The Balance between "% Circling" and "MacCready Speed to Fly"

[Scott Alexander[_2_]]

I have been told over the years two consistent things: when trying to
achieve the fastest possible cross country speeds, 1. Try to minimize
circling as much as possible and 2. Fly the true maccready airspeeds
between thermals.


But these two things contradict. The first couple days I flew at
Perry this year, I flew true maccready airspeeds between thermals.
The lift was strong, 5-7 kt thermals to 6-7,000 feet. So I flew about
70-80 between thermals. My average speed was then in the upper 40's.
Then for the later part of the contest, in the same weather
conditions, I delibertly did everything I could to reduce my %
circling. I flew 50-70 kts between thermals and drastically reduced
my circling. I had one flight with 15% circling and my resulting
average speed was 52 mph. So by slowing down I had a faster speed.

I understand that thermal density plays a big part in deciding what
speed to cruise at. If the thermal density is lower, it may be best
to stretch out the inter thermal glides to stay in the altitude power
band. I fly faster when the lift sources are obvious, gaggle, clouds,
birds, smoke. But when the lift is invisible, I slow it down to
stretch out my glide. I also slow waaay down if there's no good
landable fields insight....for safety.

So is there a balance between % circling and true maccready speed to
fly? What are some things that can flex your true maccready speeds up
or down?

[T8]

On Jul 4, 8:29*am, Scott Alexander >
wrote:
> I have been told over the years two consistent things: when trying to
> achieve the fastest possible cross country speeds, *1. Try to minimize
> circling as much as possible and 2. Fly the true maccready airspeeds
> between thermals.
>
> But these two things contradict. *The first couple days I flew at
> Perry this year, I flew true maccready airspeeds between thermals.
> The lift was strong, 5-7 kt thermals to 6-7,000 feet. *So I flew about
> 70-80 between thermals. *My average speed was then in the upper 40's.
> Then for the later part of the contest, in the same weather
> conditions, I delibertly did everything I could to reduce my %
> circling. *I flew 50-70 kts between thermals and drastically reduced
> my circling. *I had one flight with 15% circling and my resulting
> average speed was 52 mph. *So by slowing down I had a faster speed.
>
> I understand that thermal density plays a big part in deciding what
> speed to cruise at. *If the thermal density is lower, it may be best
> to stretch out the inter thermal glides to stay in the altitude power
> band. *I fly faster when the lift sources are obvious, gaggle, clouds,
> birds, smoke. *But when the lift is invisible, I slow it down to
> stretch out my glide. *I also slow waaay down if there's no good
> landable fields insight....for safety.
>
> So is there a balance between % circling and true maccready speed to
> fly? *What are some things that can flex your true maccready speeds up
> or down?

Must read classic:

http://www.betsybyars.com/guy/soaring_symposia/72price.html

[Tony[_5_]]

On Jul 4, 7:56*am, T8 > wrote:
> On Jul 4, 8:29*am, Scott Alexander >
> wrote:
>
>
>
> > I have been told over the years two consistent things: when trying to
> > achieve the fastest possible cross country speeds, *1. Try to minimize
> > circling as much as possible and 2. Fly the true maccready airspeeds
> > between thermals.
>
> > But these two things contradict. *The first couple days I flew at
> > Perry this year, I flew true maccready airspeeds between thermals.
> > The lift was strong, 5-7 kt thermals to 6-7,000 feet. *So I flew about
> > 70-80 between thermals. *My average speed was then in the upper 40's.
> > Then for the later part of the contest, in the same weather
> > conditions, I delibertly did everything I could to reduce my %
> > circling. *I flew 50-70 kts between thermals and drastically reduced
> > my circling. *I had one flight with 15% circling and my resulting
> > average speed was 52 mph. *So by slowing down I had a faster speed.
>
> > I understand that thermal density plays a big part in deciding what
> > speed to cruise at. *If the thermal density is lower, it may be best
> > to stretch out the inter thermal glides to stay in the altitude power
> > band. *I fly faster when the lift sources are obvious, gaggle, clouds,
> > birds, smoke. *But when the lift is invisible, I slow it down to
> > stretch out my glide. *I also slow waaay down if there's no good
> > landable fields insight....for safety.
>
> > So is there a balance between % circling and true maccready speed to
> > fly? *What are some things that can flex your true maccready speeds up
> > or down?
>
> Must read classic:
>
> http://www.betsybyars.com/guy/soaring_symposia/72price.html

I recall a talk a few years ago at the convention by Carl Herold
titled something like "If you fly MacCready you will lose". Pretty
rough title that drew a lot of people to the talk. The main idea was
that you have to base your speed to fly on your actual climb rate, not
the vario reading or the average of the vario reading but what is
actually happening to the glider. Also, by staying higher for longer
you take advantage of true airspeed benefits of flying in thinner
air. At the time Carl was flying pretty impressive cross countries in
his Nimbus while circling something like 10% of the time, or less.

Typically in the Cherokee I set the ring at about half of what the
Vario is saying and figure that is a good number. I think that % of
time circling is more important than speed between thermals. No
forward movement is going to kill average speed a lot more than being
off by 5 or 10 mph for a little while. I've been working a lot more
on minimizing my circling time on recent flights.

I noticed that on your second day at Perry your average climb was
actually down in the 250 fpm range but your interthermal speeds seemed
about the same as the day before when the average climb was over 450.
I'm no expert but I'd guess that is why you spent 40% of the time
circling on day 2 vs 25% on day 1.

Funny, the guy flying probably the slowest glider on RAS trying to
talk about how to go fast.

[SoaringXCellence]

Scott,

You need to read Johnnie Cochran's article:

http://faculty.chicagobooth.edu/john.cochrane/research/Papers/

Down at the bottom of the page titled "Just a little faster Please"

A very good description of the ideas promoted by many excellent
pilots.

Mike

[Nine Bravo Ground]

On Jul 4, 3:55*pm, SoaringXCellence > wrote:
> Scott,
>
> You need to read Johnnie Cochran's article:
>
> http://faculty.chicagobooth.edu/john.cochrane/research/Papers/
>
> Down at the bottom of the page titled *"Just a little faster Please"
>
> A very good description of the ideas promoted by many excellent
> pilots.
>
> Mike


Generally you should not try to minimize your % circling directly by
reducing your cruise speed - it will slow you down overall. Unless of
course you are flying too fast for the conditions in the first place -
and that is the rub.

Overall, you need to set your cruise based on McCready for your
expected ACTUAL climb rate, including centering time and other
considerations such as changes in climb rate at the top or bottom of
the climb (Cochrane explains this pretty well in his papers) -
different pilots estimate this in different ways, some computers give
bottom-to-top averages as well. The net effect is slower climb rates
than you might otherwise estimate based on staring at you 30-second
averager.

Next, you need to adjust your speed to optimize the tradeoff between
theoretical cross-country speed (as estimated above) and the odds that
you might have to take a weaker than expected thermal because you got
low before you found a good one. You might do this for the whole
flight (imagine a blue day with a lot of distance between good
thermals and a not very tall lift band). This is basically trading
off optimal cruise speed for a higher probability of getting a good
climb. You will see experienced pilots often "topping up" before
heading out into suspected soft areas or pressing low and passing up
weaker lift because they know they are likely to hit stronger lift in
a few miles. It is a big exercise in estimating odds.

The last thing to remember is that the biggest contributor to speed is
to find lift lines you can follow. This can be cloud or blue streets,
convergence lines, storm shelves, wave, ridge - all allow you to make
time without going backwards. You don't make this happen by slowing
down, you make it happen by picking your path well. If thermals are
hard to center and/or if you can make sustained climbs straight ahead,
you may elect to slow down to climb straight in lift, but only under
circumstances that are supported by the "adjusted" theory described
above (accounting for circling and centering losses, probability of
"false positives", etc.) In a modern ship on a day with clouds and
some modest streeting you can cruise at 85 knots and have achieved
cruise L/Ds in the 50-60 range and % climbing in the low teens.

Lastly, keep in mind that your achieved cross country speed has very
little to do with the cruise speed you pick - within certain bounds.
If you fly 15 knots slower than McCready optimal for the entire flight
it costs you 2-3% on cross-country speed. Taking a single 3 knot climb
instead of a 6 knot climb for 2500' costs you about the same amount.

9B

[Papa3]

On Jul 4, 10:06*pm, Nine Bravo Ground > wrote:

>
> Overall, you need to set your cruise based on McCready for your
> expected ACTUAL climb rate, including centering time and other
> considerations such as changes in climb rate at the top or bottom of
> the climb (Cochrane explains this pretty well in his papers) -
> different pilots estimate this in different ways, some computers give
> bottom-to-top averages as well. *The net effect is slower climb rates
> than you might otherwise estimate based on staring at you 30-second
> averager.

>
> 9B

Just a quick note on this point. I've been informally checking with
pilots for several years after flights on our local DIY contest here
to calibrate the actual conditions against my weather forecasts.
Often times, I'll hear that it was a "great day - I was hitting
5-6kts". Post flight analysis of several traces reveal that achieved
climbs were more like 3-4kts at best. It's very clear that we don't
do a great job of accounting for our centering losses and hanging in
for too long once the lift tails off. By the way, in the good old
days before flight recorders, it seems that lift was a lot stronger.
Maybe it's weight of the FRs that's slowing things down :-)

[Andy[_10_]]

On Jul 5, 5:33*am, Papa3 > wrote:
> On Jul 4, 10:06*pm, Nine Bravo Ground > wrote:
>
>
>
> > Overall, you need to set your cruise based on McCready for your
> > expected ACTUAL climb rate, including centering time and other
> > considerations such as changes in climb rate at the top or bottom of
> > the climb (Cochrane explains this pretty well in his papers) -
> > different pilots estimate this in different ways, some computers give
> > bottom-to-top averages as well. *The net effect is slower climb rates
> > than you might otherwise estimate based on staring at you 30-second
> > averager.
>
> > 9B
>
> Just a quick note on this point. *I've been informally checking with
> pilots for several years after flights on our local DIY contest here
> to calibrate the actual conditions against my weather forecasts.
> Often times, I'll hear that it was a "great day - I was hitting
> 5-6kts". * Post flight analysis of several traces reveal that achieved
> climbs were more like 3-4kts at best. *It's very clear that we don't
> do a great job of accounting for our centering losses and hanging in
> for too long once the lift tails off. By the way, * in the good old
> days before flight recorders, it seems that lift was a lot stronger.
> Maybe it's weight of the FRs that's slowing things down :-)

I think I know where most of the weight changes in my ship have come
from...

This has been discussed before, but bears mentioning in this context:
It is not really worth spending much attention on the speed director
of your vario/computer. It takes a fair amount of attention that is
far better spent on race strategy and tactics. This applies both to
setting the McCready for general speed to fly and chasing the speed
director for localized lift/sink.

I tend to fly three speeds - 80-85 knots for "normal" conditions,
90-95 knots if it is super strong and consistent with clouds and
streeting and 70-75 knots if it is weak, I need a long glide or I am
low (this is all dry - add 5-10 knots for water). This is pretty
consistent for most racing pilots I know. Best L/D for my ship is 60
knots so there is no point in ever flying less than 70 (McCready 1).
There just isn't that much difference in glide angle (L/D of 45
instead of 47 on the factory polar) so slowing down by the additional
15% is just giving speed away. Similarly, the knee in the polar is
somewhere around 85 knots, so it has to be really strong to motivate
me give away altitude at a higher rate.

9B

[jcarlyle]

Andy, you're a Western USA pilot with an ASW-28, aren't you? The OP is
an Eastern USA pilot - he might be better off using 10 kts less for
each condition than you do (adjusting downwards for his ship, too, if
needed). Also, I fly best L/D in the East in my LS8 when things get
desperate, as quite often we're in 0 sink conditions. Flying at
MacCready 1 at 70 kts like you would be a great way to outland, here.

-John

On Jul 5, 12:31 pm, Andy > wrote:
> I tend to fly three speeds - 80-85 knots for "normal" conditions,
> 90-95 knots if it is super strong and consistent with clouds and
> streeting and 70-75 knots if it is weak, I need a long glide or I am
> low (this is all dry - add 5-10 knots for water). This is pretty
> consistent for most racing pilots I know. Best L/D for my ship is 60
> knots so there is no point in ever flying less than 70 (McCready 1).
> There just isn't that much difference in glide angle (L/D of 45
> instead of 47 on the factory polar) so slowing down by the additional
> 15% is just giving speed away. Similarly, the knee in the polar is
> somewhere around 85 knots, so it has to be really strong to motivate
> me give away altitude at a higher rate.

[Andy[_10_]]

On Jul 5, 1:09*pm, jcarlyle > wrote:
> Andy, you're a Western USA pilot with an ASW-28, aren't you? The OP is
> an Eastern USA pilot - he might be better off using 10 kts less for
> each condition than you do (adjusting downwards for his ship, too, if
> needed). Also, I fly best L/D in the East in my LS8 when things get
> desperate, as quite often we're in 0 sink conditions. Flying at
> MacCready 1 at 70 kts like you would be a great way to outland, here.
>
> -John
>
> On Jul 5, 12:31 pm, Andy > wrote:
>
>
>
> > I tend to fly three speeds - 80-85 knots for "normal" conditions,
> > 90-95 knots if it is super strong and consistent with clouds and
> > streeting and 70-75 knots if it is weak, I need a long glide or I am
> > low (this is all dry - add 5-10 knots for water). This is pretty
> > consistent for most racing pilots I know. Best L/D for my ship is 60
> > knots so there is no point in ever flying less than 70 (McCready 1).
> > There just isn't that much difference in glide angle (L/D of 45
> > instead of 47 on the factory polar) so slowing down by the additional
> > 15% is just giving speed away. Similarly, the knee in the polar is
> > somewhere around 85 knots, so it has to be really strong to motivate
> > me give away altitude at a higher rate.

Fair point.

I fly an ASW-27B in the west today, but much of my early career
(1974-85) was flying in the mid-atlantic and northeast.

I agree that if you are flying in 1-2 knot lift in the east you will
cruise slower than for 4-5 knots in the west - for a whole bunch of
reasons. That said, I don't think you gain much flying best L/D (in
my ship at least) versus Mc=1.0. Two points on L/D just isn't worth
the speed loss. And if I'm not mistaken, part of the question was
about flying even slower than best L/D (50 kts was mentioned). I've
come to believe that how you handle "survival mode" is key to doing
well on sketchy days. Part of that is not giving up - keep making
forward progress while you search for the best available lift.

9B

[jcarlyle]

I agree with you that 50 kts is too slow (the OP said he went that
slow sometimes between thermals, and slowed "waaay down" if there were
no landable fields). I also agree that one shouldn't give up, and keep
on making forward progress while searching for lift (I learned that
lesson the hard way, on several occasions).

I fly with a friend who has an ASW-27B, and he does tend to go faster
than I when we're in trouble. He explains it as "the ship just doesn't
like to go slow". It may be that I'm being too conservative - and
since you (and he) have a whole lot more experience than I, perhaps
this is another lesson I should take to heart...

-John


On Jul 5, 8:12 pm, Andy > wrote:
> Fair point.
>
> I fly an ASW-27B in the west today, but much of my early career
> (1974-85) was flying in the mid-atlantic and northeast.
>
> I agree that if you are flying in 1-2 knot lift in the east you will
> cruise slower than for 4-5 knots in the west - for a whole bunch of
> reasons. That said, I don't think you gain much flying best L/D (in
> my ship at least) versus Mc=1.0. Two points on L/D just isn't worth
> the speed loss. And if I'm not mistaken, part of the question was
> about flying even slower than best L/D (50 kts was mentioned). I've
> come to believe that how you handle "survival mode" is key to doing
> well on sketchy days. Part of that is not giving up - keep making
> forward progress while you search for the best available lift.

[Andy[_1_]]

On Jul 5, 1:09*pm, jcarlyle > wrote:
> Andy, you're a Western USA pilot with an ASW-28, aren't you?

No, that's the other Andy (me). We usually sign with our contest
numbers but if we don't you can check the email address.

Another factor related to this discussion is how long to hold onto
water ballast when caught in weak conditions. I tend to dump early to
get up and going again while others tend to hold onto ballast at all
costs in case they need it later.

Andy (GY)

[Andy[_10_]]

On Jul 11, 6:19*am, Andy > wrote:
> On Jul 5, 1:09*pm, jcarlyle > wrote:
>
> > Andy, you're a Western USA pilot with an ASW-28, aren't you?
>
> No, that's the other Andy (me). *We usually sign with our contest
> numbers but if we don't you can check the email address.
>
> Another factor related to this discussion is how long to hold onto
> water ballast when caught in weak conditions. *I tend to dump early to
> get up and going again while others tend to hold onto ballast at all
> costs in case they need it later.
>
> Andy (GY)

Other Andy (9B) here:

I've done the math on the water ballast question. The interesting
case is a save scenario (which is what I think you are driving at),
otherwise it's simply a wing loading optimization question where less
than 2-3 knots achieved climb wet argues for going dry.

In the save scenario the factors that matter are:

1) How strong the save climb is (slower argues for dumping).
2) How high you have to climb in the save to be able to reach the next
"typical" thermal (higher argues for dumping).
3) What the expectation is for climb rate post-save to finish (weaker
argues for dumping).
4) How far it is to the finish (closer argues for dumping).

The decision to hold ballast versus dumping basically involves trading
off the time lost in the save climb versus the time gained with a
faster cross-country speed post-save. If conditions are expected to
be relatively strong later and you are relatively early in a long
flight it is more likely to pay to hold the ballast. If you are
barely gaining altitude in a weak thermal with ballast on board and
you are closer to the finish you are much better off dumping.

In most of the scenarios I've run you have to have 50-75 miles left to
go to make it worth holding your water (keep in mind that on a MAT or
TAT if you have gotten in a hole you may want to extend your flight so
it might not be hard to hit that threshold).

The analysis also assumes no difference in thermal strength for a
ballasted versus dry turn radius (i.e. considering only the inherent
glider turning sink rate difference wet versus dry). If the thermals
are harder to core wet then it becomes nearly impossible to find a
case for holding your water through the save - the lost time getting
back up is just too much to overcome.

9B