flight computer ergonomics and function

[danlj]

I'm curious.

After two successive near-midairs about 5 years ago, I installed PowerFlarm and a transponder that wouldn't deplete my battery during flight.

This required, due lack of panel space, that I give up my beloved ILEC S-10 flight computer (and its remote control). An Oudie IGC and and S-80 were recommended to replace it. The change has led to some conclusions about what flight computers should do in the cockpit.

1: Speed to fly
Flight computers all seem to calculate wind, and to be able to use the glider's polar. Not all use wind to revise STF.
It seems to me that we could use two rather different STF indications:
A: expected arrival altitude at next waypoint given polar, altitude, wind, and and optionally achieved mean L/D
B: best immediate speed for optimal L/D given polar, wind, and lift/sink over the last X seconds (user selectable)

(does any current flight computer offer these sorts of enhancements?)

2: Waypoint modification
Requiring the pilot to make an alphabetic search during flight to make a waypoint change is beyond stupid (it's a dangerous distraction).
- There are too many options for naming waypoints - recalling which name it was given is sometimes impossible.
- seldom are ICAO designations used, which aren't in any case easier to recall.
It seems to me that the best UI would have the ability to quickly limit the waypoint choices and swiftly move through the limited choices by spinning a knob or touch-pressing a knob-equivalent.
One is faced with either
- the need to modify one or more waypoints of a task
- the need to create a new task, the simplest of which is direct-to
Logical ways to create a limited set of waypoint choices include:
- distance (I think everybody offers this)
- direction (bearing, heading, or azimuth)
- distance & direction
- waypoints near current leg
- waypoints near current task

Do any current flight computers offer waypoint selection direction as well as distance?

Danl J

[SoaringXCellence]

On Sunday, June 16, 2019 at 12:05:01 PM UTC-7, danlj wrote:
> I'm curious.
>
> After two successive near-midairs about 5 years ago, I installed PowerFlarm and a transponder that wouldn't deplete my battery during flight.
>
> This required, due lack of panel space, that I give up my beloved ILEC S-10 flight computer (and its remote control). An Oudie IGC and and S-80 were recommended to replace it. The change has led to some conclusions about what flight computers should do in the cockpit.
>
> 1: Speed to fly
> Flight computers all seem to calculate wind, and to be able to use the glider's polar. Not all use wind to revise STF.
> It seems to me that we could use two rather different STF indications:
> A: expected arrival altitude at next waypoint given polar, altitude, wind, and and optionally achieved mean L/D
> B: best immediate speed for optimal L/D given polar, wind, and lift/sink over the last X seconds (user selectable)
>
> (does any current flight computer offer these sorts of enhancements?)
>
> 2: Waypoint modification
> Requiring the pilot to make an alphabetic search during flight to make a waypoint change is beyond stupid (it's a dangerous distraction).
> - There are too many options for naming waypoints - recalling which name it was given is sometimes impossible.
> - seldom are ICAO designations used, which aren't in any case easier to recall.
> It seems to me that the best UI would have the ability to quickly limit the waypoint choices and swiftly move through the limited choices by spinning a knob or touch-pressing a knob-equivalent.
> One is faced with either
> - the need to modify one or more waypoints of a task
> - the need to create a new task, the simplest of which is direct-to
> Logical ways to create a limited set of waypoint choices include:
> - distance (I think everybody offers this)
> - direction (bearing, heading, or azimuth)
> - distance & direction
> - waypoints near current leg
> - waypoints near current task
>
> Do any current flight computers offer waypoint selection direction as well as distance?
>
> Danl J

XCSoar on an OpenVario, But you have to build and maintain it yourself.

[[email protected]]

Tophat, which is based on XCsoar but with a different user interface, is all about ergonomics. (Thank you Robert Dunning!) It shows you the arrival altitude at nearby airports and landable points, right on the moving map. And you can tap on any waypoint and then tap "go to" and it becomes your new destination that it guides you to. That's handy if you get low and need to divert towards a safe landing point. If you later want to resume the original task, tap the nav bar and then tap "resume task".

And if you want to edit the task and choose an arbitrary waypoint, you can have it order the waypoints by nearest distance to your current location, or by highest arrival altitude, or alphabetically.

And talking about ergonomics, you can choose the hardware device among many possible - just about anything that runs Android can run Tophat (or XCsoar). I use an old Nook e-reader because the screen is large (6 inches) and the contrast of the e-ink type screen is not affected by direct sunshine. I've arranged for it to communicate with the FLARM but that took some hardware hacking.

BTW the STF does not vary with the wind, unless it's a survival glide to a landing.

[Tango Whisky]

SeeYou Mobile.

[Per Carlin]

Your exsisting S80 and Oudie does all you require, have you read the manuals?

[Martin Gregorie[_6_]]

On Sun, 16 Jun 2019 12:04:58 -0700, danlj wrote:

> Do any current flight computers offer waypoint selection direction as
> well as distance?
>
Add LK8000 to the list. It shows the next turnpoint in the declared task
in the top left cornet. Tapping it cycles through the declared turnpoints
in task sequence, followed by HOME and the two closest landouts to your
current position.

Each item in the list automatically becomes the next selected TP when it
is displayed, so just stop tapping when your next destination is shown.


--
Martin | martin at
Gregorie | gregorie dot org

[JS[_5_]]

Yes.
Just a matter of learning the new stuff. The old isn't particularly great, just familiar.
Jim

[danlj]

On Sunday, June 16, 2019 at 2:05:01 PM UTC-5, danlj wrote:
> I'm curious.
>
> After two successive near-midairs about 5 years ago, I installed PowerFlarm and a transponder that wouldn't deplete my battery during flight.
>
> This required, due lack of panel space, that I give up my beloved ILEC S-10 flight computer (and its remote control). An Oudie IGC and and S-80 were recommended to replace it. The change has led to some conclusions about what flight computers should do in the cockpit.
>
> 1: Speed to fly
> Flight computers all seem to calculate wind, and to be able to use the glider's polar. Not all use wind to revise STF.
> It seems to me that we could use two rather different STF indications:
> A: expected arrival altitude at next waypoint given polar, altitude, wind, and and optionally achieved mean L/D
> B: best immediate speed for optimal L/D given polar, wind, and lift/sink over the last X seconds (user selectable)
>
> (does any current flight computer offer these sorts of enhancements?)
>
> 2: Waypoint modification
> Requiring the pilot to make an alphabetic search during flight to make a waypoint change is beyond stupid (it's a dangerous distraction).
> - There are too many options for naming waypoints - recalling which name it was given is sometimes impossible.
> - seldom are ICAO designations used, which aren't in any case easier to recall.
> It seems to me that the best UI would have the ability to quickly limit the waypoint choices and swiftly move through the limited choices by spinning a knob or touch-pressing a knob-equivalent.
> One is faced with either
> - the need to modify one or more waypoints of a task
> - the need to create a new task, the simplest of which is direct-to
> Logical ways to create a limited set of waypoint choices include:
> - distance (I think everybody offers this)
> - direction (bearing, heading, or azimuth)
> - distance & direction
> - waypoints near current leg
> - waypoints near current task
>
> Do any current flight computers offer waypoint selection direction as well as distance?
>
> Danl J

Comments below force a response.
1: the most efficient calibrated airspeed for progress over the ground DOES vary with the wind. A published rule of thumb is with a headwind, add 1/3 the headwind component to best glide; with a tailwind, fly at minimum sink.
Jean Marie Clement has created a beautiful graph showing that the most efficient CAS against headwind is a curve -- it looks like it's an hyperbola. It's reproduced on page 51 of Brigliadoris' Competing in Gliders.

2: The Oudie IGC manual is quite clear that ONLY polar and pre-set MacCready are used in calculating speed to fly and arrival altitude. Wind is specifically not used in their calculation, and you can of course set your MacCready to the mean climb in the last thermal if you like.

The S8x manual does not indicate whether current lift/sink is used to modify the instant STF arrows indication, nor whether wind is taken into account.. The STF *is* based on the next waypoint, polar, and (I think) MacCready. I haven't used this, and will reconfigure my S80 now that I've updated the firmware and watch its behavior.

3: Does anyone on this forum actually ever do the research before replying? Just curious. The B/S ratio applies to more than thermals...

[waremark]

While Oudie does not take account of wind in determining STF it most definitely does in determining arrival height.

Wind affects STF to reach a fixed point on the ground as high as possible after your last climb. However it does not change STF for best progress on task - fly Mcready and accept what the wind does to your ground speed.

[waremark]

"The S8x manual does not indicate whether current lift/sink is used to modify the instant STF arrows indication, nor whether wind is taken into account. The STF *is* based on the next waypoint, polar, and (I think) MacCready."

Lift/sink does drive the speed up slow down arrows. Wind does not, nor does the next waypoint. As well as polar and Mcready it also considers wing loading. When changing Mcready ballast or bug settings LX devices tell you the zero lift figures for Mcrwady STF and glide ratio.

By the way, if you want to look for a waypoint in a particular direction on your Oudie drag your finger in that direction on the screen - sort on any column heading by touching it, touch the wp you want and goto. But you surely know all this.

[Bruce Hoult]

On Monday, June 17, 2019 at 4:20:46 PM UTC-7, danlj wrote:
> Comments below force a response.
> 1: the most efficient calibrated airspeed for progress over the ground DOES vary with the wind. A published rule of thumb is with a headwind, add 1/3 the headwind component to best glide; with a tailwind, fly at minimum sink.

Not to maximise speed in continuous cross-country soaring from one thermal to another, no. That depends only on the expected lift in the next thermal and the polar.

For a final glide to a fixed point on the ground -- airport, turnpoint (where you are going to significantly change direction), mountain peak or ridge -- then yes you should change your speed roughly as you indicate.

> 3: Does anyone on this forum actually ever do the research before replying? Just curious. The B/S ratio applies to more than thermals...

Some of us do, yes.

[Mike the Strike]

MacCready theory optimises cross-country speed over a closed course with the assumption that wind speed and direction are constant throughout the course and independent of altitude. These assumptions are approximately true most of the time, so departing from MacCready speed has little benefit. However, if there is a leg where this is not true, the optimum airspeed may be different. A downwind dash would be a special case of this.

I did some numerical simulations some years ago for the case where one is flying to a turn point against a strong wind and found a faster speed into wind followed by a slower speed after the turn was marginally better. Clearly, if your computer tells you to fly at 60 knots against a 60 knot headwind, you are not going anywhere! Another issue is that you have to maintain sufficient altitude to connect with the next thermal. This may require trading off speed for altitude and experienced cross-country pilots have their own rules for dealing with these circumstances. Flight computers and associated software are imperfect at dealing with these more complex issues which are up to the pilot to solve.

Mike

[Kevin Neave[_2_]]

For thermal XC if your computer is telling you to fly at 60kts then you're
probably flying dry and have MacCready set pretty low. So no you won't get
anywhere.
You need to climb faster and set MacCready higher

The wind only affects the rate of climb you need to make progress, speed to
fly between climbs is determined by "normal" MacCready

Flying Ridges or Wave, or final gliding to a fixed point, is a different
ball game

KN



At 07:44 18 June 2019, Mike the Strike wrote:
>Clear=
>ly, if your computer tells you to fly at 60 knots against a 60 knot
>headwin=
>d, you are not going anywhere!
>=

[krasw]

Simple table that tells you how much to increase MC at certain headwind component would be enough to solve this issue.

[[email protected]]

On Monday, June 17, 2019 at 4:20:46 PM UTC-7, danlj wrote:
> On Sunday, June 16, 2019 at 2:05:01 PM UTC-5, danlj wrote:
> > I'm curious.
> >
> > After two successive near-midairs about 5 years ago, I installed PowerFlarm and a transponder that wouldn't deplete my battery during flight.
> >
> > This required, due lack of panel space, that I give up my beloved ILEC S-10 flight computer (and its remote control). An Oudie IGC and and S-80 were recommended to replace it. The change has led to some conclusions about what flight computers should do in the cockpit.
> >
> > 1: Speed to fly
> > Flight computers all seem to calculate wind, and to be able to use the glider's polar. Not all use wind to revise STF.
> > It seems to me that we could use two rather different STF indications:
> > A: expected arrival altitude at next waypoint given polar, altitude, wind, and and optionally achieved mean L/D
> > B: best immediate speed for optimal L/D given polar, wind, and lift/sink over the last X seconds (user selectable)
> >
> > (does any current flight computer offer these sorts of enhancements?)
> >
> > 2: Waypoint modification
> > Requiring the pilot to make an alphabetic search during flight to make a waypoint change is beyond stupid (it's a dangerous distraction).
> > - There are too many options for naming waypoints - recalling which name it was given is sometimes impossible.
> > - seldom are ICAO designations used, which aren't in any case easier to recall.
> > It seems to me that the best UI would have the ability to quickly limit the waypoint choices and swiftly move through the limited choices by spinning a knob or touch-pressing a knob-equivalent.
> > One is faced with either
> > - the need to modify one or more waypoints of a task
> > - the need to create a new task, the simplest of which is direct-to
> > Logical ways to create a limited set of waypoint choices include:
> > - distance (I think everybody offers this)
> > - direction (bearing, heading, or azimuth)
> > - distance & direction
> > - waypoints near current leg
> > - waypoints near current task
> >
> > Do any current flight computers offer waypoint selection direction as well as distance?
> >
> > Danl J
>
> Comments below force a response.
> 1: the most efficient calibrated airspeed for progress over the ground DOES vary with the wind. A published rule of thumb is with a headwind, add 1/3 the headwind component to best glide; with a tailwind, fly at minimum sink.
> Jean Marie Clement has created a beautiful graph showing that the most efficient CAS against headwind is a curve -- it looks like it's an hyperbola. It's reproduced on page 51 of Brigliadoris' Competing in Gliders.
>
> 2: The Oudie IGC manual is quite clear that ONLY polar and pre-set MacCready are used in calculating speed to fly and arrival altitude. Wind is specifically not used in their calculation, and you can of course set your MacCready to the mean climb in the last thermal if you like.
>
> The S8x manual does not indicate whether current lift/sink is used to modify the instant STF arrows indication, nor whether wind is taken into account. The STF *is* based on the next waypoint, polar, and (I think) MacCready.. I haven't used this, and will reconfigure my S80 now that I've updated the firmware and watch its behavior.
>
> 3: Does anyone on this forum actually ever do the research before replying? Just curious. The B/S ratio applies to more than thermals...

From the Oudie IGC manual: . Arrival altitude (Arrival) = Expected Arrival altitude at the selected waypoint calculated by taking the
distance, MC, wind, glider polar, bugs and ballast into account

"As a published rule of thumb is with a headwind, add 1/3 the headwind component to best glide; with a tailwind, fly at minimum sink.Can you clarify what you mean by this, when you are flying min sink downwind, everyone else will be passing you, fast!

The S80 can search waypoints by nearest now, it's a feature that was added a year or two ago due to request.
The Oudie can sort points based on distance, and it will keep this sorting method until you change it. It can also filter the entire database in a variety of ways, very useful for general flying.

Good luck.

[Mike the Strike]

On Tuesday, June 18, 2019 at 7:43:16 AM UTC-7, krasw wrote:
> Simple table that tells you how much to increase MC at certain headwind component would be enough to solve this issue.

Unfortunately not.

The problem of optimum MacCready speed in wind relates to both wind shear and the relative horizontal velocity of the thermals. With a uniform atmosphere in which the wind speed is constant with height and in which thermals move at the same speed as the wind, standard MacCready theory applies. In these ideal conditions, there is no advantage in changing speed upwind or downwind.

In the case where thermals move more slowly than the wind (the more general case) or where they don't move at all (wave), the optimal speed will generally be faster upwind and slower downwind. This is what many experienced cross-country pilots have intuitively figured out in the real world.

Mathematical analyses of these have been done by several folks, including John Cochrane and Branko Stojkovic. Long enough ago to be forgotten.

The solution of speed to fly with wind is not easy, since the effects of wind shear and thermal speed are variable and may not be known well enough for a satisfactory general mathematical solution.

Mike

[[email protected]]

On Tuesday, June 18, 2019 at 5:34:40 PM UTC-4, Mike the Strike wrote:
> On Tuesday, June 18, 2019 at 7:43:16 AM UTC-7, krasw wrote:
> > Simple table that tells you how much to increase MC at certain headwind component would be enough to solve this issue.
>
> Unfortunately not.
>
> The problem of optimum MacCready speed in wind relates to both wind shear and the relative horizontal velocity of the thermals. With a uniform atmosphere in which the wind speed is constant with height and in which thermals move at the same speed as the wind, standard MacCready theory applies. In these ideal conditions, there is no advantage in changing speed upwind or downwind.
>
> In the case where thermals move more slowly than the wind (the more general case) or where they don't move at all (wave), the optimal speed will generally be faster upwind and slower downwind. This is what many experienced cross-country pilots have intuitively figured out in the real world.
>
> Mathematical analyses of these have been done by several folks, including John Cochrane and Branko Stojkovic. Long enough ago to be forgotten.
>
> The solution of speed to fly with wind is not easy, since the effects of wind shear and thermal speed are variable and may not be known well enough for a satisfactory general mathematical solution.
>
> Mike

Thank you Mike. But a saving grace is that the achieved XC speed for some range around the optimum STF does not vary very much, exactly because it is an optimum: it is the top of a roundish hump in the curve. In that sense a "simple table", or even a rough guess, of the correction, if any, for the wind, is "enough". Lacking precise data on the local shear etc, fancier methods wouldn't do any better.

And, to paraphrase UH, the left-and-right controls are far more important (for task speed) than the up-and-down. Deviate to where the air is half a knot sweeter and you'll get there faster, even if your STF is 10 knots off.

[Mike the Strike]

Indeed, the wind question is mostly academic as we rarely fly long cross-country tasks in wind speeds that make a big difference. Identifying convergence (and divergence) lines and dealing with long distances between thermals are more useful problems to solve. MacCready theory optimises your speed between thermals but doesn’t tell you what to do when you get low or can't find a thermal (John Cochrane wrote a paper on this too). Perhaps this is something that smarter flight computers could address?

Mike

[krasw]

On Wednesday, 19 June 2019 00:34:40 UTC+3, Mike the Strike wrote:
> On Tuesday, June 18, 2019 at 7:43:16 AM UTC-7, krasw wrote:
> > Simple table that tells you how much to increase MC at certain headwind component would be enough to solve this issue.
>
> Unfortunately not.
>
> The problem of optimum MacCready speed in wind relates to both wind shear and the relative horizontal velocity of the thermals. With a uniform atmosphere in which the wind speed is constant with height and in which thermals move at the same speed as the wind, standard MacCready theory applies. In these ideal conditions, there is no advantage in changing speed upwind or downwind.
>
> In the case where thermals move more slowly than the wind (the more general case) or where they don't move at all (wave), the optimal speed will generally be faster upwind and slower downwind. This is what many experienced cross-country pilots have intuitively figured out in the real world.
>
> Mathematical analyses of these have been done by several folks, including John Cochrane and Branko Stojkovic. Long enough ago to be forgotten.
>
> The solution of speed to fly with wind is not easy, since the effects of wind shear and thermal speed are variable and may not be known well enough for a satisfactory general mathematical solution.
>
> Mike

I'm aware of all this. You either fly towards fixed spot on the ground (final glide), or towards cloud or thermal. We can approximate that your home airfield does not move with the wind at all and thermal moves at wind speed (which is does not, of course, but this is approximation). Final glide to fixed spot you would increase mc to take wind into account, glide to next thermal you would (mostly) not as it will be blown towards you at wind speed. This simple idea turned into mc value to increase or subtract would get you 99% of the theoretically optimum glide speed. We do not have to worry about exact figure as we can maintain certain airspeed at maybe 5% accuracy. Accuracy above that has only academic interest.

[Tango Eight]

On Wednesday, June 19, 2019 at 8:09:12 AM UTC-4, krasw wrote:
> On Wednesday, 19 June 2019 00:34:40 UTC+3, Mike the Strike wrote:
> > On Tuesday, June 18, 2019 at 7:43:16 AM UTC-7, krasw wrote:
> > > Simple table that tells you how much to increase MC at certain headwind component would be enough to solve this issue.
> >
> > Unfortunately not.
> >
> > The problem of optimum MacCready speed in wind relates to both wind shear and the relative horizontal velocity of the thermals. With a uniform atmosphere in which the wind speed is constant with height and in which thermals move at the same speed as the wind, standard MacCready theory applies. In these ideal conditions, there is no advantage in changing speed upwind or downwind.
> >
> > In the case where thermals move more slowly than the wind (the more general case) or where they don't move at all (wave), the optimal speed will generally be faster upwind and slower downwind. This is what many experienced cross-country pilots have intuitively figured out in the real world.
> >
> > Mathematical analyses of these have been done by several folks, including John Cochrane and Branko Stojkovic. Long enough ago to be forgotten.
> >
> > The solution of speed to fly with wind is not easy, since the effects of wind shear and thermal speed are variable and may not be known well enough for a satisfactory general mathematical solution.
> >
> > Mike
>
> I'm aware of all this. You either fly towards fixed spot on the ground (final glide), or towards cloud or thermal. We can approximate that your home airfield does not move with the wind at all and thermal moves at wind speed (which is does not, of course, but this is approximation). Final glide to fixed spot you would increase mc to take wind into account, glide to next thermal you would (mostly) not as it will be blown towards you at wind speed. This simple idea turned into mc value to increase or subtract would get you 99% of the theoretically optimum glide speed. We do not have to worry about exact figure as we can maintain certain airspeed at maybe 5% accuracy. Accuracy above that has only academic interest.

(Telling the experienced guys what they already know...) If you have enough wind that there is the motivation to use something other than a MC 2 - 3 (kts) final glide value, you can't (sensibly, at least) plan a skinny final glide under any circumstances. If you have that much wind in the working band, you will have wind shear that has to be considered (but can only be guessed at) and you most likely will have to at least consider the possibility of wave lift & sink. The academic interest is simply a mathematical puzzle, nothing more, no utility.

I like the fact that there's still a role for a smart, experienced PIC :-).

best,
T8

[RR]

Flying in wave conditions with 50kts of wind, I got "low" and the computer indacated a marginal glid home. I dug out until I had 3000 ft over glide (MC 2). Considering the was more than a slight chance of wave sink on the way home, I thought I should crank up my MC as one might do on a thermal flight to insure you are looking at info with some increased margin. Rather than lower my arrival, it increased it to 5000 ft. A fence post case to be sure, but the point for this thread is, you can try fiddling with your MC when flying in wind to see if you can see a difference in arrival height. I had never seen it "reverse" like this in thermal conditions (might happen between 0 and something low) but you can try.

I have always used the rule of thumb for glide into the wind, but in 50kts you can run out of fingers to count on...

[Brian[_1_]]

On Monday, June 17, 2019 at 5:53:34 PM UTC-6, waremark wrote:
> "The S8x manual does not indicate whether current lift/sink is used to modify the instant STF arrows indication, nor whether wind is taken into account. The STF *is* based on the next waypoint, polar, and (I think) MacCready."
>
> Lift/sink does drive the speed up slow down arrows. Wind does not, nor does the next waypoint. As well as polar and Mcready it also considers wing loading. When changing Mcready ballast or bug settings LX devices tell you the zero lift figures for Mcrwady STF and glide ratio.
>
> By the way, if you want to look for a waypoint in a particular direction on your Oudie drag your finger in that direction on the screen - sort on any column heading by touching it, touch the wp you want and goto. But you surely know all this.

Agreed the Push/pull arrows on the S80 do account for lift and sink. I am disappointed that the Speed tape display does not account for lift and sink, it simply displays the calm air speed to fly speed, which is of little value on an active display. The S80 tells me that when I set the MacCready Value.

Brian

[[email protected]]

On Thursday, June 20, 2019 at 7:31:37 AM UTC-4, RR wrote:
> Flying in wave conditions with 50kts of wind, I got "low" and the computer indacated a marginal glid home. I dug out until I had 3000 ft over glide (MC 2). Considering the was more than a slight chance of wave sink on the way home, I thought I should crank up my MC as one might do on a thermal flight to insure you are looking at info with some increased margin. Rather than lower my arrival, it increased it to 5000 ft. A fence post case to be sure, but the point for this thread is, you can try fiddling with your MC when flying in wind to see if you can see a difference in arrival height. I had never seen it "reverse" like this in thermal conditions (might happen between 0 and something low) but you can try.
>
> I have always used the rule of thumb for glide into the wind, but in 50kts you can run out of fingers to count on...

It is convenient to assume that a higher MC setting is "more conservative", but clearly that is not always the case. Presumably the "MC" for the purpose of a final glide just meant that your computer assumed you will glide at the still-air MacCready airspeed for that MC setting. But the computer also took the wind into account when computing your achieved glide ratio and thus arrival altitude. Thus when you set a higher MC the computer read it as a higher airspeed, which was advantageous in this case of flying into a strong headwind. Thus it displayed a higher arrival altitude.

I think in this situation one should first decide on the speed to fly (would be nice if the computer would help you with that choice, but it may not). With a 50-knot headwind obviously you'd pick a rather high speed. Then adjust the MC until the displayed STF fits your choice of speed. Then see what it says for predicted arrival altitude, and add a safety margin to that in feet of arrival altitude, not in MC setting.