A little over 40 years ago an event occurred which stongly influenced my
thinking about spinning accidents. At the time I was employed as Deputy
Chief Instructor at a large, full time gliding club in the UK. On the day
in question I was flying the tow-plane. One of our club members, a fairly
experienced pilot with about 400 hours and a Gold Badge (very experienced
in 1972), was flying in a club glider (Ka6e). During his approach to land,
he changed his mind about where on the airfield he wanted to finish and
entered a turn at very low level over the middle of the airport. He spun
off the turn and the glider was comprehensively destroyed. I was in the
pattern at the time and landed alongside the wreckage within less than two
minutes of the accident.

The pilot had already extricated himself from the pile of firewood and was
standing next to it, apparently uninjured. I jumped out of the tow-plane
and walked up to him; what he said is stamped unforgettably on my brain.
His first remark was rather flippant, "I'm sorry, I seem to have parked it
rather untidily." The next remark was very revealing (remember this was
only about 2 minutes after he had spun in), "I can't understand what
happened, there must have been something wrong with the elevator, I kept
pulling back on the stick but the nose wouldn't come up."

Five minutes later he realised he had spun in. When he wrote out an
accident report the next day, he stated that he had entered a spin at low
altitude with insufficient height to recover and had apparently forgotten
that he did not recognise the spin at the time.

He had been trained, about 8 years earlier, in a regime which did include
spinning and recovery in pre-solo training, but there was no requirement
for anything like a BFR or annual check so long as he remained current. In
all probability he had not seen a spin from inside the cockpit for a number
of years. Little wonder that he did not recognise it instantly.

When a glider starts to spin, its acceleration downwards is somewhat less
than that of an object in free-fall and it stabilises after about one turn.
That first turn takes about 4 seconds and breaks down thus (Figures
measured on tests I conducted, mainly in a Puchacz): 1st second, height
loss about 20 feet, pitch down 20 - 30 degrees (recovery, by just moving
the stick forward will lose another 20 - 30 feet); 2nd second pitch down to
about 40 degrees, height loss total about 50 feet (recovery by moving the
stick forward and perhaps some opposite rudder will lose another 75 - 100
feet); 3rd second total height loss about 80 feet, pitch down about 50
degrees (recovery, opposite rudder and stick forward height loss an
additional 150 - 200 feet); 4th second, pitch down about 60 degrees, height
loss a bit more than 100 feet, spin now fully developed (recovery, full
opposite rudder stick forward, height loss an additional 200 - 250 feet).

If you spin at 300 feet and recognise and initiate recovery in 1 or 2
seconds you will get away with it, if it takes you 3 seconds you might just
get lucky, more than 3 seconds and you're gonna crash - hard. I should add
that my friend in the accident described above was probably somewhat below
100 feet when he spun, so he had even less time to save himself and only
about 3 seconds from departure to impact.

The FAA system, which does not require any spin training until training as
an instructor, cannot be expected to produce pilots who will recognise an
unintentional spin, purely from description, in only one or two seconds, so
departures near the ground are highly likely to result in spin-ins. Even
pilots trained under the UK system (which does include spinning and
recovery practice as part of the pre and post solo training), can't be
expected to recognise an unintentional spin that quickly if they haven't
seen and practiced one for months or years. The only thing that will work
is frequent practice and only instructors who are teaching spinning
regularly are really likely to get enough.

The pre-stall symptoms that warn of a stall, are normally readily apparent
when tou are doing a deliberate stall and looking for them, they are not so
obvious when the stall is not intended or expected, and attention is
elsewhere (on centering on a thermal or sorting out where you are going to
land, for example).

Glider pilots often tend to fly very much by attitude with the ASI as a
secondary reference, when you are very close to the ground quite small hill
or even just a row of trees can make the nose look further down than it is.
Below 500 feet AGL, glance at the ASI every 3 - 5 seconds.

At 04:10 02 September 2012, John Sullivan wrote:

At altitude thermals flow generally vertically relatively unrestricted.
At birth, even on a perfectly flat surface, thermal air must
transition from a flat, shallow disk shaped zone feeding in from
360 degrees, crashing in, upwards ,which introduces a rotational
component. Add orographic features and wind effects to these
forces occurring in such a short period of time, over a relatively
small area, and the air is very chaotic indeed.

At 00:08 02 September 2012, Andrew wrote:
I also have wondered how an experienced pilot can spin in from
low thermalling. It must be a full spin, with the resulting steep
recovery dive, that causes a spin-in accident: a stall doesn't lose
much height. Even the most spin-eager gliders I've flown (Dart
17
and Puchacz) always signalled an approaching stall in plenty of
time to stop an unintended spin developing. I've never
accidentally spun while thermalling at normal altitudes, not once
in
my several thousand hours of gliding, so I wonder why it would
be
more likely to happen low down.

One explanation might be that pilots are very stressed when
circling low, and simply don't fly as well as usual. Or maybe they
are circling unusually tightly, perhaps in a small thermal. I suspect
experienced pilots would not make these mistakes.

So I think that John Cochran's comment of August 28th may be
right. There may some unknown, unexpected risk when low-
and-
slow that catches pilots out, even the best ones. Maybe small,
strong, bubbly thermals exist low down, that can perhaps
suddenly stall one wing? That would produce an immediate,
uncontrollable, violent roll, somewhat like a flick manoever,
without any advance warning signals. Not technically a spin, but
probably ending the same way: a steep dive with insufficient
height. If true, that's a risk that no amount of pilot skill can
prevent, except by adopting the sensible rules I was trained
with:

1. never thermal below pattern altitude, and
2. always fly at approach speed below pattern altitude.

I have had one personal experience that supports John's
suggestion: after a normal thermal flight in the midwest, on a
day
with light winds, I was on a normal final approach with wings
level,
at normal approach speed (60kts). At about 100ft agl, without
any
warning, my starboard wing was pushed rapidly and smoothly
upwards, and despite immediate full opposite control input, I was
put into a steep bank, I'd estimate close to 45 degrees. The
surge
vanished as fast as it had arrived. After it stopped, I was able to
level the wings, correct the heading, and made a normal landing
further down the runway. It totally surprised me. I assume a
narrow thermal bubble lifted off under the starboard wing just as
I
passed. I estimate the surge lasted about three seconds, so at
60kts it must have been about 300ft long. I would not have
believed it, except that it happened to me. I'm sure it would
have
been much harder to cope with, if I had not been flying at
approach speed. Perhaps if I'd been flying slowly, at a higher
angle of attack, the surge might have stalled the starboard wing.
This roll event was also seen by an experienced pilot observer on
the ground, who said he was astonished to see it, and inquired
about it after I landed.

A roll upset like this has only happened to me once, so
(thankfully) its clearly a very rare occurrence, and maneageable
at approach speed. If such bubbles are baby thermals, they are
probably only in small areas, miles apart, and short lived, so
would also be rarely encountered. However a pilot who is
attempting to thermal low is presumably intentionally over an
area
where baby thermals are forming, so may have a higher chance
of encountering such an effect.



At 17:08 28 August 2012, John Cochrane wrote:

One point not reiterated yet here -- the atmosphere down low
is
very
different from what you're used to at 2000 feet and above. It's
easy
to say "I haven't unintentionally stalled /spun in a thermal in a
thousand hours. How much of a dope do you have to be?"

A short list of what's different down low: The atmosphere is
much more
turbulent. Thermals, such as they are are much smaller. In this
layer,
many small punchy thermals will start. Many will die. The ones
we use
up higher consist of many little parcels of hot air that have
coalesced. Most thermals are either short lived, or basically
unworkable to a modern glider. You're in the boundary layer
where wind
is being affected by the ground, so there is wind-induced
turbulence.
Punches of strong lift/gust followed by sink when you make a
half turn
will be the norm.

The ground picture will be totally different to the pilot. If you
turn
downwind at altitude, you don't notice that much. If you turn
downwind
at 300 feet, all of a sudden the ground will rush by and, this
being a
high stress moment, you may pull back. Just as the gust you
turned in
fades, or the thermal turns to sink. And when the canopy fills
with
trees going by at 70 mph, the urge to pull back will be really
strong.
You may push forward to recover at altitude, but it's really
really
hard to do with the ground coming up fast.

So, just because you've never unintentionally spun at altitude
does
not mean your chances at 300 feet are the same.

Not to raise a tired subject, but why we give out contest points
for
thermaling at 300 feet or below remains a puzzling question to
me.

John Cochrane