying with the survivors in the years ahead.

[Eric Janson]

Here's a poorly flown departure - very close call.

https://avherald.com/h?article=504d75c7&opt=0

This is someone who has passed a type rating but can't fly manually. At least not without FD's.

Interesting comments above about aircraft equipment and failure flags.

I would hope that people would be doing practice approaches in VMC conditions to get familiar with the indications - especially the failure indications.

When my company started flying RNAV approaches - I did several in VMC conditions to see what it looked like and how we ended up relative to the runway. Glad I didn't have to do one in IMC conditions until I'd had some practice.

[‘Bob’]

Generally with an ILS approach you are coming from underneath and with lateral deviation, which helps to avoid this issue. With RNAV it tends to be straight in a lot of the time, although still from underneath.

Lateral deviation can screw you on an ILS as well if they cut you in too close or keep you too high. You wind up above the slope and it won’t capture. An LPV has quite a wide range where it will capture the slope even when off track. Most LPV approaches are flown the same way into busy airports with open STARs.

One quirk of LNAV approaches is that the advisory glideslope doesn't seem to start until the FAWP. Before FAWP you'll just get lateral deviation, and the glidepath needle will be perfectly centred no matter what your height is. I can't remember if the glideslope is flagged in that case or not.

This shouldn’t be the case. You should have an LPV glide path when the IWP-FAWP leg is active or within 2NM of the FAWP at the same time it changes to LPV scaling. The worst case scenario is not having a baro-VNAV path to get you there but in that case you just use the step down altitudes.

[co-joe]

Here's a poorly flown departure - very close call.

https://avherald.com/h?article=504d75c7&opt=0

This is someone who has passed a type rating but can't fly manually. At least not without FD's.

Interesting comments above about aircraft equipment and failure flags.

I would hope that people would be doing practice approaches in VMC conditions to get familiar with the indications - especially the failure indications.

When my company started flying RNAV approaches - I did several in VMC conditions to see what it looked like and how we ended up relative to the runway. Glad I didn't have to do one in IMC conditions until I'd had some practice.

The decision to hand fly with no FD at 2 am in a black hole is sort of questionable. Wouldn't the 787 give a GPWS Don't Sink aural alert in that case? Wouldn't the altitude alert be going off as well? The best part is they didn't tell anyone. Oops.

[CpnCrunch]

This shouldn’t be the case. You should have an LPV glide path when the IWP-FAWP leg is active or within 2NM of the FAWP at the same time it changes to LPV scaling. The worst case scenario is not having a baro-VNAV path to get you there but in that case you just use the step down altitudes.

This is with an LNAV approach. I just checked on the sim to see what the problem is. The issue is with an LNAV, the LNAV+V doesn't activate until the SDWP, whereas with LPV it activates at the IWP. On an LNAV approach I expected the glideslope to become active at the IWP, but the GPS is still in TERM mode and the glideslope is flagged right until the descent point. I known it's a non-precision approach the glideslope is purely advisory, but it's still confusing compared to precision approaches where the glideslope is always active long before you reach the descent point.

You can easily get too high if you don't come down to the minimum step down altitude at the IWP.

Compared to the examples here it's a minor issue, as long as you're paying attention to the minimum altitudes shown on the chart. Both these examples were precision approaches.

[pelmet]

Whenever you are doing an approach and you think you are a great pilot because the needles are staying perfectly centered....be suspicious.

Alitalia lost a DC-9 in Zurich when a faulty glideslope resulted in a centered needle without a flag.

https://www.facebook.com/watch/?v=1929238810609593

Generally with an ILS approach you are coming from underneath and with lateral deviation, which helps to avoid this issue. With RNAV it tends to be straight in a lot of the time, although still from underneath.

I went through this video for the first time(subsequent to making my post) and found some interesting information about the accident. I had never read the report..

One thing not resolved in the video was the question about the outer marker. There was disagreement on whether they had passed it.

But I also discovered what appears to be a critical fatal decision made by the captain which was not looked into further by the video(by the way, I have looked at part of the report and as expected, there was more to the accident as they got a GS comparator light).

Go to 5:40 of the video. It shows the GS discrepancy. Then the captain switches both indications to the #1 nav which is the faulty receiver(Actually, it was 4 indications because the ADI's have a GS display with each pilot having their two displays get a #1 and a #2 source).

That selection right there was deadly as all indications went to the faulty nav receiver. Of course, it is just a video but it is a great 10 second example of a quick decision with no discussion that can lead to an airliner crash after all the other required items lined up(radio failure/GPWS fault/MSAW not installed/Crossing altitude not checked/memo from the manufacturer not disseminated).

So why did he select #1. Neither indication had a GS flag. How do you know that the one that is centered is the correct one. All you know is that they are disagreeing and at least one of them must be faulty. Obviously, neither should be used(especially in cloud in hilly terrain) until things can be clarified. But if you insist on using one of the GS indications, try the one that will be less likely to create hazard which is the one that will have you at a higher altitude. And while there can be a tendency to not pay much attention to the GS crossing altitude, you would really want to make sure you did whenever there has been some sort of a discrepancy like this.

In this case, they just went to Nav 1 and assumed it was correct.

The accident kind of reminds me of the 737 in Resolute. The copilot wanting to go-around due to some sort of discrepancy in the instrument indications while the captain insists on continuing down.

https://www.sust.admin.ch/inhalte/AV-be ... 1457_e.pdf