Q-talk 66 - GROUND DIRECTIONAL STABILITY PROBLEMS
- Category: Q-Talk Articles
- Published: Friday, 31 October 1997 06:11
- Written by David Gall
- Hits: 1862
The following is an excerpt of a report that David Gall has written. The original report is about eight pages long and too long to be printed in this newsletter. But the information that David discusses has proven to be very useful to us.
If anyone is interested in receiving the full length version of David's report, it can now be found online at:
I will state flatly that there's ample evidence in the pages of Quicktalk/Q-TALK (no single-data-point hyperbole here) to indicate that stock plans-built airplanes have ground directional stability problems; that the majority of fixes proposed over the years, although incorporated by many with often positive and repeatable results, have not directly addressed the problem; and that a few fixes have had legitimate success in correcting the problem, but have been largely overlooked and lost in the hoopla surrounding the promotion of certain other "factory" recommendations, specifically, the T-tail, reflexor and "ground angle of attack" campaigns.
The correct fix for the ground directional instability of these airplanes, I believe, is to give them a front-end alignment, just as you'd give your car a front-end alignment if it began heading for the weeds uncommanded. Some minor adjustments to the tailwheel can also contribute positively, which I'll cover, but the primary culprit is the main wheel alignment. If built according to the plans the alignment is plain, flat wrong.
Notice that I've said nothing about aerodynamics. No incidence change, reflexor setting, enlarged rudder or tacked-on T-tail will ever compensate for a wheel alignment problem. Frankly, I find it a bit perplexing that the designers of these planes would even think to look for aerodynamic fixes for such an obviously wheel-related problem. No one, to my knowledge, has ever complained about an overt lack of directional stability or control authority in the air; in fact, the Q-birds have been highly praised over the years by numerous writers for having good control harmony and response. If that changes when the airplane is in contact with the ground, why would anyone not go directly to the point of ground contact as the most likely culprit?
Granted, there's been much discussion over the years about wheel alignment on Q-birds. For the most part it has centered on toe-in vs. toe-out, with toe-out emerging as the apparent winner. However, there's more to wheel alignment than just toe.
On the subject of camber, from "Race Car Vehicle Dynamics" by Milliken and Milliken, published by SAE International, p. 46:
"In accordance with SAE terminology ... the camber is positive if the wheel leans outward at the top relative to the vehicle, or negative if it leans inward."
"In racing circles, tilt of a wheel is universally referred to as 'camber', with the sign conventions following SAE as above. The effect of camber on the tire forces and moments actually depends on the angle between the tire and a perpendicular to the ground - as opposed to the angle between the tire and the chassis reference..."
"In general, a cambered rolling pneumatic-tired wheel produces a lateral force in the direction of the tilt. When this force occurs at zero slip angle, it is referred to as 'camber thrust'."
"Camber angle to the road surface is one of the fundamental variables that determine tire performance ..."
"Camber also works like steer: When a tire is cambered it tends to pull the car in the same direction in which the top of the tire is leaning. A simple way to think about this is camber-steer force equivalence ... For bias-ply tires ... 1.0? of camber is equivalent to about 0.2? of steer (5:1). From this simple rule of thumb, it can be seen that static negative camber will require toe-out to keep the wheels from fighting each other."
"The amount of static toe on the front will depend on other suspension parameters such as ... ride and roll steer, compliance steer ..., and camber (both static and dynamic with ride and roll motion). Minimum static toe is desirable to reduce rolling resistance and unnecessary tire heating/wear that will be caused by the tires working against each other."
Wait a minute: what's this "ride and roll steer, compliance steer" stuff? Well that's the change in steering angle, camber and toe as a result of the geometry of the axle as the suspension moves through its range of travel. Of note to us is the fact that the camber and toe can be affected absent the steering links of a steerable axle. On the Q-birds the flexibility of the canard in both bending and twist conspire to aggravate the built-in inboard (negative) camber (of a plan-built plane) and to initiate inboard toe as the load on each wheel increases. What does all this mean? Consider for a moment ...
FORCE ANALYSIS - INSTABILITY
Our baseline airplane will be a stock, plans-built Q-2/Q-200. When this plane rolls down the runway, the built-in inboard camber and almost neutral toe allow the main gear tires to generate forces, each inboard toward the center of the airplane, that oppose one another. Another tire characteristic described in the literature is that of generated forces being generally proportional to the load on the tire, so the forces on the left and right gears balance.
Now, consider a crosswind gust from the left: the airplane is "heeled" over to the right slightly due to the new side forces on it. This increases the load on the right main tire and decreases the load on the left main tire. The changed loads allow the canard to flex differently, bending and twisting more on the right and relaxing on the left. The right main tire generates more inboard force due to the increased load on it, but more, the inboard camber and toe are increased due to canard flex, amplifying the effect. At the left canard tip the opposite conditions prevail, and the reduced inboard forces of the left main tire are further attenuated by the geometry moving toward a more 0-0 camber-toe setting. The additive resultant of the two front tires' forces is a strong force to the left, just as though the airplane had been equipped with steering and the driver had turned the wheel to the left.
But wait, there's more! This force to the left acts just like steering, so the plane starts to head for the weeds to the left. This is called a turn, and, as any turn, it generates centrifugal force. Since the turn is to the left, the centrifugal force acts to the right, but more importantly, it acts through the Center of Gravity (CG) of the airplane which is somewhere above the surface of the runway. The tires' resultant force acts at the runway surface to the left and the centrifugal force acts at the CG to the right; the resultant rolling couple tends to roll or "heel" the airplane to the right -- in this case, further to the right than the initiating crosswind had already heeled it. So the airplane's response to the initial disturbance is such as to amplify the initial disturbance; this is a textbook definition of instability.
CONTROL RESPONSE - REVERSAL
Hold on, now, we're almost done with this part, but first we've got to consider the pilot's role in all this. After all, it's not the machine but the man/machine entity, which must be stable to be useful. When the crosswind first hits, the plane veers to the left. What is the pilot's reaction? How about a good boot full of right rudder, that ought to do it. The Q-bird's tailwheel swings to the right -- wait! Which way does the Q-bird's >tail< go when this happens? To the left, of course. We can go out to the hangar and see this without even opening the canopy.
After the tailwheel is displaced to the right, the tailwheel, impotent though its reputation may be, does generate some force to the left ... acting at the surface of the runway ... below the CG ... creating a rolling couple to the right ... ARRGHH!! Our tiny tailwheel, in attempting to alleviate the veer to the left, has actually exacerbated the situation and further propelled us to certain doom! (If you don't see that, re-read the two preceding paragraphs.) Naturally, being pilots, we're going to apply even more right rudder with even more deleterious results. It's no wonder the tailwheel has a reputation for being ineffective and "skidding" just when we need it most.
"... Maurice Olley was the first to discover the critical speed beyond which some vehicles become divergently unstable."
"At the 'critical speed' the car becomes divergent, that is, a small steering input results in very large (theoretically infinite) responses in terms of path curvature, yawing velocity, lateral acceleration, or vehicle slip angle."
Hmmm. Sounds like a ground loop to me. The test goes on to describe situations requiring no steering input such as a crosswind or bump causing a small disturbance to the car's path that also result in divergence. In light of my explanation of a few paragraphs back, I think there is no need to further analyze the Q-bird landing gear for potential causes of divergence; there are ample reports in the pages of Quicktalk/Q-TALK which will attest to the speed-related nature of the airplanes' instability, even including the admonition to new pilots to practice taxiing slowly at first, then at ever-increasing speeds until they are comfortable. This amounts to little more than training the on-board computer to act as an active stability-augmentation system! Repetition makes the task secondary so that the flight test task can become primary.
The speed-related nature of some of the reported accidents may not at first be apparent to the casual peruser of back issues. Consider, though, that the speed in question is >not< airspeed, but ground speed. Suddenly, what may have been a mysterious occurrence on a calm day becomes an anomaly in the familiar home-drome airport normally has a ten-knot wind down the runway. On the day of the accident, even if the pilot made a liftoff or touchdown at lower than normal airspeed, that may have been a higher than normal ground speed nonetheless! Or, consider the first arrival of the experienced Q-bird pilot at an airport of an elevation higher than he's used to. Although all other conditions may be identical to those of familiar haunts, including the indicated airspeed at touchdown, nevertheless, the increased density altitude causes a higher than normal ground speed at touchdown which may put the airplane in its divergently unstable zone: wipeout!
Well, I've probably lost and/or angered somebody with these words. That is not my intent, but is, perhaps, inevitable. I've not edited or rewritten to convince or persuade, only to record. I've shared this with the intent of reporting my own, personal conclusions on this topic, and am open to reason, thoughtful rebuttal, but I'm pretty well convinced that I'm not exactly preaching to the choir, either. This is not an exhaustive analysis nor is it a flight test report so is subject to revision, but I believe that it is essentially grounded in the theories of ground vehicle dynamics as presented in the referenced text, which is more than I can say for any 'analysis' presented so far in advocacy of T-tails, reflexors or "ground angle-of-attack".
At gross weight you will want to see each axle sight 2" above and 2" forward of the opposing axle. This sighting will give you about a 1/2 degree of positive camber and toe out.
David J. Gall, Vero Beach, FL
David has asked that we acknowledge the contributions of Mike Dwyer, Bob Malechek and Dave Naumann in their reports to the QBA of their trials with the camber issue.
Since David wrote this article there has been a Q-2 and a two Q-200's modified using David's theory of positive camber. I changed my wheel alignment on my Q-200 after 10 hours of very interesting landing rollouts. Michel Royer had purchased a flying Q-2 and modified his alignment prior to his first flight. Al Kittleson of Dallas made changes to his camber after about 15 hours. Michel, Al and myself saw significant improvements in the way our planes handled after the change was made. These are only three cases, so there needs to be more information gathered to confirm David's theory. One of the best things about this modification is that it's fairly easy to do. In the next issue, I'll write up the procedure that I went thru to realign my wheels.
You can order a PDF or printed copy of Q-talk #66 by using the Q-talk Back Issue Order Page.