Ventus 3F - Fatal

On Friday, July 20, 2018 at 6:15:40 PM UTC-7, WB wrote:
On Friday, July 20, 2018 at 6:30:07 PM UTC-5, Jock Proudfoot wrote:
National Transportation Safety Board
Aviation Accident Preliminary Report

The glider was equipped with a FLARM electronic flight collision
alerting device, which records flight track, altitude, and airspeed.
About 11 seconds prior to impact with the ground, the glider entered a
left hand descending spiral at an airspeed of 48 knots, which continued
to ground impact.

https://app.ntsb.gov/pdfgenerator/Re...atorFile.ashx?
EventID=20180710X25659&AKey=1&RType=HTML&IType=FA
.

Doesn't really sound like a stall/spin from low altitude. Sounds more like the behavior of a glider trimmed for an off-field landing with no one at the controls.

WB: In what way does it not sound like stall/spin? Secondly, what do you mean about trimmed for an off-field landing? I've never heard about trimming differently for an on-field landing vs an off-field landing?

Separate remark: Flarm does not record airspeed as stated in the report. It could be that airspeed was inferred from the spiral descent dynamics; but that seems more complicated than suggested by the brevity of this preliminary report.

Flarm obtains aircraft data from GPS and an internal static pressure sensor. Speed would be calculated from GPS position change and possibly altitude change yielding either 2D or 3D groundspeed, not airspeed.

Possibly wind aloft could be estimated from changes in groundspeed on different tracks and used to produce an airspeed, but any gusts would make a hash of such an estimate.

On Saturday, July 21, 2018 at 5:19:42 AM UTC+1, George Haeh wrote:
Flarm obtains aircraft data from GPS and an internal static pressure sensor. Speed would be calculated from GPS position change and possibly altitude change yielding either 2D or 3D groundspeed, not airspeed.

Possibly wind aloft could be estimated from changes in groundspeed on different tracks and used to produce an airspeed, but any gusts would make a hash of such an estimate.

If you download the igc file from the Flarm into a program such as SeeYou, you will get a decent estimate of both ground speed and airspeed, presumably the latter calculated from the GPS ground speed, allowing for density altitude. I am assuming that the investigator was able to do this.

I examined Renny's previous two logged flights in the V3 at Moriarty and he maintained air speeds in excess of 60 knots on the descent to landing.

Given the good handling characteristics of the V3, I find it hard to believe that an experienced pilot like Renny misjudged his approach and entered a stall/spin and that some degree of pilot incapacitation seems more likely.

However, rather than speculate with incomplete information, perhaps we should wait for the final report.

Mike

I would have interpreted the mention of 48 knots as being a descent rate shown on the data recorder file (= around 880ft/11seconds).

On Saturday, July 21, 2018 at 2:15:40 AM UTC-7, wrote:
I would have interpreted the mention of 48 knots as being a descent rate shown on the data recorder file (= around 880ft/11seconds).

The report clearly says airspeed not descent rate. Default Flarm setting is 4 second fixes so 11 seconds represents at most 3 fixes. 11 seconds from normal flight to impact indicates a very low altitude no matter how you interpret it.

On Saturday, July 21, 2018 at 3:34:58 PM UTC+1, jfitch wrote:
On Saturday, July 21, 2018 at 2:15:40 AM UTC-7, wrote:
I would have interpreted the mention of 48 knots as being a descent rate shown on the data recorder file (= around 880ft/11seconds).

The report clearly says airspeed not descent rate. Default Flarm setting is 4 second fixes so 11 seconds represents at most 3 fixes. 11 seconds from normal flight to impact indicates a very low altitude no matter how you interpret it.

The previous two igc files of Renny's I looked at have the default 4 second recording interval. In these, it appears that he arrived back at Moriarty high and perhaps spoilered down, entering the pattern at around 1,000 to 1,500 feet AGL. In both of these landings, he maintained plenty of height and speed through his base and final turns, slowing down only on final just before touchdown. Of course, we have no idea what happened on his last flight, but his previous flights in the V3 show no signs of sloppiness that might result in loss of control.

Mike

On Saturday, July 21, 2018 at 4:34:58 PM UTC+2, jfitch wrote:
On Saturday, July 21, 2018 at 2:15:40 AM UTC-7, wrote:
I would have interpreted the mention of 48 knots as being a descent rate shown on the data recorder file (= around 880ft/11seconds).

The report clearly says airspeed not descent rate. Default Flarm setting is 4 second fixes so 11 seconds represents at most 3 fixes. 11 seconds from normal flight to impact indicates a very low altitude no matter how you interpret it.

FLARM does -not- report or calculate airspeed. The device only calculates positions from GPS fixes and has a pressure sensor to enhance that information, but any airspeed or groundspeed calculations are only speculative approximations from the GPS data in the logs. From the GPS fixes of itself and those transmitted from nearby aircraft it then calculates threat levels and warns you appropriately if necessary.

On Saturday, July 21, 2018 at 10:02:16 AM UTC-5, wrote:
On Saturday, July 21, 2018 at 4:34:58 PM UTC+2, jfitch wrote:
On Saturday, July 21, 2018 at 2:15:40 AM UTC-7, wrote:
I would have interpreted the mention of 48 knots as being a descent rate shown on the data recorder file (= around 880ft/11seconds).

The report clearly says airspeed not descent rate. Default Flarm setting is 4 second fixes so 11 seconds represents at most 3 fixes. 11 seconds from normal flight to impact indicates a very low altitude no matter how you interpret it.

FLARM does -not- report or calculate airspeed. The device only calculates positions from GPS fixes and has a pressure sensor to enhance that information, but any airspeed or groundspeed calculations are only speculative approximations from the GPS data in the logs. From the GPS fixes of itself and those transmitted from nearby aircraft it then calculates threat levels and warns you appropriately if necessary.

Another good reason to go to a 1 sec recording interval. Memory is cheap, why do people not use the best resolution our flight recorders offer?

On Friday, July 20, 2018 at 10:19:01 PM UTC-5, Steve Koerner wrote:
On Friday, July 20, 2018 at 6:15:40 PM UTC-7, WB wrote:
On Friday, July 20, 2018 at 6:30:07 PM UTC-5, Jock Proudfoot wrote:
National Transportation Safety Board
Aviation Accident Preliminary Report

The glider was equipped with a FLARM electronic flight collision
alerting device, which records flight track, altitude, and airspeed.
About 11 seconds prior to impact with the ground, the glider entered a
left hand descending spiral at an airspeed of 48 knots, which continued
to ground impact.

https://app.ntsb.gov/pdfgenerator/Re...atorFile.ashx?
EventID=20180710X25659&AKey=1&RType=HTML&IType=FA
.

Doesn't really sound like a stall/spin from low altitude. Sounds more like the behavior of a glider trimmed for an off-field landing with no one at the controls.

WB: In what way does it not sound like stall/spin? Secondly, what do you mean about trimmed for an off-field landing? I've never heard about trimming differently for an on-field landing vs an off-field landing?
Separate remark: Flarm does not record airspeed as stated in the report.. It could be that airspeed was inferred from the spiral descent dynamics; but that seems more complicated than suggested by the brevity of this preliminary report.

11 seconds just seems like a very long time to be in a spin or spiral dive. 11 seconds suggests adequate altitude and enough time for an unimpaired pilot to recognize the problem and perform a recovery. I am under the impression that most stall/spins to impact start below 200 feet and usually the glider only makes a partial rotation before impact. Just guessing, but wouldn't a spiral dive of 11 seconds duration involve shedding parts before impact?

Regarding trimming for an off field landing: If one is landing in a smaller than usual space, one might trim the glider to fly at the minimum safe approach speed. In my 301 Libelle, I trim for the pattern speed that I want, usually about +-60 knots depending on wind, and slow up on final. If I am landing in a tight field, I might trim down around +-50 knots if it's calm. In my 20, the trim is left to float and the glider trims out based on flap setting (until one gets to the "jeezus!" notch). I've never flown a Ventus of any flavor, and don't know if the Ventus drivers typically even bother to set trim or let it float. As someone else pointed out, 48 knots is also a plausible thermalling speed.

Answering the "separate remark" question: My first thought was that the NTSB report was sloppily written when it stated "airspeed" since the PFlarm wouldn't know airspeed. However, I read that PFlarm will record airspeed if it is being fed that information from another instrument like a modern complex vario. So maybe the NTSB did get airspeed.

It is just terribly frustrating that we keep losing skilled and experienced people to these puzzling crashes. Scares the crap out of me to be honest.

Please fly safely.

One last time... FLARM does not need, nor does it record airspeed, for a simple reason: airspeed cannot be used to calculate closure rates between your aircraft and others, or ground obstructions. It works entirely from GPS fixes. It only uses a pressure sensor to enhance altitude data, but does not strictly need it. To my knowledge, no FLARM devices have inputs for airspeed.

Read the manual, folks. It's all documented there, in great detail.