Logo
Builders' Plan Gallery  |  Hip Pocket Web Site  |  Contact Forum Admin  |  Contact Global Moderator
February 23, 2018, 03:07:56 PM *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with email, password and session length
 
Home Help Search Login Register
Pages: [1]   Go Down
Print
Author Topic: CAUSE OF "HUNTING" IN FLIGHT FOR CL MODELS  (Read 538 times)
0 Members and 1 Guest are viewing this topic.
OZPAF
Titanium Member
*******

Kudos: 47
Offline Offline

Australia Australia

Posts: 3,726

Topic starter


Ignore
« on: January 18, 2018, 11:42:46 PM »

"Hunting" ie a consistent undulating flight path, instead of a rock steady flight, is quite noticeable with some control line models.

I have seen references - particularly with stunt models - to allow a little slack in the elevator horn pivot, and this implies that the stability may be too high and/or that the handle/bellcrank(bellcrank too small for the handle) ratio is too high?

I experienced it on a small semi scale Sopwith Camel which used a small(approx. 2") bellcrank with a 5"' approx. spacing at the handle. I reduced the hunting by increasing the elevator horn size and reducing the arm of the elevator drive on the bellcrank. It was still there even with the much reduced throw.

I tend to feel that I should have reduced the handle spacing to much closer match the bellcrank size.

I'm sure this has been handled somewhere but I haven't been able to find a decent explanation, anmd my curiousity will not let it rest.

Thanks

John



Logged
perttime
Gold Member
*****

Kudos: 11
Offline Offline

Finland Finland

Posts: 306



Ignore
« Reply #1 on: January 19, 2018, 09:49:11 AM »

I'm no expert but some causes are mentioned on stunthanger:

"A little slop in the elevator typically will not cause an airplane to hunt.  Slop between the flaps and bellcrank definitely will."

"Another thing that will cause hunting is sticky controls.  The controls at the leadouts should be very free to the extent that they will move under their own weight!"

"Hunting is usually the result of the flaps and elevator not being aligned straight with each other.  A bit of slop in the controls can help solve this to a degree,  allowing the control surfaces to self- align with each other in the air if they weren't off much.  Misalignment  of the wing centerline with the stab and thrust line can also cause this. "

"I had my first Crosswind doing that, tried all the things listed here but while at the nats, Crist Rigotti told me to put a tad of down thrust in the engine.
I cut a couple of tabs off a .010 sheet I had, put them under the forward mounting screws and presto -no more hunt."

"... Flite Streak. I used a 3/32" piano wire pushrod (old school) and it was flexing when up elevator was applied. So I installed a pushrod guide and that stopped the flexing. However the hole for the pushrod guide was to "tight" and friction created a hunting problem. I had to enlarge the hole just a little bit and the majority of the hunting problem went away.

But I still had a small tendency to hunt. I had used piano wire leadouts with 3/32" brass tubing in the adjustable leadout guide. Somewhere on this forum I read that this could cause a hunting problem and apply some oil down the solid piano wire leadouts. Sure enough it took care of the hunting issue."
Logged
OZPAF
Titanium Member
*******

Kudos: 47
Offline Offline

Australia Australia

Posts: 3,726

Topic starter


Ignore
« Reply #2 on: January 19, 2018, 06:24:39 PM »

Thanks PT. The model I was referring to didn't have flaps and the controls were all free moving with a braced elevator push rod.
The mystery remains.
Surprisingly and fortunately it doesn't seem to be an issue with TR and speed(?) models. The handles in TR have a very low line spacing to match the small bellcranks(?) perhaps as well as to avoid over controlling.
The higher centrifugal force would also make friction in the control system less of an issue as well.
Well I'll put the issue in amongst the list of ponderables.
Thanks again.
John
Logged
greggles47
Gold Member
*****

Kudos: 11
Offline Offline

Australia Australia

Posts: 274




Ignore
« Reply #3 on: January 20, 2018, 12:52:13 AM »

Speed and tr models are also susceptible to hunting. I believe it to be an incompatibility between cg and leadout placement.
Of course I could be wrong. :-)
Logged

There must be some way out of here, said the joker to the thief,
There's too much confusion, I can't get no relief
mike
Bronze Member
***

Kudos: 4
Offline Offline

United Kingdom United Kingdom

Posts: 86




Ignore
« Reply #4 on: January 20, 2018, 03:24:46 AM »

As a non CL flyer and a non Control-Engineer, here's my thinking on a possible cause is this.
(I am assuming it's not a pilot induced oscillation (PIO) - subconscious correcting inputs applied out-of-phase.)

Imagine a rigid arm held horizontal with the handle neutral.  If the model rises above this plane, the angle of the lines to the handle will apply down elevator.  The model descends through the horizontal plane and up is applied.  You have a potential oscillation if the motion isn't damped enough or the returning force isn't reduced to match the existing damping.
The motion sounds a bit like a Phugoid but that is a relatively low frequency normally on full-size aircraft - period of a few seconds. (Google it for an explanation)

Moving the lines closer together on the handle will reduce the gearing between handle movement and elevator, reducing the returning force.  That might fix it but it will reduce control movement - might be a problem if you need the sensitivity.  (If you make the bell crank width narrower in proportion too, the gearing is the same as before)

Or hold the handle on its side to remove the effect?  I'd try that first - easiest way to see if you're on the right track.
Logged
OZPAF
Titanium Member
*******

Kudos: 47
Offline Offline

Australia Australia

Posts: 3,726

Topic starter


Ignore
« Reply #5 on: January 21, 2018, 03:29:48 AM »

Thanks Greggles and Mike. That's interesting re the problem still occurring in TR and speed Greggles. I was sure that the TR models I'd seen were rock steady - however the pilot may know otherwise. At the speed of TR and racing, it would be hard to pick up the undulations outside the circle unless they were substantial

Yes I did have a bit of success by reducing the overall gearing, however control as you mention Mike was then the problem. It was still flyable but lacking in any sort of response.

It seems like only a very far forward CG - say 20% or less may be the only way to achieve the necessary damping. The CG would be well in front of the AC of the wing and thus it would seem that as for example with the model climbing, the moment of the wing lift would tend to oppose the increase in AOA due to the model rising above wrist level.

Together with a decent tail volume this may be the answer. This is what you were referring to Greg I think.

Thanks - I could be off the rails here but your comments have sparked this idea and very forward Cg's(to me as a RC glider type as well and a bit of FF) are common in CL. Too much auto pilot can be a bad thing.

No wonder Stunters have such large tail areas!

John
Logged
Big G
Bronze Member
***

Kudos: 1
Offline Offline

United Kingdom United Kingdom

Posts: 83



Ignore
« Reply #6 on: January 21, 2018, 06:57:06 AM »

I agree 100% with Greggles on this. In the early 1980's I 'volunteered' to fly a piped OPS 29 Class B team racer, which hunted. The cg was well forward. It was a very unpleasant model to fly, apart from the huge rotational speed, and after flying a final at the British Nationals I knew I'd been in a race. I didn't build the B racer, but I have a suspicion that the cg was too far forward because the nose had been lengthened to balance the weight of the tuned pipe, most of which is well behind the cg. Strangely, I never had this problem with any speed model I built (not that many) whether piped or not. Could be that the wing-incidence on the B racer wasn't what it should have been, as well.

Nowadays I just sports-fly stunters, but always make sure that the cg isn't much further forward than the plan, or if not building from a plan my 'eye-balling' of the model. Correct CG and lead-out position needs to be correct, which is why most of my stunt models have adjustable lead-outs. Only test-flying, though, will point the way.

G
Logged
OZPAF
Titanium Member
*******

Kudos: 47
Offline Offline

Australia Australia

Posts: 3,726

Topic starter


Ignore
« Reply #7 on: January 21, 2018, 07:38:07 PM »

Thanks BG. I agree the CG can be too far forward and cause problems as well. As a RC Glider flyer I'm used to the CG being well behind the AC (25%) of the MAC - usually 35-37% plus.

The main reason for this with RC gliders is to minimise tail trim loads - but RC models don't have a built in auto pilot system as do CL models.

All the CL plans I have seen show the CG in front of the AC - which at last now might make sense to me. it's all relative though - you  wouldn't want to have the CG too far forward as this places a heavy continuous trim load on the tail. I would expect as I mentioned before that 20% of the MAC - 5% in front of the AC may be a reasonable starting point.

Placing the lead outs in the right place would then be another issue - matching the Cg and model weight, speed and line length.

Thanks all.

John
Logged
LOUCRANE
Bronze Member
***

Kudos: 5
Offline Offline

United States United States

Posts: 42



Ignore
« Reply #8 on: January 25, 2018, 06:35:36 PM »

The two main reasons for "hunting," IMHO, are PIO and sticky controls.

"PIO" Pilot Induced Oscillations. More later...

Incidentally, years ago we noticed that models set up with no slop in elevator linkage, which DID 'hunt' tended to 'hunt' less after a bit of flying time. In that time, vibration grew the elevator horn hole to excess size and hunting reduced. Why? Apparently slop allowed slight pushrod movement before the space was taken up and the elevators began to move. The flaps, without slop, moved a small angle first, in other words. That provided a slight increase in lift before the elevators caused a pitch change. Symmetrical airfoils zero lift line is at their chord line, meaning NO lift when in that attitude. The slight addition of effective wing Angle of Attack helped maintain lift before elevator pitch change. It was an undesirable way to get level flight, but it did work. The problem? The slop kept growing, and that was NOT desirable.

CLPA (Stunt) fliers lately, the past 20 years and more among the best of them, stress smooth, free and snug linkages inside the model, and a perfect, instinctive 'neutral' handle setting, i.e., NO minor inaccuracies in where your hand instinctively relaxes to 'neutral' input. "Hunting" seems to appear almost invariably when a flier tries to maintain level flight. Level flight should NOT need an unnatural handle 'hold.' When your relaxed hand 'misses' that smooth effortless setting, minor corrections can be made too large for a steady level path. That requires another corrective control input, usually too quick, and often, again, too large. PIO. PIO can increase and has destroyed a few prototypes of new full-size aircraft.

Free controls? Construction articles for stunters often stress that the control surfaces should be free enough to fall from their own weight. For a flapped model, weight of the flaps and of the elevators may cancel out in large part... even completely. It is easy enough to confirm complete free motion other ways. (If they nearly cancel, and the leadouts are cable, can the surfaces be moved easily by pushing the cables?)

"Free" also means NO binding against any element between the leadouts and the control horns. Do the leadouts impinge on ribs, formers, or any other structure? The pushrods on anything? Cable saws, designed to saw, are very effective, and leadout cables can try to saw through balsa, foam or plywood it bears on. If cable cuts into any structural piece, that piece can act as a clamp to a degree, making control motion less consistently smooth and dependable.

Snug, for want of a better term, means NO slop at all! The internal linkages should be stiff enough to not 'bow' between bellcrank attachment points and control surface horns. E.g., stiff materials like arrow shafts or small diameter carbon fiber tubes. Or guides along their lengths to snub any bowing tendency for less rigid materials under compression loads. Also, under vibration, slop between of pushrods and 'horns' can hammer out greater space where there should be none. With bushed pushrods or ball-bearing connections nowadays, slop growth is just about prevented. We know to set elevators to a slight hint of 'down' elevator at neutral and thrust line pitch adjustments to do away with many problems.

Back to PIO. I wonder how many CL fliers ever think about how they turn to face the model around a lap. That motion should be as steady as a great ballerina's torso while she dances across the stage on toe tips. If we lump around, awkwardly, nearly stumbling at times, we transmit unpredictable motions to the handle. Art Adamisin, I believe, suggested that new CL fliers try the toe-behind-heel stepping often seen used by racing fliers. It is deliberate, a quickly instinctive thing, and works in the other direction while flying inverted. It is smooth, because it is a practiced move.

Sure, CG and leadout placement can have an effect, and some of it is binding of leadouts against their guides.

Just a thought or two... It might be too late to correct a "hunting" model. Longer control horn arms and narrower space between lines at the handle can reduce response rate to handle inputs. Such 'gearing' can make a model feel unresponsive. Learn to love that; you can usually turn your handle further to compensate.   

Hunting is not helpful in any CL event. Steering up and down unnecessarily is unwelcome in traffic, AND slows a racer. Control inputs to adjust the altitude of speed models likewise. Combat models must be steady enough to the hand that watching your model  is less crucial - you need to find and cut the other guy. PIO on a scale model has claimed too many.  Thousands of hours of  intensive detailed benchwork, but too nervous to fly it before a meet? It has happened too often... Stunt, CLPA, is judged for the smoothness and precision of the level, round tracks and how extreme corners are executed.

PIO and "hunting" are NOT the CL flier's friend...
Logged

/LOU
OZPAF
Titanium Member
*******

Kudos: 47
Offline Offline

Australia Australia

Posts: 3,726

Topic starter


Ignore
« Reply #9 on: January 26, 2018, 03:19:27 AM »

Thanks Lou - particularly for the extra info on PIO and handle neutral position. I agree that PIO and friction would have an impact but still feel that the model will need to be damped aerodynamically and this could be possible with the typical lift force balance with CL models, where the CG is well in front of the wing AC.
However all the other areas you mention should also be allowed for.
I'm sure Bill Netzeband would have covered this in his articles in the 60's - 70's.
John

Logged
Heikki K
Copper Member
**

Kudos: 0
Offline Offline

Finland Finland

Posts: 4



Ignore
« Reply #10 on: February 16, 2018, 03:25:36 PM »

Some help to PIO and determining if there is a hunting effect made by insufficient distance between lines on handle would be holding your handle not upright, but palm down or half-way down. When the wind is hard, but steady, this is the way I keep the level flight flying against the wind, horsing backwards, with your knuckles up gives a smooth "down" control, and, vice versa, horsing forward flying downwind.
Logged
OZPAF
Titanium Member
*******

Kudos: 47
Offline Offline

Australia Australia

Posts: 3,726

Topic starter


Ignore
« Reply #11 on: February 16, 2018, 06:23:44 PM »

Interesting - thanks Heikki.
John
Logged
faif2d
Platinum Member
******

Kudos: 27
Offline Offline

United States United States

Posts: 1,093


Sun came up I was here to see it = good day



Ignore
« Reply #12 on: February 16, 2018, 09:19:27 PM »

A couple of experiences that I have had.  The first is a 1/2a combat wing with a VERY aft cg.  It would hunt up and down in about 10 ft swings.  I just stood there and let the handle shorten and lengthen the up and down lines to self correct the flight path.  It was very uncomfortable until I decided it was not going to crash and then it was just boring until the full bladder ran out.  There was no controlling the amplitude of the swings.  The other is from 1962.  A friend and I attended the navy nats in Chicago.  There was a man at the stunt circles that had a twin fuselage flight streak that I was quite taken with.  I purchased 2 kits and we built the plane in the motel room at night.  I used two K&B .35 as that was the only motor that I had 2 of.  It was very nose heavy and the line pull was very strong.  I went to .018 x 70 ft lines and it was still very heavy but the big thing was that it was very insensitive to control input.  It would stunt but you had to give full up and down to get it to maneuver.  Wish I knew then what I know know.
Attached files Thumbnail(s):
Re: CAUSE OF "HUNTING" IN FLIGHT FOR CL MODELS
Logged

I used to like painting with dope but now I can't remember why!    Steve Fauble
Pages: [1]   Go Up
Print
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!