3Aerodynamic Theory

3.1Apparent-wind geometry at skiff speeds

Apparent wind is "the vector resolution of the TRUE WIND and the BOAT COURSE" S19; it is the only wind the sails, the telltales and the crew experience. In a 49er, boat speed is a large fraction of true-wind speed on every leg, so the apparent wind is always well forward of the true wind and often stronger than it. On a bicycle on a still day the only wind there is comes from the rider's own speed, and it comes from dead ahead. Ride across a light side breeze and the wind on the face swings forward of it, and the harder the rider pedals the further forward it swings and the stronger it feels. The 49er lives permanently in that condition, on every leg, and the sails have no access to the breeze the coach boat feels sitting still. Most of what the crew perceives as wind behaviour is therefore a mixture of true-wind change and boat-speed change, and the purpose of this section is to separate the two and to keep the resultant stable.

3.1.1Apparent-wind angle as the primary steering-and-trim reference. The sails generate force as a function of their angle to the apparent wind, not to the true wind, so the apparent-wind angle is the reference against which sheet and course are set. Upwind the helm steers to the jib luff telltales, a direct reading of the local apparent-wind angle at the entry. Downwind the reference is the gennaker luff. Ingham's observation that on an asymmetric boat "The apparent-wind angle basically remains at 90 degrees" S27S28 is made about sportboats rather than about this one, and what transfers from it is the principle and not the number: the apparent-wind angle is held nearly constant while speed and depth vary around it. The 49er's own downwind apparent-wind angle is crew value, and it is well forward of the beam at full pace. Speed answers a change in the true wind first and the course follows once the speed is there, which is the order sections 3.1.4 and 3.6.3 work in. Compass heading is a secondary record for shift detection, not the primary trim input.

3.1.2Apparent-wind velocity changes through gusts, lulls, waves, and acceleration. Apparent-wind speed rises with boat speed only while the apparent wind is forward of the beam, and upwind it always is: there "the apparent wind speed increases and the apparent wind angle decreases, moving further forward" as boat speed rises S26. With the apparent wind square to the boat, extra speed adds nothing to its strength and only swings it forward, and further aft than that it takes strength away. The condition is the sign of the cosine of the apparent-wind angle. A rise in true-wind speed has a condition of its own and a different one, how much of the boat's speed is running straight away from the wind: while that is less than the true-wind speed, which covers everything upwind and reaching, more true wind is more apparent wind. Only deep and fast, with the boat running away from the wind nearly as quickly as the wind itself, does more true wind stop meaning more apparent wind. Both conditions hold comfortably on the beat; both are worth remembering on the run, which is the one place either of them bites. A gust therefore arrives twice: first as a jump in true wind, which raises apparent speed and moves it aft, then as an acceleration, which raises apparent speed again and brings the angle forward. The two stages are separated in time, and the gap between them is where the sail sits at its highest angle with the crew not yet moved. A lull reverses both stages. A wave impact removes boat speed, so the apparent wind drops and rotates aft at the instant the hull needs power to re-accelerate. The gust spectrum quoted in section 2.5 sets the time scale on which these cycles run.

3.1.3Separating true-wind shifts from changes created by boat speed. A change in apparent-wind angle has three candidate causes, and the correct response differs for each. The test is the pairing of angle with speed and with pressure. When the apparent wind moves aft while the boat is slowing, or forward while it is accelerating, the cause is boat speed, and the response is to hold the course and let trim follow the speed. When a change in the strength of the true wind arrives with no change in its direction, it moves the apparent wind aft in a gust and forward in a lull before the boat has had time to accelerate or slow; that is the cause the previous subsection has just described. Its sign is that the angle change arrives together with a pressure change and then reverses itself as the boat picks up speed, so it is answered with trim and body weight and not with the tiller. Only when the angle moves with no pressure change and no matching speed change, or moves the opposite way to what the speed change predicts, is the cause a true shift, and then the helm steers to it. The compass and the speed record confirm afterwards: once the helm has steered to it, a true shift shows as a changed heading at the same apparent angle with no matching speed change, whereas a speed change shows in the speed trace alone, at constant heading.

3.1.4Preserving apparent-wind stability during transitions between modes. A mode change, pointing to fast upwind or high road to low road downwind, changes boat speed and therefore rotates the apparent wind. If the sails are re-trimmed to the new course before the speed has changed, the angle of attack is wrong during the transition, flow separates and the boat loses the speed the new mode depends on. Going lower is paid for in boat speed, and the payment has to be made first: the deeper course needs more speed to hold the same apparent-wind angle, so a bear-away taken before the speed is there buys depth briefly and gives back more than it bought. Ingham's rule that "Speed before depth is essential because once the boat is going well, the apparent wind goes forward and the boat can go lower" S27S28 is the general form: speed first, then course, with trim eased or firmed progressively so that the apparent-wind angle at the luff stays inside its narrow target band throughout.

3.1.5Gold-medal standard: the apparent-wind angle survives every mode change. An observer with the video and the speed trace can find every transition the boat makes, upwind pointing to fast and downwind high road to low, and see the same thing at each: the speed moves before the heading does, the trim moves with the speed rather than after it, and the luff telltales or the gennaker curl report the same angle at the end of the transition as at the start. The trace shows no dip in speed that the new mode then has to pay back, and the crew can say, for any change of angle in the record, whether it was a shift, a velocity change or their own acceleration.

3.2Effective angle of attack and attached flow

The force of the wind on a sail "is calculated as a lifting force perpendicular to the apparent wind direction and a drag force in the direction of the apparent wind" S19; section 2.3 says what drag is and why it decides the order the mainsail is flattened and twisted in. Of the things both depend on, the one this section is about is the angle at which the sail meets the flow. The angle that matters is the effective angle of attack: the local angle between the apparent wind and the sail's chord after every distortion of the flow field has been included. The argument is that the crew manages this angle and not the sail's appearance.

3.2.3Managing effective angle of attack rather than visual sail shape alone. Gladstone's three sources of upwind power are angle of attack, depth and twist S20, and the three are not independent: depth sets the camber and twist sets how the angle of attack varies with height. A sail that looks correct from the boom may be operating at the wrong effective angle, because heel, leeway, the downwash from the jib and the local wind gradient all alter the flow the cloth actually meets. The management rule is to trim on flow evidence, telltales, leech ribbons and the pressure felt through the sheet and the tiller, and to use the visual shape as a record for repeatability rather than as the target. A recorded shape says what the crew set; only the flow evidence says what the air is doing about it.

3.2.4 Flow attachment, separation onset, and stalled-sail recognition. Attached flow follows the curvature of both surfaces from luff to leech. As the effective angle of attack rises, the leeward flow first thickens and then detaches near the leech, and the detached region grows forward until the sail is stalled, lift collapses and drag remains high.

Lift against angle of attack is a curve with a peak in it, and rowing has the same shape. A blade squared a little more to the water takes a bigger bite, up to the angle at which it stops slicing and starts dragging, and past that the stroke is heavy and the boat does not go. The sail's angle of attack has a peak in exactly that way, and on the far side of it the load is still in the sheet and the drive is not in the boat.

A thin sail has a second way to let go, and the crew steers to it every day without naming it. Meet the luff at too high an entry angle and the flow cannot turn the corner at the leading edge itself: it separates there and reattaches a little way back, leaving a small bubble behind the luff on the leeward side. That is what lifts the leeward telltale nearest the jib luff, and it is the signal the helm steers to on the free side of the groove, arriving long before anything is wrong at the leech. The other side of the groove reports on the windward telltales, which lift when the boat is pointed too high. The two failures report at different ends of the sail, so each telltale answers its own question: the leeward telltale at the luff reports the entry angle, and the leeward telltales further aft report separation walking forward from the leech. On the water a stalled mainsail feels heavy on the sheet but produces no acceleration, the boat heels without speed and the helm loads up. A stalled jib shows its leeward telltales hanging and the boat sagging to leeward despite a full-looking sail.

3.2.5Restoring attached flow through sheet tension, vang, and steering. Recovery reduces the effective angle of attack, and each control does it differently. With the vang slack, easing the sheet lets the boom lift, so the leech opens and the angle falls from the head first while the foot deepens. Easing the vang increases twist so that the upper sail unloads while the boom position holds. With the vang loaded, which is how it is carried upwind at full power, easing the sheet swings the boom outboard and reduces the angle along the whole span at constant twist. The vang condition is part of every statement about the sheet, in this chapter and in sections 2.3 and 2.5, because without it the same ease describes two different changes.

Steering up reduces the angle across both sails at once, but it spends the speed that stalling has already reduced, which is why section 2.3 puts it second in the gust order and not first. The decision rule is to use the control that acts where the stall began. For a stall from the head that is the vang, or the sheet if the vang is already slack, since then the sheet opens the leech from the head too. For a stall along the whole span it is the sheet against a loaded vang, or the helm. A whole-span control used on a stall that began at the head cures the head and takes the angle out of every section that was working properly, and that is where the cost is paid.

3.2.6Gold-medal standard: the stall is answered at the control that owns it. An observer with the video and the sheet hand in frame sees every recovery begin at the control that acts where the stall began, and sees it begin while the telltale is still stirring rather than after the boat has heeled and slowed. The crew can say, for each recovery in the session, which telltale reported it, which control answered it and whether the vang was loaded at the time.

3.3Mainsail-jib interaction

The mainsail and jib form one aerodynamic system. The jib sits in the upwash ahead of the mainsail's leading edge, and the mainsail sits in the downwash of the jib, so each modifies the flow the other meets. The gap between them, the slot, is not a nozzle that speeds the air but a region whose width and shape decide whether the mainsail's leeward flow stays attached. Trim decisions are therefore made on the pair, and a change to one sail is evaluated by its effect on the other.

3.3.1Slot geometry as a system rather than two independently trimmed sails. The mainsail's presence turns the flow ahead of it, so the jib operates at a higher effective angle of attack than its setting alone suggests and carries more of the total lift. Neither sail meets the wind that is blowing across the course. Each meets the wind the other has already turned, and each is therefore set for a wind that the other one's trim can change without anybody touching it. The jib's presence turns the flow at the mainsail's luff, reducing the mainsail's effective angle there and letting it hold attached flow at a sheeting angle that would stall it alone. The slot is the geometric expression of this exchange.

Its width along the span must stay proportionate: a slot that narrows aloft chokes the mainsail's upper leeward flow, and one that opens too far aloft wastes the jib's upper span. Trim is judged by the pair's combined flow, not by either sail's appearance.

3.3.2Jib entry angle and its effect on mainsail attachment. The jib's entry angle, the angle of its forward section to the apparent wind, sets the helm's tolerance for steering error and, through the downwash, the angle at which the mainsail's luff meets the flow. A finer entry gives higher pointing potential and a narrow groove; a rounder entry gives a wider groove and forgiveness in waves. Forestay sag makes the jib fuller S33, and in the 49er the sag is governed by rig tension, so entry is a rig setting as much as a sheet setting. The tension in the shrouds that reach the hounds, graded with the wind as section 2.2 sets out, is the coarse entry control; the crew's own tension per band is crew value.

3.3.3 Mainsail leech position and pressure recovery behind the jib. The two sails do each other two different favours and it is worth keeping them apart. What the jib does for the mainsail is bend the flow at the mainsail's luff, which lowers the suction peak there and so leaves the mainsail less pressure to recover over the rest of its leeward side. What the mainsail does for the jib is give its leech a fast, low-pressure place to exit into. The mainsail's own recovery is the part the crew controls, because air that has been accelerated over a leeward surface has to slow again towards free-stream pressure before it reaches the leech, and the leech position decides how gently it is made to do so. A leech carried too far to windward, from oversheeting or too little twist, steepens that pressure rise and the flow separates from the mainsail's after part. A leech carried too far open lets the flow off the back of the mainsail early, so the after part of the sail carries much less load and the jib loses the fast exit the mainsail was giving it. Slowing down takes distance, and a motorway exit shows what happens when there is not enough of it: a long slip road lets traffic lose speed and keep flowing, and a short one has the cars at the front still braking hard when the ones behind arrive. Closing the leech shortens the slip road. The reference the crew works from is the one section 2.3 sets, and the leech is opened or closed from there according to leech-ribbon behaviour and the helm load the boat reports.

3.3.4Balancing jib-sheet tension, jib-leech tension, and mainsail twist. Three controls set the slot's shape along its height. Jib-sheet tension sets the foot and the lower slot. Jib-leech tension, governed by the clewboard hole, which the class tuning guide sets in the middle position for winds up to about 12 knots S13, sets the upper slot. Mainsail twist sets the leeward wall of the slot.

The balance rule is that jib twist and mainsail twist change together: when the vang opens the mainsail's upper leech, or the sheet does with the vang slack, the jib's upper leech opens with it, otherwise the slot narrows aloft at the moment the mainsail is trying to shed load there.

3.3.5Slot response in flat water, chop, and across the power bands. In flat water the slot is closed to its narrowest working width, because steering error is small and the boat can hold a fine effective angle. In chop the boat's pitching swings the apparent wind at the luff, so the slot is opened, the entry rounded and the twist increased to widen the groove. The groove is the amount of steering error the setup will tolerate before it loses flow, and chop changes the entry angle without consulting the helm, so the entry is set wide enough to absorb what it does. In the under-powered band the priority is power, so depth is added, twist reduced and the slot held as closed as attached flow allows, since "More depth creates more power and better acceleration" S20. In the over-powered band the slot is opened from the top through twist, keeping the lower sails driving while the upper sails shed the load the crew cannot hold. The band names here are chapter 2's and are used in that sense.

3.3.6Coordinated helm and crew trim language for repeatable changes. Because each sail changes the other, the helm and crew adjust in an agreed sequence with a shared vocabulary. Each call names the control, the direction and the size of the step in the crew's own units, for example a fixed length of jib sheet or a marked vang position, so that a change made in a race can be reproduced in training. The order of operations is fixed: the twist controls move before the sheeting angles, and the jib follows the mainsail. The reason is that twist decides how much load the top of each sail is carrying, and the sheeting angle is then chosen for the load that is left; set the angle first and every twist change afterwards makes the angle wrong again, so the same setting has to be made twice. The reference marks on sheets and vang are calibrated at the start of each block, and any change that cannot be stated in this language is treated as unrecorded and not carried forward.

3.3.7Gold-medal standard: know reference, fast-mode, and pointing-mode trim in every major wind range. An observer holding the crew's trim sheet can, in any wind band the crew races in, watch the boat set up and read off three states: the reference trim, the fast-mode variant and the pointing-mode variant, each with its jib-sheet mark, clewboard hole, pin position, mainsheet mark, vang mark, cunningham and outhaul settings recorded as crew value. The crew moves between the three states in an agreed number of moves, and the transition is complete before the boat has sailed a stated distance. The speed and heading traces from the logger show the expected change, and the observer cannot find a band in which the crew is improvising.

3.4Spanwise loading and twist management

Twist distributes lift along the span, from foot to head, and matches the sail's angle of attack to a wind that changes in speed and direction with height. Twist is needed because the wind is stronger aloft, which gives "a stronger apparent wind and a wider apparent wind angle aloft" S21. Twist is set at the vang, and the mainsheet reaches it only when the vang is slack, so the two act on the mainsail quite differently. This section separates them, so that the crew can choose the one that changes only what needs changing.

3.4.1Twist as a control of load distribution from foot to head. Gladstone's rule, "Increasing twist reduces power; decreasing twist adds power" S20, describes the aggregate, but the mechanism is spanwise. Lift made high on the sail drives the boat no harder than lift made low, but it heels the boat through a much longer lever. A bag of shopping weighs the same hanging from the handlebars as it does in a low pannier, and the bike is far easier to hold upright with it low down. The rig sets the same problem. The question is never only how much force the sails are making, but how high up they are making it. Twist moves lift down the span by reducing the angle of attack of the upper sections, so that at a given total heeling moment the boat carries more drive.

Reducing twist loads the head and raises power and heeling moment together. The crew therefore treats twist first as the control that decides where along the span the sail works, and only secondarily as a total-power control.

3.4.2Matching twist to wind gradient and sea state. Base twist follows the gradient. Gilbert's figures for a model-yacht rig, about 4 degrees of twist close-hauled and about 12 degrees with the wind abeam S22, give the relationship's direction for any rig: required twist rises as the wind angle opens, because the apparent-wind angle aloft diverges further from that at deck level. Sea state adds a second term: pitching swings the apparent wind at the head more than at the foot, so twist is increased in waves to keep the head attached. Heel is not on this list. It lowers the effective angle of attack, but it does so much the same way up the whole span, so it is a reason to flatten the boat rather than to re-twist the sail; skiff coaching guidance treats flat as flat, plus or minus 5 degrees acceptable S36.

3.4.3Leech stability versus excessive leech tension. A stable upper leech holds its angle to the apparent wind as the boat pitches and the sheet load fluctuates. Three separate things can go wrong and only the middle state is stable, which section 2.3 names from the rig's side. Too little leech load and the leech flutterswhy. Too much and the leech hooks: past a point some way above the one that stops the flutter, more tension closes the leech, raises the effective angle at the head, carries the top batten to windward and stalls the upper sailwhy. Between the two is the working range, and it is narrow rather than absent. Past the sail's own luff curve the vang stops tensioning the leech and starts bending the spar instead, and the leech then pumps open and shut with every change of load. The tension needed to sit between the three rises with wind speed, and the tight end of the range is not the safe one to err towards. The distinction is read at the top batten and top leech ribbon: a stable leech shows the ribbon streaming most of the time with the batten near parallel to the boom; excess tension shows the ribbon stalled behind a hooked batten and a boat heeling and slowing together.

3.4.4Top-of-sail depowering without losing lower-sail drive. When the wind exceeds what the righting moment can hold, the upper sections are the cheapest place to shed load, because they heel the boat most per unit of drive. That is a statement about cost per unit, not about order. Flattening still comes first, because it removes heeling moment without giving up any part of the sail, and twist comes second because it works by giving the top away: cheapest of the things that cost something. Section 2.3 states both orders in full, the one for a rising mean wind and the one for a single gust, and this chapter uses them as they stand there. Opening the upper leech through twist reduces the head's angle of attack while the lower sail, whose drive is cheap in heeling moment, keeps its camber and angle. The Ovington manual notes that "Luff tension induces mast bend which will reduce power to the rig" S12, and Flynn states that more mast bend flattens and depowers the mainsail S33; both act along the whole span, so neither replaces twist.

Diagonal wrinkles from clew to luff show over-bending S33 and mark the limit of the flattening route.

3.4.5Using vang and sheet independently to control twist and boom position. The mainsheet pulls the boom down and in together, so with the vang slack it sets leech tension and boom angle at once, in a ratio the crew does not set, and one that shifts as the boom travels. A second lever is what gives that choice back, and the vang is it: the vang pulls the boom down only, so it sets leech tension independently of boom angle, and downwind it controls leech twist on its own, because the mainsheet no longer pulls down S34. One lever working two wheels is a scooter with linked brakes; a second lever gives the rider the choice back, and that is what the vang does here. Upwind at full power the vang is loaded to its mark and sets the twist, and the sheet then sets the boom position for the mode, so the boom can move without the leech changing. That is the condition under which the next subsection calls the sheet the first control to move in a gust: it moves the angle, and the vang has already settled the twist. The Ovington manual's warning that "it is easy to over vang the mainsail which will induce more mast bend than luff curve" S12 sets the limit: vang beyond it flattens instead of twisting.

3.4.6Dynamic twist changes in gust response and wave impact. A gust raises the apparent-wind speed and rotates it aft, so the effective angle at the head rises before the crew moves; the flat-boat response is the gust order of section 2.3, an immediate ease of the sheet, which against the loaded vang lowers the angle along the whole span rather than opening the leech, then a course adjustment and a re-sheet as the boat accelerates and the apparent wind returns forward. A wave impact removes speed and rotates the apparent wind aft while the rig unloads, so the sail is briefly at too high an angle with too little flow; the response is a small ease to re-attach flow and a steer that holds the entry angle through the pitch. In both cases the sheet moves first, because it is the fastest control that costs nothing in speed, and the vang is reset afterwards at the next opportunity rather than during the gust.

3.4.7Gold-medal standard: retain stable upper leech and immediate post-disturbance acceleration. Viewed from a coach boat or on video, the top batten and upper leech ribbon hold a constant relationship to the boom through gusts, lulls and wave sets, with the ribbon streaming for as much of the time as the band section 2.3 gives allows, which is its open end rather than the hooked end section 3.4.3 warns against, and no visible flutter or hooking. After each gust or wave impact the speed trace shows the boat returning to its pre-disturbance speed within a number of seconds the crew has set as crew value, with no second dip from a late or oversized correction. Heel stays within the flat-is-flat band of plus or minus 5 degrees S36, and the sheet hand's movements are small and continuous rather than large and periodic.

3.6Gennaker flow and projection

The 38 m² gennaker S11 is the largest surface on the boat, and the wind it flies in is very largely made by the boat itself: the crew is "looking to really build the apparent wind over the gennaker from quite early on in the wind range" S29. It is not the strongest wind the boat meets, because upwind apparent wind is stronger still; it is the wind the boat has the most say in. It works as a lifting sail with attached flow over most of its span rather than as a drag sail. Treading water shows the difference between the two. A flat hand shoved downwards lifts a swimmer for as long as the shoving lasts, and it is hard work. The same hand held at a small angle and swept sideways lifts them with far less effort, because the force now comes from water flowing past the hand rather than from the hand pushing water away. The gennaker is worked the second way. Its efficiency depends on projection, on a stable luff, and on the coupling between sheet and course that the two details below describe.

3.6.2Luff stability, projection, and driving-force generation. Projection, in the sense the crew uses the word and the sense this book uses it in, is the gennaker rotated out from behind the mainsail with its luff to windward and forward, flying in clean air rather than in the mainsail's wake. A projected gennaker carries its luff clear of the jib and mainsail, carries a leech open enough for a clean exit, and holds its shape by the pressure difference across it rather than by sheet tension alone. It is a question of where the sail is and which way it faces, not of how much cloth can be counted from any particular direction.

The sheet sets the angle the sail meets the flow at; it does not hold the shape, which the air does. Drive is lift at an apparent-wind angle far enough forward that the flow crosses the sail rather than pushing it: Ingham gives about 90 degrees for sportboats S27S28, which carries the principle that the angle is held while speed and depth vary, and the 49er carries its own angle, crew value, forward of that. A luff that collapses and refills breaks the flow each time, so luff stability is the first requirement. So is a full hoist: a head that has not reached its mark lets the top of the sail fall away to leeward, out of the clean air it should be flying in, which from the boat looks very like a trim fault and is not one.

3.6.3Relationship between sheet tension, luff curl, and steering angle. Sheet, luff curl and course form one control loop. Easing the sheet reduces the angle of attack and lets the luff curl; steering lower increases the angle and re-attaches it. The target is a slight, continuous curl at the entry, held by a coordinated ease and bear-away as the gust arrives, and by the bear-away alone as the boat accelerates and the apparent wind moves forward, since an ease then would push the same way the moving apparent wind already does.

The curl is an edge detector. It is the only sign the sail gives that the luff is at its limit, so a sail showing no curl is reporting nothing at all and leaves the limit somewhere unknown. A working curl is not under-trim, and the two are told apart by sign rather than by feel. The curl is a small, continuous fold at the entry that comes and goes while the leech stays where it was. Under-trim has the leech rotating up and the head falling away to leeward, which is the failure the Ovington manual names: "Do not under trim the spinnaker as this will allow the leech to rotate up" S12. That unloads the sail and destroys projection, and it is the state section 2.6 is referring to when it says the gennaker is never under-trimmed. In extreme conditions the loop is inverted and the gennaker oversheeted to slow the boat, against the risk of pitchpoling when the boat ploughs into the back of a wave S29; this is the one home of that exception, and section 2.6 refers to it.

3.6.4Gold-medal standard: the curl is held through a gust and the gust is taken in depth. An observer on the leeward bow quarter sees a small curl appearing and disappearing at the luff throughout the leg, never a collapse and never a luff that is still for long stretches. Through a gust the trace shows speed building first and the heading following it down, and at the end of the gust the sail is back at the same curl on a lower course rather than at the same course. The crew can say for any ease in the record whether it was made to restore flow or to convert speed into depth.

3.7Aero-hydrodynamic coupling

Steady sailing requires the driving force to equal the resistance and the heeling moment to equal the righting moment S19. The sails supply the force, the hull and foils supply the resistance and the side force that reacts the sails' lateral component, and the crew supplies the righting moment. The four are one system: a change in sail force changes heel, heel changes leeway and foil loadwhy, and foil load changes the drag the sails must overcome. This section sets out the coupling and the decisions it forces.

3.7.1Sail-force vector and its effect on heel, leeway, and foil load. The sail force is the sum of lift perpendicular to the apparent wind and drag along it S19. Resolved into the boat's axes, a small forward component drives and a large lateral component heels and pushes sideways. The daggerboard and rudder generate side force only when the boat moves somewhat sideways S19, so the lateral component is reacted through leeway. Most racing boats sail upwind in medium breeze with 3 to 5 degrees of leeway S23, which is a general figure for racing boats and not a 49er figure; the crew's own measured leeway is crew value. A tyre works on the same terms. A car in a corner has its front wheels pointing slightly inside the line the car is actually travelling, and that small difference between where the wheel points and where it goes is what produces the grip. The board is the tyre and leeway is that difference. A boat crabbing not at all is a boat whose board is making no side force. Foil load rises with the lateral component and induced drag with foil load. Heel tilts the vector: part of the lateral force becomes a downward push on the hull, the effective angle of attack falls, and drive per unit of sail force drops.

3.7.2Matching sail power to available righting moment. Righting moment is crew weight applied over the longest available lever S25, and it is fixed by crew weight and wing width once both sailors are on the wire. Sail power is the variable. In the under-powered band the crew adds power, through angle of attack and depth and by taking twist out; in the over-powered band the crew removes power in the rising-mean order of section 2.3 until heel returns to the flat band. Low trapezing dampens roll and keeps the sail plan upright S25, so body position is treated as part of the righting-moment budget, not as a correction applied after trim.

3.7.3Using aerodynamic depowering to preserve hydrodynamic efficiency. Excess heel costs more than the drive it wastes.

A heeled hull presents an asymmetric shape to the water, the daggerboard's effective vertical span shortens and its induced drag rises, and the increased leeway pushes the foils' angle of attack toward their own stall. Depowering the rig before heel builds keeps the foils working at low leeway and the hull on its designed lines, where the forward V keeps the bow from pounding and the forward chine turns water down for dryness and lift S14. The rule is that the sails are depowered to protect the hull and foils, and the cost of doing so is measured in speed and height together, not in sail area given up. It is done before the heel builds rather than after, because once the foils are loaded and the boat is over, nothing that happens at the rig returns the height already given away.

3.7.4Preventing excessive rudder angle caused by sail-plan imbalance. A small, steady rudder angle is wanted, because a rudder carrying a little lift makes side force the daggerboard would otherwise have to make at a greater leeway angle. Angle beyond that is drag with no driving return, and it arises when the sail-force and foil-force centres are misaligned so that the boat turns unless held.

Raking the mast aft "creates weather helm" S32, and the North Sails target of 3 to 5 degrees of weather helm in moderate conditions S32 is a general figure for racing boats rather than a 49er figure; the crew's own is crew value. Beyond that band the sail plan is out of balance, usually because the mainsail is over-powered relative to the jib or because the boat is heeled.

Which of the two it is can be read from when the load appears. Helm load that comes only in the gusts is heel or an over-powered mainsail moving the centre of effort, and it is answered on the leg with twist and sheet. Helm load that is there all the time with the boat flat is a standing fault in rake or in the clewboard hole, and it is answered between races. Too much rake shows as a tiller loaded all the time; too little shows as a helm that goes dead and a boat that slides sideways without reporting it. Reading one as the other is how a crew comes to re-rake a mast to cure a trimming problem. The vang release before the bear-away, which section 2.5 sets out, is the same principle applied to a manoeuvre: unload the leech before the rudder has to fight it.

3.7.6Gold-medal standard: optimise sails, hull, foils, and crew as one system. An observer watching the boat through a wind band sees heel held within the flat band, leeway consistent with the foils working at low angle, rudder angle small and steady, and the crew's body height on the wire changing before, not after, the trim changes. When the mean wind rises the depowering sequence follows the order of section 2.3, flattening, then twist, then angle, with shroud tension between races; when a single gust arrives it follows that section's other order, sheet, then helm, then re-sheet. The speed and heading traces show no period in which the boat is both heeled and slow. The crew can state, for any observed adjustment, which hydrodynamic quantity it was made to protect, and the logged data confirm that height and speed moved together rather than one at the expense of the other.

3.8Aerodynamic performance process

Aerodynamic theory is only useful if it can be tested, and testing requires a reference, a comparison, a record and a decision. This section defines the process that turns trim experiments into settled knowledge. It relies on the loggers and recording equipment the class rules permit, on video, and on the crew's own observations, and it treats every conclusion as provisional until it has been reproduced. The crew's own logged values are the baseline throughout, and the process exists to make those values trustworthy.

3.8.1Establishing reference trims before each training block. A block begins by fixing the reference trim for each wind band: rig tension, clewboard hole and pin position, sheet marks, halyard marks, vang mark, cunningham and outhaul positions, recorded as crew value. Rake is recorded with them, as a logged value rather than as a band setting, for the reason section 2.1 gives. Rig tension is checked against the crew's own logged readings with the mast stepped and no sails, crew value on each wire, taken on the same gauge every time; section 2.7 sets out why the gauge's kilogram scale is not valid for the wire this boat carries and why repeatability is the only property this check actually needs. The reference is sailed first in every session so that the day's conditions are characterised against a known state. Any change to be tested is defined as a single departure from the reference, so that its effect can be attributed. The reference exists so that every test has one variable in it and one thing to undo; two changes made on the same morning can produce a boat that feels better and nothing that can be put back on its own.

3.8.2Comparing modes against benchmark boats or performance baselines. Two-boat testing is the primary comparison: a benchmark boat holds the reference trim while the test boat sails the variant, in the same water and for long enough that the wind's cycles average out, and the roles are then swapped. Without a benchmark boat the comparison is against the crew's own baseline, built from log files as Njord builds polar targets and gain-and-loss comparisons S31. Published 49er polars by wind band were not found, so the crew's own targets by band are the reference, recorded as crew value. GPS data from an Olympic-class regatta show mean velocity and VMG increasing with wind velocity on both legs S30. A run is therefore comparable only with another run in the same wind, which is why every baseline is indexed by wind speed before it is compared. What a band means is set out in section 1.2 and in section 2.6: a band is the stretch of wind over which the same quantity is the one in short supply, and the crossovers are the crew's own logged values rather than fixed knots. The wind itself is sampled from the coach boat, which section 16.4 makes responsible for the observations the crew cannot take from the boat. How a run's wind speed is obtained, at what height and over what interval, is not settled anywhere in this book, and until it is, a baseline indexed by wind speed is only as repeatable as whatever the crew did that day.

3.8.3Using video, speed traces, headings, and crew observations together. No single record explains a gain. Speed and heading traces from a logger such as the Vakaros Atlas 2, which logs at 2, 5 or 10 Hz and reports VMG and shift tracking S79, or the Sailmon MAX, which records boat speed, heading, heel and pitch S80, show what happened; video from the coach boat shows the sail shapes and body positions at the time; the crew's observations supply the reason the change was made and what it felt like through the sheet and the tiller. The analysis aligns the three on a common clock, and a finding is accepted only when trace, picture and account agree. All three depend on what may be carried. Class rule C.5.2 permits camera and sound recording equipment on board S3; whether a speed logger may be carried is a separate question, and not one that rule answers.

3.8.4Distinguishing repeatable gains from one-off environmental effects. A speed difference in a single comparison may come from the variant, from a difference in wind between the two boats, from a wave set, or from the wind's own cycle, in which gusts carry 30 to 40 per cent extra speed S50. Repeatability is established by reversing the boats' positions and roles and by requiring the gain to appear on both sides of the swap.

Two things have to be said plainly, because without them the process has no acceptance test. The first is how long a run must be. The oscillation that matters here is the directional one, which runs from about 3 minutes in heavy air to 10 minutes or more in light-medium air S50, and not the gust cycle of 6 to 12 seconds, so a run that is to average the wind out lasts minutes and not seconds. The second is how large a gain has to be before it counts. The rule the crew can apply without any statistics is this: before the variant is tested at all, the reference is sailed against itself for two or three paired runs, and the spread between those repeats is written down. That spread is the noise of the day. A variant is accepted only if it wins by more than that spread, and only if it does so in at least four paired runs, two on each side of the swap. A gain smaller than the day's own scatter is not a small gain; it is no measurement at all. A gain that appears only on one side of the course, only on one tack or only in one wind band is recorded as conditional and is tested again under those conditions before it is admitted to the reference.

3.8.5Converting findings into concise on-water cues. A finding becomes useful when it is reduced to a cue that can be spoken in a few words and acted on without deliberation. The cue names the trigger the crew will see, the control that responds and the size of the response in the crew's units: for example a stated change in top-leech ribbon behaviour that calls for a stated vang step. Cues are written in the same language as the trim calls of 3.3.6, tested in training for a full block, and dropped if the crew cannot reproduce the response under race load. The cue list for each band is kept short, because a long list is not executed under pressure.

3.8.6Gold-medal standard: every adjustment has a trigger, purpose, outcome, and evaluation method. An observer reviewing a session with the crew can take any trim change from the video or the trace and hear the crew state four things without hesitation: what they saw that triggered it, what aerodynamic or hydrodynamic quantity it was meant to change, what the trace and the sail showed afterwards, and how they decided whether it worked. Changes that fail the fourth test are logged as untested and are not repeated in racing until they pass. The trim sheet for each band carries a revision history, and every entry cites the session and the comparison that justified it.