Does STF depend on wind?
Pat Russell
"Speed-To-Fly Takes Wind Into Account." I always thought that final glides
depend on wind, but not STF. Would some expert explain this to me?
Pat
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
mike pitoniak
I am hardly an expert, but i just happened to
research the same question recently.....
It turns out that speeds to fly do not take into
account wind on interthermal cruise because the your
speed is relative to a moving airmass of which you are
part of. The goal is to optimize the probability of
intercepting a new thermal, and not of optimizing your
progress over the ground. With this in mind you can
think of new thermals as approaching you at the speed
of the wind in the airmass when you penetrate upwind,
therefore no speed increase is neccessary, as the
thermals are being blown toward you at the same speed
that you are loosing ground....they cancel out.
When the goal is to optimize your distance covered
over the ground, the next thermal is of no
consideration, as you are now on final glide. With
this in mind, the calculatons do in fact take into
account the affect of hedwing on your progress. The
actual math is ugly (in New Soaring Pilot) but the
bottom line is that you need to add approx 1/2 the
windspeed to the best l/d speed.
Took a lot of reading to get a mathmatical
explanation...if you really want it i can provide page
numbers or a fax.
hope this helps,
mike
*** Posted from RemarQ - http://www.remarq.com - Discussions Start Here (tm) ***
jonathangere
ads should say is "program bug has been eliminated - Speed-To-Fly NO LONGER
takes wind into account." This is why upgraded LNAV's correctly show
minimum altitude required into a headwind at a Mc setting slightly greater
than 0. This was already the case for the MNAV ten years ago.
Of course, altitude required depends a great deal on winds in any case.
The effect of the bug in the older software is an unintended distortion of
the Mc setting: decreased by tailwind, increased in headwind. Which in
turn causes the altitude required to not reflect the actual Mc setting.
Uncomfortably low downwind, wastefully high upwind. Scratching desperately
in zero sink eventually cures the former, and 120kt final glides by
eyeball, the later.
Jonathan Gere
Pat Russell <pat...@hotmail.com> wrote in article
<76tbge$85h$1...@nnrp1.dejanews.com>...
Robert Ehrlich
> [snip]
> If you're too high, ...
> [snip]
I think beside this "too high" case there is also a "too low" case,
when you are at a height sufficient to reach the finish line, but
not at the optimal inter-thermal speed, for some reason you cannot
continue to climb (cloud base, restricted airspace above, thermal
vanishing) and there is no hope of thermal between you and the
finish line. In this case also your speed is depending on the wind.
jonathangere
> That speed will be slower than
> the optimum interthermal cruise speed, but faster than the
> maximum distance speed.
In the strong headwind-weak lift situation (relatively speaking) we get the
insignificant case of optimum inter-thermal speed < maximum distance speed.
Out on course the confirmed optimum interthermal speed pilot somehow gets
ahead while cruising slower AND with a worse glide ratio relative to the
ground. Somewhat of a paradox.
It turns out this only happens when the optimal achieved speed through the
airmass is less than the headwind, so that it is impossible to make forward
progress. The optimal techniques loses more ground in climbing and gliding
off a given height, but takes more time doing it. The result is a less
negative achieved speed over the ground.
Jonathan Gere
Ed Davies
> I just read an advert in Soaring magazine for a computer. It says
> "Speed-To-Fly Takes Wind Into Account." I always thought that final
> glides
> depend on wind, but not STF. Would some expert explain this to me?
> I just read an advert in Soaring magazine for a computer. It says
> "Speed-To-Fly Takes Wind Into Account." I always thought that final
> glides depend on wind, but not STF. Would some expert explain this
> to me?
We had a long thread in r.a.s about a year ago on exactly this
subject and never got to a final conclusion. Here's my view:
1. STF when flying to a thermal which drifts with the wind is
not affected by the wind speed.
2. STF when flying to lift which does not drift with the wind
(e.g., a wave bar) is affected by the wind.
3. STF on a final glide when the objective is to glide as far
as possible is affected by the wind speed.
4. STF on a final glide when the objective is to reach the goal
as quickly a possible and when the pilot can choose the height
and position of the last thermal exit (the normal racing case)
is not affected by the wind (though the distance out that the
final glide will start for any given last thermal exit height
is affected by the wind).
Point 4 was the main source of contention last time. I've written
a paper on the subject which is at:
<http://homepages.nildram.co.uk/~edavies/ed729fgs/ed729fgs.htm>
The previous discussions were triggered by a question about the
use of wind speed in the Cambridge L-Nav's STF calculations.
Hope this makes sense,
Ed.
Pat Russell
Ed Davies <eda...@nildram.co.uk> wrote:
> We had a long thread in r.a.s about a year ago on exactly this
> subject and never got to a final conclusion.
I'm very sorry I missed it! I thank everyone for tolerating the
rehashing of the subject. I've had several direct email replies,
and I reviewed the books by Reichmann and Welch, Welch & Irving.
The most succinct synopsis is yours, Ed:
> 1. STF when flying to a thermal which drifts with the wind is
> not affected by the wind speed.
>
> 2. STF when flying to lift which does not drift with the wind
> (e.g., a wave bar) is affected by the wind.
>
> 3. STF on a final glide when the objective is to glide as far
> as possible is affected by the wind speed.
>
> 4. STF on a final glide when the objective is to reach the goal
> as quickly a possible and when the pilot can choose the height
> and position of the last thermal exit (the normal racing case)
> is not affected by the wind (though the distance out that the
> final glide will start for any given last thermal exit height
> is affected by the wind).
I believe that this should put an end to the debate on theory. Now for
the bit that still intrigues me: I would expect a final-glide calculator
to deal with all four of the above scenarios, yet it seems that none of
them do! They assume that one scenario fits all situations, and, apparently
one of the brands has recently switched from one scenario to another.
In the past, it was probably true that these distinctions could be ignored
because achieved average speed does not depend strongly on inter-thermal
cruise speed. But these days, with GPS, 60:1, and glides around turning
points, more precision should be expected from our instruments.
jonathangere
on final glide club would probably like the chance to switch to "Have STF
consider the wind" panic mode at the end.
I believe the manufacturers' collective wisdom is that the pilot interface
is already too complex. As a simple first step, I suggest they consider
wind for STF purposes whenever the pilot selects Mc 0. Anyone setting the
Mc to < .1kt is probably more interested in a flat glide over the ground
than speed optimization for the case of zero knot thermals!
In general though, I can't agree that we've exhausted theoretical
considerations yet. Or that instruments can reasonably cover every
interesting scenario. Would you like the optimum street crossing stategy
built in? How about two I've never seen addressed: the optimum strategy
for a perfect "graph paper" sky of intersecting streets and waves or even
for the simpler case of oblique tracks through just wave.
Lots of winter left,
Jonathan Gere
Pat Russell <pat...@hotmail.com> wrote in article
<775883$6fd$1...@nnrp1.dejanews.com>...
Martin Hellman
graphical analysis in there, with associated explanation,
finally got me to
a) Agree that STF is independent of wind when the source
of lift is drifting with the wind. AND
b) Understand why I had my mistaken view before, and
why it was mistaken.
Until now, I had a mental block that kept me wondering
why people thought STF didn't depend on wind, when it
"so clearly did." Of course, what was wrong was my
"clearly." In case my mental block is shared by others
(and I suspect that is the case, judging by the thread),
I will briefly explain what it was, and why it was wrong.
Anyone not interested can skip the following section
enclosed in ====='s.
=====================================
Wrong reasoning: If the no wind STF is less than the
headwind, then the glider will never get to its goal flying
at that speed. But it will get to the goal,
if it flies at a speed greater than the headwind. This
seemed to prove that, at least with extreme
headwinds, STF will be influenced by wind. And, if STF is
influenced by large winds, it is reasonable to suspect
that it will be affected by any wind.
Error in above reasoning: This reasoning only applies to
the final glide portion of the flight. But, as Ed points out
in his paper (and several others did in their postings), in
the phase of flight just prior to the final glide, the
glider is drifting backwards while it is climbing. So, if
the wind speed is greater than the STF,
the glider will never get enough altitude
to get to the final glide portion of the flight! During the climb, it
is getting further and further from the goal, and is doing so
"faster" than it can ever make up the distance during the final glide.
Two additional points, if anyone is still reading:
1. The STF optimization seems to assume that the pilot
can find lift of the Mc setting any time he wants, and ride
it as high as he wants -- a clearly optimistic assumption,
especially when flying through large sink holes.
I've never seen this assumption clearly stated, which
it should be. (Or am I wrong on this too?) I think this
is one reason Cambridge and others recommend using
a smaller Mc setting than the actual lift encountered.
2. When I was contemplating taking delivery of a Stemme
S10VT (yes, I know how lucky I am!), for a few crazy days
I contemplated trying to fly it over the North Atlantic, rather
than having it shipped by boat. Part of my fascination with
this idea was the realization that between the turbocharged
engine and the 50:1 glide ratio, I could stay within safe
gliding distance of land almost the whole way, eliminating one of
the major dangers of the crossing. I forget the exact numbers,
but my reasoning went something like this: The longest
overwater flight was a bit over 400 nm, so the midpoint is 200 nm,
so I need 4 nm = 26k feet (with no safety margin) to be in gliding
distance of one or the other airports. The S10VT was found
to have 400 fpm of climb at 26k feet, so this looked doable,
at least from the engine's point of view.
My next thought though, was that I'd probably be bucking
a large headwind, and that would "clearly" add to the
required altitude. To my surprise, analysis showed that
a pure headwind (or tailwind) actually _reduced_ the
required altitude. Without going through all the math
(It's not that involved, but hard to express in email.),
here is a convincing argument:
For purposes of the example, assume a flight of
400 nm between the two airports, each at sea
level, and a best glide
ratio of 50:1 at 60 kts. Then, with no wind, you
would need 4 nm of altitude at the "decision point"
(when you change your safety airport), in order
to be in gliding distance of one or the other airports
at all times. With no wind, the decision point is
also the midpoint of the flight. Should you lose your
engine at the midpoint, you could glide to either
airport at 60 kts.
Now, suppose there is a headwind of 60 kts. You can
still be within gliding distance of an airport the whole
flight, if your altitude is 4 nm. If you should lose your
engine directly over the goal airport (at 4 nm altitude),
you could still glide to either airport by flying at 60 kts
air speed. If you fly into the wind, you make no headway,
but land at the goal. If you fly with the wind, you
make 120 kts ground speed, so you glide twice as far
as with no wind, or 400 nm, the exact distance back
to your takeoff point!
This example made me suspect that, if after losing
your engine, you fly at the no-wind best L/D speed
(60 kts in this example), the required altitude is
independent of the wind -- assuming it is a pure
headwind or tailwind. A small amount of math
bore that suspicion out.
But here is the kicker: By optimizing your airspeed and
optimizing your decision point, you can clearly (and, unlike
my earlier ones, this "clearly" is right!) do better.
Thus, the required altitude to be within safe gliding
distance of land, should you lose your engine,
is actually reduced by a headwind.
The above argument only applies to a direct headwind
or tailwind. A crosswind, which gives a "headwind
component" on both out and return flights could
(and probably would if strong) increase the required
altitude to be within safe gliding distance of one
of the airports the whole flight. I wonder if the
same applies to the "independence of the MacCready
STF to headwind/tailwind," but am too tired to
figure it out. Any answers out there?
Martin Hellman
Sean
cold-water survival suit, a chase boat and full-coverage insurance <g>.
I can't say there's anything wrong with your math though basing everything
on expectations of constant horizontal winds without veering, backing,
shear, lift or sink components seems a bit optimistic.
Still, it would make for quite a good story if you make it.
Sean
--
se...@direct.ca
Martin Hellman <hel...@leland.stanford.edu> wrote in article
<369B08D2...@leland.stanford.edu>...
Pat Russell
"jonathangere" <jonath...@yahoo.com> wrote:
> As a simple first step, I suggest they consider
> wind for STF purposes whenever the pilot selects Mc 0. Anyone setting the
> Mc to < .1kt is probably more interested in a flat glide over the ground
> than speed optimization for the case of zero knot thermals!
Now that is an excellent suggestion. After all, in real life, the speed
ring setting reflects the pilot's confidence as well as the actual
weather. The more desperate you get, the more you care about distance
(and wind). I am guessing that this is the change made by Cambridge.
Richard Brisbourne
<pat...@DONTSPAMME.worldnet.att.net> writes
[generally a pretty neat analysis]
A couple of minor riders (nitpicks?)
>Until the last thermal you use before final glide. Then you
>*should* use the current thermal strength to set your wind
>independent STF.
>
[re the recommendation to set the M-ring setting lower than theoretical]
>
>There are lots of reasons why one might make this
>recommendation, but I don't think it's because of the basic
>M-assumptions. One reason for using a lower setting is that
>people tend to overestimate thermal strength. Another is
>that the penalty for flying too slow is being a bit slower
>and higher, while being too fast may mean landing out when
>you can't reach the next thermal. An interesting issue I've
>seen debated at length is the effect of wasted "thermal
>centering time" on the "correct" M-setting. It's relevant,
>but I'll leave it for another day.
>
One assumption of STF theory not often quoted is that average climb rate
is independent of interthermal cruise speed. Calculations show that a
drop of 50 fpm in average climb rate makes a lot more difference than a
10kt error in STF (or an error of 30% or so in the climb setting you
use). Get low and accept one weak thermal and all the advantage of your
efficient cruise has gone.
Another way of looking at it is that if you are too slow you get some
trade-off in terms of more choice of thermals, if you are too fast there
is no trade-off.
--
Richard Brisbourne
Richard Brisbourne
<pat...@DONTSPAMME.worldnet.att.net> writes
>>Another way of looking at it is that if you are too slow you get some
>>trade-off in terms of more choice of thermals, if you are too fast there
>>is no trade-off.
>
>the thermal strength ahead. Using classical assumptions,
>there is no advantage in going slower to be able to reach a
>stronger thermal that would otherwise be out of gliding
>range. One should just stop and refuel at the strongest
>thermal one can reach.
>
I think the operative phrase here is "using classical assumptions".
Reichmann put a very apposite worked example in "Cross-country soaring"
that looks at exactly this case. I could dig it out (if I could
remember where I left my copy) or rework it but actually I think it's
pretty intuitive.
If the strongest thermal you can reach flying at 80 kt gives a 1 kt
climb, your correct STF on classical theory isn't 80 kt, it's about 55.
(because 1 kt is all you'll get in the next thermal).
If flying at 55 kt won't just get you to that 1 kt thermal, it will get
you all the way out from under the clag to that patch of sun with a
6-knotter available, you're not going to stop in 1 kt, are you?
But if you then say "right my next thermal will be 6 kt" and set your
MacReady accordingly you won't get there....
In that case your real optimum is as fast as you dare under 80 kt.
Happens on British days with a lot of cloud cover.
--
Richard Brisbourne
JSB
> In the middle of winter, sitting in front of a keyboard, I'd
> even take the clag and 1-knotters :-)
I have had lot's of experience with "1-knotters," but what's a "clag"?
Nick Leaton
> This is an interesting point to discuss, but I don't really
> think of this as part of what I call "classical" MacCready
> theory in which the pilot knows the thermal strengths ahead,
> they are all on the course line, the pilot wastes no time
> centering, and the thermals drift exactly with the airmass.
> In that scenario, the perfect optimum is well defined, and
> average climb rate is optimized by flying the calculated
> optimum STF between each thermal. (Average climb rate is
> not "independent" of the STF, but it is optimum, which is
> better).
There is another assumption. Classical theory assumes that you can reach
the next thermal flying at any speed. If the next thermal is far enough
away, you will have to fly at a lower McCready setting.
--
Nick