just out of curiosity I went through my OLC flights from last year.
Looking at the E value and averaging it between 9 flights resulted in
an average L/D of 20.25 for my Cherokee which I flight tested in
perfect smooth air conditions at 23.5. Sounds about right.
The above describes one way to develop what I think of as a 'working L/D' for
your glider, your soaring style and 'typical conditions.' I developed my own
w/o benefit of electronics/GPS/software, and it was something over 30:1 for my
never-detailed 15-meter glider, which model Dick Johnson measured at
37.something. Given the relative crudity of my measurements, the
routinely-in-motion atmosphere through which the measurements were taken, and
the fact I 'rarely' flew at max L/D, the disparity between numbers seems
entirely sensible to me. I'd hope (expect?) anyone desiring to fly XC does
something similar before heading out while boldly depending on their glide
computer to get them home sans risk of a landout. That noted, don't mistake a
lower number (and I *have* experienced 30+ mile glides in the PM exceeding
60:1 L/D) for manufacturers' 'brochuresmanship.' In the absence of controlled
test conditions, you are NOT evaluating your ship's max L/D.
(manufacturer claimed 51.5.
> How the hell could DG claim a half percentage point in 51???? That
> represents a 1% accuracy! What nonsense!)
Um...how the hell did whomever evaluated Robert Harris' former world altitude
record determine he soared to 49,009 feet? Nine feet?!? 9/49,000 = .018%
accuracy. Bogus? Not a bit; it simply 'fell out in the wash' when doing the
evaluative arithmetic. No need to impute sinister motives to anyone...
- the Stemme around 42
> (claim approaching 50) almost comparable with the DG and the
> SparrowHawk around 29 (claim 35)
> If you haven't done these seat of the pants measurements with your
> own gliders I would suggest you all do them. It takes a little
> practice and time to do the averaging over many flights. Interestingly
> after a time and with practice you will find the measurements become
> quite consistent which suggests that they might represent something
> close to reality.
We're in agreement here, but don't mistake the numbers for anything other than
what your ship delivers, with your flying style, in 'typical conditions.' Then
be happy, don't worry!!! (Oh... and be prepared for the occasional landout,
too, I don't care WHAT your computer may have told you ' a while back!')
Regards,
Bob W.
Do you really spend all your flight time at best L/D. You must fly in
really crappy conditions. I feel sorry for you.
Andy
> Do you really spend all your flight time at best L/D. You must fly in
> really crappy conditions. I feel sorry for you.
>
> Andy
I was going to say something similar, but without the pathos.
The glide angle I most need to know is at around 80 knots dry / 90
knots wet. I rarely fly at best L/D as it is generally not useful for
X-C flying and certainly not for racing except under extreme
circumstances. Even when I am low and scratching I generally fly 10
kts above best L/D as the tradeoff between glide angle and forward
progress argues for biasing towards a higher speed in the flat part of
the polar.
When I am flying slow I don't count on achieving the quoted L/D. At
that flat a glide angle any air motion will blow the glide out of the
water. Think about a 50:1 glide over 25 miles - if you experience 500
fpm in sink for 60 seconds you will all of a sudden need an L/D of
62:1. Generally I won't fly a final glide at less that 3-4 knots
McCready. If I do that then I can set the computer to read out arrival
altitude and modulate my speed depending on whether the arrival
altitude is going up or down over time. My experience is that I need
to fly around 8-10 knots below the calculated speed to fly to arrive
at the intended altitude. With this technique it pretty much doesn't
matter how accurate the polar in my computer is.
While cruising on course my achieved L/Ds as calculated by SeeYou run
anywhere from the high 40s in to the 100s at cruising speeds of 85
knots and up. This is way above the polar so the polar's not really of
any use unless I'm trying to cross a big blue hole in which case I
revert to the technique in the above paragraph.
9B
If you do the math you'll notice you have an 8 point or so discrepancy
in all 3 of your gliders performance results.
You think that's coincidental or maybe your testing technique?
Simply put, you are comparing an average computed over all
the whole range of speeds you fly at in unstable rising/sinking
air.. with a single point of LDmax at a single speed in completely
stable air.
This doesn't work.
Read one of Dick Johnsons flight tests where he describes how
difficult it is to accurately measure LDmax, then ask yourself
why he does not use your technique...
Best L/D is just one number that has dominated marketing for gliders.
Like most things marketing it is subject to a lot of creativity....
Actual performance, how well a wing uses energy from vertical gusts, how
it climbs, how sensitive it is to contamination, whether it gets
distorted over time. All these will affect how far and fast you fly -
Best L/D is a useful "summary" but it is a generalisation and subject to
a deplorable level of hype and exaggeration.
So - Real world performance is affected a lot by wing loading, and
profile chatacteristics. Your best L/D is a nice easy number to compare
relative performance, but it is a measure in a flight regime that you
will very seldom occupy. (minimum weight, still air, cool dense air,
slow flight) In the real world it is high wing loading wherever
possible, as fast as prudent and turbulence (aka lift) is good.
Predictably the published L/D is sometimes a poor indicator of overall
performance. The DG600 is a classic example.
So - an example
My Std Cirrus has a best L/D of say 36. I have cleaned her up, sealed
everything and made her as good as she gets. However, one just about
never flies at 95km/h - so my achieved L/D in her varies between 22 and
32. What she does well is climb, particularly in rough air.
What she is bad at is dolphin flying - that wing is very rigid so no big
AoA changes please or you are out of the drag bucket and it all goes
downhill...
What she is absolutely awful at is contamination - particularly water on
the wings converts the glide performance to Ka8 standard.
The Kestrel with it's 19m wing is magnificent at 1:44 at 97kmh. Real
world final glides get me > 40. But notethat the polar is quite steep at
higher speed. So in low to moderate speed flight she is very efficient,
and will happily run at very hight L/D numbers. But on a strong day you
have the problem that performance deteriorates fast over say 170km/h. On
a weak day the Kestrel will thrash a Ventus (which also has best L/D of
1:44), but if the average climb value gets above say 3m/s - the Ventus
disappears into the distance. Better climb and same "best" L/D count for
nothing when it comes down to a drag race. Here it is wing loading and
how flat the polar is.
An extreme Example
The Bergfalke II/55 has a best L/D of 27 at around 82 km/h - the Blanik
L13 has about the same 1:28 but at a more usable 90km/h. Now, while two
seat contests were won in the 70s with the Bergfalke 3- you don't REALLY
want to go XC in either of these ladies. But if you were enthusiastic
enough to attempt it - you would soon discover the vast difference in
achievable XC performance between the two.
On 2011/01/11 11:02 AM, Chris Wedgwood wrote:
> Read one of Dick Johnsons flight tests where he describes how
> difficult it is to accurately measure LDmax, then ask yourself
> why he does not use your technique...
>
> Chris
> www.condorsoaring.com
>
--
Bruce Greeff
T59D #1771 & Std Cirrus #57
snip
> you would soon discover the vast difference in
> achievable XC performance between the two.
This posting gets my vote for "best overall view of the situation".
I routinely exceed Schleicher's 50:1 claim for my ASH 26 E by 10% to
40%, flying 15 to 20 knots higher than best L/D. That's "Mean L/D" from
SeeYou statistics. It's easy in good conditions with plenty of lift,
cloud streets, or ridge lift. If the Mean L/D drops under 50:1, it's
almost always been a bad day with lift hard to find. So, I really doubt
this L/D statistic has any value for determining a point on your
glider's polar.
It is instructive to compare your statistics for the day to another
pilot flying a comparable glider. I've been surprised at how different
they can be, particularly the number of thermals taken, how fast they
cruise on average, and the percentage of circling times.
--
Eric Greenwell - Washington State, USA (change ".netto" to ".us" to
email me)
Mean L/D is actually a very poor statistic. 1/ Mean D/L is a lot
better. Seriously now, they are very different. As you go through
lift, L/D passes through infinity and then becomes negative. 1/ Mean D/
L is much better behaved. Now, which one do our computers really
present???
John Cochrane
My mean L/D is always much better than that. I routinely make flights
of 150-250 miles with a net loss of altitude of 2000 ft or less.
That's an achieved L/D (if we want to persist in using that term out
of context) of about 450/1. Pretty impressive for a standard class
glider. The manufacturer only claims 44/1.
Why do we want to continue using the terms L/D, and best L/D, out of
context?
Andy
> If you haven't done these seat of the pants measurements with your
>own gliders I would suggest you all do them. It takes a little
>practice and time to do the averaging over many flights. Interestingly
>after a time and with practice you will find the measurements become
>quite consistent which suggests that they might represent something
>close to reality.
Hi Dave,
as others already pointed out, you are measuring an L/D that is
typical for your flying style, but not the performance of your glider.
In my AS 22-2 I usually get an L/D of 75 (!) on long flights, although
in reality its maximum L/D is only about 55.
Andreas
The "correct value" at which speed?
I suspect that you don't want the correct value, but the correct polar
in your flight computer - which, obviously, cannot be measured in the
way described.
Andreas
L/D numbers are the things dreams are made of. Max L/D is much like
breast size, it gets a lot of attentions but the important thing is
what you do with it.
Ron
This flight computer I used required 3 pieces of speed/sink rate data
to represent the polar. Best L/D, speed at best L/d and V2.
Sorry to be pedantic but we need to be careful how we throw these
terms around. L/D is Lift divided by Drag which has little to do with
distance covered for altitude lost in cross country soaring. D/H has
far more to do with pilot skills selecting lift, avoiding sink and
decision making than with aerodynamics. It's unfortunate some glider
software misuses the term L/D when they mean D/H
L/Dmax is a one-dimensional metric used as a shorthand to evaluate the
aerodynamics of a glider design. I think that is what the OP was
asking about. While it's true L/Dmax is not commonly used in cross
country soaring, it is nonetheless a useful metric for aerodynamicists.
>This flight computer I used required 3 pieces of speed/sink rate data
>to represent the polar. Best L/D, speed at best L/d and V2.
This what all flight computers do - they convert these three points
into an approximated polar curve.
But how do you get these three points by the method described by Dave?
Andreas
Well, I'll bite. Here's what I do for the planes I fly:
First, I've been fortunate enough to fly planes that have Johnson
reports
available. I know there are probably Akaflieg reports as well, but I
haven't
seen those. I carefully extract the data points from Dick's polar
charts and
correct them for my flying weight (unfortunately always considerably
higher
than Dick!). I input the adjusted values into my PDA (which just
wants
the sink rate at 3 airspeeds, rather than the numbers listed above).
Finally,
I set the Polar Potential via experiment. Typically I'll set it to
90% and then
see how well my final glides work out. If I have a bunch of altitude
left over
on a glide then I'm doing better, and I'll increase the potential. If
I tend to fall
below glideslope a lot then I'll decrease the potential. For the most
part I've
wound up with values around 90% or 92% (which probably means I need to
work harder at tuning up the planes I fly).
Essentially this is a refinement of the beginner approach to glide
slopes:
take the published value and divide by 2 as a safety factor. I divide
by something
closer to 1.1 and usually make it home just fine. The times I've had
to
break off have been because I was below glideslope to begin with.
-- Matt
2 is a bit harsh!!
I use 32.8:1 plus circuit height no matter what I'm in :-)
It's about right for a PW5 (or K6), about your factor of 1.1 for a
Libelle or Grob Twin, and maybe 25%-30% pessimistic for a Janus or
DG1000.
Those are also in order (and I think reasonable proportion to) the
cost and embarrassment and inconvenience of breaking each of those
gliders!
It also happens to be 100 ft per km, which doesn't require a computer
to calculate.
-John
The numbers the OP gets from his PDA, and what SeeYou provides on the
Statistic page, are labeled "L/D", so that's probably why we were using
them. SeeYou also provides "Mean L/D" separately for rising air (it's
negative) and sinking air. So far, no Mean L/D beyond about 80, so
infinity is safe from me!
>
> My mean L/D is always much better than that. I routinely make flights
> of 150-250 miles with a net loss of altitude of 2000 ft or less.
> That's an achieved L/D (if we want to persist in using that term out
> of context) of about 450/1. Pretty impressive for a standard class
> glider. The manufacturer only claims 44/1.
An extreme example of why just flying around doesn't yield very good
polar measurements.
>
> Why do we want to continue using the terms L/D, and best L/D, out of
> context?
I do it out of habit, history, and continued use of the term by the
manufacturers and other pilots. I don't know why the manufacturers
continue to use it, but I'm going to guess: habit, history, and
continued use of the term by their customers.
--
Eric Greenwell - Washington State, USA (change ".netto" to ".us" to
email me)
- "
Sounds like you have some very nice gliders, Dave! I'm pretty new to
soaring, but isn't the manufacturers L/D really a "best case" kind of
number, meaning when the moon is in the seventh house of aquarius and
all the elements are aligned in ultimate perfection? I've heard the L/D
can be dropped by all manner of things like water droplets on the wing,
bug splatter, and the stern look of a angry red-tailed hawk. Still, it's
good to know the real numbers. As it is, it looks like you can just
estimate that reality is a percentage (all three of yours are between 82
and 84%) of dreamland L/D numbers from manufacturers. Also, and I'm just
wondering here, since I don't really know, could it have something to do
with the way you fly? Or how much weight you're carrying? CG location?
Wouldn't there be many factors?
--
Sparkorama
I don't think the misuse has anything to do with manufacturers, if you
mean glider manufacturers. They use the term best L/D in its proper
context. The fidelity of the number is a separate discussion.
I think a lot of the blame for the misuse of "L/D" lies with SeeYou
flight analysis software. There are far too many people that believe
anything that appears on their computer screen.
In any case the knowledge of the glide angle achieved in any segment
of a flight is meaningless without taking into account the wind
velocity and average ground speed for the flight segment. If those
and the glider polar are known then something can be deduced about the
behavior of the air mass for that flight segment. Similarly if they
are known, and the air mass movement is known, then something can be
deduced about the glider polar.
So how about it SeeYou people. Can you please stop using the term "L./
D" out of context?
Andy
In general - to be usefully able to predict performance they all try to
match actual performance against a polar curve (L/D graph) - which they
calculate by taking at least three points on the polar and doing a fit
to these points. Clearly the maximum value is significant so they want
the speed and quantum of minimum sink at measured minimum sink + a
higher number (preferably in the cruise speed range) + a sink rate at
minimum speed or close to it. Then the resulting graph sort of relates
to the actual performance - it gets complicated and bumpy for ships with
flaps, and some airfoils have kinks and bulges in their graph.
So in all cases the polar curve gets estimated - it is a model - all
models are false, some models are useful. In this case the polar model
is a useful approximation of glider performance under standard
atmospheric conditions, at a specific wing loading and speed.
How well that matches to your aircraft, your conditions and your flying
style varies. But at least the flight computer can give you a place to
start.
--
Exactly. L/D/speed are used to establish one point on the polar curve.
I don't think anyone knows how accurate flight computers are in
predicting performance but they are probably more helpful than looking
out the canopy and guessing.
Actually, SeeYou Mobile does measure and display your "current L/D" (over
the ground) based on the measured sink rate and distance over the ground
over - averaged over some period of time (20 seconds?). On a low wind day
it could (perhaps) be used to determine the L/D at specific airspeeds. You
could fly approximately upwind at a given speed for a minute, read the L/D,
then do a 180 degree turn and do it again for another minute. Write down
both numbers and average them. Then do it at a different airspeed. Of
course, your altitude could throw off the numbers a lot. The L/D over the
ground would always been correct, but the set airspeed would not be the same
as your speed over the ground if you are at 10,000 feet (for example).
The "Current L/D" window in SeeYou Mobile is very powerful because you can
compare it with the required L/D to see how you are doing on the way to a
goal. I remember being impressed with it one day while flying in the
Minnesota Soaring Club's SZD Junior. The Junior is a wonderful glider, but
it has thick wings and doesn't penetrate wind very well. I was fighting a
20+ knot headwind to get to my destination and could see the destination
clearly. I noticed that my required L/D was onlly 20. The Junior can
theoretically perform at a 35:1 glide ratio, but with the strong headwind my
measured "Current L/D" was 12. I liked that the number was an actual
measured performance number, not an estimate based on a previously measured
wind and the entered approximate polar data. I knew for a fact that I
needed another thermal - and I found one. Cool feature!
Best Regards,
--
Paul Remde
Cumulus Soaring, Inc.
"BruceGreeff" <Br...@wird.co.za> wrote in message
news:igp3kk$u58$1...@news.eternal-september.org...
You are right something has become lost in this discussion. The OP
stated "Like you all I have questioned what the best L/D of my 3 (now
1)
gliders...were in reality suspecting that real L/Ds would be lower
than the manufacturers published values". He goes on to discuss how
measurements of glide performance during cross country flights showed
that his measured L/D did not match the claimed best L/D for his
gliders.
Several people have pointed out that the achieved glide angle on a
typical cross country flight has little relationship to best L/D since
XC flights are typically not made at best L/D speed and cross country
flights are typically not made in a motionless air mass.
It should be obvious that the ratio of distance flown to altitude lost
on a complete XC flight, or on any segment of an XC flight, is not a
measure of glider performance but a measure of the combination of
glider performance, air mass characteristics, and pilot skill. It is
my opinion that using the term L/D to describe this ratio is
misleading. Also, comparing this number with a manufacturer's claimed
best L/D tells you absolutely nothing about the validity of the
claimed best L/D value unless the primary objective of the flight, or
flight segment, was to fly at best L/D speed.
The title of the thread is "How to simply determine the L/D of your
glider". Several experienced XC pilots have pointed out that using
reported L/D data from typical XC flights is not a valid way to
determine the L/D of a glider.
None of the above denies that useful performance data can be derived
from making constant airspeed speed runs in a still air mass. None of
the above denies that a flight computer or post flight analysis
software can be useful in processing the data. None of the above
denies that the value of the ratio (distance flown)/(altitude lost)
achieved in a properly conducted flight test or by computer modeling
is properly described as "L/D". None of the above denies that the
maximum value of that ratio is properly described as "Best L/D".
Andy
It's very interesting that you used the term "L/D" three times, and
every single time it was an incorrect usage. What you were referring
to was in fact the distance traveled in losing a certain amount of
height, or in other words the achieved glide ratio.
And yet the ONE time that you could have correctly said "L/D" you said
"glide ratio" instead!!
A given aircraft at a given weight and in similarly clean condition
will always perform at a fixed L/D (lift/drag) at any particular speed
and G loading. It is an aerodynamic property.
The actual glide ratio achieved will depend on head or tail wind, lift
or sink encountered. As well at the air mass movement it is dependent
on the glider's L/D ratio, but it IS NOT an L/D ratio. It is a glide
ratio.
Just look at some ridge flights on the OLC in SeeYou......"mean L/D"
is often 100 or 200 are even 9999........!!!
Meaningless....
Cookie
In principle, no ... (L/D)max is a non-dimensional ratio, so air density
effects and IAS/CAS/TAS effects cancel out*. It's a constant and is
determined by the product of the zero-lift (profile) drag coefficient and
the induced drag factor.
Changes to the profile drag coefficient (dirty wing, flaps, airbrakes,
undercarriage etc) or the induced drag coefficient (flaps, winglets etc)
will affect the max L/D, but as Bruce says, for a given aircraft at a
given weight in a similarly clean condition at a given speed and g-loading
it will be a constant - as long as by 'speed' you mean IAS.
> It paaers we may be close to the root of much confusion. Most
I am sort of partial to the annual battery powered leaf blower
discussion, myself...
Zero One
Sure, lots of things affect it, but it's fairly close if you're
talking IAS which takes altitude, temperature and humidity into
account fairly well in that they affect the ASI in about the same was
as they affect the wings.
It may only be 21 outside (possibly "70" to you) right now but I can
assure you that this is what passes for "summer" in these parts, not
winter.
now THAT'S funny!!!