Aeronautics 101

Left turns at takeoff

by Alan Brown

This article was originally written some time ago to explain the phenomena of left turns at take-off, but has come to the fore again with an article on `The Anti-torque Device' which appeared in the October 2009 issue of Model Aviation. The device used is a weight mounted near the right wingtip of the airplane to create an out-of-balance situation to oppose the tendency of the airplane to rotate left. Not surprisingly, a number of letters were written to the editor, notably one from the distinguished aeromodeler, Frank Tiano, in the December issue of the magazine,  commenting on this approach. The intent of this article is to look at the phenomena involved, and the possible techniques to overcome them.  

Propellers which spin anti-clockwise when viewed from the front generate forces which frequently cause  aircraft to veer leftward while taking off. As there are at least three different phenomena at work during this operation, I thought it might be useful to categorize them, and to look at some of the techniques used to counteract them.

The first phenomenon is the torque generated by the engine. Newton's laws about action and reaction tell us that there will be equality between the turning moment generated by the propeller and that in the opposite rotational direction by the rest of the airplane. When the aircraft is sitting stationary on the ground, the torque can only be resisted by the main landing gear, and obviously the wider that it is spread, the easier it is for the aircraft to resist the turning moment. This torque does not vary substantially with aircraft speed; it is primarily a function of engine power setting. 

The next phenomenon is the rotational effect of the slipstream on the surfaces of the airplane. The predominant effect relates to the fuselage and the vertical surfaces. Slipstream effects on the wing and tail surfaces tend to be opposite to torque effects. However, the effect of slipstream on the vertical tail depends on the latter's vertical position. If it is above the thrust line of the engine, then a side force is generated which will turn the aircraft to the left, and if it is below the thrust line, then it will try to turn the aircraft to the right. The latter is helpful, the former is not. Remember that yawing to the left generally induces rolling to the left, which is in the same direction as the torque forces. We don't usually put the fin below the thrust line because that makes it difficult to rotate the airplane at take-off, so the top mounted fin is usually bad news. Those of you who have built free flight power competition aircraft know that we often mount the wing very close to the engine on a high pylon. This pylon gives a side force pretty much on the c.g. of the airplane, so doesn't induce much yaw, but it does produce a fairly healthy side force (being very close to the propeller) which results in a strong rolling moment opposed to the moment induced by the engine torque. Careful selection of pylon area allows us to trim the airplane to climb in a spiral either with or against the torque. Note that the spiraling effect of slipstream diminishes significantly with distance back from the propeller, so the effects of a pylon close to the propeller are much more significant than those of a vertical fin and rudder some distance downstream. So the effect of slipstream on a fin and rudder combination is probably the least significant of the three phenomena of interest.

The third phenomenon is a bit more esoteric, and relates to tail draggers versus tricycle geared aircraft. It also, however, relates to either type of aircraft as it rotates to get lift as it takes off. This probably needs a picture, so I hope I can sketch something that's intelligible.

This supposedly shows an engine thrust line moving at an angle to the flight path. As shown, this would be typical of a taildragger running along the ground, but note that it also applies to an aircraft flying at a high angle of attack, which generally means slowly, or to an airplane which is rotating as it takes off.

The nearer blade to us is going upward if we are looking at the left side of the aircraft and the further blade is going downward. If the airplane's flight path and thrust line are at say 10 degrees to each other, then the nearer blade will be at 20 degrees less incidence than the further blade and so will generate a lot less thrust. The net thrust from the propeller will thus appear to come from a point which might be an inch or so to the right of the actual thrust line. The steeper the angle between the thrust line and the flight direction, the greater will be the effect. Let's see how this translates to a typical World War I fighter. Engines rotated relatively slowly and so propellers were large, which meant long landing gears and steep thrust line angles. A large propeller will also have a larger thrust offset in addition to the larger angle between ground and thrust line. No surprise that these airplanes were notoriously bad ground handlers.

Let's follow one of these aircraft through takeoff. First, we're standing at the end of the runway. There's no forward speed so air gets pulled through the propeller parallel to the thrust line. No effect of thrust line angle yet or for a little while until forward speed builds up. The torque on the airplane is just a function of engine rpm and there is almost no resistance from the flying surfaces (wings, tail, fin) as yet. The main reaction is from the landing gear. The side forces from the slipstream are again proportional to engine thrust and have very little to do with forward speed. So at the beginning of takeoff, slipstream side forces and landing gear torque reaction dominate.

As we accelerate, the effect of the thrust line angle starts to take over and the airplane begins to see more offset thrust, which means more left turning moment. At this point there still isn't very much aerodynamic force available from the flying surfaces as we are below stall speed.

Finally, we get the tail off the ground and the thrust line is now parallel to the direction of flight. The inclined thrust line effect goes away and the only forces tending to turn the airplane left are torque and slipstream. By this time however we've got quite a bit more aerodynamic forces from the flying surfaces (remember force is proportional to speed squared) to counteract these two effects. We've got it made - right?

Not quite! We finally have to apply up elevator to get enough lift to get off the ground - remember our World War I heroes didn't have the benefit of the huge thrust/weight ratios that we have today in the model world - and suddenly we're back into the large offset thrust moment, probably at full power, caused by the misalignment of the thrust line and the flight direction. However, that doesn't last long as the airplane quickly starts to pull up at a fairly moderate angle of attack as its speed builds up. Now there's another however! If we start to rotate at too low a speed and the aircraft is near or past the stall angle, it can hang in this position, not get off the ground, and veer to the left. Sounds familiar? The immediate solution (theoretically!) is to either put in down elevator, reduce incidence and increase speed before rotation, or cut power and start again while applying massive right rudder. All too often the solution is (1) panic, (2) apply lots of power, (3) stall airplane while pivoting on left wing tip, (4) re-kit airplane.

Tail-dragging pattern aircraft generally don't have these problems for two major reasons. First, their attitude to the ground is much flatter - propellers are relatively small and landing gears are short - and their thrust/weight ratios are high, so they generally fly right off the ground without much rotation.

Several trim techniques have been used to offset these effects, the most common being right side-thrust (for a propeller rotating anti-clockwise when viewed from the front). This is particularly common on World War I type models, which tend to sit fairly high with a steep thrust line inclination. As the WW I airplane engines were very heavy with respect to the rest of the airplane, they tended to be very close-coupled, which exaggerates further any side-thrust angle requirements. 

Even World War II airplanes needed something of the sort, as they got substantially higher thrust/weight ratios, but still had the tail dragger configuration. The Blackburn Firebrand went from prototype to production via doubling the vertical fin area and installing it with substantial side angular trim.

You can see it even on this small scale drawing. Note that if you want to make an accurate scale model of the Firebrand, the fin offset is consistent with British engines, which rotate in the opposite direction to American engines and to our model engines!

Also this technique works better for a naval aircraft than for a normal land-based aircraft, because both take-off and landing approach are made at high power settings to allow for go-around from an aircraft carrier.

The Macchi MC.200, probably the best Italian WW II fighter, had the left wing a little bit longer span than the right wing, as shown below.


If you look carefully, you'll see that the left (port) aileron has one more rib than the right one.

Several aircraft have had increased incidence on the normally down going wing. This seems to me to be a particularly bad approach, because it's inviting the higher angle of incidence wing to stall before its partner, and thus spin the airplane to the left.

The wing asymmetries are less effective than side thrust, because they rely on the airplane getting up to sufficient speed to make the flying surfaces effective. The offset fin is somewhere between the two, because it does counter the effect of propwash to some extent.

Note also that inputting lots of downthrust negates to some extent the difference between thrust line alignment and flight path at high angles of attack. It is very common to see both downthrust and sidethrust recommended to aid in counteracting the left turn tendencies at take-off. However, one should be careful with downthrust. Its primary purpose should be to maintain unchanged trim with power setting.


So set the downthrust to achieve this desirable feature, and then add sidethrust as necessary. The disadvantage of any of these lateral imbalance techniques is that they are still present at higher speeds, when they are much less necessary, and they tend to work in the wrong direction when the aircraft is inverted.

Let's now try to summarize what approaches we prefer to take care of the `left turns at take-off' phenomena. The first item to note, and Frank Tiano infers this point in his letter to the editor, is that different approaches are appropriate for different kinds of model airplanes -- and different skill levels of pilots.

If we are flying aerobatic model aircraft, either scale models (IMAC, for example) or pattern competition aircraft, then it is of the utmost importance that the models be set up to maximize their aerobatic performance. This means almost always precise location of the c.g both fore and aft, and laterally. Almost every aerobatic flyer will tell you that he wants completely symmetrical lateral balance. Also, he or she will tell you that both downthrust and sidethrust will be determined by how the airplane flies in maneuvers, and not by take-off and landing considerations.

So our aerobatic pilot will go with the downthrust and sidethrust determined for good aerobatic flight, which fortunately is generally in the same direction as required for take-off, and have a completely balanced airplane laterally both in terms of c.g. and aerodynamic surfaces. The take-off and landing will require control surface manipulation by the pilot, who is usually fairly competent.

A trainer generally will have a tricycle landing gear with a fairly wide track to mitigate the various effects mentioned earlier, and may also have built in sidethrust and downthrust to make take-offs even easier for the novice pilot.

The area of greatest difficulty for the model airplane pilot is, as mentioned above, WW I and WW II airplanes. My own view is that the best solutions for these airplanes, particularly when they are both tail-draggers and have narrow track landing gear, is with a combination of downthrust and sidethrust. The downthrust should, as mentioned earlier, be determined by the amount required to give minimum trim changes between high and low power settings in normal flight, and the sidethrust should be enough to make the pilot feel comfortable on the take-off run.

Note that excess sidethrust may make the airplane uncomfortable to fly in normal flight, and certainly will affect inverted flight characteristics. However, the nice thing about using sidethrust and downthrust to help you in take-off and landing is that they are naturally reduced as the power is reduced, and so have no bad asymmetric effects, particularly in idle throttle or dead-stick landings. On the other hand, any technique which makes the airplane intrinsically asymmetric either by weight transfer or aerodynamically will give rise to problems in low power situations.

However, if you were a Sopwith Camel pilot, you might want to trade some of the inconveniences mentioned above for being able to get off the ground at all!