| Author |
 Topic  |
|
eric.carden
Aquilotto
USA
220 Posts |
 Posted - 03/02/2011 : 19:49:59
|
Hi,
I’m going back and forth on which option to choose for “Landables Value” (Config Menu 11, “Appearance”). The options are “ArrivalAltitude” and “EfficiencyReq”. Each option has some good points, but neither seems to relieve me as much as possible of some tough in-flight mental calculations. My primary interest in LK is for helping me (in my hang glider) cross some large areas with few or no landing options, so this “landables value” is the most important feature in LK to me.
The arrival altitude option accounts for the wind (speed and direction) and for the glider’s polar. This is great, but an estimated 400’ arrival altitude means something entirely different to me when only a couple of minutes away from the field than when 20 minutes away. In the former case, I’d have little hesitation to go for it, but in the latter case, I’d probably want more altitude. Therefore, I’d need to do some “mental processing” in flight to come up with the right decision based on my distance from the field.
The “EfficiencyReq” option is nice in that it naturally accounts for varying distances from destinations, but it doesn’t help me account for wind effects and makes no use of my polar data. So my reach upwind may be only 4:1, while my reach downwind might be 20:1. Hang gliders, with lower airspeeds and poorer high-speed performance, are very much affected by the wind – much more so than are gliders/sailplanes.
So how about a new (third) option that combines the best of both of these, accounting for wind effects, my polar, AND my distance from the field? And while we’re at it, how about we also account for the average sink expected ahead? Here’s what I have in mind…
In a similar way to that in which LK already tracks the average gliding L/D over a user-set time period (“average” value) or since the last thermal (“cruise” value), LK could track the average sink found in the air “recently” (however you want to define “recently” – or leave it up to the user). LK already calculates “netto” vario values for snail trail purposes, so the same logic could be used here – just over much longer time periods (i.e., minutes instead of seconds). By the way, the netto value would be much more accurate in this case for users (like me) with only GPS altitude available to LK, given the much longer period over which to measure an altitude change. Glides are typically made at fairly constant airspeeds, so I think it would be good enough to just compare endpoints over a time period and average the airspeed between these two points (as opposed to keeping a running log of instantaneous airspeeds).
The only new thing might be coming up with a way to track the distance covered THROUGH THE AIR during the time period. (The above L/D calculations just use distance over the ground.) That might not be that difficult, though. I suppose LK could check the airspeed every X seconds, calculate the air distance covered at that speed during those X seconds assuming that constant airspeed, and add that to a running total distance. Given this through-the-air distance, the altitude lost, and the amount of time; LK could calculate the average airspeed and sink rate and then could use the existing netto vario logic (i.e., look to the polar) to calculate the average vertical speed of the surrounding air during the period.
Once this between-thermal netto value is calculated, it could be applied to the existing “ArrivalAltitude” value if desired (maybe a user option). But I think the ideal displayed value is a third option… maybe called “ExtraSink”. With this option, the value displayed would be the “extra” sink (relative to the above-calculated netto value) one could tolerate while on glide to the field and make it there at exactly his safety altitude (AGL). The value would still appear next to the field name as soon as LK currently puts it there (as soon as the field is reachable – NOT considering sink along the way). Then assuming the ambient netto value is -0.5 (0.5 kts sink), the label/value would first appear as “FIELD:-0.5”, indicating that to reach “FIELD”, I’d need to find 0.5 kts LESS sink than I’ve found recently on glide. Once LK determines that I could make the field assuming 0.5 kts of sink all the way there, it would read “FIELD:0.0”. Then as I climbed higher or drifted closer to the destination, it would start showing positive numbers (e.g., “FIELD:0.5”), indicating that I can make it there even if I find MORE sink than I’ve found on glide recently. Sweet!
And while I’m dreaming… I’d like to see the “landables value” toggle between two or more options. For example, I’d like to see the values toggle every few seconds between arrival altitude (assuming it accounts for ambient sink) and “extra sink”. I might even like to see L/D, too (a 3-way toggle!). And it might be good to add a user option for toggle timing, too.
Using LK as it is now, I’m sure I’ll be trying to do math in my head when deciding whether to go on glide over a large unlandable area. I’d much rather the computer do it for me, of course, freeing my mind up for other things… like finding more lift!
Thanks for considering, and let me know if there’s anything I can do to help make this happen.
Eric |
|
|
Coolwind
Moderator
Italy
8957 Posts |
Posted - 04/02/2011 : 00:27:53
|
Uhm. Very complicated, Eric. Lets start from the bottom.
We can add a third option for landables values: both LD and Arrival Altitude.
Of course, page 2.2, 3.1 etc are there for this reason but for paragliding (do you hang glide, alas Delta, or paraglide eric?) I understand we want a page with all in there.
Now, last year I also did an experiment by using the average LD to forecast arrival altitude (in this case compensated for wind). Results were very similar to what we have normally, so no reason to keep it.
This year we do have something new to test: equivalent MC.
They are all guesses, and we cannot forecast the air we have in front of us, or the wind.
I shall ask our Calculation Man (Lucas "atan") if he can work out a value that combines LD and distance and arrival altitude on a scale . Best Alternate is working like that already.
paolo
|
 |
|
|
MPusz
Aquila imperiale
Poland
683 Posts |
Posted - 04/02/2011 : 14:48:47
|
I also had similar request from a very famous glider pilot. He said that he does not like Final Glide bar because for him 20m above final glide line is much different when he is 20km or 2km from the finish line. He wanted to see some scaling (i.e. relative to the glide line angle) for the Final Glide strip taking into consideration the distance from the destination. So in example if your needed glide angle to the finish line is 10 degrees more/less than needed than print the same length of the strip no matter how many meters you are above or below the glide line.
Maybe someone will invent and define a new unit for that, that I believe is similar to what Eric is asking for? |
|
|
|
Coolwind
Moderator
Italy
8957 Posts |
Posted - 04/02/2011 : 15:33:38
|
I am glad to know that this idea was shared among many people, Mat.
Like you say, 25 reqE at 40km distance is very different than 25 reqE at 10Km !!
stand by..
|
|
|
|
MPusz
Aquila imperiale
Poland
683 Posts |
Posted - 04/02/2011 : 16:27:41
|
| I am fine with Efficiency (as it describes your glide angle) but I tend to have problems with altitude that is independent from the distance to destination. |
Edited by - MPusz on 04/02/2011 16:28:16 |
|
|
|
2xa
Pulcino
Sweden
18 Posts |
Posted - 04/02/2011 : 20:21:29
|
Is this a possible solution?
Safety altitude equals the highest value of set safety altitude or Alt diff Mc safety – Alt diff Mc=0
Alt diff Mc 0: Altitude difference at Mc setting of 0
Alt diff Mc safety: Altitude difference at the abort / safety MacCready setting
Example. The glider LD at MC=0 equals 40. Glider LD at safety MacCready setting equals 35.
Fixed safety altitude 100 m.
Case 1. Glider is 30 km from landing field. Alt diff Mc 0 = 714
Alt diff Mc safety = 625. Alt diff Mc safety – Alt diff Mc=0 = 89. Safety altitude = Set safety altitude. In this example set to 100m.
Case 2. Glider is 60 km from landing field. Alt diff Mc 0 = 1714
Alt diff Mc safety = 1500. Alt diff Mc safety – Alt diff Mc=0 = 214. Safety altitude is set to 214m.
214 > 100.
I know there is safety MacCready setting but I don’t understand the use of it.
|
|
|
|
Coolwind
Moderator
Italy
8957 Posts |
Posted - 04/02/2011 : 20:32:49
|
Safety altitude? We are not talking about safety altitude. It cannot be changed by the calculator!!
Beware that LD is meaning lift to drag, relative to air mass. GR is distance / altitude, relative to ground and thus including wind. |
|
|
|
eric.carden
Aquilotto
USA
220 Posts |
Posted - 04/02/2011 : 21:33:27
|
Yes, it's a bit complicated, but I have hope that it's possible and that it would significantly improve LK's predictions of whether fields are within reach. If no one's interested in tackling this development, I may dust off my C programming skills and give it a try myself - with your blessing, of course. (Of course, it would probably go much more quickly and smoothly to just consult for someone already set up for development and familiar with the LK code.) I believe I could provide all the math/calculations/logic pretty quickly, but getting a development environment set up, learning my way around the LK code, etc. could take a while.
Adding the third option as you suggested would be a slight improvement - as long as it toggles between the two, not showing both at the same time, which would take up too much precious screen space. Adding this option, though, wouldn't do anything to address the main issue - sink ahead on course.
I've flown paragliders in the past, but I don't any longer. Now I just fly hang gliders. And you're right, HG/PG pilots are best served by instrumentation that requires no touching in flight. Ideally it wouldn't even require LOOKING! :-) By the way, LK might get more interest from hang glider pilots if it called their gliders "hang gliders" instead of "deltas". I've been flying HGs for 15+ years now and never heard anyone refer to one as a "delta" (except when discussing the original gliders that have been obsolete for decades) until I saw LK. Using this terminology might give HG pilots the idea that HGs were given serious consideration in LK development.
In the experiment you mentioned that you did last year, did you do real thermal-soaring FLIGHT experimentation or just experiment using the simulator on the ground? If the testing was only done on the simulator, then the results don't surprise me, since the simulator doesn't allow one to specify ambient sink. I'd like to learn more about this experiment, though, in order to avoid possibly duplicating efforts.
I understand that we can't KNOW what the horizontal (wind) and vertical (lift/sink) motion of the air ahead is, but I do think we can make educated guesses that help our flight computers give us more accurate, and thus more useful, information. For example, LK already "predicts" what the wind ahead will be when calculating arrival altitudes. It predicts that it will be the same as recently measured. And I think that's a fine thing - much better than just assuming there will be NO wind. LK predicts the sink ahead, too. However, when it does so, it just predicts that there will be NONE. I suspect that that's a significantly less-than-ideal approach. If possible and potentially helpful, why not apply to the question of SINK ahead the same good approach proven useful for the question of WIND ahead? Here's an example to illustrate how much impact a little inter-thermal sink can have on my glide...
I'm 11.9 miles upwind of and 3K' above my safety altitude at a field on the other side of a large remote, unlandable area. I have a 20 mph tailwind, and LK tells me that I'm in position to (just barely) make it to the field at my safety altitude (LK shows an arrival altitude in the single digits). I've found an average of 100 fpm sink (a realistic-enough number) between thermals in the last half hour or so, and the same 100 fpm sink awaits me all the way to my field. I'm a moron, though, and I just blindly trust LK and go on glide over dinosaur country towards my field. My Flytec 6030 notices the sink and audibly guides me to the ideal speed to fly to get the farthest reach. I dutifully comply, but I peek at LK after a couple of miles and see that LK says my field is no longer reachable. Oh no! No other fields are in reach, so I'm in trouble and start praying for lift. I find none and am down to my safety altitude four miles short. My safety altitude (700') gets me an extra 1.8 miles, and I crash into the remote forest 2.2 miles short of my intended field.
Granted, I'm not bold (stupid?) enough to just go for it like in this example, but I believe it would be much better to assume the inter-thermal sink will stay the same from glide to glide than to assume there will be NONE on the next glide. I can, of course, study scenarios like the one above and boost some safety factors to cover all possibilities (unnecessarily limiting my X-C potential in the process), but why not just improve LK's predictions... if possible? :-)
By the way, a similar but less ideal approach might be in order to test the water on this one. A user-adjustable (in flight) setting similar to the manual McCready setting could be added. This setting could be the expected amount of sink ahead, and LK could use it when calculating arrival altitudes. Think of it as another safety factor. For those of us with barometric pressure netto varios, we could take the value we see there on glides and manually enter/adjust it in LK in flight. The pilot should be able to tweak this number during a given flight until LK's predictions get better (no problems like in the above example). This certainly isn't a hands-off, no-thinking-required solution (like the original one I offer), but it would be a substantial step in the right direction. And depending on how useful this proves in the real world, we'd have a better idea whether it's worth going all the way with the "fully automatic" approach.
Thanks,
Eric |
|
|
|
eric.carden
Aquilotto
USA
220 Posts |
Posted - 04/02/2011 : 21:42:48
|
Clarification... When I concur that the approach I propose is indeed a bit "complicated", I mean that it's a bit complicated to implement/develop. For the pilot, though, if implemented like wind consideration, it wouldn't change the program interface one bit. It would be completely transparent to pilots except that they would notice (if it works and they're observant) that LK's arrival altitude predictions improved!
Eric |
|
|
|
Coolwind
Moderator
Italy
8957 Posts |
Posted - 05/02/2011 : 00:55:33
|
Eric, I already made it like you say. It is called Equivalent MC, set AutoMC on, and you will get arrival altitudes that are based on what you are actually doing in the sinking or lifting airmass. Equivalent MC is the same used by paragliding instruments...
(btw thanks for suggestions about HG, I shall change delta terminology!)
|
|
|
|
eric.carden
Aquilotto
USA
220 Posts |
Posted - 05/02/2011 : 20:24:47
|
Interesting... Yes, my Flytec 4030 and 6030 give an "equivalent" MacCready setting, just showing me on screen what climb rate I appear to expect in the next thermal based on my current choice of airspeed. I just modify my speed till I see the climb rate I really expect, and voila, I'm flying the ideal speed-to-fly. I played around with this some in LK (simulator), and it appears to work as with the Flytecs except that LK appears to average values over a few minutes, where the Flytecs seem to average over just a few seconds. (Just an observation... not that I have a preference...) I experimented (in LK) with the effect on arrival altitudes (especially while thermaling, which is my most important scenario) and found that they're indeed affected by the MacCready setting. I have just one question:
Does LK do netto vario calculations on glide and then use those values when predicting arrival altitudes, or does it just use the MacCready setting to find the speed to fly and assume that on the next glide I'll fly that speed and sink at the polar-indicated sink rate (not adding in sink in the air mass)? I can't test/prove this using the simulator, since I don't think the simulator allows me to cause the air mass as a whole to sink. If LK just uses the MacCready setting and the corresponding polar-derived sink rate, then the sink in the air mass is still unaccounted for. If LK does use the netto vario calculations here, then yes, indeed. It would appear that LK already does more or less what I want.
This is technical stuff, and I'm probably asking the question poorly, so let me attempt an example to clarify it. Imagine I'm flying along in air that's sinking at 100 fpm. I don't see any good signs of lift ahead, so I fly at a MacCready setting of zero in order to reach as far as possible. Because of the sinking air, though, and my need to speed up to get through it with minimal altitude loss, my sink rate is about 120 fpm higher than if flying at a MacCready setting of zero in vertically neutral air. I get lucky, find a thermal, and start circling. LK continues to show a MacCready setting of zero as I climb. While climbing, do LK's arrival altitude calculations assume I'll make my next glide at my best glide speed of 28 mph with a neutral-air sink rate of 200 fpm (L/D = 12), or do they assume I'll make my next glide at 30 mph and a sink rate of 320 fpm (L/D = 8) like on my previous glide? Both of these methods account for the MacCready setting, but only the latter takes into account the sink I found on my previous glide.
Thanks,
Eric |
|
|
|
Coolwind
Moderator
Italy
8957 Posts |
Posted - 05/02/2011 : 21:23:12
|
Eric, I have spent a great deal of time writing the manual. Everything is in there. The EqMC is on the website, new features link.
|
|
|
|
eric.carden
Aquilotto
USA
220 Posts |
Posted - 14/02/2011 : 22:04:38
|
Thanks. Yes, thank you for the work on the manual. I refer to it often and intend to make some contributions myself soon. Neither the manual nor the web site (including the description of the new EqMC feature/function) quite answer my question, though. I downloaded the source code, and after several hours of digging, I found that LK doesn't do quite what I'm looking for.
I want to know, while circling, how likely I am to make it to a distant field across a large unlandable area. In this scenario, I'll want to assume MC=0, as I'll need every mile of glide I can get to cross the dinosaur country. In this scenario, I wouldn't want LK to use the EqMC setting. After all, I may have flown fast (to a good cloud, maybe) on my last glide but want to fly at best glide speed on the next glide - in order to maximize my reach across dinosaur country. When circling just about to go on a glide like this across dinosaur country, I'd like to see/know how much sink I could handle on the glide and still make it. That's the most useful info I've been able to think of, anyway.
Also, even if LK is using the EqMC setting to predict arrival altitudes while circling, this doesn't mean that LK is predicting arrival altitudes based on the amount of sink in the air on the last glide. It's a bit tough to get my head around this concept, but it goes like this... Let's say that on my last glide, I flew such that EqMC was pretty steady at 0.0 (maximizing my glide across the ground). I noticed big sink in the air, though, and was thus flying way faster than usual for a 0.0 setting. So the EqMC setting can be zero, even when there's huge sink in the air. So if LK predicted arrival altitudes assuming I'd fly this EqMC setting on the upcoming glide - without predicting big sink all the way there - I'd come up way short!
I've begun working on the code to offer a "MaxOkaySink" option for "Landables value", which will display the maximum average amount of sink that can be found on the upcoming/current glide and still make it to the landable waypoint at the desired safety altitude (while still considering the current or safety MC setting). I'm close, but it's taking quite a bit of time to learn my way around the LK code, set up a development environment, get my head back into C programming, and learn the new stuff (not in C) in C++ that I've never seen (I'm new to C++). It's good to have a "project", though! And I'm eager to finish by Spring, when the best three X-C months (by far) of the year begin.
Eric |
|
|
|
rimaroc88
Falchetto
102 Posts |
Posted - 14/02/2011 : 23:28:00
|
what about those two ideas for new values:
"arrival altitude" divided by the "distance to the airfield"
or
"efficency to reach the arrival altitude" minus "Req efficiency"
The second choise look better. And it could be compared to the best L/D or safety L/D of the glider/paraglider.
There is still to define the MC used to calculate this "Eff to reach the arrival altitude": current or safety or equivalent MC? Maybe the one in use... |
|
|
|
eric.carden
Aquilotto
USA
220 Posts |
Posted - 15/02/2011 : 00:07:38
|
Hi,
While the first idea tackles the issue of the effect of varying distances from targets, it doesn't address the effect of the wind adequately. For example, if flying the ideal speeds-to-fly, I'll (in my hang glider) get about 3X the ground speed when heading DOWNwind in a 20 mph wind as when flying UPwind in a 20 mph wind. So a displayed "500 ft/mi" would mean very different things to me depending on the wind along a route to the destination. I'd feel 3X as good about that number if the field were downwind of me than I'd feel if it were UPwind of me. To put it more clearly, I could "survive" 3X as much sink heading downwind as heading upwind given the same "Arrival Altitude / Distance" number.
I must not understand what you meant by your second idea. I think that LK's term "Req efficiency" means exactly "efficiency to reach the arrival altitude", so the way I read your second suggestion, the value would always be zero! :-) (Set me straight. I know this can't be what you meant.)
Thanks,
Eric |
|
|
|
Michel
Pulcino
79 Posts |
Posted - 02/03/2011 : 10:23:50
|
| Can we test a value for the relative glide angle (RelE.dif) being (current average efficiency-requested efficiency to destination)/(requested efficiency to destination*0,01) and see if we like it? |
|
|
|
Topic |
|
|
|
new "landables value" option?