search

weglide / bugtracker

WeGlide platform bug tracker.
https://www.weglide.org
11 stars 1 forks source link

Coach shows very low (wrong) average gliding speed #282

Closed JohnCJ2 closed 3 days ago

JohnCJ2 commented 1 month ago

The coach function shows consistently very low average IAS for my glide segments. The graph, however, shows an unlabeled speed value for each segment that seems to match my calculations.

To Reproduce Steps to reproduce the behavior:

  1. Use my flight on May 11, 2024 (John Johnson, Tucson Soaring Club)
  2. Look at the gliding segment 21:30:51 - 21:42:43 (~12min duration, 20nm distance)
  3. See that the graph shows 102kn (unlabeled) and simple calculations yield ~100kn (distance/time)
  4. See that the coach function says the average speed was only 69kn: You were flying the glide phase from 21:30 to 21:42 with an average indicated airspeed of 69 kn. Based on an average MacCready of 559 fpm and the net air mass movement of 7 fpm, the optimal glide speed is 95 kn according to MacCready theory. Try to fly faster between thermals.
  5. See similar discrepancies in other glide segments.

Expected behavior A reply with an explanation of my confusion or a response to correct this if it is a bug.

Screenshots Attached JJohnson glide speed error 5-11-2024

Device (please complete the following information):

morhall commented 1 month ago

Not someone from WeGlide, but saw this and read through your report. Keep in mind that the speed shown on the graph is Groundspeed. At 10-13k you're also going to see true airspeeds that are ~20% higher (2%/1000ft) than your indicated. So TAS + a tailwind could put you at 100kts over the ground with 70ish kts indicated. Not knowing what your winds were, their number seems a bit low, but not completely out of the realm of possibility.

I have felt like the numbers were generally a bit low as well, but the data probably tells a better story than my memory.

JohnCJ2 commented 1 month ago

Hi Morgan,

Thanks for the response. I do have TAS and GS side-by-side on my flight computer and keep an eye on them mainly to sanity check the calculated wind dir/speed. On the example glide, they were within a few kts with a crosswind on that segment. I really think I'm flying fast enough - maybe even too fast because it feels like I'm clawing back up into the lift band too often. Since I end up slower compared to my peers, I'm doing something wrong. On the other hand, WeGlide says ALL of my glide segments are too slow and that just doesn't make sense to me since some are pushing deep into the yellow only slowing to 90kts as I drop into lower altitude bands. I'm in the middle of a local event with some fast pilots so I'll do some comparisons. I'm also in the 18m nationals so should have plenty of opportunities to learn more there. I do wish that WeGlide (and flight computers) published the formulas used for their derived metrics. thx, JJ

On Mon, May 27, 2024 at 10:59 PM Morgan Hall @.***> wrote:

Not someone from WeGlide, but saw this and read through your report. Keep in mind that the speed shown on the graph is Groundspeed. At 10-13k you're also going to see true airspeeds that are ~20% higher (2%/1000ft) than your indicated. So TAS + a tailwind could put you at 100kts over the ground with 70ish kts indicated. Not knowing what your winds were, their number seems a bit low, but not completely out of the realm of possibility.

I have felt like the numbers were generally a bit low as well, but the data probably tells a better story than my memory.

— Reply to this email directly, view it on GitHub https://github.com/weglide/bugtracker/issues/282#issuecomment-2134403457, or unsubscribe https://github.com/notifications/unsubscribe-auth/BIODHPS7BAHXER3QBJ6RYMLZEQMMNAVCNFSM6AAAAABHTFKYQGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZUGQYDGNBVG4 . You are receiving this because you authored the thread.Message ID: @.***>

morhall commented 1 month ago

Sounds like you've got reasonable data available real-time if you've got the GS and TAS displayed. If you're using an LX 90X0, I also recommend the GS - TAS infobox which gives you your headwind/tailwind component at a faster update rate than their actual "HW/TW" box.

WeGlide is doing a pretty amazing job with the data that they have. The main thing is that they don't have "puckerfactor" or "needtonotlandout" settings. So they only know your polar data and your actual achieved climb rates. This is usually why i'm "slow" according to the coach. Sure, I stopped for 6 knots and flew MC4, but I also had 35 miles of unlandable terrain under me.

As for going faster, the simple "no shit" answer is thermal less. You'll find that the speed you fly between thermals is less of a predictor of overall speed than the %thermal time on any given flight. Sounds like this is where your own analysis is already taking you. John Cochrane has an article on speed to fly with numbers run for an ASW-27 and MC4 dry is only about 50kts/hr average speed if I remember right. Cruising at over 80, but stopping to climb often means only 65% of the time is in cruise or whatever the math works out to.

MC theory is an excellent tool, but often flying a bit slower will keep you in the strongest lift band and allow you to skip climbs entirely. Especially if you've got convergence or streeting types of energy lines where you can fly 2-3x longer between needing a circling climb. If you are finding yourself low and working 3kts to get back into the 6kts band above, that's where changing your working band can help. Assuming the day/area is predictable like that. Given that you're in AZ, I wouldn't be surprised if the 6kts+ band is too far apart to reach at MC6. So you might be faster flying MC4 so that you reach the next climb in the 6kts band. But WeGlide will tell you that you are flying too slow because you flew MC4 and only stopped for MC6+ climbs.

Maybe some day they can add thermal strength bands into the coach analysis. But that is kind of like proving a negative. If you stay high, you'll never show that the thermals weren't strong from the ground and you'll still appear slow.

Anyhow, good luck at the contest. Comparing traces through the same task and air is a great way to learn. So is asking the fast guys. Hopefully everyone is open to sharing their technique.

Morgan

On Tue, May 28, 2024 at 7:01 AM JohnCJ2 @.***> wrote:

Hi Morgan,

Thanks for the response. I do have TAS and GS side-by-side on my flight computer and keep an eye on them mainly to sanity check the calculated wind dir/speed. On the example glide, they were within a few kts with a crosswind on that segment. I really think I'm flying fast enough - maybe even too fast because it feels like I'm clawing back up into the lift band too often. Since I end up slower compared to my peers, I'm doing something wrong. On the other hand, WeGlide says ALL of my glide segments are too slow and that just doesn't make sense to me since some are pushing deep into the yellow only slowing to 90kts as I drop into lower altitude bands. I'm in the middle of a local event with some fast pilots so I'll do some comparisons. I'm also in the 18m nationals so should have plenty of opportunities to learn more there. I do wish that WeGlide (and flight computers) published the formulas used for their derived metrics. thx, JJ

On Mon, May 27, 2024 at 10:59 PM Morgan Hall @.***> wrote:

Not someone from WeGlide, but saw this and read through your report. Keep in mind that the speed shown on the graph is Groundspeed. At 10-13k you're also going to see true airspeeds that are ~20% higher (2%/1000ft) than your indicated. So TAS + a tailwind could put you at 100kts over the ground with 70ish kts indicated. Not knowing what your winds were, their number seems a bit low, but not completely out of the realm of possibility.

I have felt like the numbers were generally a bit low as well, but the data probably tells a better story than my memory.

— Reply to this email directly, view it on GitHub < https://github.com/weglide/bugtracker/issues/282#issuecomment-2134403457>,

or unsubscribe < https://github.com/notifications/unsubscribe-auth/BIODHPS7BAHXER3QBJ6RYMLZEQMMNAVCNFSM6AAAAABHTFKYQGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZUGQYDGNBVG4>

. You are receiving this because you authored the thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/weglide/bugtracker/issues/282#issuecomment-2135295531, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAFHHKQMQV3GWAVBOIP7ZALZESE5JAVCNFSM6AAAAABHTFKYQGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZVGI4TKNJTGE . You are receiving this because you commented.Message ID: @.***>

JohnCJ2 commented 1 month ago

Great feedback, Morgan. Thank you! I think you're right about weglide (and reality) in terms of using a pure MC calculation. I definitely need to unlearn my hang gliding 'turn in every thermal and top them out' past. thx, JJ

On Tue, May 28, 2024 at 8:48 AM Morgan Hall @.***> wrote:

Sounds like you've got reasonable data available real-time if you've got the GS and TAS displayed. If you're using an LX 90X0, I also recommend the GS - TAS infobox which gives you your headwind/tailwind component at a faster update rate than their actual "HW/TW" box.

WeGlide is doing a pretty amazing job with the data that they have. The main thing is that they don't have "puckerfactor" or "needtonotlandout" settings. So they only know your polar data and your actual achieved climb rates. This is usually why i'm "slow" according to the coach. Sure, I stopped for 6 knots and flew MC4, but I also had 35 miles of unlandable terrain under me.

As for going faster, the simple "no shit" answer is thermal less. You'll find that the speed you fly between thermals is less of a predictor of overall speed than the %thermal time on any given flight. Sounds like this is where your own analysis is already taking you. John Cochrane has an article on speed to fly with numbers run for an ASW-27 and MC4 dry is only about 50kts/hr average speed if I remember right. Cruising at over 80, but stopping to climb often means only 65% of the time is in cruise or whatever the math works out to.

MC theory is an excellent tool, but often flying a bit slower will keep you in the strongest lift band and allow you to skip climbs entirely. Especially if you've got convergence or streeting types of energy lines where you can fly 2-3x longer between needing a circling climb. If you are finding yourself low and working 3kts to get back into the 6kts band above, that's where changing your working band can help. Assuming the day/area is predictable like that. Given that you're in AZ, I wouldn't be surprised if the 6kts+ band is too far apart to reach at MC6. So you might be faster flying MC4 so that you reach the next climb in the 6kts band. But WeGlide will tell you that you are flying too slow because you flew MC4 and only stopped for MC6+ climbs.

Maybe some day they can add thermal strength bands into the coach analysis. But that is kind of like proving a negative. If you stay high, you'll never show that the thermals weren't strong from the ground and you'll still appear slow.

Anyhow, good luck at the contest. Comparing traces through the same task and air is a great way to learn. So is asking the fast guys. Hopefully everyone is open to sharing their technique.

Morgan

On Tue, May 28, 2024 at 7:01 AM JohnCJ2 @.***> wrote:

Hi Morgan,

Thanks for the response. I do have TAS and GS side-by-side on my flight computer and keep an eye on them mainly to sanity check the calculated wind dir/speed. On the example glide, they were within a few kts with a crosswind on that segment. I really think I'm flying fast enough - maybe even too fast because it feels like I'm clawing back up into the lift band too often. Since I end up slower compared to my peers, I'm doing something wrong. On the other hand, WeGlide says ALL of my glide segments are too slow and that just doesn't make sense to me since some are pushing deep into the yellow only slowing to 90kts as I drop into lower altitude bands. I'm in the middle of a local event with some fast pilots so I'll do some comparisons. I'm also in the 18m nationals so should have plenty of opportunities to learn more there. I do wish that WeGlide (and flight computers) published the formulas used for their derived metrics. thx, JJ

On Mon, May 27, 2024 at 10:59 PM Morgan Hall @.***> wrote:

Not someone from WeGlide, but saw this and read through your report. Keep in mind that the speed shown on the graph is Groundspeed. At 10-13k you're also going to see true airspeeds that are ~20% higher (2%/1000ft) than your indicated. So TAS + a tailwind could put you at 100kts over the ground with 70ish kts indicated. Not knowing what your winds were, their number seems a bit low, but not completely out of the realm of possibility.

I have felt like the numbers were generally a bit low as well, but the data probably tells a better story than my memory.

— Reply to this email directly, view it on GitHub < https://github.com/weglide/bugtracker/issues/282#issuecomment-2134403457>,

or unsubscribe <

https://github.com/notifications/unsubscribe-auth/BIODHPS7BAHXER3QBJ6RYMLZEQMMNAVCNFSM6AAAAABHTFKYQGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZUGQYDGNBVG4>

. You are receiving this because you authored the thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub < https://github.com/weglide/bugtracker/issues/282#issuecomment-2135295531>,

or unsubscribe < https://github.com/notifications/unsubscribe-auth/AAFHHKQMQV3GWAVBOIP7ZALZESE5JAVCNFSM6AAAAABHTFKYQGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZVGI4TKNJTGE>

. You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/weglide/bugtracker/issues/282#issuecomment-2135584422, or unsubscribe https://github.com/notifications/unsubscribe-auth/BIODHPW7SJVETOAMYXJBDCLZESRNXAVCNFSM6AAAAABHTFKYQGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZVGU4DINBSGI . You are receiving this because you authored the thread.Message ID: @.***>

morhall commented 1 month ago

Classic HG moves. I too came from HG and had to unlearn take em all, top em all.

On Tue, May 28, 2024 at 21:29 JohnCJ2 @.***> wrote:

Great feedback, Morgan. Thank you! I think you're right about weglide (and reality) in terms of using a pure MC calculation. I definitely need to unlearn my hang gliding 'turn in every thermal and top them out' past. thx, JJ

On Tue, May 28, 2024 at 8:48 AM Morgan Hall @.***> wrote:

Sounds like you've got reasonable data available real-time if you've got the GS and TAS displayed. If you're using an LX 90X0, I also recommend the GS - TAS infobox which gives you your headwind/tailwind component at a faster update rate than their actual "HW/TW" box.

WeGlide is doing a pretty amazing job with the data that they have. The main thing is that they don't have "puckerfactor" or "needtonotlandout" settings. So they only know your polar data and your actual achieved climb rates. This is usually why i'm "slow" according to the coach. Sure, I stopped for 6 knots and flew MC4, but I also had 35 miles of unlandable terrain under me.

As for going faster, the simple "no shit" answer is thermal less. You'll find that the speed you fly between thermals is less of a predictor of overall speed than the %thermal time on any given flight. Sounds like this is where your own analysis is already taking you. John Cochrane has an article on speed to fly with numbers run for an ASW-27 and MC4 dry is only about 50kts/hr average speed if I remember right. Cruising at over 80, but stopping to climb often means only 65% of the time is in cruise or whatever the math works out to.

MC theory is an excellent tool, but often flying a bit slower will keep you in the strongest lift band and allow you to skip climbs entirely. Especially if you've got convergence or streeting types of energy lines where you can fly 2-3x longer between needing a circling climb. If you are finding yourself low and working 3kts to get back into the 6kts band above, that's where changing your working band can help. Assuming the day/area is predictable like that. Given that you're in AZ, I wouldn't be surprised if the 6kts+ band is too far apart to reach at MC6. So you might be faster flying MC4 so that you reach the next climb in the 6kts band. But WeGlide will tell you that you are flying too slow because you flew MC4 and only stopped for MC6+ climbs.

Maybe some day they can add thermal strength bands into the coach analysis. But that is kind of like proving a negative. If you stay high, you'll never show that the thermals weren't strong from the ground and you'll still appear slow.

Anyhow, good luck at the contest. Comparing traces through the same task and air is a great way to learn. So is asking the fast guys. Hopefully everyone is open to sharing their technique.

Morgan

On Tue, May 28, 2024 at 7:01 AM JohnCJ2 @.***> wrote:

Hi Morgan,

Thanks for the response. I do have TAS and GS side-by-side on my flight computer and keep an eye on them mainly to sanity check the calculated wind dir/speed. On the example glide, they were within a few kts with a crosswind on that segment. I really think I'm flying fast enough - maybe even too fast because it feels like I'm clawing back up into the lift band too often. Since I end up slower compared to my peers, I'm doing something wrong. On the other hand, WeGlide says ALL of my glide segments are too slow and that just doesn't make sense to me since some are pushing deep into the yellow only slowing to 90kts as I drop into lower altitude bands. I'm in the middle of a local event with some fast pilots so I'll do some comparisons. I'm also in the 18m nationals so should have plenty of opportunities to learn more there. I do wish that WeGlide (and flight computers) published the formulas used for their derived metrics. thx, JJ

On Mon, May 27, 2024 at 10:59 PM Morgan Hall @.***> wrote:

Not someone from WeGlide, but saw this and read through your report. Keep in mind that the speed shown on the graph is Groundspeed. At 10-13k you're also going to see true airspeeds that are ~20% higher (2%/1000ft) than your indicated. So TAS + a tailwind could put you at 100kts over the ground with 70ish kts indicated. Not knowing what your winds were, their number seems a bit low, but not completely out of the realm of possibility.

I have felt like the numbers were generally a bit low as well, but the data probably tells a better story than my memory.

— Reply to this email directly, view it on GitHub <

https://github.com/weglide/bugtracker/issues/282#issuecomment-2134403457>,

or unsubscribe <

https://github.com/notifications/unsubscribe-auth/BIODHPS7BAHXER3QBJ6RYMLZEQMMNAVCNFSM6AAAAABHTFKYQGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZUGQYDGNBVG4>

. You are receiving this because you authored the thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub < https://github.com/weglide/bugtracker/issues/282#issuecomment-2135295531>,

or unsubscribe <

https://github.com/notifications/unsubscribe-auth/AAFHHKQMQV3GWAVBOIP7ZALZESE5JAVCNFSM6AAAAABHTFKYQGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZVGI4TKNJTGE>

. You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHub < https://github.com/weglide/bugtracker/issues/282#issuecomment-2135584422>,

or unsubscribe < https://github.com/notifications/unsubscribe-auth/BIODHPW7SJVETOAMYXJBDCLZESRNXAVCNFSM6AAAAABHTFKYQGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZVGU4DINBSGI>

. You are receiving this because you authored the thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/weglide/bugtracker/issues/282#issuecomment-2136490949, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAFHHKXTWCXADTSDD5JYPY3ZEVKTFAVCNFSM6AAAAABHTFKYQGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMZWGQ4TAOJUHE . You are receiving this because you commented.Message ID: @.***>

samuel-git commented 3 days ago

The discrepancy can be explained by the windvane in the bottom right corner indicating a 16kn tailwind and the difference between ground and airspeed pointed out by @morhall. The wind is an approximaten derived from your circling offsets. In real live scenarios we often see people flying "too slow" because they can not take the full risks in all scenarios and are cautious. But we found that the best pilots consistently hit the green areas of the coach.