I’ve tried and tried, but I can’t get Qgroundcontrol(4.0.8) to calibrate my radio (Frsky X9R+). Each time I calibrate Qgroundcontrol sees my stick input until the last stick input (pitch stick down). I’ve even tried holding the elevator stick down with a rubber band, but no matter how long the stick is down I get no response. Out of desperation I move the stick around pitch and roll. This causes Qgroundcontrol to accept the pitch input and lock it with roll.
While calibrating, the channel monitor monitor shows all the channels operating as expected at least until roll and pitch are locked together.
What do I need to do to get calibration to function properly?
Quick question. The lowest speed allowed in the parameters is 5 m/s. Is this just an arbitrary number? I’d like the aircraft to fly at 3-4 m/s if possible, I was thinking of writing in 3 m/s in the parameters regardless.
Also, @Roger_McClurg, have you tried missionplanner at all?
That’s great, thanks for posting, this will help to tweak the drag polar learning feature. Note that there’s no need to set TECS_OPTIONS to 0 if you have SOAR_ENABLE 1.
There should be no problem entering a smaller number than 5m/s. The warning is just that you are outside the “normal” range, most planes can’t fly that slow.
Can you try APM Planner for mac? if you have all wiring correct and no weird parameters I have a feeling that the problems with QGroundcontrol/software instead of arduplane itself. Also, this isnt the best place to ask this question either, I think you’d have better luck finding answers in the ground control software side:
Here is a typical track of what I get when I activate SOAR:
The small circles are me manually thermaling in a very strong thermal, small circles, staying in thermal, and drifting with the wind.
I activate SOAR near the end of the track so that plane is in the same thermal, and SOAR circles using much larger circles, and not drifting with the wind. It loses a good thermal and does not get much altitude gain even when presented with a very good thermal.
SOAR flys the plane really well, performance is not a problem at all, the only issue I am having is SOAR ability to circle small enough and stay in a thermal that always drifts with the wind. Any suggestions ???
I could see the same as @JetPilot, and maybe a SOAR_RAD and a SOAR_LIM_ROLL could help, because circling in thermals is much different to just loitering.
Circling in thermals normally is not so ‘efficient’ in matters of gliding or sinking, but the small radius is necessary to stay in the thermal lift - and so it’s efficient.
Thank You for the recommendation Samuel. ATX_Heli told me about the WP_LOITER_RAD setting in RC Groups. I had my first success using SOAR yesterday after setting WP_LOITER_RAD to 7 meters. I do not know how I missed that setting, but that was the problem. The glider would actually turn into a thermal and circle in it now. Yesterday was a really bad thermal day since there was cloud cover, so it is hard to judge how well it did as I was having trouble finding good consistant lift manually also… But SOAR did find and stay in some thermals, so I look forward to seeing how SOAR will do on a normal good day. I will also try optimising NAVL1_PERIOD and report back here as soon as I get a chance to try it again.
it shows turns at around 25° bank angle at an airspeed of roughly 11 m/s.
calculating the turn radius using
r = v**2 / g * tan φ
where
r = turn radius [m]
v = TAS [m/s]
φ= turn radius [deg]
i did hit a turn radius of roughly 25 m. if i had loitered at the default WP_LOITER_RAD of 60 m at that given airspeed, i would have had to use a bank angle of 11°. on the other hand, if i had made full use of the default 45° LIM_ROLL_CD, i most likely would have had to increase airspeed to compensate for the added load factor (~1.4 times) and avoid a stall. with the airspeed raised to 14 ms and a bank angle of 45°, i would have achieved a turn radius of ~ 20 m, only marginally tighter than the turns on 25° with 11 m/s, at the cost of noticeably less efficiency and and higher load.
so my assumption is that while reducing the default loiter radius will likely improve thermal performance and centering relevantly, increasing the 45° bank limit will most likely not add noticeable benefit while increasing the risk to exceed the airframe’s load limits at compareably low airspeeds. this might be applicable to a range of typical 2 - 3 m foam gliders, but vary noticably on different plane types and sizes.
i appreciate thoughts and corrections on this.
I need to purchase an airspeed sensor for my new project. Is there a consensus as to which type analog or digital is better, or does it come down to personal preference?
@Quaxwilly yes separate parameters specific to soaring would be good. I am planning a code re-structure that will make this more natural.
@JetPilot great, glad it’s working better for you now. Looking forward to your flights when conditions improve!
@vierfuffzig Your analysis is correct. The “optimum” bank angle depends on many factors - wing loading and drag characteristics of the glider, thermal strength and also thermal radius. In full size 45° is often a rule of thumb. At smaller scale, a lower bank is better unless you are in a really tight bubble.
@Roger_McClurg I have used analog or digital with success. Currently I’m using this unit . They also sell one with an integrated compass that might be very useful.
I agree with basti @vierfuffzig .
In very weak thermals, I find it even more efficient to fly uncoordinated curves with more rudder in order to keep bank and airspeed as low as possible and not to risk a tip stall.
Great discourse @Samuel_Tabor and @vierfuffzig on optimizing the radius, bank angle and speed parameters. So for my ~2M foamie, I gather from Basti’s results that 25M radius, ~25 degree max bank angle and ~11m/sec speed sound like the sweet spot? Also, can we get (or do we already have) separation of those three key parameters from Soaring context vs general purpose flight? I’d like to have a higher bank angle possible outside of auto-thermaling, but guide Soaring logic to keep if relatively “flat and slow” when coring thermals. Maybe also someday we have a 0 to 1 parameter that influences how much to use rudder vs bank to core the thermal.
Looking forward to results from @JetPilot’s next big air day.
I agree, this makes logical sense.
During the first tests three years ago I had already reported this to @Samuel_Tabor.
Whenever it finds a thermal and switches to Loiter, the development of evolution is out of phase with respect to the point where the thermal actually develops in height.
In the first turn it is normal, but from the second onwards the Loiter point should shift better.
The code should calculate the evolution of the thermal not only about the wind direction but also by attempting to center it by calculating the points where the glider goes up and the others where it goes down during a complete turn.
The central point of the Loiter should be recalculated at each turn based on this too.
Thermals tend to drift with the wind, but they do not drift with the wind perfectly There is always wind shear, and other factors that change how much the thermals drift in relation to the wind, and it changes depending on the day… Some days, I find that as I circle in the air mass, and have to offset significantly into the wind every few turns as the glider tends to get blown out of the thermal with the wind.
CONCLUSION: It is not posslbe to stay centered in a thermal based on wind speed and wind direction. You must measure lift in each sector of the circle, and adjust your circle to go towards point higest lift.
Hi,
Is there a way to limit the maximum altitude with this new soaring version?
I’m flying below a control zone. Currently with Plane 4.0, not yet the new soaring code.
But even if the plane reaches the max. altitude in circle mode and switches back to auto, the thermals are often so strong that the plane still gains altitude in auto mode.
I manually have to activate speed brake (crow) to stay below the altitude limit and keep the speed slow enough.
So I’m thinking about a max. altitude with a max. speed and active break (crow) to stay below these two parameter.
There is always wind shear, and other factors that change how much the thermals drift in relation to the wind, and it changes depending on the day… Some days, I find that as I circle in the air mass, and have to offset significantly into the wind every few turns as the glider tends to get blown out of the thermal with the wind.