Ridge thermals severely interfere with wind over terrain

[wickid]

Anyway, I just want to make sure we agree on the methodology. If so, Ill try make some better tests

No... look at my email

[havet865]

I fly with a HAWK vario IRL. Photo attached below. Wind 195/37, my track is 100. So almost 90 degrees. I'm only getting 2.9 m/s netto. Which almost matches your bottom screenshot. So yes, it works correctly in Condor3. Condor2 was WAY to overpowered on ridgelift.

100% agree with this, there is no way a ridge would give you +2m/s or +3m/s with 15km/h wind perpendicular to the ridge in real life.

Condor 2 was overpowered… it was a big issue for task setting as putting a wind speed > 20km/h would give you « unlimited » varios on the ridge… you would never have to stop or slow down, and everyone was flying at VNE.


Wind breeze needs to be added for flying low on ridges, but apart from that, the new system is 10 times better.

[Vertigo]

No... look at my email

Thats really cool. Wonder if you shouldnt make that public with the author's consent.

However, what i see is windspeed dropping off, which I have no reason to question and I have no idea of the magnitude or complexity. My issue is not that with x km/H wind in the notam (not even sure if thats supposed to be ground wind or at what altitude) I would expect Y kmH wind or Z m/s lift on some ridge. Thats completely beyond the scope or my intuition.

My issue is the hawk measures (by your account, only the horizontal component of) windspeed based on the air the glider is moving in, so that is after those effects are accounted for and simulated. No matter how complex those physics, Pythagoras still holds true, you can calculate the wind angle based on the horizontal and vertical vectors. If anything the simulation shows air moving even more "perfectly" following the slope angle than I would have assumed, and that angle is not what I seem to see in condor for steep slopes by just adding up horizontal and vertical wind vectors. Calculating the actual wind speed on a three dimensional slope is way above my pay grade, but the relation between angle, horizontal and vertical speed, is not complicated and doesnt need supercomputer modelling. Its simply Pythagoras. if the resulting lift is correct, and I didnt completely misjudge the slope angle, then maybe the hawk is showing too much wind (or not just the horizonal component)?

[wickid]

No... look at my email

Thats really cool. Wonder if you shouldnt make that public with the author's consent.

However, what i see is windspeed dropping off, which I have no reason to question and I have no idea of the magnitude or complexity. My issue is not that with x km/H wind in the notam (not even sure if thats supposed to be ground wind or at what altitude) I would expect Y kmH lift on some ridge. Thats completely beyond the scope or my intuition.

My issue is the hawk measures (by your account, only the horizontal component of) windspeed based on the air the glider is moving in, so that is after those effects are accounted for and simulated. No matter how complex those physics, Pythagoras still holds true, you can calculate the wind angle based on the horizontal and vertical vectors. If anything the simulation shows air moving even more "perfectly" following the slope angle than I would have assumed, and that angle is not what I seem to see in condor for steep slopes by just adding up horizontal and vertical wind vectors. Between angle, horizontal and vertical speed, one of them has to be wrong. if the resulting lift is correct, then maybe the hawk is showing too much wind (or not just the horizonal component)?

You assume you are right next to the ridge. Where in fact you are a few 100 meters away from the ridge where the flow is more horizontal. If you look at your AGL height at a typical position for ridge soaring you'll see that you are 200 to 300 AGL mostly. Assuming a 45 degree slope, you are the same distance horizontally away. Most mountains don't have an average slope of 45 degrees. They are flatter than that. That makes your horizontal distance even further away.

Condor also simulates wind flowing around spurs and into bowls in the ridge. So the wind is not at all points 90 degrees to the slope, even though you set it to be 90 degrees in the general wind. The local wind varies. Both with terrain and the wind variation setting. It is a very dynamic calculation. If you want to do a good test you have to make a scenery with a perfectly straight 45 degree slope and no terrain upwind. I've been doing several test flights in different locations today. And the steeper the slope, the further upwind I have to be to get good lift. Which makes sense as the air just doesn't turn 90 degrees at the bottom of a cliff.

Anyway, as Antoine indicates, and from my personal experience the lift is about what is expected in RL conditions at similar windspeeds. Uros is a meteorologist in his day job and he flies in the mountains almost weekly both with paragliders and gliders. So I'm sure he knows what he is doing.

To get back on the OP's issue. I've tried some weather settings with northerly winds (around 30 to 40 kph and thermals activity at normal). I'm finding soarable ridges even lower down in the valleys around Lienz-Nikolsdorf in Slovenia3. Thermals do indeed interfere with the ridge lift. The higher the activity, the worse the ridgelift gets. I don't see that as totally unrealistic as thermals do interfere with local windfields. Strong thermals can and do make the windsock rotate 180 degrees in calm winds in the flatlands. It is far from as strong as Condor2. But not totally unusable as some claim.

I bet when we get valley breezes it is going to be even more of a mess in the mountains trying to find good lift

[Vertigo]

You assume you are right next to the ridge.

I dont make any such assumption, I looked at the numbers given by the hawk vario. It gives me 2 numbers: horizontal and vertical speed, from which I derived a third: wind angle. Or the other way, based on a guestimated slope angle and measured horizontal wind speed, I derive the vertical speed. Im open to discussing which of those numbers might be incorrect or why, but the relationship between them requires no assumptions and was established a few centuries before computational fluid dynamics

[wickid]

You assume you are right next to the ridge.

I dont make any such assumption, I looked at the numbers given by the hawk vario. It gives me 2 numbers: horizontal and vertical speed, from which I derived a third: wind angle. Or the other way, based on a guestimated slope angle and measured horizontal wind speed, I derive the vertical speed. Im open to discussing which of those numbers might be incorrect or why, but the relationship between them requires no assumptions and was established a few centuries before computational fluid dynamics

Oke, you assume the flow should be perfectly parallel to the slope. Which it certainly isn't at the position you are normally flying when ridgesoaring. It only is just above the surface. And you are deep in the windgradient there, so the speed isn't that high.

[Vertigo]

Did a slightly more "scientific" test. I flew this ridge from the green to the blue arrow, making screenshots every few seconds, and then later noting horizontal windspeed and netto vario:



Average was 32.2 KmH wind and 3.8m/s netto (=13.68 KmH, ignore the wrong unit in the spreadsheet).
That works out to an average angle of 22.9 degree


I dont have a PhD in physics, and I didnt use any supercomputer time on this, but that doesnt seem quite right to me, and not in line with the simulation you showed me. Seems to me either the vertical speed is too low, or the indicated windspeed is too high, but the resulting angle is too shallow.

[wickid]

Did you take into account the direction of the wind. If the wind is not 90 degrees to the slope, your calculation is off.

[Vertigo]

I put the wind as close to 90 on the ridge as I could (that its not perpendicular to my plane crabbing against the wind shouldnt matter), there was no variability in wind speed or direction in the notam, the sim does vary the wind speed, but the direction in the screenshots did not vary by any visible amount.

sure, its not perfect, but I think its a bit too far off to be handwaved away.

edit: wait, I have the wind direction as numbers, I could try correct for that.

edit bis: not worth it. Almost all between 103 and 105 with one outlier to 106. Fair to say the ridge bends more than that.

[Vertigo]

If any landscape makers are bored; how much work would it be to make a test landscape with a single perfect 25/45/60 degree sloped mountain?

[Vertigo]

To get a feel for how far off we are, this shows what the expected average lift from the vertical wind component would be for varying angles, assuming the hawk's reported horizontal wind speed component is correct:


Red is what I measured in condor, somewhere between green and yellow would seem more correct angles to me. This is by no means a perfect test, but thats a pretty substantial difference.

Again though, I have no idea if the problem is with the reported horizontal (wind) component or the vertical (lift) component or the angle condor's physics engine comes up with. But at least one of them seems off in this test, and in most "snapshots" I looked at.

[wickid]

Your diagram is oversimplified. The wind curves from horizontal to parallel to the slope while in a wind gradient (reducing in strenght as the height above terrain reduces) It is not as linear as you draw. It is also only valid it the wind is perfectly 90 degrees to the slope.

[Vertigo]

(reducing in strenght as the height above terrain reduces)

Im using measured wind strength, at the exact (and only) place the glider can measure it, Im not trying to calculate what the wind strength ought to be or how it varies across heights or inclinations; I have no idea what that should be, Ill leave that to supercomputers and those much smarter than me. But if you give me 2 numbers, I can draw right triangles

As for the wind not being perfectly perpendicular to the slope or the slope curving a bit here and there; sure, that might explain a bit of it and I wouldnt expect a perfect match. Though the slope changing direction by a few degrees ought to work in both directions and average out, in the end the wind has no where to go but up and I would expect average wind speed, wind direction and avg slope angle to be close enough. Ill do the math later how much the wind/slope angle should be off to explain the measured discrepancy, I dont believe it will come close to explaining this.

[wickid]

Dude... you are not getting it... at the point your measuring the angle is correct. It is what you measure it to be. It is more flat further away from the ridge and steeper closer to the ridge... what is so hard about that.

[Tom]

Hey, how about we wait for Uros to pop in on this one, as mentioned he is a "Senior meteorologist iirc he works with the Slovenian FAA" and he is the guy behind the weather system, couple that with a huge amount of mountain flying experience both with Paragliders and Gliders, the ridge and mountain lift system for 3.0 went under a ton of testing. Every real life pilot with real frequent experience in Mountain flying I have spoken to has said that the current form is as close to real life that they have seen in a sim and this is the first time I have seen anyone saying that its not realistic etc.

I am sure he will when he gets a chance he will drop in with his thoughts on this etc.