Thread Tools
Old Mar 08, 2012, 05:40 AM
Taurus Flyer is offline
Find More Posts by Taurus Flyer
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
4,249 Posts
Idea
Gliders, measurement of slip while turning.

Daniel Bernoulli
"...there is no philosophy which is not founded upon knowledge of the phenomena,....................."


Gents,
I did fly the real KA8 in the past and know how important our thread was, glued on the outside of the canopy. The slip indicator.

Flying the Scale ASK18 (4,3m) model in the past I never did use a mixer, so left hand rudder and right hand ailerons and no feedback of my input only used my experience of controlling a real one.


knowledge of the phenomena

Bringing the ASK in the air again in the near future I already did make a electronic thread replacement, a vane slip sensor.
Combined with a data logger and other signals it will be possible to analyse my pilot skills and to optimize a rudder mixer I already made for a Horten Glider.
The Horten was controlled with small dragrudders to prevent slip.

I Could have made all kind of sophisticated instrument but the vane seems to be the easiest way.

Most important is the mechanical layout of a vane sensor and so a picture to show. The ball bearings are mounted IN the vane, the shaft is fixed with the electronic part inside the fuselage.

Second picture the MODIS 4 channel analogue and 2 digital to store in the memory.

Third picture, an old vane sensor inside with optical method. LED, 2 LDRs and amplifier. Case was of a servo with 2 ball bearings.

Flying of the ASK18 will take a while, the glider is still in the shop for update and repair.

Of course these sensors I also can use as a AOA sensor.

Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Last edited by Taurus Flyer; Mar 08, 2012 at 05:46 AM.
Reply With Quote
Sign up now
to remove ads between posts
Old Mar 08, 2012, 09:20 AM
Brandano is offline
Find More Posts by Brandano
Registered User
Joined Oct 2004
3,349 Posts
Well, since you have the vane alerady fitted you can mix its signal with the rudder axis input, so that flying without touching the rudder will keep the plane coordinated, but you would still be able to intentionally induce a sideslip to slow down for a landing.
Brandano is offline Find More Posts by Brandano
Reply With Quote
Old Mar 08, 2012, 12:39 PM
Taurus Flyer is offline
Find More Posts by Taurus Flyer
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
4,249 Posts
Primary I want to know the relationship of airspeed, G load, elevator- and aileron input for needed rudder deflections.
Knowing this a next step can be fine tuning or modify of a mixer I already have. This can be a simple solution because no external sensors are needed.

One reason is regulations, think about XC flying in which all control functions have to be done by the pilot.

A second reason not starting with a control loop is to prevent unwanted oscillations or hysteresis, this can happen without notice and can be a loss of energy and forces the need of continuing observation and guarding of the system.. Determining of the optimal control parameters will not be easy either, also because the plane is flown over a wide speed range and in a lot of dynamical circumstances.

We will see in the future, also combinations are possible, but first enlarge the knowledge of the phenomena by observation.

Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Reply With Quote
Old Mar 09, 2012, 10:15 AM
aeronaut999 is offline
Find More Posts by aeronaut999
Registered User
The Willamette Valley, Oregon
Joined Dec 2008
1,388 Posts
Slip discussion thoughts / links

Sounds interesting Cees

Related thread:

"Curving relative wind, how much slip is ideal in turns, yaw strings videos"
http://www.rcgroups.com/forums/showthread.php?t=1597991

Includes link to this interesting article:

http://www.wisoar.org/Documents/Holi...Efficiency.pdf

Arguably, some slip is desirable in a turn, especially in a long-spanned aircraft with dihedral, using ailerons for roll control. Especially as measured at the nose of the plane (since the relative wind is curved, then if the flow were straight at the vertical fin or even at the CG, then it would be somewhat sideways at the nose, in the slip direction. A straight yaw string at the nose would mean that a yaw string back at the vertical tail or even the CG would show a slight skid.)

So some of the main things that cause slip are--
Adverse yaw from rolling motions
Adverse yaw deflected ailerons even when bank angle is constant
The fact that in a constant-banked circle, the outboard wingtip moves at a higher airspeed, and thus tends to experience more drag, than the inboard wingtip, yawing the nose toward the outside of the turn.

Some of the advantages of a bit of slip are--

Interacts with dihedral to create a rolling-out torque so less (or no) rolling-out aileron deflection is needed to hold bank angle constant in sailplane, etc. Aileron deflection may create more drag than a bit of slip, especially if the fuselage is sleek and slender.

Allows fin to generate some yawing-in torque, reducing the amount of inside rudder we need to apply. Rudder deflection may create more drag than a bit of slip, especially if the fuselage is sleek and slender.

Even just keeping the fin 100% streamlined to the flow means that some slip will exist further forward on the aircraft. This may be more efficient than keeping some point near the CG or nose tangent (streamlined) to the flow. Especially if the fin is big and the fuselage is streamlined and sleek, like a modern sailplane. Certainly in no case would it be optimal to keep the extreme nose tangent (streamlined) to the flow-- on a slab-sided glider like a Schweizer 2-22 we might prefer to keep the center of lateral area tangent (streamlined) to the flow-- this generally lies aft of the CG....

Steeping back to some more fundamental stuff that applies to all aircraft even with big boxy fuselages--

One way to think of it-- fundamentally, slip has nothing do with G-loading or pitch input or turn rate, it's not caused by the "wrong amount" of turn rate for the bank angle or the "wrong" amount of centrifugal force etc, as some of the books teach us. It's simply due to real, tangible adverse yaw torques from real, tangible aerodynamic effects that act to yaw the nose out of line with the flow. Ultimately this does end up creating an aerodynamic sideforce that affects the turn rate so that it is "wrong" for the bank angle, but that is not the fundamental cause of the slip.

Langeweische was wrong on this one ( see http://www.aeroexperiments.org/critiques.shtml#SAR )

Some other writers were too ( see http://www.aeroexperiments.org/critiques.shtml )

Thought question-- could easily be explored experimentally-- in a constant-banked turn, does using the elevator to pull some "extra" G's beyond normal for the bank angle, or "unloading" to zero G's, create an immediate, marked sideslip? I've found generally not-- fundamental cause of slip is not the "wrong amount of lift for the bank angle" as some books teach-- though undoubtedly changing the lift vector does have some small effect on the various aerodynamic torques that tend to yaw the nose this way or that... in a very extreme wing-over with near-vertical bank and near-zero G-load "over the top" there is some noticeable tendency for sideslip unless the pilot gives some bottom rudder--

Please keep us posted on your results, thanks

Steve

Quote:
Originally Posted by Taurus Flyer View Post
Primary I want to know the relationship of airspeed, G load, elevator- and aileron input for needed rudder deflections.
Knowing this a next step can be fine tuning or modify of a mixer I already have. This can be a simple solution because no external sensors are needed.

One reason is regulations, think about XC flying in which all control functions have to be done by the pilot.

A second reason not starting with a control loop is to prevent unwanted oscillations or hysteresis, this can happen without notice and can be a loss of energy and forces the need of continuing observation and guarding of the system.. Determining of the optimal control parameters will not be easy either, also because the plane is flown over a wide speed range and in a lot of dynamical circumstances.

We will see in the future, also combinations are possible, but first enlarge the knowledge of the phenomena by observation.

Cees
aeronaut999 is offline Find More Posts by aeronaut999
Last edited by aeronaut999; Mar 09, 2012 at 02:34 PM.
Reply With Quote
Old Mar 09, 2012, 01:48 PM
aeronaut999 is offline
Find More Posts by aeronaut999
Registered User
The Willamette Valley, Oregon
Joined Dec 2008
1,388 Posts
PS the yaw string video in the thread I linked above suggests that we can do pretty well simply with electronic mixing of rudder to aileron and nothing fancier.... the situation is a little more complicated in a more scale-like sailplane which tends to have shorter fuselage in relation to span, so tends to show a little more slip in a constant-banked turn so you would need to over-ride the mix a with little inside rudder even if you are holding some outside aileron, if you want to keep slip to a minimum- it would be a pretty interesting project to an automatic slip eliminator going on the sailplane but one's biggest problem might be that it would now act like a glider with zero dihedral, since dihedral only makes roll torque during sideslip. That might not be such a good thing...

Keep us posted on the results of your observations.... Steve
aeronaut999 is offline Find More Posts by aeronaut999
Last edited by aeronaut999; Mar 09, 2012 at 02:24 PM.
Reply With Quote
Old Mar 09, 2012, 01:55 PM
aeronaut999 is offline
Find More Posts by aeronaut999
Registered User
The Willamette Valley, Oregon
Joined Dec 2008
1,388 Posts
Suggestion-- mount the sensor on a mast near the CG not on the canopy-- the forward placement (ahead of CG) plus the location on a curved canopy both cause the slip measurement to be exaggerated as described here

http://www.wisoar.org/Documents/Holi...Efficiency.pdf

Especially if the plan is to collect data for use later in a glider without the vane, why not get the best data you can even if the installation is a little draggier?

This picture may help show why the location forwards or aft along the fuselage where you measure slip, makes a difference-- the relative wind is curved in turning flight

http://www.av8n.com/how/htm/yaw.html#sec-long-tail-slip

Steve
aeronaut999 is offline Find More Posts by aeronaut999
Last edited by aeronaut999; Mar 09, 2012 at 05:29 PM.
Reply With Quote
Old Mar 09, 2012, 02:22 PM
aeronaut999 is offline
Find More Posts by aeronaut999
Registered User
The Willamette Valley, Oregon
Joined Dec 2008
1,388 Posts
Quote:
Originally Posted by aeronaut999 View Post

Thought question-- could easily be explored experimentally-- in a constant-banked turn, does using the elevator to pull some "extra" G's beyond normal for the bank angle, or "unloading" to zero G's, create an immediate, marked sideslip? I've found generally not-- fundamental cause of slip is not the "wrong amount of lift for the bank angle" as some books teach-- though undoubtedly changing the lift vector does have some small effect on the various aerodynamic torques that tend to yaw the nose this way or that... in a very extreme wing-over with near-vertical bank and near-zero G-load "over the top" there is some noticeable tendency for sideslip unless the pilot gives some bottom rudder--
* Breaking it down into very practical terms-- should we mix rudder to move with elevator deflection as well as ailerons, or ailerons only? Probably ailerons only. Some of the errors perpetuated in some of the pilot training material tend to suggest the opposite.

* The wingover case described above is a special unusual case where the aircraft is deliberately maneuvered in a way that makes the nose suddenly yaw down at the top of the wingover-- we have created a sudden demand for yaw rotation so this yaw torque must be provided be either rudder, or by sideslip blowing against the fin or equivalent surfaces -- in a more normal situation where say you are flying at constant bank angle and airspeed and then suddenly shove the stick forward to "unload" to zero G's, the forward stick movment creates little demand for increased yaw rotation-- in fact the turn rate slows, rather than increases-- so we don't expect to see sideslip here-- if anything, yaw rotational inertia should tend to cause a very slight, temporary skid not slip, as the aircraft tends to maintain the original, higher yaw rotation rate for a second or two even as the rate of curvature of the flight path decreases?-- (maybe not actually detectable, can't say I've ever seen this happen in actual flight, and I have looked)-- anyway the lack of a marked sideslip in this situation reinforces the case that slip is not caused, fundamentally, by the "wrong" amount of lift for the bank angle or the "wrong" turn rate for the bank angle... this is an area in which the some of the full-scale flying community could benefit from some re-education--

* Don't mean to preach to any particular person on all this, it's just a topic that interests me ... looking forward to your observational results... Steve
aeronaut999 is offline Find More Posts by aeronaut999
Last edited by aeronaut999; Mar 09, 2012 at 02:49 PM.
Reply With Quote
Old Mar 09, 2012, 05:11 PM
Taurus Flyer is offline
Find More Posts by Taurus Flyer
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
4,249 Posts
I hope it is not needed to read all your stories Steve, we have our own educational system in the Netherlands, This thread I only started to show and to give other people ideas, for me it is not a discussion.

Gents,

Arspeed I need to know for research and for the ASK I will design an airspeed transmitter especially for low values of speed.
The electronic output signal I also will use for the variometer for TE compensation and Netto (Vario) calculation.
Te tube already has a place on the nose of the plane.

The total drag will not be much more than the old ventury.

,
Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Reply With Quote
Old Mar 09, 2012, 05:12 PM
BMatthews is online now
Find More Posts by BMatthews
B for Bruce
BMatthews's Avatar
The 'Wack, BC, Canada
Joined Oct 2002
14,010 Posts
In a turn the CG of the model is passing through a circular path. Because the plane is built straight one end or the other of the fuselage and tail surfaces will always be flying with some amount of slip. On top of this the inner side of the wing is seeing a lowered airspeed compared to the outer.

So to compensate two things need to happen to achieve a truly coordinated turn.

First is that the wings have to have to compensate to some amount of aileron that equalizes the lift on the two sides despite the differing speeds. We can do this with a slight amount of aileron to both alter the camber and angle of attack over the span such that the lift is again balanced.

Second is that some rudder must be used to trim out the side slip. The tail of the plane in a turn is going to be seeing the air coming from an angle from the outside of the flight path in a turn. This is caused both by the distance of the tail behind the CG and it's location outside of the flight path as well as by the rotational rate change as the model turns while in the turn. To avoid the vertical tail trying to turn the model out of the circular flight path we need to "feather" the vertical surface to the local airflow such that there is no lift either inward or outward. We can do this by deflectung the rudder to the inside of the turn so that the combination of camber and angle of attack change due to adding the camber with the rudder produces a "zero lift" camber and angle so the tail isn't trying to push the model into our out of the turn any longer.
BMatthews is online now Find More Posts by BMatthews
Reply With Quote
Old Mar 09, 2012, 05:18 PM
aeronaut999 is offline
Find More Posts by aeronaut999
Registered User
The Willamette Valley, Oregon
Joined Dec 2008
1,388 Posts
See post #6 especially, just a suggestion. It's just a topic of special interest to me, comments were aimed at anyone interested, not trying to suggest you needed educating. Looking forward to seeing your results. Steve

PS did you get to cloud fly in sailplanes in the Netherlands? A friend of mine did quite a bit of that (decades ago) in the Netherlands before doing more for an NCAR research project in the US in wave/ incipient cunim. I wish it were legal for the average pilot to do that here in the US without the expense of the full 6-pack of IFR instruments which takes all the fun out it.... maybe when there are drones everywhere we'll all rely on transponders not see-and-be-seen and we'll be allowed to go fly in clouds without having to be in the IFR traffic control system...

Steve
aeronaut999 is offline Find More Posts by aeronaut999
Last edited by aeronaut999; Mar 09, 2012 at 05:27 PM.
Reply With Quote
Old Mar 09, 2012, 06:05 PM
aeronaut999 is offline
Find More Posts by aeronaut999
Registered User
The Willamette Valley, Oregon
Joined Dec 2008
1,388 Posts
Quote:
Originally Posted by BMatthews View Post
To avoid the vertical tail trying to turn the model out of the circular flight path we need to "feather" the vertical surface to the local airflow such that there is no lift either inward or outward. We can do this by deflectung the rudder to the inside of the turn so that the combination of camber and angle of attack change due to adding the camber with the rudder produces a "zero lift" camber and angle so the tail isn't trying to push the model into our out of the turn any longer.
I see your point but--

Technically, if a point near the CG is to be tangent to the curving flow, or even if the vertical fin is to be tangent to the curving flow, the rudder must be deflected enough to produce quite a bit of yawing-in torque. At least in the case of a long-spanned aircraft. Because the outside wingtip flies faster and experiences more drag than the inside wingtip, which produces a yawing-out torque that makes the plane tend to fly in a bit of slip. Maybe better to allow at least a few degrees of slip, even as measured way back at the vertical fin, so the fin gets "put to work" and generates some of the yawing-in torque and the rudder doesn't have to do at all. But this idea is only valid of the fuselage is slender and non-draggy in a sideways flow. Also the whole argument assumes that the ailerons aren't deflected so much that they create an adverse yaw torque (yawing-in torque) that overpowers the yawing-out torque from the difference in airspeed between the wingtips. This and other ideas are discussed in more detail in the links I posted above.

A plane shaped like this (link below) has little distance from nose to "tail", so there's relatively little tendency for the "tail" itself to "feel the curving flow" and yaw the nose to the outside, but the outboard wingtip still tends to experience more drag than the inside wingtip, so the nose still tends to yaw toward the outside. In a steady constant-banked turn, the yaw string is deflected some degrees toward the outside or high side of the turn. Adding a (larger) fixed vertical fin on the end of the "keel" or tail boom would likely reduce the slip angle, not increase it, in my opinion, because the higher drag of the outboard wingtip is the main thing creating the slip. But no matter how big the fin is it could never reduce the slip to zero because of the curving flow you described, for that we would need a rudder, deflected enough to overcome the increased drag from the outboard wingtip....

http://www.aeros.com.ua/structure/rw/index_en.php
aeronaut999 is offline Find More Posts by aeronaut999
Last edited by aeronaut999; Mar 09, 2012 at 06:42 PM.
Reply With Quote
Old Mar 09, 2012, 10:27 PM
BMatthews is online now
Find More Posts by BMatthews
B for Bruce
BMatthews's Avatar
The 'Wack, BC, Canada
Joined Oct 2002
14,010 Posts
The thread seemed to be about what a plane is like in a fully coordinated turn. Not about allowing some slip for some reason.

Yes there's going to be some difference in drag from inside to outside. But if the ailerons are used with a little "top" aileron to effectively reduce the camber and angle of attack on the outside then the reduction in angle and camber should produce a matching reduction in drag on this faster wing. And a similar but opposite effect would occur on the inner panel. Would this cancel out any shift in the center of drag? I'm not sure. It may on a case by case basis or it may not. If there were some off set residual drag than the fin would need to produce SOME side force in what ever direction was needed to counteract this. But first of all the rudder would still need to deflect to produce a feathered or zero lift angle to match the circular path and amount of offset due to the factors. If it needs to also be set slightly off that zero lift angle to deal with a slight drag couple in either direction then so be it. But that's a separate concern.
BMatthews is online now Find More Posts by BMatthews
Reply With Quote
Old Mar 10, 2012, 06:00 AM
Taurus Flyer is offline
Find More Posts by Taurus Flyer
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
4,249 Posts
Gents, read post 1 and 3 again.

"...there is no philosophy which is not founded upon knowledge of the phenomena,....................."

{Daniel Bernoulli}


and for that I need data collecting when the ASK is ready to fly.

An example of a mixer I prefer to use is shown in picture 1, the mixer of a Horten glider.
The main structure of such a mixer is hardware determined (analogue), 8 amplifiers are mounted on the backside of the board, the parameters are adjustable for fine tuning.
When I analyse the relationship of needed control inputs and measurable signals for this specific scale glider (ASK18) I can evaluate if it is possible to make a mixer as for the Horten and if it is worth the effort.

Interesting to know, the Horten IV did have mechanical mixers, so there is nothing new in my philosofie!!!!
A note is shown in picture 2 the output signals in degrees of deflection. "Flap" to read as control surfaces for control of pitch, roll and yaw.

After such a mixer is installed, adjusted and is working properly all measurments can be deleted. For example also the vane of the side slip measurement to reduce drag.

So data logging is needed two times of the project, to analyse and for tuning.

If results are not as expected? Not interesting to think or discuss about that and a waste of time too in a learning curve!


Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Last edited by Taurus Flyer; Mar 10, 2012 at 06:08 AM.
Reply With Quote
Old Mar 20, 2012, 06:44 PM
Taurus Flyer is offline
Find More Posts by Taurus Flyer
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
4,249 Posts
Gents,

A short note about my project. I wrote:

Daniel Bernoulli
"...there is no philosophy which is not founded upon knowledge of the phenomena,....................."

IDEA?
Never trust one airspeed measurement, always use a back-up method!


A picture of development of a hot wire airspeed sensor, also shown in the thread about XC flying.
With the "hot wire" method it is possible to measure ultra low values of airspeed, so for example to determine slip near the stall speed of the ASK18 or even below? I doubt if this is possible (with enough accuracy) with the differential pressure method, (the general used method), so part of the experiment is to use both, "Prandtl'schen Staurohr" AND "Hot Wire".
Electronic circuits are used to linearize the signal and compensation of ambient temperature. Special attention I have for power consumption of the system and calibration.

Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Reply With Quote
Old Mar 20, 2012, 07:23 PM
aeronaut999 is offline
Find More Posts by aeronaut999
Registered User
The Willamette Valley, Oregon
Joined Dec 2008
1,388 Posts
Good stuff....

Some more airspeed sensors (scroll down to "speed probes") http://www.brauniger.com/english/pro...cessories.html
Designed for low speed (paragliding)

It will be interesting to see if/ how you design the mixer to flip back and forth between the 2 required modes (rudder against aileron in steady circle with no bank angle change, but rudder with aileron whenever bank angle is changing) -- or just design for one or the other mode and do the rest by adding extra inputs w/ your thumb...

But as you say, data first... keep us posted thanx
aeronaut999 is offline Find More Posts by aeronaut999
Last edited by aeronaut999; Mar 20, 2012 at 07:51 PM.
Reply With Quote


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Question turning circles while doing rolls? AnbuPunk 3D Foamies 20 Aug 24, 2010 03:40 AM
Cool "Gliders readied to slip 'the surly bonds of Earth' " Sailhigh South Bay Soaring Society 3 Dec 06, 2008 12:57 PM
Discussion Glitch while turning - Lama V3 dbar2 Coaxial Helicopters 8 Dec 10, 2006 02:29 PM
Turning the EVO 9 on While on the Charger? aplphabandaid Radios 19 Nov 28, 2004 04:45 PM
Turning email notification completely off while keeping "my threads" jonferran Site Chat 1 Mar 22, 2004 05:43 AM