as350 B3 Accident in Alaska

[Discap]

My friend was the "safety pilot" in this accident.  His brother is a missionary pilot (as is his dad) currently flying in Uganda.  Brother asked my opinion on what happened and what role if any the FADEC played.  I have read the report and it is obvious that they were attempting an auto to the beach to take a break.  They did not make it.  I have no AS time and don't have any idea how low Nr can go before control problems set in.  I noticed that for some reasons there were several roll-on/roll off events prior to entering the water.

I am not trying to assign blame or criticism etc.  I would like to know how low the reported Nr really was and if the FADEC makes any difference to recovery.  The safety pilot had lots of B2 time but very little B3.  As you will see this was a brand new helicopter being brought home after purchase/training.

Thanks in advance

Bill

CEN18FA391.pdf

[kona4breakfast]

There's a time and a place for everything. Practice autos from ~500' with kids on board in the middle of nowhere ain't the time or place. The FADEC might've saved them from their poor judgement and execution had they managed to leave it alone once they got it back to the FLIGHT detent. Your friend might not have realized the engine was pulled down but they both should've known to get the collective down at least. I can tell you from a SIM at least that it's really hard to keep the black on bottom when the rotor is significantly drooped on the AStar. Luckily I started from ~2000' and I was in a simulator.

[Discap]

So is it safe to assume that the 254 Nr is below the redline for this ship.  Also what is the FLI that they address?

[Nearly Retired]

Discap, 254 is waaaaay below the NR operating range for that model.  The power-off lower limit is 320 rpm.  You get that dad-blamed annoying Astar warning horn if it falls below 360 rpm.  So 254 is crazy-low...count-the-blades low.

(EDIT: The "FLI" they refer to is the First-Limit Indicator.  Airbus started putting these gauges in their helicopters back when the company was called Aerospatiale.  Basically, you get one gauge that can display engine temperature, torque, or rpm.  Whichever parameter gets close to its limit, that is the parameter that will be displayed.  So it might be showing your torque...or temperature; you have to read the fine print to find out which.)

With some accidents, it's really hard to fathom how or why they occurred.  This one in particular leaves us scratching our heads.  On one hand, we had a "safety pilot" who had quite a lot of experience in the AS-350.  In the other seat was a dual-rated pilot (with less than 100 r/w hours) who just came through the factory Astar checkout, was probably the very proud owner of a new aircraft.  And even if he didn't have very much total flight time (1,100 hours), he was probably pretty sharp on Astar emergency procedures and such.

On the other hand, we have two guys who decide to initiate a practice autorotation in, as kona4breakfast points out, a very bad place.  I would guess that one of the pilots probably had to pee, and so decided to set 'er down to, umm, "stretch their legs."  Yeah, sure.  Come on, we've all done that.  And, since they were going to land anyway, hey, why not do a practice auto on the way down?  Must've seemed like a good idea at the time...  I wonder which one of those guys suggested it?

The NTSB report shows them closely paralleling the long, straight shoreline with probably a long stretch of exposed sand.  So they SHOULD have been within gliding-distance.  The pilot/owner initiated the throttle roll-off.  The report seems focused on the FLI (first limit indicator) but at this point it's irrelevant.  When the throttle is in IDLE, the *ONLY* important item is rotor rpm.  In this case, that figure decreased and decreased until it was seriously low.  Sooo...no "collective down"....no "cyclic back"...no nothing but a little reduction in power-pedal.  And so we have to ask: WHAT THE HELL WERE THEY THINKING??  Wasn't either of them watching the tach?  Unfathomable!  Rotor speed is life, no?  Don't we all know that?

There was approximately eighteen seconds between throttle roll-off (at 10:57:35:25) and impact (at 10:57:53:50).  Not a whole lot of time, but time enough for the "safety pilot" to realize that the entry to this practice auto was bad.  The procedures for this do not change from single-engine helicopter to single-engine helicopter.  If it's f'-ed up, you bail out of the maneuver early.  And clearly...CLEARLY!...the entry to that auto was f'-ed up.  I, for one, cannot believe that a Robbie-trained pilot would *not* immediately push the collective all the way down.  I, for one, cannot understand how those two pilots could listen to that damn Astar horn going off and not *DO* something about it!  (Is the low-rotor horn in an Astar even mutable?)  How do you *not* pitch for best-auto speed??  Was the "safety pilot" asleep?  I mean, no offense intended Discap, but as the kids say, WTF!

At 10:57:42, or less than ten seconds from impact, the "safety pilot" finally decided to get in the game.  He grabbed the controls and, according to the NTSB report, it appeared that both pilots were manipulating the controls at the same time.  Really?  Was this one of those situations, like...

"I've got the controls." 

"No, I got it."

"No, I've got it!"

"That's okay, I've got it!"

"I SAID I"VE GOT IT, DAMMIT!"

Aaaaaaaand, crash.

 

Or did the "safety pilot" take over and the owner/pilot was just following through?  That's even worse to imagine, because when the "safety pilot" got on the controls, he evidently did little or nothing to recover the main rotor rpm.  Getting the collective all the way down might've helped, as would have hauling bac on the cyclic and loading the rotor.  Maybe some quick S-turns?  Something?  None of those things were attempted.  The owner/pilot did cycle the throttle a couple of times, but never returned it and left it in the FLIGHT position.  Again, unfathomable!  The NTSB report does not address why the power did not appear to come back in and bring the rotor back up.  

Look, we all know that every "practice" auto may turn out to be a "for real" auto and you have to be prepared for that eventuality.  And so if you roll the throttle up and nothing happens, well, no big deal, you do a full touchdown.  You expected that.  As you are screaming toward the ground with no rotor rpm, that's not the time to be dicking with the throttle.  I'm sure they did not intend on doing a full touchdown, auto-with-ground-run on a sandy beach.  But why on God's green earth would you initiate a practice auto when you weren't even in a position to glide to the beach?  Again, unfathomable some more.

Helicopters with the type of engines we call "free-turbine" are sometimes deceptive.  In a flat-pitch descent, there may be little difference in indications between IDLE throttle and FLIGHT.  This is especially confusing in a 206B; not much changes and you CANNOT tell what position the throttle is in without looking at it.  You do *not* always get a needle-split in a 206B - and I don't even think the Astar has a dual-tach.  Oh, and not sure about the Astar, but maybe the pilots were assuming that they would see "some" indication on the engine instruments that the throttle was back up at FLIGHT, but did not.  And maybe that was normal.  And so maybe the "safety pilot" said, "Cycle that damn throttle again!"  But when you initiate a simulated auto at around 500' agl, there ain't a bunch of time to troubleshoot stuff.  And if you're really sharp in 350-B2's and now you're in a B-3e, maaaaaaaybe you should get a little ground school in the thing before you blast-off for Alaska.

Eighteen seconds.  Seems like a long time if you watch the second-hand on an analog clock.  But those seconds get eaten up quick, and there's no going back.  At some point during that 18 seconds they were committed to crashing - nothing they could do about it from then on.  I've said it before and I'll say it again: Helicopters are "easy" to fly (once you get the hang of it), but they are always super-easy to crash!  Even when you're doing something as "simple" as a little practice power-recovery to a deserted stretch of beach, the results can be fatal.  As this one was.  The helicopter wreckage can tell us "something" about the accident, but it cannot tell us "why."  That, we may never know.

My advice?  If you're going to land somewhere off-airport to take a lea...err, "stretch your legs," just do a textbook landing: high/low recon, *normal* approach into the wind, etc.  If you're bound and determined to impromptu, stupid sh*t, do it at an airport.  So if you do goof-up and kill yourselves, at least they'll be able to find the wreckage and the bodies fairly quickly.

Edited April 10, 2020 by Nearly Retired
Add info on FLI

[Spike]

It's been a few years since I've flown an B3 but as I recall, there is a limitation restricting manipulation of the"twist-grip" during flight. Specifically, the twist-grip is not a throttle. It's a switch with two positions. Flight and Idle. Again, if my memory is not failing me, the only time you can switch the grip from flight to idle, in-flight, is during "training" with a "qualified" flight instructor.  Why? First gen B3's had issues with the old style throttle/switch combo which caused a few training accidents while practicing auto's. AIRBUS, then American Eurocopter, was done with it and gave the switch manipulation a limitation.... No more goofin around.....

In the end, from the information in the report, I agree with the previous posts. Furthermore, accidents like this one are reasons why rich private owners are given a bad rep and, an unfortunate early demise......

The saying goes; the good news is, we know why we're crashing helicopters.... The bad news is, we keep making the same mistakes..... This appears to be one of them....

IMHO.......

Edited April 11, 2020 by Spike

[Nearly Retired]

That's interesting, Spike.  How then does one do a practice autorotation in a B3e?

[SBuzzkill]

I had a friend who overtorqued an aircraft trying not to hit the ground while recovering a botched terminate with power autorotation.  Army guys out on a planned training flight over a big runway.  Part of the procedure is to confirm the throttle is open, and while doing so the “student” got confused and rolled the throttle back off when they were about 400 feet AGL.  Even pilots with good training do screwy stuff, and there’s not much time to fix it. 

My speculation in this case is that the pilot dropped the collective to arc in to the beach, and subconsciously cut the throttle.  Maybe muscle memory from 180 auto training or something.  Once things started going wrong, got task saturated and neither of them had time to fix it.  Eventually there was nothing they could do.

Edited April 13, 2020 by SBuzzkill

[helonorth]

Strangest accident ever. Why did the safety pilot sit there for 18 seconds while the RPM decayed and do nothing? The FLI was at 1.25 to 2, right up to impact. Nobody ever attempted to try to fix the low RPM with the throttle or lower the collective. You have two jobs during an auto: make your spot and RPM. The spot is optional. We all know RPM is not. Boggles the mind. The one guy had over 15,000 hours. 

Edited April 11, 2020 by helonorth

[iChris]

On 4/10/2020 at 5:46 PM, Spike said:

 if my memory is not failing me, the only time you can switch the grip from flight to idle, in-flight, is during "training" with a "qualified" flight instructor.  Why? First gen B3's had issues with the old style throttle/switch combo which caused a few training accidents while practicing auto's. AIRBUS, then American Eurocopter, was done with it and gave the switch manipulation a limitation.... No more goofin around.....

 

On 4/11/2020 at 9:22 AM, Nearly Retired said:

That's interesting, Spike.  How then does one do a practice autorotation in a B3e?

It appears your memory hasn't failed you. At least that was the way it was before Airbus.
Maybe the qualified flight instructor requirement part came in later manuals.

 

Edited April 12, 2020 by iChris

[Discap]

HeloNorth Why is NTSB focused on the FLI.  Nearly Retired said that it displays all kinds of faults.  What is the significance of it going from 1.25 to 2?

Thank you guys for being so polite and professional during this discussion.  I have not told the family about this site. but I think I will at this point.  I think they will understand the discussion.

 

Bill

[Nearly Retired]

Discap: "HeloNorth Why is NTSB focused on the FLI.  Nearly Retired said that it displays all kinds of faults."

Not "faults," but parameters.  Whichever power parameter is approaching a limit, *that* will be the one the needle is showing.  Could be torque first, but not necessarily.

[helonorth]

The FLI appears to show the collective position was never really changed. It had to be indicating torque. You could probably survive an auto with no flair or cushion but not letting the RPM get that low, depending on altitude. They also got into low G during the descent. Maybe once the RPM got too low, the bottom really fell out. At some point, no amount of throttle or collective is going to get the RPM back. 

Edit: Now that I think about it, does torque change with collective inputs during an auto? I don't think so but the report says the pilots made almost no movements to the collective.

Edited April 14, 2020 by helonorth

[SBuzzkill]

It would change if the throttle was moved to flight and RPM was low, wouldn't it?

Edited April 15, 2020 by SBuzzkill

[Spike]

The FLI indicates Ng, T4 and TQ (in laymen’s terms: engine speed, temp and power). The FLI takes all three of these parameters and combines them into one indicator. However, all 3 are displayed numerically to the right of the FLI on the same indicator. The FLI does not indicate rotor RPM. That is displayed in its own indicator.

What the report tells us about the FLI is; #1; the engine is running and #2; its at a very low power setting.  As report and others have pointed out, the lack of collective input (lowering) once the TWIST GRIP was turned off, the RPM decreased which was worsened by the 8 degree nose down attitude and subsequent high speed.  A bad combination. Pete Gillies authored the “Cyclic back” article many years ago so pulling the nose up when entering into an auto is not new information.  

Thank you iChris for the clarification.  

[HawaiiCFII]

Just a note. The throttle on the dual fadec versions of the B3 has two positions. Flight and idle. It snaps into position to flight after you start rolling from idle. If you roll it on with a gentle pressure it will move itself in to the flight position, via a spring force, at some point after you start rolling it. So it's in a detent and also being held there by the spring system. 

You cannot control the speed that the engine spools up by rotating it slowly or quickly. There is no manual control of the engine with the throttle if the fadecs fail. It's essentially a microswitch and the same function could be performed with an on/off toggle switch.

I don't know if there is actually a spring in the throttle system but thats exactly what is feels like.