winch manual

8/6/2019 Winch Manual

1/68

THE GLIDING FEDERATION OF AUSTRALIA

WINCH LAUNCHING MANUALINCORPORATING AUTO-TOWING

GFA Winch Launching Manual Issue 1, 1998


2/68


WINCH LAUNCHING MANUAL

Issue 1, 1998

Published by:

The Gliding Federation of AustraliaBuilding 130, Wirraway Road,

Essendon Airport, Victoria 3014,Australia

ABN 99 008 560 263

Phone: (03) 0379 7411Fax: (03) 9379 5519

Email: [email protected]

Copyright Gliding Federation of Australia, 2007
mailto:[email protected]:[email protected]:[email protected]


3/68

iTable of Contents

1. INTRODUCTION ................................................................................................. 12. BASIC WINCH LAUNCH PRINCIPLES .............................................................. 1

2.1 The stages of a winch launch ....................................................................... 1

3. GLIDER CONSIDERATIONS ............................................................................. 33.1 Hook positions and launching characteristics .............................................. 3

3.1.1. The shoulder release hook .................................................................... 33.1.2. The belly release hook ........................................................................... 43.1.3. Unwanted pitching moments .................................................................... 43.1.4. Porpoising on the launch ....................................................................... 53.1.5. Other unwanted moments ........................................................................ 63.1.6. Automatic back-release ............................................................................ 63.1.7. Predicting likely launch characteristics ..................................................... 73.1.8. Useful tips ................................................................................................ 73.1.9. Final reminders ..................................................................................... 8

3.2. Winch-launching loads and launch-speed boundaries ................................. 83.2.1. Winch-launching loads .......................................................................... 83.2.2. Wing bending relief ............................................................................... 93.2.3. The relevance of bending relief to winch-launching .............................. 93.2.4. The purpose of the weak-link .............................................................. 103.2.5. Launch-speed boundaries .................................................................. 11

4. PILOT CONSIDERATIONS .............................................................................. 134.1. Correct launch technique ........................................................................... 13

4.1.1. Ground-run and separation ................................................................. 134.1.2. Initial climb .......................................................................................... 144.1.3. The maximum speed placard and the initial climb .............................. 154.1.4. Full climb ............................................................................................. 164.1.5. The maximum speed placard and the full climb .................................. 174.1.6. The effect of the weight of the wire ..................................................... 174.1.7. The release ......................................................................................... 174.1.8. Winch-launching in crosswinds ........................................................... 174.1.9. Launch speed signals ......................................................................... 184.1.10. Kiting ................................................................................................... 18

4.2. Incorrect launch techniques ....................................................................... 184.2.1. Holding the glider down ...................................................................... 184.2.2. Allowing the glider to self-rotate ....................................................... 19

4.2.3. The Kavalierstart .............................................................................. 204.3. Launch failures - Philosophical and training considerations ....................... 204.3.1. Introduction ......................................................................................... 204.3.2. The mental approach to launch failures .............................................. 20

4.4. The physical characteristics of launch-failures ........................................... 224.4.1. Cable breaks ....................................................................................... 224.4.2. Land ahead or circuit? ........................................................................ 254.4.3 Inertial factors ..................................................................................... 264.4.4. Preference for landing ahead .............................................................. 274.4.5. Briefing versus training ....................................................................... 274.4.6. The non-manoeuvring area ................................................................. 27

4.4.7. Engine failures .................................................................................... 285. WINCH DESIGN CONSIDERATIONS .............................................................. 28



4/68

ii5.1 General ...................................................................................................... 285.2. How much power? ..................................................................................... 29

5.2.1. Low-powered winches ........................................................................ 295.2.2. High-powered winches........................................................................ 29

5.3. Car retrieve or self-lay? .............................................................................. 30

5.4. Protection of personnel .............................................................................. 316. AIRFIELD REQUIREMENTS FOR WINCH-LAUNCHING ................................ 32

6.1 Strip length ................................................................................................. 326.2 Strip surroundings ...................................................................................... 326.3. Strip layout and cable safety ...................................................................... 336.4. Diagram - Recommended strip layout and safety precautions ................. 34

7. CABLES AND ANCILLARY EQUIPMENT ........................................................ 347.1. Cable types ................................................................................................ 34

7.1.1. Solid wire ............................................................................................ 347.1.2. Stranded cable (wire rope) .................................................................. 367.1.3. Polypropylene rope ............................................................................. 36

7.2. The cable end ............................................................................................ 377.2.1. Trace................................................................................................... 377.2.2. Drogue parachute ............................................................................... 387.2.3. Weak link ............................................................................................ 387.2.4. Types of weak-link .............................................................................. 397.2.5. Rings................................................................................................... 41

8. SIGNALLING SYSTEMS .................................................................................. 428.1. Visual signalling systems ........................................................................... 43

8.1.1. Lamp signals ....................................................................................... 438.1.2. Bat signals .......................................................................................... 438.1.3. Wing-waggling .................................................................................... 44

8.2. Aural signalling systems ............................................................................ 448.2.1. Field telephone ................................................................................... 448.2.2. Radio .................................................................................................. 458.2.3. Particular requirement when using radio for launch signals ................ 458.2.4. Special precaution for operators of all winch-launch signalling systems

459. WINCH-DRIVER TRAINING ............................................................................. 46

9.1. Selection of suitable winch-drivers ............................................................. 469.1.1. Mechanical aptitude ............................................................................ 469.1.2. Solo pilot or not? ................................................................................. 46

9.2 The training syllabus - normal procedures ................................................. 479.2.1. Preparing for launching ....................................................................... 479.2.2. Safety precautions and emergency equipment ................................... 509.2.3. Laying the cables ................................................................................ 519.2.4. Launching ........................................................................................... 51

9.3. Abnormal procedures ................................................................................. 559.3.1. Cable breaks ....................................................................................... 559.3.2. Engine failures .................................................................................... 569.3.3. Excessive drift on the launch .............................................................. 57

10. AUTO-TOWING ................................................................................................ 5810.1 Pilot technique - table of differences .......................................................... 58

10.2. Driver technique - table of differences .................................................... 5910.3 Auto-tow vehicle requirements ................................................................... 59



5/68


iii

10.4 Auto-tow variations .................................................................................... 6010.4.1. Pulley launching, 1st variation ............................................................ 6110.4.2. Pulley launching, 2nd variation ........................................................... 61

10.5 Auto-tow safety .......................................................................................... 6210.5.1. Pilot safety .......................................................................................... 62

10.5.2. Driver and ground-crew safety ............................................................ 6210.5.3. Overall safety ...................................................................................... 62


6/68

1

1. INTRODUCTION

The aerotowing of gliders is only permitted by pilots whose competence forthe task has been assessed and who hold a Glider Towing Permit. The

Permit is subject to certain recency requirements.

A Glider Towing permit is issued by a Delegate of the Civil Aviation SafetyAuthority (CASA) under Civil Aviation Regulation (CAR) 149 only when thedelegate has assessed the pilot as competent in glider towing operations.

2. BASIC WINCH LAUNCH PRINCIPLES

The basic principle of a winch launch is simple. The glider is attached to acable which is wound back into the winch at such a speed that it provides theglider with flying speed. Most winches are fitted with powerful V8 petrol

engines and automatic transmissions. Although popular in Europe, dieselwinches are rare in Australia, as are manual transmissions in thisapplication.

As the winch accelerates the glider toward its safe launching speed, theglider is flown in such a way that it follows a gradually steepening flight path,gaining height rapidly until it is almost overhead the winch, whereupon thecable is released and the glider goes on its way.

This is simplified description, but it will suffice for a starting point. For thoseto whom a picture is worth a hundred words, the following diagram may help.

Minimum strip length (GFA requirement) 1200 metres.

2.1 The stages of a winch launch

The guideline given in the diagram, that a glider should achieve about one-third to one-half of the strip length as its launch height, is notional. The exactheight will vary with pilot and winch driver technique, as well as with windvelocity and aircraft characteristics.



7/68

2Referring to the winch launch diagram, it will be noted that the flight pathfollowed by the glider in the early stages is progressively steepened asheight is gained. This should be a smoothly-executed process, with theaccent on the word progressive. There should be no steps in the processof transitioning from the separation into the full climb and no sudden

changes of climb-angle at any time.

This graduated early stage of a winch launch is a crucial point and a grossdeviation from a graduated profile, either in the form of climbing too steeplytoo close to the ground or making a sudden change to climb-angle, is thebiggest single cause of winch launch accidents all over the world.

That said, the profile does not have to be followed with millimetre accuracy.There is a built-in tolerance to enable pilots to make the minor errors whichare inevitable in learning a new skill, without them being at risk. Providedthat the recommended climb profile is followed in principle, and the glider is

never allowed to climb too steeply at a height from which recovery from afailure cannot be made, there is no reason why winch launching shoulddemonstrate a higher accident rate than any other kind of launching.

Sometimes a launch is criticised for going too EARLY into a steep climb,when what was really meant was that it was too LOW into the climb. If alaunch accelerates rapidly, the glider will attain climb speed rapidly and willtransition through the early stages quickly. In this case the glider mayindeed be early into the climb, but because it achieves a safe height quickly,it is quite safe provided the progressive principle is followed. If it is allowedto enter the full climb at too low a height, the progressive principle is

violated and the process tends to become unsafe.

To put a figure on it, a glider should not be established in the full climb below200 feet, but should still be in the progressively changing process below thatheight.

Apart from this early stage, which must be managed properly in order toensure safety, there is no other significant risk attached to launching bywinch.

Safe winch-launching results from successful collaboration between the pilot

and the winch-driver, the methods outlined in this manual representing goodpractice leading to such collaboration and thus to safe and successfullaunching.

Winch and autotow launching have been used for the launching of gliders inAustralia since the late 1920s. The use, practices and technology haveevolved by experience in those early years to become standardised. Thismanual reflects the standardisation and expertise developed over somethinglike 5,000,000 winch launches in Australia over a period of 60 years



8/68

3

3. GLIDER CONSIDERATIONS

3.1 Hook positions and launching characteristics

When winch-launching a glider, an ideal situation is that the gliders attitude

should be able to be controlled accurately by the pilot without any unwantedpitching moments from the glider, or from a combination of the glider and thecable pull.

3.1.1. The shoulder release hook

Some years ago, designers sought to place the release hook as close aspossible to the gliders centre of gravity (CG), in order to avoid unwantedpitching moments. The CG is somewhere just below the wing roots of mostgliders, a most inconvenient place for mounting a tow-hook.

It is obviously impossible to mount a tow-hook on the centreline of the gliderin such a place (unless the pilot held the cable in his teeth!), so thedesigners arranged for two hooks to be fitted, one each side of the fuselage.This in turn necessitated a V-shaped trace from the launching cable, dividingon each side of the nose and attaching to the hooks on each side of thefuselage.

Blanik using shoulder releases for winch-launching (Oerlinghausen, Germany)

This so-called shoulder position for tow-hooks is theoretically very good,because it imparts little or no pitching moment to the glider under winchacceleration and during the climb. However, for obvious reasons, thesystem demands that both hooks release simultaneously when the pilot pullsthe release knob, and history has shown that this was not always the case inthe past. The resulting asymmetric releasing problems brought the systeminto disrepute in Australia, although it is still in use in some countries.

A further factor was the difficulty in designing an automatic back-releasingmechanism which would also operate symmetrically at all times.



9/68

43.1.2. The belly release hook

Instead of using shoulder releases for winch-launching, the accepted methodis a single hook placed under the belly of the glider. As a general principle,winch launching should only be carried out using the belly hook, never the

nose hook. The position of belly hooks varies from one glider type toanother, some being way back under the pilot, others being further forward.

3.1.3. Unwanted pitching moments

Any belly-mounted hook must of necessity be displaced some distanceunderneath the gliders CG. Therefore, unlike the shoulder-release system,a belly-mounted hook will, due to the gliders inertia, impart a nose-uppitching moment about the gliders CG under winch acceleration during thetime when the glider is substantially level and not in a climbing attitude (e.g.during the ground-run and initial separation). The factors affecting the

amount of pitch-up which will be imparted to the glider, and the extent towhich any such pitch-up will be able to be controlled by the pilot, are :-

1. How far in front of the gliders CG the hook is mounted. The further backthe hook, the more the pitch-up tendency.

2. How far underneath the gliders CG the hook is mounted. High-winggliders, with high CG positions, tend to be the worst offenders.

3. The rate at which the glider is accelerated at the start of the launch. Thegreater the acceleration, the greater the tendency to pitch up.

The diagram above (reproduced with acknowledgement to Sailplane andGliding) shows the effect of winch pull at the start of the launch. The line ofthe pull is very low with respect to the gliders CG, imparting a strong pitch-up moment.



10/68

5

This diagram (again with acknowledgment to Sailplane and Gliding) showsthe situation in the climb. The line of the winch pull now passes much closer

to the CG and the pitch-up tendency is considerably reduced. In addition,acceleration has now ceased and the glider is at a steady speed, thereforeinertia is no longer a factor.

Any pitching moment which may be present is resisted by a lift force from thehorizontal stabilizer. A short tail-arm lessens the effectiveness of thehorizontal stabilizer in controlling any pitch-up which does occur. Winch-drivers launching gliders like this (Ka8, Pirat, Junior and plenty of others)should be briefed to take it easy and not accelerate too rapidly.

Some degree of pitch-up tendency imparted by a belly-hook is not

necessarily a bad thing, as the pilot will need quite a lot of down elevator tocontrol the gliders initial launch profile. Therefore if the cable breaks or theengine fails, the pitch-up moment is removed and the pilot already has downelevator to restore the glider to its normal nose attitude for a safe landing.However, it is important that the hook is not placed so far back that the pitch-up tendency is uncontrollable. It is also important that pilots realise theimplications of their weight on pitch-up tendencies. The lighter the pilot, themore marked the pitch-up tendency.

If the belly hook is mounted a long way forward, there will be little or nopitch-up as the launch accelerates. In this situation the pilot may well have

to use some up elevator to obtain the required launch profile. If a failureoccurs, the glider will be in a nose-up attitude, but this time with the elevatorup instead of down. The pilot will need to react very quickly if the glider is tobe restored to the required attitude for a safe landing.

3.1.4. Porpoising on the launch

Mounting a belly hook too far forward creates another problem. From aboutthe half-way point of the launch onwards, as the cable pull becomesincreasingly downward, the pilot needs a lot of back stick to keep the noseup. The horizontal stabilizer reaches a critical angle and stalls the wrong

way round (i.e. inverted) in the downwash from the wing. This sets up apitching oscillation, or porpoising, which grows progressively more violent if



11/68

6the pilot keeps the stick back. It is a most unpleasant phenomenon, but canbe fixed immediately by easing the stick forward a little. Some gliders (e.g.Slingsby T31, Schneider Boomerang and Super Arrow) do this sort of thingall the time and such gliders are known for not getting much height on winch-launches. This has led to a modification to move the hook back a little in

some designs. Naturally such a modification should not be done withoutairworthiness approval.

Both extremes of hook position should be avoided by basic design. A hookmounted too far back (especially on a deep fuselage) could result in such astrong nose-up pitch that it would be uncontrollable and the glider would bepulled into a steep climb near the ground whether the pilot wanted it or not.Too far forward and it might not want to climb at all and would leave the pilotexposed to the risk of having a lot of up elevator applied if a failure occurrednear the ground.

A designer chooses the belly hook position carefully and the glider is flight-tested on the winch in all permissible CG positions. It will only obtain aCertificate of Airworthiness for winch launching if it shows no dangeroustendencies over the entire range of CG positions and launching speeds.

3.1.5. Other unwanted moments

Some gliders are fitted with a nose-skid as part of the main undercarriageassembly. In this case the belly-hook has to be offset to one side, as theskid occupies the space where the hook would normally be mounted.Examples of this kind of glider are short-wing Kookburra, K7, K8 and ASK13.

Even though the hook may be only about 10cms off the centreline of theglider, a surprisingly large swing can be produced if the winch acceleratesvery rapidly at the start of the launch. Although this is usually quitemanageable, many pilots will recall leaving the ground with quite a lot of leftrudder in a Kookaburra when being launched by a ham-fisted winch-driver.Pilots flying gliders with offset hooks are advised to be well prepared to putsome effort into keeping straight on take-off.

3.1.6. Automatic back-release

There is an important design feature built into belly hooks. This is anautomatic over-ride, or back-release mechanism, which will function if theglider flies too far overhead the winch. Thus, even if the main releasemechanism is disabled inside the glider (for example, if the connectionbetween the release knob and the belly hook should break, not entirelyunknown), the back-release mechanism will function and automaticallyrelease the launching cable when the glider gets overhead or nearly-overhead the winch. If there is such a thing as a foolproof device, the back-release mechanism is as close as we are going to get.

Nose-mounted aerotowing hooks generally do not have back-release

mechanisms, although there are exceptions such as the Blanik and IS28B2.It is best to assume that a back-release is not fitted to ANY nose hook and



12/68

7on that basis to ensure that you never attempt a winch launch using one ofthese hooks. If you do, your last line of defence against a defect in the hookor its installation is removed.

3.1.7. Predicting likely launch characteristics

A pilot can predict the likely launch characteristics of a glider by standingback from it and eyeballing its general shape and hook position. A deepfuselage (implying a rather high centre of gravity) and aft-mounted hookequals a strong pitch-up on launch. Check out the tail-arm too - a short-coupled glider like a Pirat, which also has a deep fuselage and a very aft-mounted hook, is likely to be quite a handful on a winch take-off, especiallywith a lightweight pilot aboard. Make sure winch-drivers are briefed not togive such gliders very rapid acceleration. A shallow fuselage and hook notso far back will not produce so strong a pitch-up on take-off.

3.1.8. Useful tips

Use of soft cushions

Ensure that no soft cushions are placed behind pilots for winch-launching.Under acceleration, such cushions compress and allow the pilot to moveback in the cockpit. This has three effects:

4. It moves the CG back, maybe only slightly but enough to cause trouble.

5. The pilot may involuntarily move the stick back as he/she moves back.

6. The pilot may be unable to reach the release knob after moving back.

Minimum pilot weight

If a pilot is on or near the minimum permissible weight to fly the glider, it iswise to add some extra cockpit ballast, especially if the glider is known to bedemanding in its pitching characteristics.

Careful attention to seating position

Choose the seating position carefully. There is merit in positioning the pilot

one notch further forward than appears to be necessary, anticipating a smallamount of movement under acceleration. Similarly the rudder pedals couldbe brought back one notch.

Awareness of poor harness design

Some gliders (e.g. Std Cirrus and its variants) may allow the pilot to slidebackwards and upwards along the seat once the glider is established in thefull climb. This is a function of the very smooth seat and the particulardesign of the seat harness. The effects are the same as for compressibleseat cushions. A simple modification, such as fitting some Velcro to the seat

pan to increase friction, may help to prevent this disturbing tendency.



13/68

83.1.9. Final reminders

Attempting a winch launch on the nose (aerotow) hook forfeits the safety-netof the automatic back-release mechanism and needs considerable upelevator to maintain a climb. The pilot is very vulnerable if a cable or winch

failure should occur in the early stages. Dont try it.

It is the PILOTS responsibility to ensure that the launching cable is attachedto the correct towhook. Pilots should not try to blame the ground-crew if anerror is made.

Take care in setting up the cockpit exactly as required to ensure safety inwinch-launching.

Talk to the Instructors Panel and Committee about throwing out any softcushions which have been used as cockpit padding.

3.2. Winch-launching loads and launch-speed boundaries

Winch launching is potentially stressful for a gliders structure. The reason isnot particularly obvious and needs a detailed explanation.

We start by considering a glider about to carry out an aerobatic manoeuvrein free flight, in this case a loop.

3.2.1. Winch-launching loads

In free flight, when a glider is pulled up into a looping manoevre, the wingsproduce much more lift than they usually do. This is at the command of thepilot, who firstly dives the glider for excess speed (thus producing more lift byincreasing the airflow over the wings) then eases the stick back to pull thenose up (thus producing even more lift by increasing the angle of attack ofthe wings). The glider therefore has a good supply of lift to enable it to flyaround the curved manoeuvre without stalling.

As the wings produce this greatly increased amount of lift, they naturallybend upwards under its influence. This places considerable stress on allcomponents in the wings, but especially at the wing roots, where all theforces generated in flight accumulate. The bending stresses at the wingroots of a glider during aerobatics are considerably higher than in non-aerobatic flight.

In the situation described, the pilot feels G loads as the manoeuvredevelops. This feeling is normal in all aerobatic manoeuvres and the valuesof G typically felt by a glider pilot will be in the order of 2 to 3. This meansthe pilot is feeling 2 to 3 times his/her own weight (we are all at 1 G in normalcircumstances). Anything more than 2 to 3 is usually unnecessary and Gvalues of over 4 may be regarded as somewhat excessive in gliders.



14/68

93.2.2. Wing bending relief

It follows that, if the pilot feels 2 to 3 times his or her own weight during amanoeuvre, so does every component of the glider. Every part of the wings,fuselage and tail of the glider experiences this G force.

As the wings are bent upwards by the increased lift, the G forces (pulling inthe opposite direction) try to straighten them out again as the manoeuvredevelops. Thus the G forces produce a measure of bending relief for thewings, which is very useful in protecting the structure from beingoverstressed.

With lift pulling in one direction and being directly opposed by G forcespulling in the other direction, it will be seen that aerobatics, although morestressful than non-aerobatic flight, are quite capable of being absorbed bythe glider, as G forces tend to prevent the gliders wings from being bent

beyond their limit. This is obviously only true if the glider is rated foraerobatics and is flown within its placarded limitations.

3.2.3. The relevance of bending relief to winch-launching

What has all this to do with winch-launching? Well, there are two similaritiesbetween a winch-launch and a glider doing aerobatics. One is the excessspeed (over and above, say, min sink speed), the other is a higher thannormal angle of attack (again relative to the min. sink situation). On a winch-launch the speed is at least 1.3 Vs and is usually higher than that, and theangle of attack in the full-climb phase of the launch may be as high as 9 or

10 degrees. The combination of the two produces plenty of lift, enough togive a peak rate of climb during the full climb part of somewhere between2,000 and 3,000 feet per minute (20 to 30 knots).

During a winch-launch, a glider is tethered by its belly, being attached to thelaunching cable by the belly hook. Tethering the glider prevents it frommoving into a curved looping manoeuvre, which it would naturally want todo if it were untethered. As the glider is not following a curved path around aloop, there are no G forces being produced and this is borne out by the factthat a pilot feels no G during a winch launch.

If there is no G, there is no bending relief for the wings. They are producingabout the same amount of lift as they would for a 2 to 3 G loopingmanoeuvre, but the looping manoeuvre is being prevented by the cable andthe glider follows a more or less straight climbing path. Thus the wingsbend, and are not being straightened out again by any G forces.

The net result of this is twofold :-

1. The bending moment around the wing roots is unrelieved and istherefore rather high.

2. The pilot feels no additional G forces and tends not to realise that thewings are being subjected to these higher bending moments.



15/68

10Refer to the two diagrams overleaf.

3.2.4. The purpose of the weak-link

Because of these two points, it would not be difficult for the pilot to produce

bending moments about the wing roots which are higher than those intendedby the designer. Excessive bending moments may be produced by flying theglider in excess of its placarded maximum launch speed or by pulling backexcessively hard on the stick, or by a combination of the two.

To guard against this eventuality, it is a feature of the certification of glidersthat a weak link be fitted in the launching cable for the protection of theglider against overstress. The maximum breaking strength of the weak linkis placarded in the cockpit and repeated on an external notice near the bellyhook. Failure to use a weak leak of the specified maximum breakingstrength compromises the certification of the glider and, more importantly,

puts the pilot at risk through overstressing of the glider to the point ofstructural failure. The weak link fulfils the same function as a fuse in anelectrical system.

Free flight, pulling up into a looping manoeuvre. About 2Gs-worth of lift beingproduced, tending to bend the wings upwards. 2Gs-worth of wing-weight, creating

a relieving force tending to straighten wings out again.



16/68

11

In full climb on a winch-launch. About 2 Gs-worth of lift tending to bend wingsupward. Because glider is tethered by its belly and thereby prevented from entering alooping manoeuvre, there is only 1Gs-worth of wing-weight trying to straighten themout again. In addition, the cable is being pulled into the winch, creating a downwardforce which is acting centrally on the fuselage and is reacted by up elevator. The

overall result is a more severe bending moment about the wing roots.

3.2.5. Launch-speed boundaries

3.2.5.1. Maximum speed

The permissible maximum speed for winch-launching is quite restricted andis usually the lowest of all the speed limitations imposed on the glider. Thereason, as explained in the previous section, is because the glider istethered by its belly, thus allowing more wing-bending to take place than innormal flight. Limiting the maximum speed during winch-launching puts anupper limit on the amount of wing-bending which is likely to occur. Themaximum winch-launch speed (known as Vw) is placarded in the glidercockpit.

A word of clarification is appropriate in relation to the maximum speed limit.We know that this limit is imposed for structural reasons, specifically wing-

bending. The conditions conducive to wing-bending are at their worst in thesecond half of the launch, where cable-pull is added to gravity and the pilotmay need a lot of back stick to produce enough lift to counteract thecombination of the two forces and keep the glider climbing. If the maximumplacard speed is exceeded at this point, enough force will be generated tobreak the weak link. If a weak link stronger than specified is being used, oreven worse if no weak link at all is fitted, the glider can easily beoverstressed. The maximum placard speed must never be exceeded oncethe glider has passed about the half-way mark of a launch.

There is a bit more freedom lower down. At the beginning of a launch, withthe cable-pull almost parallel with the ground, there is very little extra load onthe wings over and above the loads applied in normal flight. Even at cable-



17/68

12to-ground angles up to about 25 and the glider in full climb, it can be shownthat the loads remain acceptable. Exceeding the maximum placard speed isreally not very harmful during these stages and it is acceptable to tolerate asmall excess (say 10%), progressively steepening the climb angle while thewinch-driver sorts himself out and adjusts the speed. If excess speed

remains when the full climb has been attained, the standard too fast signalcan be given or the launch abandoned if it appears to be really getting out ofhand.

Glider pilots sometimes (too often?) climb steeply close to the ground,offering as an excuse one of three reasons, viz:-

1. I had plenty of speed, so it was safe to climb steeply. WRONG!

2. I am experienced and current, and would instantly recognize a failure.WRONG!

3. I had too much speed, so I pulled back to try to kill it. WRONG!

Climbing excessively steeply near the ground is never acceptable, whateverthe excuse. It is much safer to allow small inroads into the maximum launchspeed in the early stages, while controlling the climb angle so that itsteepens progressively in the normal way, than to pull the nose up close tothe ground.

History shows that pilots do not recognize launch failures as quickly as theythink they do. The known delay in realising that a failure has occurred can

prove fatal if the nose is so high, and the ground so close, that recoverycannot be made in time.

3.2.5.2. Minimum speed

The permissible minimum speed for winch-launching is not placarded and isbased on a value of 1.3 times the stall speed (1.3 Vs). The stall speed to beused is the one applicable in the configuration in which the glider is beinglaunched. As well as varying with glider weight, stall speed also varies withflap settings in the case of gliders fitted with these devices.

Going back for a moment to the cable-pull discussion in the previous section,this has an effect on the minimum speed too. In the early stages of thelaunch, with the cable-pull nearly parallel with the ground, the wings are notloaded much above normal flight and the stalling speed is pretty much thesame as in normal flight too.

As the launch progresses and the cable pull approaches closer and closer tothe vertical, the loads on the wing increase and the stalling speed increasesaccordingly. Near the top of the launch the increase can be as much as30%, which takes the stall speed to 1.3 Vs. This just happens to be thespeed we choose as our minimum speed to fly a winch-launch.

It follows that the logical action to take is to reduce the back-pressure on thestick as the top of the launch is approached. In fact we go even further and



18/68

13modern winch-launch training calls for a gradual forward movement on thestick towards the top of the launch. This is in contrast to the stick back inthe guts technique of past years, where it was considered the right thing todo to get the last inch of height out of the launch. In reality, it is probablyonly an inch that will be gained, because the biggest proportion of the total

height obtained on a winch-launch is gained in the early stages, immediatelyafter the full climb has been attained. This is not an argument forexcessively early rotations into full climb, but for trying to persuade pilots thatclutching the stick back in the guts at the top of the launch will not make a jotof difference to the height gained.

For those pilots who might still be tempted to use this old-fashionedtechnique, think about this. Even if the risk of stalling on the wire isconsidered to be worth taking at that height, and even if you dont reallybelieve that you wont get a bit of extra height out of it, think about the poorold winch-driver when the glider breaks the weak link and sends hundreds of

metres of very badly-behaved spring-steel wire hurtling downwards. Even ifthe winchs protective devices work and the driver is unharmed, it is likelythat he will spend the next hour sorting out the mess the pilot created. Hewill almost certainly nominate you for the clubs pain-in-the-arse award afterthat experience.

4. PILOT CONSIDERATIONS

4.1. Correct launch technique

The correct launch technique allows the glider to separate from the ground ina natural flying attitude and then (provided the speed is above the minimim1.3Vs) to enter a graduated climb profile, becoming steeper as height isgained.

4.1.1. Ground-run and separation

The natural flying attitude on the ground varies from type to type, as doesthe means by which the pilot uses the controls to adopt that attitude.

4.1.1.1. Taildragger gliders

Most gliders of taildragger layout are already at or very close to the naturalflying attitude on the ground and the only action needed by the pilot is to beready for separation to occur and anticipate any minor corrections that maybe needed to either prevent an excessively steep climb developing or givesome assistance to the glider to enter a gentle climb, depending on anumber of factors such as rate of acceleration, etc.

Some pilots prefer to raise the tail off the ground during the ground run, thenuse a small back movement of the stick to get the glider into the naturalflying attitude, following which it will usually lift off cleanly. The advantage ofdoing this is that enables the pilot to feel at what point during the ground-run

the elevator becomes effective. The disadvantage is that it requires a fairdegree of finesse to return the glider to the natural flying attitude. It can also,



19/68

14if mishandled only slightly by a pilot who over-reacts to the nose going downtoo far, cause the glider to over-rotate into an excessively steep climb attoo low a height, especially if the acceleration is rapid.

The essential point is that the glider should leave the ground in a natural

attitude and should not be allowed to steepen its climb until the speed isconfirmed to have risen to the minimum safe value.

4.1.1.2. Nosedragger gliders

Gliders of nose-heavy layout, will usually self-rotate into a tail-down attitudeunder winch acceleration. If the acceleration is fierce, the tail might bangdown very hard. The pilot should anticipate this and start the take-off runwith the stick well forward in this type of glider. When the actual acclerationstarts and the pilot has felt what it is like, the control actions can be refined toget the glider into the natural flying attitude.

The same comments apply to the nosedragger as for the taildragger, withrespect to the glider leaving the ground in a natural attitude and only beingallowed to steepen once the speed has increased to the minimum 1.3Vs.

4.1.2. Initial climb

Once the glider is clear of the ground, differences in undercarriage layout areno longer relevant and all gliders are treated identically for the remainder ofthe launch.

In the initial climb phase the pilot controls the glider in such a way that theclimb angle progressively increases as height is gained, the speed beingsomewhere between the minimum of 1.3 Vs and the maximum placardedvalue.

The natural tendency of the glider at this stage of the launch will varydepending on its design features, the CG position (mainly a pilot weightconsideration), the rate of winch acceleration and whether or not there is awind-gradient. Some gliders will try to self-steepen, in which case the pilotwill need to resist this tendency and restrain the climb angle to achieve thenecessary graduated profile. Other designs may be more reluctant tosteepen and such gliders may need a little help from the pilot to steepen theclimb in order to achieve the required profile.

The main point is that the initial climb profile must be positively controlled bythe pilot, unless by happy coincidence the glider is following exactly thecorrect profile of its own accord. This is not often the case, but unfortunatelytoo many pilots allow the glider to do its own thing without exercising therequired amount of control over it.

The more height you have, the steeper you can afford to be. If the pilotsteepens the climb progressively as height increases, and if the speedremains at a safe value, the initial climb technique is correct. Keep doing itthis way.



20/68

15Glider attitude accurately controlled. Donot climb steeply at too low a height.Make changes to climb angle smoothlyand progressively, gradually steepeningthe climb as height builds up.

Speed greater than 1.3Vs,permissible to exceed theplacarded maximum by about10% at this point in thelaunch

Correct initial climb

4.1.3. The maximum speed placard and the initial climb

Previous sections of this manual referred to the maximum placardedwinch/auto speed (Vw) and the fact that this speed is intended to protect thestructure against being overstressed on the launch. The point was madethat the loads on the glider during the early part of the launch are not muchmore than in level, untethered, flight.

If too much speed is noted during the initial climb, it is better to continuegraduating the climb until the glider is established in the full climb at a safeheight than to climb the glider too steeply near the ground in an attempt toload up the winch engine. When the glider is safely established in the fullclimb, the pilot can then take stock and see whether action (e.g. signals,abandoning the launch) is then necessary. The only proviso here is that thespeed has not become intolerably high during the initial climb, say more than10% higher than Vw. If the speed does get intolerably high, the decision hasto be made whether to give a too fast signal before the full climb orabandon the launch there and then.

If the launch is abandoned during the initial climb phase, two things apply:

1. The winch/car driver may not be able to feel the glider on the end of thewire in the early stages of the launch before the full climb has beenreached. If the launch is abandoned, the driver may not realise itstraight away and may not cut the power promptly. Drogue parachutes(if fitted) can become quite unstable under these conditions and theirbehaviour is unpredictable. If the drogue decides to oscillate, the glidermay become entangled.

2. The winch/car driver may not be able to see the glider because ofground clutter and/or strip curvature. This adds to the lack of feel andoffers another opportunity for entanglement in a cable drogue.



21/68

16All things considered, it is best to stay on the wire if possible (provided ofcourse that the speed is not trending towards too SLOW, or is notoutrageously fast), and grade the climb carefully until the driver can see andfeel the gliders presence, then assess the options and act accordingly.

4.1.4. Full climb

The full climb presents no great difficulties. Direction is monitored by lookingeach side of the gliders nose (clouds are handy for this too), climb angle isdetermined by looking under the wing and the wingtips are checked in turn tosee if the glider is level laterally (but see 4.1.8).

It is quite safe to climb steeply once the glider is properly established in thefull climb. The principal height gains are made in this early part of the fullclimb, so get the maximum benefit out of it by optimum utilisation as soon asyou safely can. You wont be able to make up for any losses higher up the

launch.

Speed should be kept between the minimum (1.3Vs) and the placardedmaximum, the region known as the working speed band. Glider speed isbasically determined by the winch-driver, pilot technique making relativelylittle difference. There are exceptions to this, such as a very low-poweredwinch, where pulling back on the stick results in engine revs decreasing andthe speed decaying. However, there are not many low-powered winches leftnowadays and it is a mistake to think that launch speed can be controlled inthis way.

Rather, the opposite is the case. Pulling the stick further back in the fullclimb when being launched by a powerful winch can result in the speedactually increasing. This is the arc of a circle argument familiar to water-skiers, where following a line outside that taken by the ski-boat will cause theskier to increase speed because of the longer distance which has to betravelled.

It is good practice on the part of the pilot to develop the habit of relaxing theback-pressure on the stick if the speed falls, and increasing the back-pressure if the speed rises, provided of course the speed remains within theworking speed band.

As height is gained and we pass the half-way point of the launch, we enterthe region where there is an increasing downward pull on the cable. Thisresults in an increase in stalling speed and an increasing amount of stresson the wings. From this point onwards, we need to be more conscious of theloads applied to the glider and ensure that speed limits are conscientiouslyobserved.

When approaching the top of the launch, it is prudent to start relaxing anyback-pressure on the stick, translating this into a forward movement as thenose is pulled level with the horizon by the cable. This ensures a clean

release and minimises any chance of breaking the weak link and/or cable.



22/68

17The top of the launch is the most likely place to break the cable or weak link.Cable pull combines with gravity to maximise the strain in the wire. Faultypilot technique, in the form of keeping the stick right back, will very likelycause a break at this point. This is the kind of break which is very hazardousto winch-drivers as the spring-steel wire descends upon him.

4.1.5. The maximum speed placard and the full climb

There is no leeway in the maximum winch-launch speed in the full climb.

4.1.6. The effect of the weight of the wire

The nominal mass of 3.15mm range 2 spring steel wire is 0.0611 kgs permetre. The weight of a 1500 metre length of wire is therefore 92 kgs. By thetime the glider is near the top of the launch, there will be somewherebetween 500 and 700 metres of wire hanging from its belly, which amounts

to an extra weight of 31 to 43 kgs added to the fuselage.

The figures for the slightly thinner 2.8mm range 2 spring steel wire arebetween 24 and 34 kgs for the same wire lengths. For 5mm 7 x 19 wire rope(stranded cable), the figures are between 48 and 67 kgs.

4.1.7. The release

At the top of the launch the winch driver will close the throttle quitenoticeably and at this time, one of two things will happen. Either the cablewill back-release or the launch will become noticeably slack, with a fall-offin speed and no more upward progress. In both cases, the pilots action is toestablish a normal flying attitude and pull the release twice.

Use of the back-release in normal launching is quite acceptable. However,pilots must ALWAYS be taught to pull the release twice, to be certain thecable has gone.

4.1.8. Winch-launching in crosswinds

The maximum height on a winch-launch is obtained if no correction is madefor crosswinds and the glider is allowed to drift with the wind. However, thistechnique will earn the pilot no friends and may be dangerous if there arepowerlines in the vicinity. It is normal practice to make some kind ofcorrection for crosswinds.

This can be done in one of two ways. The gliders wing can be lowered intothe wind and a touch of opposite rudder applied, thus effectively sideslippingthe glider into the wind. The alternative is to lower the wing and apply thesame rudder, altering the gliders heading so as to offset the drift. Thesecond one is probably the tidier of the two and it is usually more effective,especially in strong crosswinds.

Crosswind correction should be applied as soon as the glider is clear of theground. If you wait until the full climb, it will usually be too late and the cable



23/68

18will probably drift outside the airfield after release, the winch-driver beingpowerless to prevent it.

4.1.9. Launch speed signals

Too slow

If the speed falls below the optimum but has still not fallen to the 1.3Vsminimum, the too slow signal may be given by lowering the nose of theglider and rocking the wings from side to side with coordinated aileron andrudder. If speed does not increase, release.

Too fast

If the speed is showing an upward trend and does not appear to bestabilising, a too fast signal may be given by yawing the glider from side to

side, taking care not to let the secondary effect of rudder produce any rollwhich may confuse the winch driver.

4.1.10. Kiting

Because of the hazard to the general public, the kiting of gliders on a winchlaunch (i.e. the paying out of cable in order to increase launch height) isprohibited.

4.2. Incorrect launch techniques

4.2.1. Holding the glider down

We know that a good launch starts by climbing gently away from the groundafter separation, the gliders climb angle (or flight path, if you prefer) beingprogressively steepened as height is gained.

We have discussed the pitfalls inherent in climbing too steeply too close tothe ground. There is no need to labour that point any further here.

Another common fault in learning the winch launch is to separate normally,then hold the gliders nose down in an excessively flat attitude immediatelyafter separation. This is usually done in error, while a pilot develops the

feel for the correct attitude and how to achieve it. Such errors in learningare quite understandable and instructors and winch-drivers have no greatproblem in overcoming them.

However, there are some pilots who genuinely think that holding the gliderdown, then rotating into the climb, is the correct technique to use. This is amistake, for the following reasons:-

1. The winch-driver cannot feel the glider in the early stages of the launch,because negligible load is being applied to the winch. He/she thereforedoes not know whether too much or too little power has been applied.



24/68

192. The winch-driver usually cannot see the glider because it has not risen

above ground clutter. This is especially true if the background consistsof high ground or trees or in the mirage effect typical of summerconditions.

3. In most cases, it will almost certainly guarantee that the glider willgrossly overspeed.

4. If a climb is then suddenly commenced, the excessive speed andconsequent sensitive elevator control will make it very difficult for thepilot to control the gliders attitude accurately and a very steep climb mayoccur very suddenly, especially if the glider is known to be pitch-sensitive and/or has a very aft-mounted hook.

5. A combination of 3 and 4 above will probably result in a low-level cablebreak due to the sudden application of load to the cable.

It will therefore be obvious that a smooth, progressive transition into the fullclimb is preferable to the stepped technique, from both the pilots and thewinch-drivers viewpoint.

4.2.2. Allowing the glider to self-rotate

Another common problem relates to the tendency of some gliders to pitch-upof their own accord under the influence of winch acceleration. We know thereasons for this from the first section on glider considerations.

A pilot who fails to counteract this tendency, by restraining the climb anglewith forward pressure or even a definite forward movement on the stick, willfind the glider climbing more and more steeply as speed builds up. This willoften occur before sufficient height has been gained for a steep climb angleto be considered safe.

The tender trap here is that it feels so natural for the glider to be entering theclimb of its own accord that the pilot may not smell a rat and take someaction to restrain it. Some of these self-steepenings are rather subtle andmust be monitored very carefully by the pilot. Others are very obvious andyou would have to be an idiot not to detect that things are getting out of

hand.

Each launch is a little bit different from the one before. Winch-drivertechnique, wind velocity, wind-gradient are just some of the factors makingfor variety in launches. There is no alternative but to closely monitor theclimb profile of EVERY winch take-off and not assume that the next one willbe similar to the last one.

The tendency to allow the glider to take charge of the pilot during this criticalphase of flight is known to have been responsible for a number of accidents,some of them fatal, in Australia and overseas. The Germans, the most

experienced winch-launch pilots in the world, call this particular error lettingthe winch take-off happen to the pilot.



25/68

204.2.3. The Kavalierstart

We look to the Germans again for an apt expression to describe excessivelysteep winch take-offs. The Kavalierstart term was coined after a very badrun of winch-launch accidents in Germany in 1995, during which there were

32 accidents in this phase of flight, 12 of which were fatal.

A Kavalierstart means simply treating the winch take-off in a cavalier (=arrogant, or an accident cant happen to me) manner.

The remedial treatment for pilots who do this kind of winch take-off is to bringthe matter to their attention. This should be repeated, ad nauseam ifnecessary, until the pilot gets the message. It doesnt matter who brings it tothe pilots attention. It doesnt have to be an instructor, in fact it is probablymore effective coming from a pilots peers. It is a serious mistake to letthese things go without comment.

If the pilot does not get the message and persists in using the Kavalierstarttechnique, that pilot is living on borrowed time and will eventually have anaccident which will probably be serious or fatal.

If it proves impossible to reason with the pilot and neither advice nor checkflights do the trick, it is kinder to refuse to launch that pilot until he/she seesreason than to turn a blind eye and run the risk of a terrible accident.

4.3. Launch failures - Philosophical and training considerations

4.3.1. Introduction

No system is foolproof. Aircraft engines fail, pilots run aircraft out of fuel andthey also make simple errors of judgement or skill. Similarly winch problemsof various kinds occur from time to time.

Winch cables break from time to time. Winch engines also run out of fuel, just like aircraft engines do. Such occurrences should not in themselvesresult in a serious accident, as all the failure cases are well-known andtraining strategies are in place to deal with them. Nevertheless accidents dooccur and we are forced to ask why.

The answer is twofold. Part of it lies in human factors, the nut behind thewheel. The other part lies in the training system itself and whether thisprovides an adequate level of protection for a pilot when somethingabnormal occurs.

In this section, the philosophical approach to launch failures will beexamined, as will all aspects of the training of pilots.

4.3.2. The mental approach to launch failures

In the early days of light aircraft, when aero engines were rather unreliable,there were many failures, but the good side of the story was that almostevery forced landing was successful because the pilots got plenty of



26/68


27/68

22The time taken to recognise a launch failure can be readily confirmed bycheck flight, where the failure can be contrived under controlled conditions.Without going into detail of all the various failure cases, which are thesubject of another section of this manual, it is sufficient to say here that, tobe safe, pilots must accept that there will be a delay between a failure

occurring and the pilot realising that it has occurred.

4.3.2.2. Expecting a failure

This is a key point. A skilled pilot always EXPECTS a failure on each take-off. Nowadays this extends to incorporating the expectation of a launchfailure into the pre-take off check, the O in CHAOTIC standing for Optionsas well as Outside.

Expectation of a failure on each launch creates a mental preparedness todeal with it, the primary action and the various options having been

considered before take-off. If the failure occurs, there is no mental barrier toovercome; if it doesnt occur, the pilot can feel pleased with a successfullaunch.

Only if a pilot is expecting a launch failure can the reaction time be reducedto manageable proportions. This is known as defensive flying and it is theanswer to safety management in winch-launching.

Expecting a failure is a key point. Always be prepared for a launchfailure. No-one is immune from the time delay inherent in detecting a

failure and reacting to it.

4.4. The physical characteristics of launch-failures

Launch-failures fall into two main categories, cable-breaks and engine-failures. They both rather obviously result in termination of the launch, thecable-break being an abrupt occurrence, the engine-failure perhaps beingmore subtle.

4.4.1. Cable breaks

Regardless of the type of wire, cable or rope being used, breaks can occur

suddenly and randomly. Solid wire is prone to fatigue due to constantbending and straightening and gives no warning of impending failure.Stranded cable (wire rope) gives a bit more of a hint by revealing loosestrands or an incipient birds nest, but this can only be of use as a warning ifit happens to occur where someone on the field can see it. The suddennessof cable failure, together with the lack of warning, accounts for why themental approach to launch failures places so much emphasis on alwaysbeing prepared and never assuming that it wont happen to you.

The other aspect to consider is that the weak link may break. This isdesigned to protect the structure, but the result is a sudden failure of the

launch which is identical in all respects to a cable-break.



28/68

23Because the speed margin above the stall is small and the climb angle maybe steep, the pilot must react promptly to a cable-break and lower the nosesmartly to at least the approach to land attitude and allow the speed tobuild up to a safe value. The two types of cable-break to consider are asfollows:

Low-level

If the failure occurs low down, below say 200 feet, the climb angle will not be(or should not be) very steep and the lowering of the nose is not an extrememanoeuvre. The speed should not decay very much during the pushoverfrom the climb to approach attitude and there is no great inertial problem toovercome, as the glider does not have to climb to the apex of a steep hillthen come down the other side to build up speed. This is the very reasonthe glider is not allowed to climb steeply during this phase. There is,however, one fly in the ointment, which is wind-gradient.

If there is a wind blowing, there will be a wind-gradient, the severitydepending on the wind-strength. As the glider climbs, it enters an increasingwind strength and this effectively adds to the airspeed on the launch. This isa good bonus at this stage of the launch. However, if the cable breaks atthis point and the gliders nose is lowered, it has to descend through thewind-gradient and the speed will start to fall again. The pilot will need tocontrol the gliders attitude very carefully to strike a balance betweenmaintaining an adequate margin of airspeed, yet not diving into the ground.Do not open the airbrakes until an adequate speed has been regained andthey should not be opened at all if the glider is very low, or you will be

rewarded with a heavy landing.

WHAT THE WIND-GRADIENT GIVETH ON THE WAY UP, IT TAKETHAWAY ON THE WAY DOWN.

High-level

The full climb stage of the launch is characterised by a very high climb rate,typically in excess of 2,000 ft/min (20 knots). Height is obviously gained veryrapidly and it is quite safe to climb steeply during this phase, provided ofcourse that the speed is safely within the working band.

If a cable-break occurs during this phase, the bad news is that the pilot hasto take prompt and positive action to ensure that the gliders nose attitude ischanged from the steep climb attitude to the approach to land attitudenecessary for re-establishing a safe speed. The good news is that there issufficient height to do this safely, so there is no need to rush things. Promptand positive is not the same thing as panic-stricken.

There are however two additional factors, jointly the most important of alland persistently responsible for causing winch-launch accidents year afteryear. The factors are inertia and time. Study the diagram below.



29/68

24

From the time of the cable-break (at the left of the diagram) and theimmediate application of forward stick to commence the recovery to theapproach to land attitude, a minimum of 5 seconds may elapse. This isdue to the inertia of the glider (there is no significant lag in an airspeedindicator) and it is essential to let the gliders speed stabilise at the new,nose-low, attitude, and CONFIRM that the build-up in speed has occurred,before opening the airbrakes or attempting to manoeuvre. Remember thatthe 5 second recovery time is ADDED to the detection/reaction time.

If an attempt is made to manoeuvre during the curved part of the diagram, in

other words while the glider is short of speed and has not been allowed tostabilise in the nose-low attitude, there is a likelihood of loss of control,usually a spin. There are many accidents on record from this cause andthey are all entirely preventable.

To provide a fail-safe situation, a pilot is recommended to TREAT ALLCABLE-BREAKS AS LAND-AHEAD CASES IN THE FIRST INSTANCE,regardless of the height at which they occur. If this is done, it means twothings.

1. The recovery action becomes a CONDITIONED RESPONSE and the

pilot will always react automatically to a cable-break in the safestpossible way.

2. The glider will always have sufficient speed to manoeuvre safely, if itbecomes apparent after a moments analysis of the actual situation thata modified circuit is in fact possible.

Given that the entire elapsed time for the recovery manoeuvre is severalseconds, there is ample time for the pilot to release the remaining piece ofcable attached to the glider and this action must always be carried out aspart of the conditioned response mentioned in 1 above.



30/68

25When the above actions have been carried out, the glider is as ready to landsafely as it ever will be. It is now time to refine the judgemental aspects ofthe launch-failure to determine just where it is most appropriate to land it.

4.4.2. Land ahead or circuit?

This is one of the most crucial decisions a pilot will ever be called upon tomake. However, the fact that it is crucial does not mean that it must bedifficult. It is purely a function of training. Refer back to section 4.3,philosophical and training considerations.

Many accidents result from a decision to turn immediately after a cable-break, with the intention of joining some kind of circuit. The glider, short ofspeed because of its own inertia, does not have enough energy to attemptsuch a manoeuvre unless the correct recovery action has been taken andenough time has elapsed for it to take effect. The pilot, who may be under a

fair bit of stress at this time, turns the glider as a kind of reflex action beforethe required amount of time has elapsed. It is worth examining how thisreflex action became embedded in the pilots mind, to come to the surface atthe worst possible moment.

There are several possibilities, any or all of which may be relevant.

1. An obsession with getting back to the launch point, based either on afeeling of shame associated with landing anywhere else on the field(real pilots always get back to their take-off point) or an underlyingfeeling that the launch point is familiar and safe, anywhere else being

strange and threatening.

2. A feeling that landing down the field will cause much inconvenience andwill prevent launching another glider for some considerable time.

3. The pilot lacks confidence that he/she can safely land ahead in thespace available.

4. The pilot may feel that, although a landing ahead might be possible, itwill need full airbrake and he/she has never done that before.

With respect to the first two points, a feeling of shame at landing anywhereother than at the launch-point is often a function of peer-pressure. Peoplewho ridicule any pilot for putting safety before convenience have a lot toanswer for. Apart from remonstrating with the pilots friends who apply thiskind of pressure, the only way to proof a pilot against behaving in this wayunder pressure is through correct and conscientious training. It is importantto realise that it may take more training than we have traditionally allocatedto this sequence.

The third point, lack of confidence to land ahead in the space available, ispurely a function of training. If a pilot has never had to do this exercise

during training, it is not surprising that confidence will be lacking when the



31/68

26real thing occurs. The answer is obvious and again depends on proper andconscientious training. Convenience does not enter into it.

The fourth point, fear of using full airbrake, is more common than might bethought. It is again a function of proper and conscientious training.

Both the lack of confidence in landing in the space available and the fear offull airbrake can be covered in a useful simulation exercise, not related tolaunch failures but providing the necessary background to create confidenceand remove fear. The simulation can be carried out during an ordinaryroutine circuit, where the instructor will define a space to be landed in, withparticular emphasis on defining the end-of-roll, and then require the pilot toapproach as closely to this area as he/she dares on final approach, withoutusing any airbrake. It will get to the stage of looking impossibly high andsteep, to the extent that a gross overshoot seems inevitable. It is at thispoint that the approach looks remarkably similar to what a pilot sees when a

cable breaks in the full climb at, say, 400 feet, and the pilot has just pitchedthe nose down.

The airbrakes are now fully opened and the nose lowered to counteract theincreased drag, whereupon the rate of descent builds up to the extent that atask which initially looked impossible is now looking much more feasible. Asthe exercise is completed and is preferably repeated a number of times, thepilot finds that he/she can in fact land in spaces previously thoughtunavailable. And all because the pilot had never done a full airbrakeapproach before, or at least not within recent memory.

It should be stressed to pilots that they are not expected to carry out everyapproach this way. It is another string to their bow if they should ever need itin an abnormal situation. As well as launch-failure, it could be beneficial inan outlanding, for example.

Although this exercise is very successful in most gliders, there are a fewtypes (e.g. Salto, early Kookaburras) which have such feebleairbrakes/spoilers that its effectiveness is not as great. Briefing on this pointshould be a function of type conversion.

4.4.3 Inertial factors

The notion that airspeed indicating systems have a significant built-in lag is afallacy. On a Daily Inspection, if someone blows gently into the pitot head,the effect on the ASI needle is instantaneous. The lag in speed indicationsduring the recovery from a cable-break is not caused by the instrument, it iscaused by the glider. Provided there is no slip or skid present (these cancause pitot/static errors), if the airspeed indicator says you are slow, youARE slow. Do not manoeuvre, or open the airbrakes, until you have thecorrect indication.



32/68

274.4.4. Preference for landing ahead

Numerous accidents have occurred over the years because pilots haveelected to make a turn immediately following a cable-break. There are noaccidents on record which have been caused by pilots choosing to

land ahead.

This record speaks for itself. It is always preferable to land ahead if possibleand it is possible on more occasions than pilots think. The answer lies inproper training, so that a pilot has the confidence to land ahead and avert anaccident, rather than turn and lose control.

4.4.5. Briefing versus training

When a fatal spin off a 400 ft cable-break occurred in 1990, the Bureau of AirSafety Investigation (BASI) contacted a number of winch-launch clubs to

discuss their launch-failure training procedures. It had a very surprisingoutcome. Half the clubs questioned in this survey admitted that they did notcarry out live simulations of cable-breaks on the launch, but relied on thedive - pullup - pushover exercise described earlier. There was no evidencethat the club involved in the accident was in this category, but the informationwas startling nonetheless.

It has already been stressed (section 4.3.2.) that live training is necessaryfor proper coverage of this exercise and is in fact a GFA requirement. Thesame principle applies to the use of words instead of actions during alaunch.

To explain this words instead of actions phrase, some instructors say totheir students during a launch where would you go if the cable broke nowor tell me the last time you would be able to land ahead. This techniquemight seem to work quite well, but this is often because the student is tellingthe instructor something he knows he wants to hear. The technique in facthas no training value whatsoever unless the pilot has experienced somereal or simulated failures and can relate to what the glider can actuallyachieve. It is unfair to pilots to have them believe that words are as effectiveas actions in preparing them for possibilities like launch-failures. It isprobable that pilots supposedly trained in this way harbour an underlying

fear of launch-failure throughout their lives and they are likely to performpoorly when it happens to them in earnest.

4.4.6. The non-manoeuvring area

Although not strictly a failure case, a similar judgemental problem mayconfront a pilot who gets caught in this area. The non-manoeuvring area(NMA) is an area of sky which, if entered by the glider during a launch,results in the glider being unable to land ahead within the airfield if a launch-failure occurs, but the glider is also too low to make a safe turn-back into acircuit. In practice this means that a launch with very poor acceleration,

taking a glider a long way down the field without much height being gained,is the likely cause of such a situation.



33/68

28The obvious answer to the NMA is not to let the glider get into it. If a launchis not accelerating quickly enough, get off before there is any risk of enteringthe NMA.

The shorter the strip, the more acute the NMA problem. The GFA strip

minimum of 1200 metres is designed specifically to prevent NMA problemsfrom occurring.

4.4.7. Engine failures

When a cable breaks, the pilot does not have to make a decision on whethera failure has occurred - the launch stops dead and thats that. An enginefailure may be like that too, but it is also possible for it to be quite different. Avapour-lock in the fuel system on a hot day, for example, may causecoughing and spluttering of the engine, giving the pilot a slow decay ofairspeed and forcing a decision on whether this is the early stage of a failure

or a misjudgement of speed by the winch-driver. In the full climb, with thenose pointing up at about 40, it is essential that a pilot is trained to flydefensively to counteract this kind of problem.

Flying defensively means (a) detecting that a problem has occurred and (b)reacting correctly to the problem, erring on the side of conservatism andsafety if there is any doubt. We know that most pilots do not detect aproblem as quickly as they think they will, and this introduces a delay whichcan be critical if the problem is itself difficult to detect. Engine failures oftenfall into this category.

Referring back to the curved-line time/speed diagram; it was based on asudden failure, which forced the issue very clearly in the pilots mind.Imagine how much more difficult it would be if the failure is slow and difficultto detect. In the slow engine-failure case, more than any other, a pilot has tobe alert, suspicious and pessimistic in order to stay safe in winch-launching.

Expect the worst and you wont be disappointed when it happens.

YOU SHOULD FEEL PLEASANTLY SURPRISED WHEN YOU REACHTHE TOP OF EACH LAUNCH

5. WINCH DESIGN CONSIDERATIONS

5.1 General

Most Australian winches are designed to the NTA* standard. They come inall shapes and sizes, single-drum, double-drum, petrol, diesel, automatic andmanual. There is no standard Australian winch, just as there is no standardGerman or British winch.

The tendency in recent years has been towards a steady increase in the sizeof winch engines and the amount of power they produce. This is nosurprise, because we have seen a comparable increase in the weight ofgliders over the same period. It stands to reason that a six-cylinder side-



34/68

29valve Dodge is going to have more trouble launching a Twin Astir than it didlaunching the old Kookaburra.

5.2. How much power?

5.2.1. Low-powered winches

In the early days of winching, low-powered engines were common, mainlybecause they were readily available. Gliders tended to be light at that time,single-seaters grossing at about 300 kgs and two-seaters at about 400 kgs.These gliders also had old-fashioned wing-sections and low stalling-speeds,all of which suited them to being launched at relatively low speeds. Low-powered winches had little trouble launching such gliders.

As glider weights increased (the Blanik, for example, has a gross weight ofabout 500 kgs and later two-seaters are much heavier than this), these

winches started to groan a bit as they launched them. Modern laminar-flowwing sections and higher stalling-speeds worsened the problem. It wasfound that the glider pilot could control the speed of the launch - pulling backon the stick reduced the speed, easing off a bit allowed the speed toincrease.

From the drivers point of view, there was not much skill or finesse attachedto driving these winches, as much of their time was spent running flat-outand all the driver did was open the throttle and wait for the glider to appearabove the horizon.

The low-powered winch was responsible for instilling in pilots and winch-drivers some habits which were found to be quite inappropriate when appliedto the higher-powered winches which were a logical development to copewith the increasing glider weights.

Low-powered winches have nothing going for them. It is to be hoped thatthe last of these devices went to the scrap-heap years ago.

*No Two Alike

5.2.2. High-powered winches

As more and more high-capacity V8 engines found their way into wreckersyards and had plenty of life left in them after they were removed frompranged vehicles, they started to appear in winches, in most cases withautomatic transmission. They transformed the launching capacity of theaverage club winch, in particular their ability to handle heavy gliders in lightwind conditions.

They also cope with the density-altitude problems which occur on hotsummer days. For example, Bond Springs, NT, the site of the Alice SpringsGliding Club, is 2,400 feet above sea-level. A 35 summer day increases the

effective altitude of the strip to just over 5,000 feet. At that altitude, glidersgobble up a little more distance before getting airborne (slightly higher true



35/68

30airspeed) and the engines power output is down a bit. The higher poweredengines have enough in reserve to handle this sort of thing, which is a fact oflife at many Australian clubs during the summer.

The biggest problem with high-powered winches occurs when the power is

misused to produce too much acceleration at the start of the launch. Ofcourse, this is not the fault of the winch. Just because there is plenty ofpower on tap, it doesnt have to be used all at once.

Problems which may occur due to excessive acceleration are:-

1. Uncontrollable nose-up pitch, due to a combination of glider inertia andhook position.

2. Sideways swing on gliders with an offset hook.

3. Backward movement of the pilot, especially if soft cushions are in use.4. Banging of the tail in nosedragger gliders, risking structural damage.

The power of the winch should be applied in such a way that the glider doesnot have an excessively long ground-run, but is not accelerated so rapidlythat any of the above-listed problems occur. This is a function of winch-driver training.

When full climb is established, the old-fashioned technique of controlling thespeed from the cockpit no longer applies in the same way. It is still true that,if the speed falls, easing the stick forward a little will cause it to rise again.This is the correct initial response to a falling speed on any winch launch.However, if the speed is too high, pulling the stick back will not have thesame effect as it had on a low-powered winch.

On a high-powered winch, pulling the stick back is likely to cause anincrease in speed, because of the water-ski phenomenon mentioned insection 4.1.4. Pilots and winch-drivers should clearly understand this.

There is every reason to fit a powerful engine to a winch, provided that it isused properly. If a change has been made from low to high power, someeducation of winch-drivers and pilots will be necessary to ensure a continuedsafe operation.

5.3. Car retrieve or self-lay?

If a double-drum winch is in use (there are no more than two drums on anyAustralian winch at present), there is a choice of how the cables are laidbetween launches.

Some clubs use a vehicle to pull the cables back to the launch-point from thewinch after each pair of launches. This is effective and is the only methodwhich can be used if the ground is a bit rough.



36/68

31If the ground is smooth enough for a heavy vehicle to be driven on it, a self-laying winch can be used and is popular in some clubs, mostly in SA and theNT. After each pair of launches, the winch is driven to the launch-point, bothcables securely anchored and the winch driven back to its own end. Thismethod has the advantage that there is no wear of the cables during the

retrieve. The downside is that it is slower in operation than using retrievecars. Clubs which use two self-laying winches report very high launch-rateswhile benefiting from the advantage of the self-laying system.

There is probably no point in using the self-laying system with a single-drumwinch, as it will be too slow for anything but the smallest clubs with only oneor two gliders. However, if car retrieve is used, a single-drum winch can bealmost as quick in operation as a two-drum winch.

5.4. Protection of personnel

We have on record the death of a winch-driver due to the wire coming intothe winch cab through the cage when a cable-break occurred. We havelearned the lesson that a cage alone may not be sufficient to protect thedriver when things go wrong - in some winch designs the additionalprotection of polycarbonate or safety glass is needed too.

The more the wire can be isolated from the occupants of the winch (by basicdesign), the better. The German Tost winch, for example, is completelyenclosed, the drums being underneath the bodywork and the wires notseeing the light of day after they enter the front of the winch via the rollers. Itis not difficult to provide some covering for the wire in most winch designs,

but plenty of them are wide open, with the end of a broken wire able to flailaround causing all kinds of trouble.

A means of cutting the cable in an emergency must be provided. With effectfrom 1st January, 1999, this must be some form of guillotine capable ofbeing operated from inside the cab. Until that date, as an absoluteminimum, a pair of large bolt-cutters must be provided, to enable the winchdriver to cut the cable in the very unlikely event of it failing to release fromthe glider. The winch must not go out on the field without means of cuttingthe cable.

An earthing spike is a useful, if not essential, item of equipment. If athunderstorm approaches a winch-launch site, flying should cease before thestorm gets too close. There is no point in repeating an exercise alreadycarried out over a hundred years ago by Benjamin Franklin, who flew a kiteinto the base of a thunderstorm, a key being carried by the kite, a wirepassing the current back to the ground. It proved the point that theressparks in them thar clouds. As a storm approaches, a surprising amount ofstatic electricity can build up in the winch and it is prudent to earth this staticbuild-up. The alternative is to use the winch-driver for the purpose whenclimbing in and out of the winch, a function he/she will not enjoy performing.An earth spike is a much better idea.



37/68

32If the winch is to be used for training, both occupants of the cab shouldhave:-

(a) equally good protection against the ingress of flying cable and otherodds and ends; and

(b) equally good visibility from the cab to enable the glider to be kept directlyin sight by driver and tutor during the entire launch. It is not goodenough to rely on rear-view mirrors to attempt to train new drivers, assome clubs have done in the past.

6. AIRFIELD REQUIREMENTS FOR WINCH-LAUNCHING

6.1 Strip length

The minimum strip length for winch-launching is 1,200 metres (3,940 feet).

The purpose of this minimum length is to minimise the effects of the non-manoeuvring area (NMA), to create a buffer against density altitudeproblems in summer and to discourage any tendency to enter excessivelyst

winch manual

Documents