TSB Report: Mid-Air Collision Saskatchewan May 2012

[KK7]

Mid-Air Collision
Piper PA-28R-200 Arrow, C-GLAJ
and
Lake LA-4-200 Buccaneer, C-GFCH
St. Brieux, Saskatchewan, 8 nm W
12 May 2012

http://www.tsb.gc.ca/eng/rapports-repor ... 2C0053.asp

SUMMARY

The privately-registered Piper PA-28R-200 Arrow (registration C-GLAJ, serial number 28R‑7435312) was approaching St. Brieux, Saskatchewan, on a flight from Nanton, Alberta, with the pilot and 2 passengers on board. A privately-registered Lake LA-4-200 Buccaneer amphibian (registration C-GFCH, serial number 786) was en route from Regina to La Ronge, Saskatchewan, with the pilot and 1 passenger on board. At approximately 0841 Central Standard Time, the 2 aircraft collided about 8 nautical miles (nm) west of St. Brieux and fell to the ground at 2 main sites about 0.5 nm apart. Both aircraft, which were being operated in accordance with visual flight rules, were destroyed and there were no survivors. There was no post-crash fire and the emergency locator transmitters did not activate.

...

FINDINGS

Findings as to causes and contributing factors:

1. Both aircraft arrived at the same point and altitude at the same time, which resulted in a mid-air collision.

2. The converging position of the 2 aircraft relative to each other, coupled with physiological vision limitations, likely rendered visual detection extremely difficult. As a result, the available reaction time was reduced to a point where collision avoidance was not possible.

3. The left ailerons and part of the wings from both aircraft were shorn off in mid-air during the collision. This would have rendered both aircraft uncontrollable, and would have precluded either aircraft from recovering after the collision.

Findings as to risk:

1. Aircraft operating in visual flight rules conditions are at continued risk of collision when the see-and-avoid principle is relied upon as the sole means of collision avoidance.

Other findings:

1. The design and operating features of the collision avoidance systems in the aircraft involved in this occurrence are such that they can inadvertently be set to detection parameters resulting in insufficient warning time to pilots.

[costermonger]

They've now released two reports in less than a week that bluntly say "see and avoid isn't good enough".

[lownslow]

"1. Both aircraft arrived at the same point and altitude at the same time, which resulted in a mid-air collision."

What OTHER conditions can possibly result in a mid air collision? Do TSB investigators/writers get paid by the word or something?

LnS.

[Gogona]

"1. Both aircraft arrived at the same point and altitude at the same time, which resulted in a mid-air collision."

What OTHER conditions can possibly result in a mid air collision? Do TSB investigators/writers get paid by the word or something?

LnS.

+1!
You read my thoughts.

[FlyGy]

http://www.cbc.ca/news/canada/saskatche ... -1.2649793

[Illya Kuryakin]

"1. Both aircraft arrived at the same point and altitude at the same time, which resulted in a mid-air collision."

What OTHER conditions can possibly result in a mid air collision? Do TSB investigators/writers get paid by the word or something?

LnS.

No shit Sherlock. Shakes head....
Illya

[Big Pistons Forever]

Too bad neither aircraft had a airframe parachute system like the Cirrus....

[Shiny Side Up]

Too bad neither aircraft had a airframe parachute system like the Cirrus....

Still doesn't magically save you from this kind of thing.



Still best to be aware and keep your eyes out the window. Note that the see and avoid in this case would have been difficult, but not impossible.

[Big Pistons Forever]

Too bad neither aircraft had a airframe parachute system like the Cirrus....

Still doesn't magically save you from this kind of thing.



Still best to be aware and keep your eyes out the window. Note that the see and avoid in this case would have been difficult, but not impossible.

The fuselages of both aircraft were not significantly damaged right after the aircraft collided, and so in this case it is highly likely that an airframe parachute system would have saved all the occupants.

This, in my mind is the nightmare scenario, because the reality is nobody keeps a a full look out for every minute of every flight.

[Shiny Side Up]

This, in my mind is the nightmare scenario, because the reality is nobody keeps a a full look out for every minute of every flight.

The problem isn't that the lookout isn't every minute, its because a lot of the time you'd be lucky if the crew wee looking out the window, even 50% of the time. Compound this with all of the other things that pilots do to significantly increase their chances of putting themselves in close proximity to other aircraft.

I think if you time it, you might total up just over a minute that this guy spends looking outside.

[jeta1]

http://www.cbc.ca/news/canada/saskatche ... -1.2649793

While we all sympathize with the widow of deceased passenger Eric Donovan and mother of the deceased 11 year old boy, I doubt this $6M lawsuit against the estates of the two dead pilots will go anywhere. The TSB Report stated "There is no indication that either an aircraft malfunction or the weather contributed to this occurrence", which essentially eliminates two of her claims. She may get an out of court settlement from the two pilots' insurance companies, which is somewhat typical of such a tragic outcome. Thank you for posting. Now this may drag a while.

[Mr. T]

Doesn't really mention anything about radio's, if either had any, and if so, what frequencies they were tuned to. It still amazes me, in this day and age, we are still able to fly aircraft completely nordo...

[Blakey]

In my opinion, the TSB missed an opportunity to point out an important lesson here. If you read the report and pay particular attention to the illustration of the aircraft colliding, you can see that turning away from an aircraft that is VERY close to you is not the best way to avoid a collision. To me, it looks as if the pilot of the PA-28 saw the Lake at the last moment and tried to turn away to avoid the collision.

Turning the aircraft separates you from the danger but it does so VERY slowly and at the expense of greatly increasing your vertical cross-section. In this case, I believe that the turn caused his wing to rotate downward into the path of the Lake's wing. If you fly a lot of close formation, this is drummed into you from day one - "do not turn away from a possible collision, separate yourself vertically". The preference is always to go under the other aircraft as that is the direction in which you can proceed fastest due to gravity but in certain cases, I believe this to be one of them, up can work too. Have a look at the reconstruction illustration and ask yourself what the outcome would have been if the pilot of the PA-28 had pulled up, wings-level instead of trying to turn away. No doubt there would still have been a collision but perhaps not as disasterous a result.

Colonel Sanders states all of this, far more eloquently than I can in this thread: http://www.avcanada.ca/forums2/viewtopi ... =3&t=96009

[white_knuckle_flyer]

These are the kinds of accidents that I think about when I read threads / posts about laying off the radio and just open your eyes. I don't see how, and TC seems to agree, you can expect to avoid an accident of this kind just by keeping a really good lookout. No one keeps a 100% effective lookout, no matter what they promise. You couldn't enjoy flying if you did nothing but scan. And as this accident points out, even a good lookout may not keep you safe.

Some accidents are just that....accidents. And as such, they are an unfortunate confluence of factors with clearly undesireable results.

This is why I don't see how I can expect to turn off the radio and depend upon nothing but my eyes. Looks to me like I need the radio, my eyes, a TCAS and some luck.

Which leads me to ask this question...likely one which has been asked before.

Should a nervous newb invest in some collision avoidance hardware ? if so, what kind and how much ?

[Big Pistons Forever]

One day I was flying between Prince George and Watson lake. A beautiful summer day, no cloud and vis about 15 miles in a light haze. I make a point of using flight following and going by Williston lake, Edmonton center called me about traffic at 2 O'clock sqawking 1200 and at what appeared to be on a collision course and at the same altitude. At 2 miles still no contact so Centre recommended a turn.

I finally saw the airplane at less than a mile. The airplane was a Cirrus and if I had not been alerted and then turned we would have had at best had a near miss, at worse a collision would have occurred . I asked Centre if there was any other aircraft around and he said we were the only contacts below the flight levels he was painting for 50 miles in any direction.

It is a big sky until it is not.......

[CpnCrunch]

I don't generally use flight following, and I'm not sure if it's even possible in the Vancouver area. Considering that they can't even manage the class C traffic on sunny days, it seems doubtful they would be able to keep an eye on VFR traffic in uncontrolled airspace.

I don't usually broadcast much/any on 126.7 or other enroute frequencies...partly through laziness, partly because I just want to enjoy flying, and partly because I doubt the value of transmitting my position to the entire province. I do usually listen out though.

It would be nice to have a PCAS, but I'm to cheap to spend $1000 on it. TCAS is completely out of the question.

I guess I just accept the small possibility of air-air collision as one of the risks of flying and try to keep a good lookout.

[Shiny Side Up]

I don't see how, and TC seems to agree, you can expect to avoid an accident of this kind just by keeping a really good lookout.

A good lookout is just part of it, and I might add something most pilots don't do, so until we have people doing it its really hard to say how avoidable these types of accidents are. In my small time in aviation I can count about a dozen close calls. In Every one of them, the other party was doing something stupid - in addition to obviously not looking out. The holes in the swiss cheese align. Half of those were people burning through the traffic pattern unannounced and seemingly uncaring. The remainder also were pilots deliberately putting themselves in harms way. Frig, I know a pair of pilots who played a dangerous game of chicken over a dispute in circuit procedure. Silly stuff. "Me first" is probably ultimately the cause of most of these.

Mid air collisions don't happen strictly by chance. If this one was thoroughly investigated we probably could come up with a dozen things these pilots each could have done that would have prevented this accident, and things that one could do to substantially decrease the chances of it happening to themselves.

Either way even if this is the thing that keeps you up at night, take a good look at the CAPS deployments and why they happened. Only 2 out of the 55 or so were due to mid air collisions. That's less than 4% of all the reports. Both of which it might be added were completely avoidable (of note, not just Cirrus chute deployments, but a lot of mid airs with chute deployments happen to involve glider activity - coincidence?) The lion's share happen when people lose control flying VFR into IFR, and just plain losing control in VFR. As a new pilot what should you be more worried about that you're lacking in?

If you got a $1000 to spend, what is going to be the most productive use to make you a safer pilot? Hint: gizmos don't automatically confer safety.

[Stubby Phillips]

I don't think it has been mentioned here yet but if you really want to be seen out there take a look at one of the pulsing landing light systems that are available. I have seen them installed on a fleet of aircraft I was involved in and they make an amazing difference. That little black speck in the distance just pops to your attention.

[white_knuckle_flyer]

One day I was flying between Prince George and Watson lake. A beautiful summer day, no cloud and vis about 15 miles in a light haze. I make a point of using flight following and going by Williston lake, Edmonton center called me about traffic at 2 O'clock sqawking 1200 and at what appeared to be on a collision course and at the same altitude. At 2 miles still no contact so Centre recommended a turn.

I finally saw the airplane at less than a mile. The airplane was a Cirrus and if I had not been alerted and then turned we would have had at best had a near miss, at worse a collision would have occurred . I asked Centre if there was any other aircraft around and he said we were the only contacts below the flight levels he was painting for 50 miles in any direction.

It is a big sky until it is not.......

If he was at your two o'clock and converging, it sounds to me like somebody was at the wrong flight level, unless you were both uner 3000'. What was the deal ?

[CpnCrunch]

One day I was flying between Prince George and Watson lake. A beautiful summer day, no cloud and vis about 15 miles in a light haze. I make a point of using flight following and going by Williston lake, Edmonton center called me about traffic at 2 O'clock sqawking 1200 and at what appeared to be on a collision course and at the same altitude. At 2 miles still no contact so Centre recommended a turn.

I finally saw the airplane at less than a mile. The airplane was a Cirrus and if I had not been alerted and then turned we would have had at best had a near miss, at worse a collision would have occurred . I asked Centre if there was any other aircraft around and he said we were the only contacts below the flight levels he was painting for 50 miles in any direction.

It is a big sky until it is not.......

If he was at your two o'clock and converging, it sounds to me like somebody was at the wrong flight level, unless you were both uner 3000'. What was the deal ?

Both going west...that's the problem with the semicircular rule.

[Shiny Side Up]

If he was at your two o'clock and converging, it sounds to me like somebody was at the wrong flight level, unless you were both under 3000'. What was the deal ?

I hate to say it WKF, but you're falling into the trap of assuming people are coming at you. You have to be able to think about the problem of traffic in more logical terms. In this instance for a contact at your 2 o'clock position to be on a collision course, if they're flying roughly the same speed, they need to be almost going the same direction as you. Only if they're travelling faster than you are (an unknown since BPF didn't specify) would they be on a head on collision course - and consequently probably at a wrong altitude. Though using a bit of logic, your cirrus cruises at around 140-150 Kts, for someone to be on a collision course with BPF and going roughly an incorrect altitude for their track they almost need to be going twice his speed. If that was the case BPF would have been flying a Cub, cruising in the ~70 knot range. That means that BPF should have been fully in the Cirrus's frontal arc -but then we all know how awesome Cirrus pilots are at looking out the window. For some reason I doubt this was the case. The more likely scenario being that BPF and the Cirrus were cruising at roughly the same speed, or BPF at a slightly overtaking speed.

Now before you get too freaked out, remember that a convrging intercept, as opposed to a head on or t-bone intercept happens at quite a bit slower closure rate.

Either way, one should note that BPF used one of the other tools in his arsenal of collision avoidance to prevent and take action early. He's a clever fellow that way.

[Sidebar]

Here's the graphic from the TSB report showing the tracks of the two airplanes.

[Sidebar]

And here's what the report said about altitude and direction of flight.

PA28 eastbound

C-GLAJ ... was last observed by ATC radar ... at an altitude of approximately 4400 feet asl on a track of 064° true ....

Lake northbound

The available information indicated that the aircraft was operating at an altitude of 4500 feet asl, which was the correct altitude for direction of flight ... on a direct track of 352°T for La Ronge.

Ground elevation was about 1800 ASL, so they were about 2700 AGL.

[Colonel Sanders]

Fun fact #1: Aircraft must be at the same altitude to collide

Fun fact #2: Aircraft at your altitude will be on the horizon.
Aircraft below you will be hidden in the ground clutter and
may be very difficult to see. Aircraft above you will be above
the horizon and are much easier to see.

[Shiny Side Up]

Fun fact #2: Aircraft at your altitude will be on the horizon.

This is the part people don't seem to get Colonel. You don't have to search the whole sky when your eyes are out the window, some parts of the sky are going to be way more important than others. In this accident since both airplanes cruise at similar speeds, each should have had the other in view from spotting range all the way until collision. In front of the plane, not behind wings, not underneath the wings. Roughly at 45 degrees off of the nose. Closure speed would have been about 180 Knots, that means at a reasonable spotting distance assuming reasonable visibility there would have been about 2 minutes from visual contact to impact.

I'd be curious to know how many gizmos were on board each aircraft. Hand held GPSs, headsets with bluetooth, iPads, etc.