Announcing the Guardian 2D/3D Stabilizer - Page 933 - RC Groups
Shop our Airplanes Products Drone Products Sales
Thread Tools
This thread is privately moderated by billpa, who may elect to delete unwanted replies.
Apr 14, 2017, 02:54 PM
Doug
d0uglas's Avatar

prepping for my first guardian flight


I think I've got the flight modes and gains more or less right, but once I try to put my plane in the air I may be in store for some negative surprises not anticipated in the basement. To mitigate risk of disaster, how does this sound?

In addition to the gain knob, I set a switch to crank any gain all the way off, an off switch (also effectively giving me a fourth mode). For the 3D non HH mode which I am most interested in using (I mostly just want to fight wind and turbulence, and then maybe screw with 3D HH), I dialed down my DR on all axes so that the full travel of the control surfaces matches the full travel of the sticks, rather than the 3D mode amplifying and for example giving me full rudder with only half a stick push.

I have flown this plane with no gyro assistance at all, so my plan is to take off in the 3D mode, that off switch not engaged, and once I've got some altitude I'll proceed to dial up the gain until something bad happens, at which point, depending on how scared I am, I'll either twist the knob back or hit the off switch. Oh, I also have the trainer button set to turn on 2D HH mode.

So for someone who's a little gunshy regarding using a flight controller, is that the way to go for the first flight? Thanks.

Doug

edit: Never mind, I stuck to the plan and landed safely.
Last edited by d0uglas; Apr 14, 2017 at 06:47 PM.
Sign up now
to remove ads between posts
Apr 14, 2017, 07:26 PM
ENE@10
RidgeMan's Avatar
It seems that the Guardian overrides my rate setting... when I go to "low rate" I'm still getting full deflection, whenever the Guardian is in 2D or 3D... do I have something set to high.. I do have "Direct Rate 3D Control" checked... not that I fully understand that setting... gains to high? Any advice appreciated.

RidgeMan
Apr 14, 2017, 10:23 PM
Registered User
Wintr's Avatar
Testing in 3D with direct rate on the ground, or in 2D, will give larger control deflections than expected. In 3D mode, direct rate will deflect the control until the aircraft rotates at the commanded rate, and, on the ground, it will not rotate, so the G keeps increasing the control deflection. The logic is different in 2D, but the result is the same - you command a given angle (roll or pitch), and the G deflects the controls until that angle is reached. In theory, if you command an 80 degree bank, then roll the model by hand, it should return the control to neutral as 80 degrees is approached. This test didn't work for the elevator when I tried it, so I can't confirm it will pass this test. I may have had Center Stick Stabilization Only mode enabled, though, so will have to check and try again.
Last edited by Wintr; Apr 14, 2017 at 10:29 PM.
Apr 14, 2017, 10:48 PM
Rick
Quote:
Originally Posted by d0uglas
....I dialed down my DR on all axes so that the full travel of the control surfaces matches the full travel of the sticks, rather than the 3D mode amplifying and for example giving me full rudder with only half a stick push....
Quote:
Originally Posted by RidgeMan
It seems that the Guardian overrides my rate setting... when I go to "low rate" I'm still getting full deflection...
First, let me suggest reading the New User's Guide on my blog page here, it address many questions: https://www.rcgroups.com/forums/member.php?u=354219

By design the Guardian will use as much deflection as it needs to achieve the intended result up to the limits set on the servo tab in the PC app. This is not controlled by your tx travel/rate settings, although there can be some influence. This often confuses new users but it's only a concern if these deflections cause binding. The default limits correspond to ▒400Ás or 100% on a Spektrum, Futaba and some other radios, 80% on Jeti, FrSky and some others. You should check to be sure there is no binding, if so, the Guardian limits need to be reduced. Otherwise you're good to go.

Don't be concerned if the deflections you see with the Guardian on differ from when it's off. In flight the G only uses the deflection it needs to achieve the result. It often differs from G off, this is normal and just shows it's working. In operation several factors affect the deflection, primary being the gain settings. Bottom line, follow the step by step in the New User's Guide and don't worry about it.
Last edited by choochoo22; Apr 15, 2017 at 03:27 AM.
Apr 15, 2017, 07:21 AM
ENE@10
RidgeMan's Avatar
Thanks! Will do.

Quote:
Originally Posted by choochoo22
. Bottom line, follow the step by step in the New User's Guide and don't worry about it.
Apr 15, 2017, 07:34 AM
ENE@10
RidgeMan's Avatar
Thanks! Will do. ... Great explanation on your blog!!

RidgeMan

Quote:
Originally Posted by choochoo22
. Bottom line, follow the step by step in the New User's Guide and don't worry about it.
Apr 16, 2017, 05:14 PM
Rick
Quote:
Originally Posted by Wintr
...In theory, if you command an 80 degree bank, then roll the model by hand, it should return the control to neutral as 80 degrees is approached. This test didn't work for the elevator when I tried it, so I can't confirm it will pass this test.....
You're right, the reality doesn't quite match the theory. On my test plane with 80░ target, all gains max, and default 100% travel and DR (▒400Ás), roll should neutralize at ~80░. Actually I see about 100░ bank angle, a little past vertical.

Although the bank comfortably works past vertical, pitch gets wonky at that point and with these settings looks like it will carry through a loop. Pitch worked more as expected with a little lower rate. This could be by design as not neutralizing vertical pitch just brings it through a loop and back to horizontal where the same behavior on roll could cause more problems. Just speculating.
Apr 17, 2017, 04:51 PM
Rick
Reconsidering my last post I remembered that this plane has a Lemon rx which tends to amplify the tx commands. Measuring the actual pulse width at the rx shows the default settings are producing ▒456Ás instead of the tx commanded ▒400Ás, or about 114%. Applying 114% of 80░ means the ailerons should hit neutral at about 91░. So the estimated observation of 100░ isn't really that far off and probably within a reasonable margin of error, this being a fairly imprecise target.

On the other hand, the pitch on this plane is still at the default 60░ target. Even applying 114%, it should still hit neutral well before vertical. At lower rate settings it seems to behave as expected except it hits a higher pitch angle than predicted. At a 60% rate it seems to hit neutral at about 70░ Expectation would be about 60░ x 60% x 114% = 41░. At 75% it seems to hit neutral at about 85░, just short of vertical. At 100% it never hits neutral.

The elevator behavior differs from the ailerons. While the ailerons seem happy to work essentially the same math ignoring vertical, the elevator seems to give vertical some special consideration. If the plane passes vertical before the elevator reaches neutral, the deflection will stop decreasing and begin to increase. Also, when pitch hits vertical the ailerons start trying to roll the plane over. Basically 2D becomes unworkable when the plane approaches vertical pitch, which makes some sense actually considering it's main intent is to level the plane. I've no explanation for why it seems to exaggerate the pitch target.

One other note; when testing this, you should first hold the plane level for a bit then move quickly to find the neutral point. If you hold the plane at a steep angle for any length of time the accelerometers continuously re-define "down" which can interfere with observations.

Conclusions? Ailerons behave pretty much as described in my blog write up. Elevator doesn't. I'm not sure of the reasons for the difference. I think the reason this hasn't been discussed/noticed a long time ago is that it really doesn't matter very much. People use 2D because they want the plane held near level. It's necessary to bank the plane to make turns and keep it in the flying field so people pay attention to 2D banking. Yanking large amounts of elevator is inconsistent with the general objective of keeping the plane near level so people don't do it and the behavior described above just doesn't come into play.
Apr 17, 2017, 09:23 PM
Registered User
Wintr's Avatar
With center stabilization only disabled, I started with the fuse level; the wing isn't on, so no aileron observations. I began to increase pitch and move the stick back at more or less the same time, and once the fuse reached about 50-60 degrees, the elevator remained at rough neutral, unless I increased stick displacement. Holding the same stick position from when neutral was reached at the 50-60 degrees, the elevator remained about neutral all the way to vertical. Of course, if speed is an issue, my test wasn't that fast. Increasing stick movement raised the elevator, while relaxing it lowered it. Crossing 90 degrees, things got less predictable. Clearly, the G is not meant to hold verticals in 2D mode. Too bad - it could have been a nice feature.
Last edited by Wintr; Apr 17, 2017 at 09:36 PM.
Apr 19, 2017, 01:17 PM
Just thumbing through...
victapilot's Avatar
Quote:
Originally Posted by Wintr
With center stabilization only disabled, I started with the fuse level; the wing isn't on, so no aileron observations. I began to increase pitch and move the stick back at more or less the same time, and once the fuse reached about 50-60 degrees, the elevator remained at rough neutral, unless I increased stick displacement. Holding the same stick position from when neutral was reached at the 50-60 degrees, the elevator remained about neutral all the way to vertical. Of course, if speed is an issue, my test wasn't that fast. Increasing stick movement raised the elevator, while relaxing it lowered it. Crossing 90 degrees, things got less predictable. Clearly, the G is not meant to hold verticals in 2D mode. Too bad - it could have been a nice feature.
Good idea! You might want to request this for the next version on the thread here:

https://www.rcgroups.com/forums/show...ou-want-to-see
Apr 19, 2017, 03:59 PM
Registered User
Wintr's Avatar
I haven't tried yet, but I would think 3D-HH would be better than 2D for what is a 3D maneuver. If it worked in 2D, it would be the lazy way to do a harrier, though. There is still the aileron issue in 2D, though.

Anyone here try a vertical or harrier with the G helping?
Apr 19, 2017, 08:47 PM
Michael
Yes, both, 3DHH. Guardian handles it quite nicely. I gradually reduce the Master Gain so that I am learning control, but have a momentary PANIC button that forces 2D, 40% Master Gain.

Michael
Quote:
Originally Posted by Wintr
I haven't tried yet, but I would think 3D-HH would be better than 2D for what is a 3D maneuver. If it worked in 2D, it would be the lazy way to do a harrier, though. There is still the aileron issue in 2D, though.

Anyone here try a vertical or harrier with the G helping?
Apr 21, 2017, 04:29 PM
Gnd warning sensor needed
GWPSneeded's Avatar
ET saves the day,I never realized how good it was. I just painted a pattern on the wings of my Fun Cub and then brought it to the field. I use the ET in 3d mode, it was a windy day. I did my normal ground check visual rudder. evaluator, ailerons, all looked normal. Took off for a normal flight no issues, dropped to full flaps for landing and landed with no problem. Only after I landed did i realize that the right flap never deployed because a small amount of paint made it stick. My mistake of not checking all surfaces before flight, but was amazed that the ET handled the major imbalance and save my plane from a possible very unhappy ending. I appreciate that the ET allows me to fly in not the best conditions but a normal land in such a configuration in the wind blow my mind. ET thank you!
Apr 21, 2017, 08:06 PM
Rick
Yes, it's very good at that. Have you seen this:

One Wing Short (0 min 20 sec)
Apr 21, 2017, 09:46 PM
Gnd warning sensor needed
GWPSneeded's Avatar
That quite a video a little more then my issue, ET definitely works


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Can I control Guardian stabilizer with DX6i? njwxyz Eagle Tree Systems 9 Jun 02, 2014 12:17 PM
Discussion GPS 2D Plots incomplete, 3D OK jackerbes Eagle Tree Systems 1 Oct 23, 2011 09:42 PM
Discussion Premier 1st generation, Gen 2 Vector unit, 2D or 3D motor unit lanternfly RC Paragliders and Kites 0 Oct 05, 2011 09:42 PM
Data 2d or 3d wing ribs? Marty7n Beginner Training Area (Aircraft-Electric) 56 Sep 22, 2011 04:06 PM
Discussion 3View to 3D to 2d techasist CAD/CAM 0 Mar 27, 2011 12:19 AM