Energy Management in the Descent

This article was written by Steve Hall, VATNZ Pilot and Controller.

Section 1: Introduction

During my time as a VATNZ controller I have seen many "interesting" and creative descents and arrivals.  The aim on this brief tutorial is to assist you in  understanding how to manage the descent of your aircraft.  This is primarily aimed at the beginner to intermediate pilot who perhaps is having a few issues especially when it comes to flying the "heavy tin" aircraft.

The basic principals of flight are the same for all fixed wing aircraft so the information and suggestions here can be applied to all fixed wing aircraft from small Cessna to a Boeing 747.

In order to understand what is happening to the aircraft in a descent let us first revisit the basic principals of flight.

Section 2: The four forces

The four forces acting on an aircraft in flight are:

• Lift
• Weight
• Thrust
• Drag

We can represent each of the four forces by vectors.  A vector has both direction and magnitude (in other words the length of the vector determine the amount of force present).

ThrustWell this one is a no brainer.  Thrust is created by the propulsive unit.  It could be a propeller, jet or a combination of the two (turbo-prop).  The force acts along the centre axis of the engine (crankshaft in a piston engine) and centre shaft of a jet and is called the thrust line.

Drag
Drag is the rearward acting force which resists the forward movement or the airplane through the air.  It either has to be overcome somehow (engine) or used to our advantage to help slow us down and make a manageable descent and glide angle.

Every part of the aircraft which is exposed to the air while the airplane is in motion produces some resistance and contributes to the total drag.

Total Drag may be classified into two main types:

• Induced Drag
• Parasite Drag

Induced Drag:
Ironically induced drag is the by-product of generating lift and is related to the aircraft's angle of attack.

The higher the angle of attack the great the induced drag.  The airflow around the wing is deflected downward, producing a rearward component to the lift vector which is induced drag.

Higher angles of attack are used at slow speeds so induced drag is greatest at slower speeds and reduces with increasing speed.

Parasite Drag:
This in itself comprises of many different forms of drag but suffice to say that it is the sum of things such as the aircraft's frontal area, streamlining or lack thereof and skin friction.  You can simulate the effects of parasite drag by holding your hand out of the window of a moving car.  The drag is considerably less with your hand held out with the palm facing downwards than it is with it facing the airflow.  That rearward force is drag.

All drag acts parally to the relative airflow (relative wind).  Parasite drag increases with increasing speed therefore, for a given speed the Total Drag (Dtot) on an aircraft is the sum of induced drag and parasite drag.  We can control drag with airspeed, speed brakes (spoilers on the wing), prop pitch, reverse thrust in flight (some aircraft), flaps, side slipping and retractable undercarriage.

Lift:
This is a science in itself and without going into too much detail it can be considered as the upward force created by an airfoil when it is moved through the air.  Generally it is the wing that produced lift but other parts of the airframe can contribute to lift as well.  I can recall that a popular training aircraft in the 70's was the Victa Airtourer.   A wicked beast, it had a large bubble canopy that was claimed to produce 30% of the total lift!

An airfoil is the cross section of a wing, as if you cut the wing thru in fore and aft plane.  The lift generated by a specific airfoil is not uniform.  Some lift is generated at the beginning of the airfoil and it increases sharply to about a third of the way along and then it reduces again towards the trailing edge.  The resultant position of all that varied lift is referred to as the Centre of Pressure and is located approximately a third to almost 40% of the way back along the airfoil from the leading edge.  The Centre of Pressure moves back and forth with angle of attack.  As the angle of attack increases it moves progressively forward until at the poiunt of stall it moves rapidly rearwards.  It is this rapid movement reard (as well as reduction in lift) that causes the nose of the aircraft to lwoer and thus regain flying speed (a desirable design feature I'm sure you'll agree).  The amount of lift generated by the wing depends upon several factors:

• speed of the wing through the air
• angle of attack - see above
• the lift coefficient - simply the shape of the cross section of the airfoil.  Generally the thicker the wing the more lift it can produce
• wing area
• the plan form of the wing - i.e. tapered wing, sweep back or sweep forward
• the density of the air

For a typical aircraft we can change the first five:

• Speed - We can vary the speed of the aircraft by adjusting nose attitude or power setting.
• Angle of Attack - We can vary the angle of attack with a combination of power and elevator control and also by using flaps or leading edge slats or flaps.
• Wing Area - We can change the wing area by using fowler flaps (they extend out and down so increasing the wing area).  The F1-11 and F15 fighters can sweep the wing.
• The lift co-efficient - We can change using flaps, slats and spoilers.
• Wing Area - Wing area we can control once again using fowler-type flaps.  Take a look at a modern airliner and you will notice that the trailing edge flaps are slotted.  A B747 has a triple slotted trailing edge flap.  This places a whole series of small 'wings' into the airflow and so generating more area therefore more lift.

From a generalised point of view we can say that:

Life = Angle of Attack x Airspace.

Both lift and drag increase to the square of airspace.

That means for a given angle of attack if we double the speed then the lift and drag will increase by a factor of 4.  This becomes more relevant later on when we look at energy management techniques.

Weight:

A no brainer here.  This is the product of the mass of the aircraft and the effect of gravity.  The only way we can control weight airborne is to burn or dump fuel.  Jettisoning baggage or passengers is frowned upon within the aviation industry.   Weight acts through a point called the Centre of Gravity.

The Centre of Gravity is determined by how we load the aircraft.  The further aft we place the load the further aft the C of G goes.  We are specifically interested in where the CG is located in relation to the wing.  We usually express the CG as a percentage of the MAC (e.g 23% MAC).  MAC stands for Mean Aerodynamic Chord.  In this case the CG positioned 23% along the chord from the leading edge (about 1/4 of the way back).  The chordline is a line drawn between the leading and trailing edges.

If you have a tapered wing or sweptback wing then the chordline length is going to vary depending on the spanwise position.  Therefore a mean chordline is established (sort of a declared average) and is used when determining the position of the CG and the COP.

The CG and COP have a coupling effect on each other and so it is important that the CG is located within a defined narrow band.  Undesireable flight characteristics can result otherwise.  Once an aircraft is loaded the CG is onlt really varied by fuel burn although this is minimal, as the fuel tanks are usually in the wing and have little effect on the CG.  Some aircraft such as the Concord use longitudinal fuel transfer to vary the CG within very fine limits to be capable of supersonic flight.

Let us take a look at an aircraft in straight and level flight:

Here our aircraft is in steady straight and level flight.  Note that:

• Thrust equals Drag
• Lift equals Weight

This is called being in a state of equilibrium.  Notice that the lift vector is located behind the weight vector.  This is referred to as the lift/weight couple.  The effect of this is to create a slight turning moment in an anti clockwise (nose down) direction.  Also the thrustline is located slightly below the drag line (exaggereted here) which is a slight clockwise moment.  The resultant anti-clockwise turning motion is counteracted by the down force generated by the tail plane.  This is why it is there!  In most aircraft with low thrust lines, increasing thrust causes the nose to pitch up and decreasing the thrust causes the nose to pitch down (because the resultant turning moment favouring the thrust/drag couple).  This is particularly desirable if we lost power as if we were to do nothing the aircraft will automatically pitch nose down towards the gliding attitude rather than pitching up towards the stalling attitude.

What happens if we want to fly faster?

To fly straight and level faster we need to apply power to increase thrust.  Doing this does two things:

1. Increases the thrust vector (causes acceleration)
2. Increases lift (remember lift increases at the square of the speed)

If we did nothing the aircraft would fly faster but would also climb.  In order to remain level we need to reduce the amount of lift generated.  We do this by reducing the Angle of Attack.  Remember from above: Lift = Angle of Attack x Speed.

To fly level lift must equal weight.  If we increase speed then we can see from the simple formula  that we need to reduce the angle of attack to keep lift constant.  We do this by lowering the nose with the elevator and then trimming out the resultant control column force with forward elevator trim.  As the aircraft gains speed the drag also increases as the square of the speed so our drag vector will eventually equal the thrust vector and the aircraft will stop accelerating.

What about flying slower?

Well generally the reverse applies.  The thrust is reduced which makes the drag vector larger than the thrust vector.  The drag slows the aircraft down until a point is reached whereby thrust equals drag and a state of equilibrium is reached.  However, we need to take a closer look at this:

As speed reduced the total drag reduces until the induced drag component takes a hold and drag increases again.  So in order to maintain level flight at slow speeds we need to increase thrust again to equal the value of the drag.  Coincidently the bottom of the orange drag curve is called min drag speed and is useful when determining aircraft endurance and holding speeds.

Section 3: Forces in the Descent

In a jet we generally close the thrust and carry out a glide descent.  This is what we will look at here.

Here are the four forces that we looked at in straight and level flight, acting in the descent (assuming a constant glide angle, speed and rate of descent).  The obvious question that comes to mind is if the thrust is reduced why doesn't the aircraft just slow down because drag is much larger than thrust?

Let us start by looking the lift vector.  When we lowered the nose for the glide the lift vector (which acts 90 degrees to the relative airflow) has been inclined forward.  Because the lift vector has tilted it has two components:

1. Vertical Component of Lift (opposing weight)
2. Horizontal Component of Lift

The vertical component of lift is less than the weight.  Ah.. so the aircraft should accelerate down? Wrong!  The reason is because the drag component also has a vertical component to it which acts in the ame direction as the vertical component of lift, which when added to it is exactly equal to weight.  The red 'R' vector in the diagram now equals the weight vector.

Now let's look at the thrust and drag.  When we reduced power for the descent thrust was less than drag.  If we just held level flight the aircraft would slow up and if left unchecked the aircraft would stall.  To maintain speed we lower the nose.

Now the weight vector has two components that we are interested in:

1. Acting opposite to lift and;
2. Acting in the same direction as thrust.

We call the latter component the Forward Component of Weight.  This FCW (when added to the residual thrust from the idling engines is equal to drag).  In other words, we are using the weight of the aircraft to provide the 'thrust' to counter the drag.

Let's take it to its extreme... a vertical dive.

Initially the Weight would be far larger than Drag so the aircraft would accelerate. As the speed increased so would the Drag until eventually the Drag would equal the Weight and the aircraft would stop accelerating (Of course the ground would be a great help here too!). This is called Terminal Velocity. The greater the drag, then the slower the vertical speed shall be. Of course parachutists use this to their advantage during freefall. They control their descent rate by either being stable with their arms and legs spread out, or, if they want to descend faster (to say catch up and join a formation that is below them), then it is arms and legs in and head down. Reduced drag. Of course later on the chute is deployed creating immense drag thereby reducing their vertical terminal velocity sufficiently to allow for a safe touch down.

Glide Angle:

The glide angle is the angle formed between the lift vector and the vertical. You can also see that there is a relationship between the Lift and Drag vectors that form this angle.  The ratio of these 2 forces is known as the Lift/Drag ratio. The greater the lift, compared to drag, then the greater the LD ratio.  The greater the LD ratio then the shallower the glide angle will be. An aircraft with an LD of 12 to 1 would mean that it would glide 12 units forward for every 1 unit of loss in height. The LD ratio is a function of speed and therefore the best LD ratio speed is also the best Glide speed.

If you fly faster or slower than the best glide speed (LD ratio) then your glide angle will be steeper.

What about the effect of Weight on the descent?

If we suddenly increase the Weight then we also increase the forward component of Weight acting in the same direction as thrust. This would cause the aircraft to accelerate until Drag equaled this new forward component of Weight. Lift would also increase due to the faster speed and the vertical component of this increased lift vector would equal weight. Eventually a state of equilibrium is reached.  The interesting point is that the Glide angle remains the same. We just travel at a higher speed down the descent path. As we are traveling down the same glide path faster we must also be descending at with a faster rate of descent (Ft per min).
 
We can see this when we fly an ILS. A heavy aircraft will fly faster down the slope (and be harder to slow up) than a lighter one.
 
If we flew the descent in a heavy aircraft at the same speed as a lighter aircraft then we would have to descend earlier and therefore at a flatter angle.  So as long as an aircraft descends at the correct speed for the weight then the glide angle will be the same. The heavier aircraft will sink faster but it is also traveling forward at a faster speed to will reach the ground at the same point as the lighter aircraft descending at a lesser rate and lesser forward speed.

Section 4: Other effects on the descent

Wind:
Wind has no effect on the descent rate but it will affect the glide angle relative to the ground.  A headwind will reduce the forward speed of the aircraft across the ground. This means that it is not traveling as far across the ground as in still air so the glide angle would be steeper. To take it to the extreme an aircraft flying at 80 kts IAS into an 80kt headwind has 0 groundspeed. It would therefore descend vertically!
 
The opposite happens with a tailwind. Our groundspeed will be much higher and therefore we cover more distance across the ground as we descend thus reducing the glide angle.  This needs to be taken into account when planning the top of descent point.  It is also important when trying to fly a fixed angle descent such as an ILS approach.

Let us look at landing on Runway 02 at Christchurch Airport with a 10knot headwind as an example.

The ILS is three degrees and we will be flying it at 120 kts.  On a calm day we would be descending down the glideslope at 600 fpm (1/2 Groundspeed x 10).  With our 10 kt headwind our new groundspeed would be 110 kts.  The rate of descent required now is only 550 fpm.

If we elect to land on Runway 20 we would now have a 10 kt tailwind.  Groundspeed is 120+10=130 kts.  The rate of descent would now need to be 650 fpm to stay on the same three degree slope.  This also means that you will be approaching the ground at a much higher rate even though you are still on a three degree glide slope... Something to consider when flaring.

Thurst:
Some jet engines have a higher residual thrust than others and this will have an effect on the descent angle.  This more the idle thrust the shallower the descent angle.  Increased thrust = Decreased descent angle and rate.

Icing:
There are two things to consider here:

1. Ice accumulation.  It increases drag and also weight.  It may also affect lift if ice is on the wing.  You may need to descend faster.
2. For jet aircraft extra thrust may be required to produce sufficient bleed air to the anti-icing systems.  This will flatten out the glide angle.

Section 5: Managing the Descent

Calculating the top of descent point:
If you have an FMC then this is easy as all you have to do is program it.  Make sure that you program in what you expect to get from ATC.  Put in the STAR and the approach and check/enter speeds and altitudes at waypoints where necessary.  Avoid high speeds on approach so enter a manageable speed at the final approach point. If you have no idea what approach you expect ATC to issue at the destination then enter a speed and altitude over the destination so that the FMC can at least work out a crude TOD point and you don’t get into trouble; a sort of safety net. If you are most likely going to join straight in then put something like 140/1000 over the destination airfield or aid. If going direct to the airfield but expecting to be vectored onto a base for the approach then put in say 180/3000. Once you are given a STAR and approach then enter these in.

If your FMC is capable of it then enter your forecast descent winds. This will fine tune the descent calculations. Even if you don’t know what they are take a note of the wind value you have at your current altitude and put half that wind value against half your altitude. For example if the wind at FL360 is 240/60 then enter a wind of 240/30 at FL180.
Some FMC’s have the ability to enter an anti ice ON altitude. This is to make allowance for the extra thrust required for anti icing and the effect it will have on the profile.

If you do not have an FMC then you need to calculate your descent manually. For jet aircraft and pressurised turboprops work on 3 miles per thousand feet of height to lose.  Subtract the airport elevation from your cruise altitude and then divide by 3 to get the number of track miles until landing. If you are using the GPS then ensure that the approach is loaded and look at the cumulative distance. You will use this distance readout to monitor your TOD point.
 
If you are using just a DME then you need to do a little fudging to get a distance.  You also need to note where the DME is in relation to the runway as not all DME’s are close by. Take the NN DME in Nadi, Fiji for example.
 
Example
NZWN to NZCH at FL260 and landing on RWY20

Calculate descent distance:
Christchurch elevation is only 123 ft so is not worth considering.
26000/1000 x 3= 26 x 3 = 78
 
We need 78 nm to descend from FL260.
As the approach is more or less straight in then we will use the CH DME to initially determine our distance as it is located pretty much on the field. As we get closer use the IHW ILS DME as it is more accurately positioned for landing.
We now need to determine:

• Wind on the descent
• Aircraft weight
• Speed restrictions if known
• Icing
• Forecast turbulence (slower speed)

As a general rule add or subtract 10 % of the Headwind or Tailwind component (as required) to allow for wind. If you had a 40 kt headwind then reduce the 78 miles to 74.
 
If you are heavy descend a few miles earlier.
 
In the real world I would also allow an extra 5 to 10 miles to slow the aircraft up for the approach etc but I have noticed that the aircraft in Flt Sim descend much more quickly so I don’t bother.  If you are flying a new aircraft type or are unsure then add a few more miles.
 
So at 74 nm CH DME we commence the descent at the recommended descent Mach or IAS as appropriate.
 
What about RWY02?
Well here we need to make some assumptions to best calculate the TOD point. It is quite possible that we are going to track to YW and then vectors for the ILS 02 commencing at BU NDB.

 
YW has a DME so we will work out a TOD based on YW.
Looking at the 02 ILS approach plate we can see that from YW we will track 182 degrees for 17 nm and then turn left to BU which is 8.7 miles from touchdown. Roughly eyeballing the chart I would say that after the 17 nm from YW there would be about 4 to 5 miles until reaching BU. So lets add that mileage up.
 
17nm - YW to the commencement of base turn.
4nm - Base turn to BU
9nm - BU to touchdown (approx)
30nm in total
 
30nm represents 10000 ft of altitude so we could cross YW at 10,000ft and still land if we tracked as planned.
 
From the RWY 20 calculation we determined that 74nm is needed for the descent.
If 30 nm are needed from YW then we need 74 – 30 =44 nm from YW as the descent point. Quite a lot closer than when landing on RWY20.
 
From a practical point we may not fly out the whole 17 nm if ATC are on watch. They may vector us in closer than 17nmYW so I will make an allowance for that.
I will plan to cross YW at 7000ft and not 10,000ft. This will give me some flexibility and equates to about 9 nm.
So with this I will now make my TOD point on being 7000ft at YW.
Therefore:
(26000-7000)/1000 x 3
= 19 x 3
= 57 nm
 
Allowing for the wind, as in the RWY02 example above, I will make it 53 nm.

 
 
So at the calculated descent point we will reduce thrust and commence the descent at the recommended descent speed. For jets this would typically be in the order of M.70 to M.85 and 280 to 320 kts. Consult the documentation for your aircraft.  From here we will need to monitor the descent to ensure we remain on profile. What we are looking for is a trend away from the calculated descent path.  To do this we need to subtract 7000 from our current altitude and multiply it by 3 and add any additives we had calculated.
 
So at 18000ft we should be:
18000 – 7000 = 11000
11000/1000 x 3= 33 nm
33 + 4 = 37 nm
 
We then check our DME from YW and compare it to our mental calculation. If the distance is more than 37 nm then we are going below the ideal profile and if it is less then we are above it. If we were say 40 nm then the diff is 3 nm which equates to 1000ft of altitude. We are therefore 1000 ft below profile.
 
For an FMC equipped aircraft we would monitor the path to make sure we are on it. For a Boeing aircraft we should be indicating VNAV PTH. If it shows VNAV SPD then we are NOT on the path and we will have to do something to regain it. The path indicator on the pilots flying display (PFD) will show where the computed path is in relation to the aircraft (similar to the glidepath indication on an ILS indicator) and Progress page 2 will give a read out of the actual distance in ft from the path. Progress page 1 will also readout the track distance to the arrival airfields as well…….very useful and a good check to make sure there is no gross errors in the programmed descent.

Methods to manage aircraft energy to maintain or regain profile:
The method used to maintain or regain a profile is largely dependant on the degree of displacement from the profile, the speed of the a/c and the distance remaining until touchdown (track distance).
 
Generally (for heavy tin) above 10,000ft use speed to alter the rate of descent. From what we have learnt above if we alter the speed of the aircraft we will alter the drag and therefore the glide angle.  If we increase speed we will increase drag and the angle will be steeper.  Reducing speed will decrease drag and flatten the angle.  The amount of speed change required will be dependent on the magnitude of the variance.  There comes a point where the speed may either become unacceptably slow or fast.

Too slow
If this case add some power and maintain speed. Have a look at the before and after affects on the rate of descent using the VSI.  Continue to monitor the next few thousand feet for a trend. What you want to see is a gradual return to profile.

Too fast
If you are too fast or very high or need to get down in a hurry then extend the speedbrake. Keep it extended with the higher speed until you are back on the path or slightly below it as you will want to resume normal descent speed which will require a little slowing up and consequent shallowing out of the descent.
 
Below 10,000ft you will need to be at 250 kts unless cleared by ATC.  If you are still quite high by 10,000ft you could ask for high speed and this will help you regain profile.  Speed brake is not as effective at slower speeds.

Another method at low altitude is to reconfigure early. This means putting flap and, if necessary the gear, out earlier than would be done normally.  Reconfiguring early does 2 things:

1. Reduces closing speed
2. Increases rate of descent and glide angle

If we were descending at 240 kts we will be traveling 4 miles every minute. If we reduce our speed to 180 kts then that will be 3 miles per minute.  If we have 20 nm to run then at 240 kts we will take 5 min to cover the ground and;
at 180 kts it will take almost 7 min. This gives us an extra 2 min of descent time. Sure we will take some of that time to slow up but with the flap coming out this will not take too long. The extra drag of the flap will give us an increased rate of descent.  Another option is to lower the gear earlier and at a higher speed than normal (observe gear operating speed limitations). The extra drag created by the gear, especially at higher speed, will greatly increase the descent rate and glide angle. An extra 1000 fpm over 2 min equates to 2000 ft of altitude recovered. On a 747 dangling 18 wheels out in the breeze is a huge help!
 
In an FMC equipped aircraft be prepared to use another pitch mode if necessary. VNAV is Ok as long as all goes to plan. If you get a shortcut by ATC then enter this in the legs page so that the descent path can be updated. FLCH is a good mode for terminal airspace work changing to VERT SPD for close in fine tuning close to the ILS or other intermediate/final approach fix

Summary:
Calculate the descent distance required and apply that to a suitable DME/GPS point etc to calculate the TOD point.
 
Monitor the descent by using the 3 times the altitude rule of thumb. Add a little extra for deceleration and reconfiguring (especially for straight in approaches).
 
Be proactive in modifying your descent. Do it earlier rather than later.  Try changing speed first. The more the variance from profile the higher the change in speed required. Monitor the changes on the VSI and adjust as necessary.  Be prepared for altimeter changes descending thru the transition layer. Could be a big change when going from Flight Levels to altitude depending on QNH.
 
Use thrust and speed brake if extra action is required.
 
At lower altitudes reconfigure earlier – flap and gear.
 
The more you fly your aircraft the more you will become familiar with its operating characteristics and you will then be able to apply your own additives to the 3 times the alt rule.  Try experimenting with speed and see what affect it has on your aircraft.