Glider accident while filming commercial in 2011. NTSB Report updated

On Thursday, June 20, 2013 8:07:12 PM UTC-4, Bill D wrote:
There are a bunch of troubling things in this report. I have a lot of ground launch (auto tow) time, but it's a bit dated, so maybe I've forgotten. IIRC, our target speeds were on the order of 50MPH or 55MPH in zero wind, not 70. This was with a Grob and 2-33, but I can't imagine that 70 is right.



---------------------------



From the DG1000 flight manual:

"Recommended winch launch airspeed 110-130 km/h (60-70 kts.)."



Theoretically, the best airspeed for winch launch is that which give the angle of attack for best L/D. With the extra wing loading due to the rope pull, it would certainly be in the 60 - 70 knot range.

Right... but IIRC, the vehicle speed is less (significantly?) than the airpseed when the glider is in the aggressive climb portion of the flight profile. If I read the report correctly, it suggests the tow vehicle was briefed for 70. I have to go dig up the diagram, but the airspeed is the additive of some climb vector involving the secant or arc-tangent or some other function on my old Ti-35 that I no longer remember how to use, no? Like I said, it's been 25 years since I was doing 3-5 ground launches per day, so I could easily be off base.

With the increased interest in ground launch, I do think it's useful to revisit some of the old rules of thumb that used to be well known. Sure, we need the SOP and manuals (which are still out there), but it seems to me that it's good to review the underlying physics every now and then.

Max ground launch speed for my LAK-17a is 75 kias.

I had many simulated failures during training but only one real break in
service and I climbed away from that.

I'm sure some folks will argue, but my technique is stick full forward NOW
and then recover to controlled flight and decide whether to land (and in
which direction) or climb away. Stick full forward does not result in a
dive if you follow up with proper control inputs.

Oh, yeah... I don't think it's wise to exceed 45 degrees in the climb.


"Bill D" wrote in message
...

There are a bunch of troubling things in this report. I have a lot of
ground launch (auto tow) time, but it's a bit dated, so maybe I've
forgotten. IIRC, our target speeds were on the order of 50MPH or 55MPH in
zero wind, not 70. This was with a Grob and 2-33, but I can't imagine that
70 is right.

---------------------------

From the DG1000 flight manual:
"Recommended winch launch airspeed 110-130 km/h (60-70 kts.)."

Theoretically, the best airspeed for winch launch is that which give the
angle of attack for best L/D. With the extra wing loading due to the rope
pull, it would certainly be in the 60 - 70 knot range.

On Thursday, June 20, 2013 6:38:01 PM UTC-6, Ramy wrote:
On Thursday, June 20, 2013 5:16:45 PM UTC-7, noel.wade wrote:

On Thursday, June 20, 2013 4:55:19 PM UTC-7, Papa3 wrote:



Finally, it's hard to tell, but it sounds like this wasn't a failed recovery from an extreme nose up low altitude rope break. From the text, it appears that it may have been a stall-spin after the initial recovery - maybe trying to perform a full 180 from low altitude? At least, that's how I read it.







Anyone else getting the same picture?











This happened in my area and I was marginally involved in the aftermath (mostly second-hand, so apply some grains of salt as-necessary). Some things to keep in mind, regarding this accident:







1] It was done during the filming of a commercial, NOT a normal launch. There were goals to capture certain maneuvers and angles on-camera.



(You can provide your own conjecture about how that may or may not have affected people's thinking, actions, and/or safety-margins.)







2] The filming location was not chosen for its safety, it was chosen for its aesthetic appeal and/or possibly other factors (that I was not privy to). What I can say for a fact is that several other airports exist in the region that have much longer & wider runways, fewer obstructions, more taxiways, and more "outs" in case of an emergency. (And at least one person told me they suggested one of those other airports to the accident-pilot, sometime in the days or weeks before the accident).







4] Eyewitness reports by fellow pilots (to me) corroborate the NTSB report about the glider returning to "level" after the rope-break, prior to entering a turn/spin. You may speculate about whether the nose-down pitch recovery was done properly; or perhaps whether the pilot recovered properly but then got distracted looking at his emergency landing options. Or perhaps he lost track of the tow-vehicle and was afraid to pitch down and land on the runway area for fear of hitting the car/camera-crew. Or perhaps he had a plan but some combination of low-airspeed and/or cross-wind and/or wind-gradient and/or turbulence from the nearby trees resulted in a loss of lift on one or both wings.







Just remember its speculation. Speculating can be valuable to help us think through potential hazards and guard against them in our own experiences; and I think its worthwhile to play "what if" with accidents with a view towards making ourselves safer. Certainly lessons can be learned. But always keep in mind that we don't _know_ what was going on in that person's mind, or what aggravating factors might have tipped the situation out of "potentially unsafe" territory and into an "accident".







--Noel



Well said Noel.

So much for the "wait for the NTSB report" I keep hearing after every accident. After 2 years the NTSB report didnt tell us anything new. Since there may be new readers on RAS which did not read about this accident, discussing it further can only help.



Ramy

I thing the NTSB report gives a lot of information and a pretty clear description of what happened.

The proper recovery from a ground launch failure is to pitch over at zero G until the nose is as far below the horizon as it was above at the point of failure then recover from the dive to a normal glide at 1.5xVs. Zero G eliminates induced drag so the glider retains more airspeed. Height loss from a rope break is determined by the minimum airspeed at the top of the ballistic trajectory. This airspeed may be below the 1G level flight stall speed but the glider remains unstalled at 0G.

From the NTSB report:

"Other witnesses located adjacent to the departure runway reported that the first stage of the automobile ground launch appeared normal, and the glider became airborne within the first one-third of the runway. Shortly thereafter, about three-quarters of the way down the runway, the glider pitched to a steep nose-high attitude. As the glider ascended through about 100 – 125 feet above the ground, the rope slackened. The glider continued to ascend, and then leveled off about 200 feet above the end of the runway. Shortly after, the glider entered a steep right bank and descended into the ground.. As it descended, the glider turned approximately 300 degrees from its initial departure heading before it impacted terrain."

If this witness is to be believed, the launch failed at a steep nose-up attitude followed by a continued climb ending level flight attitude. This indicates no attempt at airspeed recovery. A turn was then initiated from what was likely a critically low airspeed resulting in a spin to impact.

I do lots of winch launching but have never even seen an auto-tow. The report says the rope length was 234 foot. Is that normal or even possible? Using a winch, at my club we get 800 to 1200 foot of launch height from 2,100 feet of winch cable - so launch height is a bit more than one third of the cable length. How on earth could you get to a safe launch height from a 234 foot rope?

Forgive me for teaching granny to suck eggs, but some readers may not have had training or experience in this. As a general comment on ground launch failures, firstly it is vital not to rotate too rapidly and not to be too steep too low. However, once in the full climb at a safe height, if a launch failure occurs, it is vital to lower the nose of the glider rapidly well below a normal glide attitude. If the nose is lowered only to a normal glide attitude after a launch failure in a steep climb, the speed may well have reduced below a safe flying speed, with the potential as here for a stall/spin. It is common to teach a rule of thumb that the nose should be lowered as far below the horizon as it was above the horizon before the launch failure.

Mark Burton, London Gliding Club, UK


On Friday, 21 June 2013 01:53:19 UTC+1, Papa3 wrote:
On Thursday, June 20, 2013 8:07:12 PM UTC-4, Bill D wrote:

There are a bunch of troubling things in this report. I have a lot of ground launch (auto tow) time, but it's a bit dated, so maybe I've forgotten. IIRC, our target speeds were on the order of 50MPH or 55MPH in zero wind, not 70. This was with a Grob and 2-33, but I can't imagine that 70 is right.







---------------------------







From the DG1000 flight manual:



"Recommended winch launch airspeed 110-130 km/h (60-70 kts.)."







Theoretically, the best airspeed for winch launch is that which give the angle of attack for best L/D. With the extra wing loading due to the rope pull, it would certainly be in the 60 - 70 knot range.



Right... but IIRC, the vehicle speed is less (significantly?) than the airpseed when the glider is in the aggressive climb portion of the flight profile. If I read the report correctly, it suggests the tow vehicle was briefed for 70. I have to go dig up the diagram, but the airspeed is the additive of some climb vector involving the secant or arc-tangent or some other function on my old Ti-35 that I no longer remember how to use, no? Like I said, it's been 25 years since I was doing 3-5 ground launches per day, so I could easily be off base.



With the increased interest in ground launch, I do think it's useful to revisit some of the old rules of thumb that used to be well known. Sure, we need the SOP and manuals (which are still out there), but it seems to me that it's good to review the underlying physics every now and then.

On Thursday, June 20, 2013 6:58:55 PM UTC-6, Dan Marotta wrote:
Max ground launch speed for my LAK-17a is 75 kias.

Don't confuse max ground launch speed (Vw)with recommended airspeed for winch launch. The DG1000 Vw is 81 knots.

I was supposed to do this commercial.
Instead the company went to the lowest bidder.

I had a much safer script for this and
it would have promoted soaring, but the
car company and the production company
had no interest in any of that. All
they wanted was the low hanging fruit,
which was a glider being jerked into
the sky by a car.

This is very typical in the film biz.

There is a lot I cannot talk about because
of the investigation.

It was a sad event that did not have to happen
and could have easily been avoided.

VI

On Thursday, June 20, 2013 7:10:16 PM UTC-6, waremark wrote:
I do lots of winch launching but have never even seen an auto-tow. The report says the rope length was 234 foot. Is that normal or even possible?

It's not normal or generally advisable but it's possible. With a rope that short, the launch must be planned so the landing is straight ahead or at least with a minimal heading change.

At the end of a day, I've used two aero tow ropes tied together to auto tow to about 300' AGL then glide across the airport infield to land on the ramp.

Disregard... looks like the whole setup was non-standard, so a lot of the usual rules of thumb go out the door. I hadn't noticed the very short rope for example...

On Thursday, June 20, 2013 8:53:19 PM UTC-4, Papa3 wrote:
On Thursday, June 20, 2013 8:07:12 PM UTC-4, Bill D wrote:

There are a bunch of troubling things in this report. I have a lot of ground launch (auto tow) time, but it's a bit dated, so maybe I've forgotten. IIRC, our target speeds were on the order of 50MPH or 55MPH in zero wind, not 70. This was with a Grob and 2-33, but I can't imagine that 70 is right.







---------------------------







From the DG1000 flight manual:



"Recommended winch launch airspeed 110-130 km/h (60-70 kts.)."







Theoretically, the best airspeed for winch launch is that which give the angle of attack for best L/D. With the extra wing loading due to the rope pull, it would certainly be in the 60 - 70 knot range.



Right... but IIRC, the vehicle speed is less (significantly?) than the airpseed when the glider is in the aggressive climb portion of the flight profile. If I read the report correctly, it suggests the tow vehicle was briefed for 70. I have to go dig up the diagram, but the airspeed is the additive of some climb vector involving the secant or arc-tangent or some other function on my old Ti-35 that I no longer remember how to use, no? Like I said, it's been 25 years since I was doing 3-5 ground launches per day, so I could easily be off base.



With the increased interest in ground launch, I do think it's useful to revisit some of the old rules of thumb that used to be well known. Sure, we need the SOP and manuals (which are still out there), but it seems to me that it's good to review the underlying physics every now and then.

On Thursday, June 20, 2013 8:00:14 PM UTC-6, Papa3 wrote:
Disregard... looks like the whole setup was non-standard, so a lot of the usual rules of thumb go out the door. I hadn't noticed the very short rope for example...

With the increased interest in ground launch, I do think it's useful to revisit some of the old rules of thumb that used to be well known. Sure, we need the SOP and manuals (which are still out there), but it seems to me that it's good to review the underlying physics every now and then.



I would suggest reading George Moores papers on winch launch. George has taken the whole subject of ground launch into the 21st century with a very deep and detailed mathematical analysis. His fresh view of the subject is very welcome. His papers can be read in the files section of the Yahoo group winchengineer and winchdesign.

If you'd like to take a look at my training materials, they can be downloaded as PDF's.

Winch pilot training guide: https://dl.dropboxusercontent.com/u/...20Rev%2017.pdf

Winch training slide show: https://dl.dropboxusercontent.com/u/...20Training.pdf

Airfield operations: https://dl.dropboxusercontent.com/u/...operations.pdf

Le jeudi 20 juin 2013 21:53:31 UTC+2, Steve Leonard a écrit*:
http://www.ntsb.gov/aviationquery/br...FA010 &akey=1



Happened to be looking through the NTSB Database and saw that they updated the report about a month ago.

It looks like they tried to more or less duplicate an old commercial by Michelin, where you saw an ASK-21 being auto-towed on a frozen lake. The glider went brutally up on a very short rope and released almost immediately. Then the car came to a braking halt just in front of an obstacle, while the glider was flying away. It was to illustrate the grip of a "Drice" winter tire... I can't find it on the web anymore. Some small stills are visible if you google "Michelin Drice" for images, but the links are dead.