Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Mechanics of flight

  1. May 10, 2016 #1

    Borek

    User Avatar

    Staff: Mentor

    OK, this is just a seriously looking subject to a probably trivial question.

    I am trying to learn to fly an RC model (Bixler 2, training/beginner level which I assume means "stable, easy to fly"). What is happening is that when I use ailerons to list the model and turn, it turns but then starts to climb. I have both ailerons and elevator under the right thumb, so there is a slight chance I do move the elevator involuntarily, but I don't think so - there is an easy to feel resistance on the stick which makes moving it in one axis quite easy.

    I am doing my best to trim the model and to keep the COG where it should be according to the manual, so I don't think it is something that trivial.

    Or perhaps such a behavior is perfectly normal and I should just counter it?

    There is also a chance the model is already unstable, as it got several serious hits (including one that broke 6 mm carbon tube wing spar).
     
  2. jcsd
  3. May 10, 2016 #2

    jedishrfu

    Staff: Mentor

  4. May 10, 2016 #3

    Borek

    User Avatar

    Staff: Mentor

    Trick is I am not losing altitude - I am gaining it.
     
  5. May 10, 2016 #4

    rcgldr

    User Avatar
    Homework Helper

    One potential issue is the ailerons have the same amount of throw both up and down. When banking the model, the downwards moving aileron moves into higher energy air than the upper moving aileron. This can cause some upwards pitch and adverse yaw, so that a right bank produces some upwards pitch and left yaw, both of which would nose the aircraft upwards. Assuming you have a computer radio, start off with the downward movement of ailerons to be about 50% of the upwards movement (this is the common setting for gliders). You could also mix left/right aileron -> left/right rudder (start with about left /right 10% rudder for left /right 100% aileron), which would compensate for not enough aileron differential and/or a vertical stabilizer that is too small.

    If you have a basic radio (no computer), you can setup the aileron control rods so that downwards motion only uses a portion of the servo arm travel, so that with ailerons centered, the servo arms are oriented towards the downwards throw, reducing downwards throw response.

    The goal is to trim the model so that roll (aileron) input only produces a roll without changing the pitch of the fuselage. You may get a barrel (spiral like) roll (since the model is trimmed for level flight, producing lift to counter gravity), but the key factor is keep the fuselage horizontal while rolling.

    The COG and elevator trim affect pitch stability. If the COG is forward, then more up elevator trim is needed to compensate, and the model pitches upwards at higher speeds and pitches downwards if too slow (due to downforce at the tail varying with speed). If the COG is too close to neutral, then the model tends to hold it's current pitch regardless of speed, which makes it difficult to land (no clue if it's about to stall until it's too late), so some pitch stability is a good idea. A radio mix of positive throttle to a small amount of down elevator (5% or less), can compensate for this, but this may also affect landings.

    The main issue when first learning to fly radio control models is orientation, when the model is headed towards you, the roll control appears to be reversed, since the models left is on your right. One trick for newcomers is to face your body away from the model (so that your chest faces in about the same direction as the nose of the model) and look back over your shoulder on an approaching model, so that the models left is on your left, and vice versa. You orient you body as if you were in the model, and look back over your shoulder as needed. You could also use a flight simulator, either a radio control one, or something like Microsoft Flight simulator, using tower view, on an aerobatic type aircraft like an Extra 300. After a month or so, basic orientation usually isn't an issue (complicated orientations, like inverted flight, take a bit longer, rudder and elevator are reversed, ailerons are normal).
     
    Last edited: May 10, 2016
  6. May 11, 2016 #5

    CWatters

    User Avatar
    Science Advisor
    Homework Helper

  7. May 11, 2016 #6

    Borek

    User Avatar

    Staff: Mentor

    Unfortunately, I have no one at the moment to help me, so I am on my own.

    I understand the part about differential on ailerons, but I don't think there is much I can do about the way ailerons are controlled. This is an preassembled ARF model, so the only thing I can change is the length of the pushrods - and I used them to trim the model as good as I could. Both ailerons are controlled by one simple channel (pretty basic radio), so even modifying the way the channel reacts to the stick will only make banking in one direction easier than in the other.

    Is it possible that the wind throws the plane off? My understanding is that once the model is airborne and flying with a reasonable speed, it should fly more or less straight and level (after all, these beginners models are designed to be stable). But hat's not what happens - that is, even if it starts nicely it flies straight and level for a second or two, but then it either starts to bank and turn or goes up - and I am more than sure it happens before I even touch the controls. After that I have to compensate and problems start - but as I have no time to observe how it reacts to the controls in a straight fly, there is no way I can learn correct reactions.

    I have plenty of experience with flight simulators and as a chemist I am trained to see reasonably well in 3D (learning about all these organic molecules pays of), so it is probably not a problem with being lost with the orientation of the plane. I have also read many tutorials and watched many videos - but they never mention nor address problems I see and have. I feel as if I was learning how to ride a bike from the helicopter tutorial.
     
  8. May 11, 2016 #7

    CWatters

    User Avatar
    Science Advisor
    Homework Helper

    Can you adjust the angle of the aileron control horns (and then change the length of the rods to return the aileron to neutral)? See...

    Differential-e1452807125448.jpg

    Google says that model might use two servos, one per aileron. Is that correct? If so an alternative to the above might be to move the servo arms on their splines and once again adjust the length of the push rod to return the aileron to neutral.

    Servo-Offset_cr-300x249.jpg

    Both images are from that link I posted above.
     
  9. May 11, 2016 #8

    CWatters

    User Avatar
    Science Advisor
    Homework Helper

    Sounds fairly normal for a beginner I'm afraid.

    Things to check..

    Centre of gravity - if too far back this can make it sensitive in pitch. A side effect can also make it seem more sensitive in other axis.
    Over controlling - make small movements on the controls!
    Control throws - Is there a "rate switch"? Some transmitters have a switch that reduces control throws making them less sensitive.
    Too windy or turbulent - are there trees or other obstructions upwind? Wait for a calm evening.
    Too much power - try reducing the throttle a bit once airborne. That might prevent excess climbing and make the controls less sensitive.

    Some clubs use a buddy box system to train pilots. This allows two transmitters to be connected together so that an instructor can take over instantly to prevent a crash. This gives pupils more flying time as ultimately that's what you need.
     
  10. May 11, 2016 #9

    rcgldr

    User Avatar
    Homework Helper

    I looked at the Bixler 2, and the images of the wing tips look different. On one of them, the wing tips are curved upwards quite a bit and should be reasonably stable on roll. On the other one, the wing tips are curved upwards only a small amount, and with the otherwise straight wing, that one will tend to randomly bank over to one side or the other due to lack of roll stability. Maybe you could post a photo of what your version looks like.

    First possible issue, does it go up and slow down before banking? If so, it's probably going too slow due to pitching upwards, and one wing will stall before the other, causing it to bank. Try trimming the elevator down a bit to see if this is what is happening.

    Assuming it's not due to going too slow, does it always bank to the same side? If so, trim the ailerons to deal with possible wing twist. If the fuselage is bent a bit, and the wings are bent upwards, then the misaligned rudder can cause the model to bank also, and rudder trim could also be used, trim rudder a bit left for a right bank or vice versa.

    If you have a large enough field, try short straight flights where you launch and land while trying to fly straight away from you, making sure to land before you run out of room or the model gets too far away to be seen easily. Use these short flights to trim the model so that it doesn't bank as much (if you're lucky, it may end up at least not banking).

    The next step or option is to fly the model left to right, then right to left in front of you, doing a stretched figure 8 or oval pattern, and turning your body so your chest points in the same direction as the nose of the model, looking back over your shoulder if the model is behind you.

    The model will tend to go up as speed increases. This is due to having COG in front of center of lift, which requires some downforce at the tail, so the model is speed sensitive. This is the nature of most models. With the prop that far up, and depending on how the axis of the motor is aligned with the fuselage, the initial response to increased power could be nose down, followed by nose up as speed increases, but it's probably oriented to minimize pitch effects related to throttle inputs.

    For the simulator, try using tower view. In my case, I had an old version of Microsoft Flight Simulator installed, and a joystick. I found that by pressing left and right arrows at about the same time, and holding both down for a brief moment, the result would be a random roll to either side, which I would correct using the joystick. In addition to tower view, I could also use a spot view looking back at the simulated airplane from in front of the airplane.

    These days, there are a few free radio control flight simulators that would be helpful.

    This is not going to do much with the problem you're experiencing now, as it's just as small effect. However, I looked up Bixler 2, and it uses a separate servo for each aileron. If you can access the servo arms without having to cut anything apart, then as suggested previously, detach and remount the two aileron servo arms so that both ailerons are pointed down a bit (like 1/4th of travel), then adjust the push rods to recenter the ailerons.

    Back to the problem, aileron differential is not a big deal at this point. You didn't mention what model transmitter you are using. Is this the default transmitter you get with the ready to fly version? If that is the HK-T6A V2, then it can be programmed using a PC, but unless you have a lap top that you can take to the flying field, it wouldn't be useful, and if the aileron servos are connected via "Y" harness to the same receiver channel, you can't use a mix for differential. If you did want to program the HK-T6A V2 you need to buy a cable (it's USB on the PC side) if the kit didn't come with one. The description mentions mixes, but not what mixes can be done. You can download the software for free. Be sure to save the original configuration first before making any changes.

    Another tip for beginners is try tapping or pulsing the transmitter sticks instead of holding them offset, as as way to make small corrections, and waiting a brief moment to see what happens.
     
    Last edited: May 11, 2016
  11. May 12, 2016 #10

    Borek

    User Avatar

    Staff: Mentor

    Today I decided to start with gliding (after all Bixler 2 is a motoglider), without throttle - and it helped. Place where I am trying is not flat, there is a bit of elevated terrain at one end, higher perhaps by one meter than the rest of the meadow (and the wind was blowing in the right direction). First - it allowed me to trim the model. With elevator perfectly in line with the horizontal stabilizer the model had a tendency to dive when flying slow, so I have modified the neutral position of the elevator. Second - I had an opportunity to see what is going on. Turns out the model starts to climb under power (which is most likely effect @rcgldr described), but after flying it as a glider for umpth time I was more or less capable of compensating this effect with the elevator after adding throttle.

    So, while am not there yet, at least today I never crashed the model hard (no new broken parts) and I found a way that allows me to incrementally build the skills required.

    Yes, I tried them last year, unfortunately at the time there were no signed 64-bit drivers (haven't checked them now) for the HK-T6A V2 (which you guessed right, cable included BTW) and I wasn't able to fly the simulation using the transmitter.

    Yes, that's the way they are connected.
     
  12. May 12, 2016 #11

    rcgldr

    User Avatar
    Homework Helper

    So, you'd have to do this mechanically by detaching and reattaching the servo arms as mentioned above. It's not a big deal at this point though.

    That's probably the best approach, only use the motor for short periods of time, then glide for a while, cycling back and forth as needed.

    As for the tendency to climb under power, you've already mentioned you can deal with this by compensating with some down elevator. Once you get a bit better you can check for elevator trim using the "dive" test. Get the model to a reasonably high altitude, and put it into a mild dive, maybe 5 to 7 degrees, then relax on the sticks to see if it pulls out of the dive on it's own, and how fast it need to go before it does pull out (assuming it doesn't dive down further still in which case you need up elevator trim). It should take 2 seconds or more to level out (and maybe climb a bit due to the speed). If it pulls out sooner, there's too much up elevator trim.

    Getting the COG in the right spot is a bit trickier. You should start off with COG at the recommended position, which is usually some distance behind the leading edge of the main wing, which you can determine by balancing the model with the wing on your thumb and fingers at the right distance behind the leading edge. There are balance tools you can buy that are a bit more accurate, or you could use a pair of pencils stuck into something, balancing the wing on the erasure end of the pencils. Once you get experienced with flying a model, the general rule is to move the COG back in small steps until the model becomes uncomfortable to fly for you (too neutral), then move the COG back forwards a bit. As you move the COG back, you'll need to do the dive test again, since less up elevator trim will be needed. With the COG back and less up elevator trim, the model will be less sensitive to speed.
     
  13. May 14, 2016 #12

    Borek

    User Avatar

    Staff: Mentor

    In the case of my model it was given as some distance from the nose - I just marked it on the fuselage and I balance it whenever I replace the battery by supporting the model on about a half inch plank pt where my mark is and seeing, whether the nose goes up or down.

    Sadly, as of today the weather doesn't cooperate, and the forecast for the next few days doesn't look good - mostly because it will be wet and raining.
     
  14. May 14, 2016 #13

    rcgldr

    User Avatar
    Homework Helper

    That's a bit unusual, since measuring from the leading edge of the wing root would be a bit easier. The point is to have the center of mass just a bit in front of the center of lift from the main wing, with negative (effective) lift at the tail to balance the model about the center of lift. This means that the model in glide mode will tend to glide at a specific speed. If it goes faster than trimmed speed, the downforce at the tail increases, pitching the model up and slowing it down. If it goes slower than trimmed speed, the downforce at the tail decreases, pitching the model down and speeding it up.

    With the motor on, the position and orientation of the propeller affects pitch (also roll and yaw). To reduce speed sensitivity with the motor on, the propeller is usually oriented to generate some downwards pitch when the motor is on. In the case of models where the propeller is mounted at the nose, the propeller is usually oriented a bit downwards (to reduce pitch effect) and a bit off to one side to reduce torque effect. Wiki link describes the effect, but without much explanation.

    http://en.wikipedia.org/wiki/Torque_effect
     
  15. May 18, 2016 #14

    Borek

    User Avatar

    Staff: Mentor

    So, today was the first time I got back home not because I broke something, but because both I have batteries went flat. I consider it an important step forward.

    Saying I got things under control all the time would be a lie, but at least I had no problems keeping the model in the air - so I can concentrate on gaining some skill in deciding in which direction it flies :wink:
     
  16. May 18, 2016 #15

    rcgldr

    User Avatar
    Homework Helper

    Have you been able to trim it so that at least when it's gliding, it glides at a near constant speed and doesn't bank to one side or the other, without having to make any corrections with the transmitter? Note - a gusting crosswind may cause it to roll, but it should self correct in 1 to 3 seconds, assuming the wing tips are curved upwards (dihedral) enough.
     
  17. May 19, 2016 #16

    Borek

    User Avatar

    Staff: Mentor

    Yes, it glides more or less nicely (I mean: while it sometimes drops off the curse, perhaps because of gusts, I don't think it has a systematic drift).

    But to my surprise it doesn't self correct - once rolled to the side it remains rolled and flies in circles. Actually at the moment that was my main problem during turns - how to end the turn with model flying in the direction I want it to. Even if I managed to keep it level it almost never ended flying exactly where I wanted it to fly. Skill to learn I suppose.
     
  18. May 19, 2016 #17

    rcgldr

    User Avatar
    Homework Helper

    Is this true for both left and right turns? Try turning the model using just rudder (assuming it has rudder). If the wing tips are curved upwards enough (and there aren't issues related to vertical stabilizer size and fuselage length), it should roll somewhat like a normal turn, even when just using rudder. If not, it will just yaw side to side without much roll (bank). Be ready to correct the model using ailerons when experimenting though.
     
  19. May 20, 2016 #18
    I have read through this thread, and am probably just missing the place where this was suggested, but in case I am not, why not try this and just see what happens:

    When you are rolling (say) left, under power, simultaneously give it a little left rudder, same right and right. Try it. Might sound obvious or downright strange depending on how you look at it :).

    ---diogenesNY
     
  20. May 20, 2016 #19

    rcgldr

    User Avatar
    Homework Helper

    To clarify, this would only be done during the time the aileron stick is moved left or right, and only if you know how the model reacts to aileron inputs. At this point, it would be simpler to bank with aileron first, then adjust rudder and/or elevator to adjust and keep the fuselage nearly horizontal if the goal is a steady turn. In the case of gliders during slower tighter turns, the lower relative speed on the inside wing tends to cause the model to keep increasing bank, which requires a bit of outwards aileron and inwards rudder for a coordinated turn.

    I'm still waiting for a response from Borek on how his model responds to rudder only inputs to get an ideal of how much effective dihedral there is.
     
  21. May 21, 2016 #20

    Borek

    User Avatar

    Staff: Mentor

    Yes, I remember. I didn't have a time for flying in the last few days. I have an important deadline on Monday, so I am in a bit of a frenzy right now (at least seems like I have things under control).

    Such tests require a steady flight, which is not something I can always manage (it still doesn't seem like I have things under control).
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Mechanics of flight
  1. Nano Flight (Replies: 9)

  2. Human Flight (Replies: 63)

Loading...