HobbyKing.com New Products Flash Sale
Reply
Thread Tools
This thread is privately moderated by billpa, who may elect to delete unwanted replies.
Old Jun 12, 2012, 11:19 AM
Registered User
ricoalonso's Avatar
USA, KS, Derby
Joined Mar 2003
1,476 Posts
Quote:
Originally Posted by FLAME747 View Post
Ok let's go back to basics

From the limited info that's available, the unit needs to be mounted on a flat surface, from experience when I used my FY 20 even though the unit was placed on a flat surface in the model, once the unit was activated it tend to pitch down, and roll to the right

Solution - 1 either use trim to keep the model level, it worker but took away from the models flight performance

Or 2 shim the unit to achieve level flight without using excessive trim

On the FMA Co-Pilot II, one can compensate for any unwanted roll, and pitch by using flight angles, so that the control surface stays level at neutral

In short, what options are available to trim out any unwanted pitch and roll tendencies with the guardian

Thanks

FCA

Aero, Al. Keep up the good work
The option in CPII where you can set the Flight Angles to compensate for the level of the sensors is done differently in the Guardian.

To initially set the level flight of the model is to install the Guardian as flat and as close to the CG of the model. Then, place the model on a test bench in the orientation as close as a level flight of your model when on the air, and have the Guardian memorize this attitude as the level flight by toggling the mode switch.
ricoalonso is offline Find More Posts by ricoalonso
Reply With Quote
Sign up now
to remove ads between posts
Old Jun 12, 2012, 11:35 AM
Your customer
Silverexpress's Avatar
Berkley, MI
Joined Dec 2009
1,408 Posts
Gyros in this application sense rotational motion and the Accelerometer senses g's. Both can also provide direction, but I believe that you'll need another sensor(s) to detect a stalled condition. I also think that two conditions must be handled by the software - 1. Preventing a stalled condition 2. Recovering from a stalled condition. Neither of which the Guardian is programmed to do.

I would settle for a stall alarm on my TX using telemetry. Wherein I can set an angle of attack, and an airspeed limit. The TX vibrates when you equal or exceed these limits, and you make the corrections - increase throttle, and pitch plane down. I think with the telemetry gadgets available today - this is more doable. The only telemetry sensor missing is an external gyro for rotational sensing, but hey the Guardian has this already - It's only missing the airspeed sensor. Sigh.

Reference:
These are just a list of info if anyone wants to delve into this a bit more. I add these things for my benefit since I tend to forget things.

http://en.wikipedia.org/wiki/Accelerometers
http://en.wikipedia.org/wiki/Gyroscope
http://en.wikipedia.org/wiki/Stall_(flight)
http://en.wikipedia.org/wiki/Fly-by-wire
http://www.eagletreesystems.com/Stan...standalone.htm
http://www.spektrumrc.com/DSM/Technology/telemetry.aspx

Just another cool video I found on systems that recover from a stall.

NASA AirSTAR testing L1 adaptive control for post-stall control (2 min 42 sec)
Silverexpress is offline Find More Posts by Silverexpress
Last edited by Silverexpress; Jun 12, 2012 at 07:33 PM.
Reply With Quote
Old Jun 12, 2012, 12:43 PM
Registered User
Joined Mar 2009
218 Posts
Quote:
Originally Posted by Silverexpress View Post
Gyros in this application sense rotational motion and the Accelerometer senses g's. Both can also provide direction, but I believe that you'll need another sensor(s) to detect a stalled condition.
v=yCRQLhZK6A8[/url]
Please excuse my shortening of your post, Silver. I just wanted to home in one of your comments and say that I agree with you.

Wing stall is a function of wing angle of attack, that is, the direction at which the wing is meeting the airstream. It has nothing to do with pitch attitude, airspeed, or direction of flight. Angle of attack is a parameter that we don't have available to us nor do any of the current stabilization systems sense angle of attack.

What I tried to say in a previous post is that a stabilization system might unintentionally drive an airplane into a wing stall while attempting to maintain a pitch attitude. Pilots are notorious for doing the same thing. That's why NTSB accident reports are replete with findings of "accidental stall".

Angle of attack measurement is a tough parameter to do on a model airplane. The sensor is usually a vane mounted on a probe sticking out of the leading edge of the wing or out of the nose. If the probe is on the leading edge of the wing, the sensor needs to be at least one chord length ahead of the leading edge. The sensor would be really tiny and delicate.

Dick
otrcman is offline Find More Posts by otrcman
Reply With Quote
Old Jun 12, 2012, 01:05 PM
Your customer
Silverexpress's Avatar
Berkley, MI
Joined Dec 2009
1,408 Posts
Quote:
Originally Posted by otrcman View Post
Please excuse my shortening of your post, Silver. I just wanted to home in one of your comments and say that I agree with you.

Wing stall is a function of wing angle of attack, that is, the direction at which the wing is meeting the airstream. It has nothing to do with pitch attitude, airspeed, or direction of flight. Angle of attack is a parameter that we don't have available to us nor do any of the current stabilization systems sense angle of attack.

What I tried to say in a previous post is that a stabilization system might unintentionally drive an airplane into a wing stall while attempting to maintain a pitch attitude. Pilots are notorious for doing the same thing. That's why NTSB accident reports are replete with findings of "accidental stall".

Angle of attack measurement is a tough parameter to do on a model airplane. The sensor is usually a vane mounted on a probe sticking out of the leading edge of the wing or out of the nose. If the probe is on the leading edge of the wing, the sensor needs to be at least one chord length ahead of the leading edge. The sensor would be really tiny and delicate.

Dick
This unit has autonomous landing capability, and so I would think that the developers have somehow solved the prevention of stalling.

http://www.uthere.com/products/ruby/..._aircraft.html

http://www.youtube.com/watch?feature...&v=i6eg_6JIxnk

So is it a high Alpha that you are referring to? What type of airframe do you think this type of stalling would likely occur in, and I'll see if I have something similar in my hanger. If so, I'll move my beta Guardian to it, but I'll need a flight plan/procedure to ensure I initiate the proper stall.

If anyone else has something they'd like to see tested, just let me know. Again, I'll need a description of what you want tested, a flight plan, and the enviromental conditions.

Reference: I came upon this video a while back, and it's got to be one of the neatest. It showcases the use of this technology to stabilize an aircraft in high alpha. There's no music.

http://www.youtube.com/watch?v=SViiqylV0lA

The paper on this research is floating around on the internet, and in it is a diagram of the flight control system. It shows the use of gyros and accelerometers. I dare guess that Dave from RCSuperpowers based all his gyro "testing" on youtube from this very document.
Silverexpress is offline Find More Posts by Silverexpress
Last edited by Silverexpress; Jun 12, 2012 at 07:35 PM.
Reply With Quote
Old Jun 12, 2012, 02:30 PM
Electric baptism 1975
DavidN's Avatar
Vernon, BC, Canada
Joined Dec 2000
2,921 Posts
Dick
If stall is solely related to angle of attack (& wing loading), then in planes with a zero wing incidence, if the 2D set up is done correctly, it should keep the wing @ zero angle of attack. However as you pointed out dropping an aileron will change the centerline of the wing section, effectively giving it a +ve angle of attack. So we still have to fly & keep the landing speed up.

Would turning down the gain reduce the amplitude of the control surface movement commanded by the Guardian?
Sorry for being long winded, thinking as I write.
DavidN is offline Find More Posts by DavidN
Reply With Quote
Old Jun 12, 2012, 03:00 PM
Gopher huntin' stick jockey
turboparker's Avatar
East Bethel, MN USA
Joined Jul 2009
12,465 Posts
One method of reducing the chances of a stall during an engine-out with an RC autopilot is to program the system to manage speed with elevator when the engine quits. This requires an airspeed indicator and prior knowledge of the plane's power-off stall characteristics (which of course will vary with density-altitude, G-loading, and AUW). Definitely far from being fool-proof - but it usually results in a stable power-off glide.

And yes - the full-scale guys solved the dilemma decades ago. NASA worked on it back in the '80s. The full-scale solution is nearly fool-proof, and it is much more complex than the method above. Interesting read: http://ntrs.nasa.gov/archive/nasa/ca...1984012520.pdf

From the paper:

"The NASA-sponsored conceptual research work resulted in one very promising generalized design approach based largely on point mass energy control considerations. The basic longitudinal point mass airplane equation of motion, solved for thrust required (TRE ) (shown below), indicates that the aircraft's energy rate is mainly controlled by thrust, while at constant thrust (Es-O), the elevator control results in equal and opposite responses of flight path angle and longitudinal acceleration (short-term D-constant). The elevator control is, in essence, energy conservative.

It follows then, that from-an energy management point of view, the thrust should be used to control the total energy and the elevator to control the desired energy distribution between the flight-path angle and acceleration, or altitude and speed.
"

Joel
turboparker is online now Find More Posts by turboparker
RCG Plus Member
Old Jun 12, 2012, 04:58 PM
Sopwith Camel's Cousin
Between my tx and crashed aircraft
Joined Mar 2006
3,803 Posts
Quote:
Originally Posted by otrcman View Post
Yes, you may find some specific examples where the stall you describe is desirable, but for the most part it is highly undesirable, especially if you are trying to keep the wings level with aileron control.

I demonstrate that sort of "stabilized stall" to my full scale flying students on a regular basis. But there are several caveats for doing it. First, we don't do it below 2000 feet AGL. Second, on most airplanes you dare not use aileron inputs to keep the airplane wings level. You have to use rudder. Depending on the airplane, and the CG when you perform the stall, it is very likely that you will have a violent roll reversal if you attempt to use ailerons alone. With a stability augmentation system such as the Guardian, the ailerons will respond to any roll motion and could exacerbate what modelers call "tip stall".

This is not a criticism of the Guardian, it's just an attempt to understand how the whole system plays out in conjunction with the basic aerodynamics of the airframe.

I agree that most 3D models still retain roll control well into the stall, but that's because the propellor flow straightens the relative wind such that the inboard wing is not really stalled in spite of the apparent high nose-up attitude. In addition, the ailerons are usually full span and immersed in propellor flow.

...
Thanks for your reply (and also comments from others). I am getting to learn and get reminded of a few things.

My plans are to put this into a light, bouncy foam 3d model, so my brain was angling in that direction (still have aileron control in a "stall" from the prop wash, crashing is not good but does not have such bad consequences).

Not to say that this is in the Guardian, but one guess at automatically trying to mitigate potential tip stalls (without additional sensors):
* detect a plane's desire to want to tip stall by detecting a sudden uncommanded (neutral or close to neutral aileron stick) angular acceleration in roll
* at that point, counter with opposite rudder (instead of opposite aileron)
flying-llama is offline Find More Posts by flying-llama
Reply With Quote
Old Jun 12, 2012, 05:14 PM
Gopher huntin' stick jockey
turboparker's Avatar
East Bethel, MN USA
Joined Jul 2009
12,465 Posts
Quote:
Originally Posted by flying-llama View Post
Thanks for your reply (and also comments from others). I am getting to learn and get reminded of a few things.

My plans are to put this into a light, bouncy foam 3d model, so my brain was angling in that direction (still have aileron control in a "stall" from the prop wash, crashing is not good but does not have such bad consequences).

Not to say that this is in the Guardian, but one guess at automatically trying to mitigate potential tip stalls (without additional sensors):
* detect a plane's desire to want to tip stall by detecting a sudden uncommanded (neutral or close to neutral aileron stick) angular acceleration in roll
* at that point, counter with opposite rudder (instead of opposite aileron)
FL,

Sounds similar to what AS3X does when it combats tip-stalls.

Joel
turboparker is online now Find More Posts by turboparker
RCG Plus Member
Old Jun 12, 2012, 05:44 PM
Registered User
LawnDartMike's Avatar
United States, OR, Salem
Joined Jan 2012
287 Posts
Quote:
Originally Posted by flying-llama View Post
....one guess at automatically trying to mitigate potential tip stalls (without additional sensors):
* detect a plane's desire to want to tip stall by detecting a sudden uncommanded (neutral or close to neutral aileron stick) angular acceleration in roll
* at that point, counter with opposite rudder (instead of opposite aileron)
What would happen when you have a wind gust that mimics the "sudden uncommanded (neutral or close to neutral aileron stick) angular acceleration in roll"?
LawnDartMike is offline Find More Posts by LawnDartMike
Reply With Quote
Old Jun 12, 2012, 07:09 PM
Registered User
Joined Mar 2009
218 Posts
Quote:
Originally Posted by Silverexpress View Post
So is it a high Alpha that you are referring to? What type of airframe do you think this type of stalling would likely occur in, and I'll see if I have something similar in my hanger. If so, I'll move my beta Guardian to it, but I'll need a flight plan/procedure to ensure I initiate the proper stall..
Silver,

I'd suggest an airplane of fairly conventional shape -- Piper Cub or P-51 style should do fine. It shouldn't be a flat foamie, though, because flat plate wings don't have the same stall characteristics as wings with airfoils.

The plan would be (in 2D mode) to establish a steady, level flight at cruise or full power. Then chop the throttle to zero power. Watch the airplane to see if it noses down in the descent or does it try to hold the nose level as if it still wants to maintain cruising flight.

If the nose remains in a level flight attitude, but the descent angle gets steep enough (10 or 15 degrees), the wing will probably stall.

DavidN,

In your post, you said, "If stall is solely related to angle of attack (& wing loading)......" You were right on the money until you added wing loading. Wing loading has little or nothing to do with stalling angle of attack. If anything, the same airplane at a lower wing loading may stall at a slightly lower angle of attack.

Think about it this way: You go out to the airport and notice a row of airplanes parked on the ramp. They are all facing the same way, nosed directly into a 10mph breeze. There is a Bonanza, a Mig 21, a Cherokee, and a Piper Cub. Which airplane, if any, is stalled in this light breeze?

Answer: Only the Piper Cub. Not because of its airfoil, or its wing loading, or the low airspeed. Only because it is sitting tail down such that the wing is at or above its critical angle of attack. If you put tufts on the wings of all these airplanes, you will see smooth airflow over the Bonanza, Mig and Cherokee because their wings are at low angles of attack. But the tufts on the Piper Cub will be showing separated airflow.

Now, are any of these planes making lift in the 10mph breeze ? Yes, all of them. But not enough to fly.

Now, imagine the wind speed increases to 60mph. Which airplane will tug at its tiedown chains, trying to fly ? You guessed it, only the Piper Cub. Even though the wing is stalled, it is still making some lift, and that's enough to overcome the weight of the airplane. The Bonanza and the Cherokee would probably try to lift off if you put a jack under their nosewheels and raised the nose 10 degrees or so. The Mig ? Forget it, wing loading is too high to make enough lift even at 60mph.

Dick
otrcman is offline Find More Posts by otrcman
Reply With Quote
Old Jun 12, 2012, 07:22 PM
Registered User
Ken41's Avatar
United States, PA, Clarksville
Joined Aug 2006
39 Posts
My thoughts are, that a fair number of RC pilots that have not learned to fly a full scale AC, or been have been in the aviation industry, do not understand that the horizontal stabilizer / elevator controls pitch / airspeed. The throttle controls altitude
(more throttle, more speed equals more lift, AC climbs, less throttle, lower airspeed, AC descends) The Ruby by Uthere has a pitot tube to sense airspeed, and can control the throttle position and elevator, hence airspeed, but the cost is about 4 times the Guardian.

When I install the Guardian, on one of the first flights, at altitude, I'll pull the throttle back and see how it is going to stall.

Ken
w3kmp
A&P 1964
ASEL 1965
Ken41 is offline Find More Posts by Ken41
Last edited by Ken41; Jun 12, 2012 at 07:24 PM. Reason: Add name
Reply With Quote
Old Jun 12, 2012, 07:25 PM
Sopwith Camel's Cousin
Between my tx and crashed aircraft
Joined Mar 2006
3,803 Posts
Quote:
Originally Posted by turboparker View Post
FL,

Sounds similar to what AS3X does when it combats tip-stalls.

Joel
Thanks for the info.
I have read that AS3X handles tip stalls,
and I would have guessed in the way described,
but I did not know that AS3X did handle tip stalls that way.
flying-llama is offline Find More Posts by flying-llama
Reply With Quote
Old Jun 12, 2012, 07:30 PM
Sopwith Camel's Cousin
Between my tx and crashed aircraft
Joined Mar 2006
3,803 Posts
Quote:
Originally Posted by LawnDartMike View Post
What would happen when you have a wind gust that mimics the "sudden uncommanded (neutral or close to neutral aileron stick) angular acceleration in roll"?
With what I just described, it would, for better or worse, still try to counter with opposite rudder (as opposed to opposite aileron).

My description is that of a simple way to try to detect and handle tip stalls, probably not the best way.
flying-llama is offline Find More Posts by flying-llama
Last edited by flying-llama; Jun 12, 2012 at 08:06 PM.
Reply With Quote
Old Jun 12, 2012, 08:06 PM
Gopher huntin' stick jockey
turboparker's Avatar
East Bethel, MN USA
Joined Jul 2009
12,465 Posts
Quote:
Originally Posted by Ken41 View Post
My thoughts are, that a fair number of RC pilots that have not learned to fly a full scale AC, or been have been in the aviation industry, do not understand that the horizontal stabilizer / elevator controls pitch / airspeed. The throttle controls altitude
(more throttle, more speed equals more lift, AC climbs, less throttle, lower airspeed, AC descends)
The Ruby by Uthere has a pitot tube to sense airspeed, and can control the throttle position and elevator, hence airspeed, but the cost is about 4 times the Guardian.

When I install the Guardian, on one of the first flights, at altitude, I'll pull the throttle back and see how it is going to stall.

Ken
w3kmp
A&P 1964
ASEL 1965
Ken,

I've been preaching this to RC pilots for years! That is how I was taught when I learned to fly full-scale. It is often a tough-sell to RCers. Interestingly, two full-scale instructors have argued this with me in other UM threads. Also - I've been looking at the Ruby with keen interest.

BTW - nice to meet another fellow ham on here!

Joel - N0NCO
turboparker is online now Find More Posts by turboparker
RCG Plus Member
Old Jun 12, 2012, 08:46 PM
Rick
United States, CA, Santa Clara
Joined Mar 2011
2,195 Posts
Quote:
Originally Posted by turboparker View Post
Ken,

I've been preaching this to RC pilots for years! That is how I was taught when I learned to fly full-scale. It is often a tough-sell to RCers. Interestingly, two full-scale instructors have argued this with me in other UM threads. Also - I've been looking at the Ruby with keen interest.

BTW - nice to meet another fellow ham on here!

Joel - N0NCO
My flight instructor provided a pretty convincing demonstration. She took control of the yoke and let me manage the throttle. Then she challenged me to accelerate, we were cruising at perhaps 80kt. If I firewall the throttle, she just pulls the elevator up, 80kt and climbing. If I kill the throttle she pushes the nose down, 80kt and descending. No matter what I do to the throttle, she can hold the airspeed with the elevator. My throttle changes, however, had a big impact on altitude. Likewise, using the elevator, she could reach any airspeed from stall to structural failure no matter what I did to the throttle. This was a very important lesson for managing the glide slope on final approach, and it's just as useful for RC models.

I think the confusion comes because this way of thinking breaks down under certain circumstances where the elevator can override the throttle. If you hold the plane in a steep dive, for instance, no amount of throttle will make you gain altitude. Also, the principal applies to steady state so transient conditions like flying a power-off loop don't abide. Some common sense is required in extreme situations.

An interesting situation occurs at cruise. If you are cruising at a steady altitude and airspeed and would like to cruise at a higher speed, you need to add throttle. This appears to violate the elevator-controls-airspeed principal, but really it doesn't. As the plane accelerates, it will start to climb and you must re-trim the elevator to maintain altitude. Looking at the situation a different way, you are trimming the elevator down to get the airspeed you want and increasing the throttle to hold altitude, elevator controls airspeed.
choochoo22 is online now Find More Posts by choochoo22
Last edited by choochoo22; Jun 12, 2012 at 10:14 PM.
Reply With Quote
Reply


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 01:17 PM
Discussion GPS 2D Plots incomplete, 3D OK jackerbes Eagle Tree Systems 1 Oct 23, 2011 10:42 PM
Discussion Premier 1st generation, Gen 2 Vector unit, 2D or 3D motor unit lanternfly RC Kites 0 Oct 05, 2011 10:42 PM
Data 2d or 3d wing ribs? Marty7n Beginner Training Area (Aircraft-Electric) 56 Sep 22, 2011 05:06 PM
Discussion 3View to 3D to 2d techasist CAD/CAM 0 Mar 27, 2011 01:19 AM