See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/277426761

Lift and Escalator Energy Consumption

Conference Paper · September 1996

CITATIONS READS

3 9,377

1 author:

Lutfi Al-Sharif
University of Jordan
129 PUBLICATIONS 557 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Design of Elevator Traffic Systems for High Rise Buildings View project

Notes in Mechatronic, Electrical and Mechanical Engineering (NIMEMEs) View project

All content following this page was uploaded by Lutfi Al-Sharif on 31 May 2015.

The user has requested enhancement of the downloaded file.

 LIFT & ESCALATOR ENERGY CONSUMPTION

LUTFI AL-SHARIF, B.Sc., M.Sc., Ph.D., C.Eng., M.I.E.E., D.B.A.

London Underground Ltd., 30 The South Colonnade, London, E14 5EU

Proceedings of the CIBSE/ASHRAE Joint National Conference,

Harrogate, UK, 29 Sept. - 1 Oct. 1996, (pages 231-239, Volume I).
ABSTRACT
This paper discusses the general characteristics of the energy consumption of lifts and
escalators, outlines the factors affecting it, and discusses the methods used to derive
daily and annual consumption.
The efficiency of the motor and the gearbox play a major role in deciding the
efficiency of the complete system, and systems which run under utilised will be running
at a low power factor. In lift system the variable voltage variable frequency drive
consumes the least energy, while the hydraulic system consumes the most energy. A
general method for deriving the daily energy consumption for lifts can be used which
depends on the type of drive, the number of starts per day, the number of floors and the
motor size. Lifts usually will regenerate power back into the supply, and the amount
regenerated will depend on the parameters of the specific system.
Due to the fact that escalators run continuously regardless of passenger demand, they
consume a fixed amount of energy if not boarded by passengers. Two methods are
outlined to calculate these fixed losses. Depending on the number of passengers using
the escalator, and whether it is running in the up or down direction, the final total energy
consumed can be calculated by adding or subtracting the two quantities.

1. INTRODUCTION
The approach towards understanding the energy consumption of lifts and escalators is based on three principles:
Outlining the general characteristics of the consumption; the factors affecting it; and the calculation of the total
energy and its cost. This paper will attempt to highlight some of these points, and point the direction in terms of
others.
At the start, it is worth mentioning the following two points:

• The efficiency of the motor and the gearbox will have an impact on the overall efficiency of the system,
and it usually depends on the size of the motor. Hence, the motor and the gearbox are the major
component through which all the energy transfer processes will take place.
• A lift or escalator system which is under-utilised will run at a very low power factor (for AC power
supplies), and this will have an effect on the cost of the electricity (depending on the tariff structure).
However, this problem depends on the tariff structure for charging for the cost of reactive power, and has
been deemed to be beyond the scope of this paper.

The main reason for needing to know the energy consumption of lifts (or escalators) is to calculate the cost of such
consumption, and to reflect that in the design of the system. However, in some cases, the energy consumption of a lift
or escalator can be used for passenger surveying purposes [1],[2].
This paper has addressed the main issues involving the topics of lift and escalator power consumption. The
full details and derivation can be found in [1,2,3] and [4].
In some cases the terms power and energy although distinct have been used interchangeably for
convenience. However, the reader has to be aware that power is the rate of transfer of energy.

1
2. LIFTS
This section discusses several topics related to the energy consumption of lift systems. There are different estimates
for the percentage energy dissipation of lifts as a percentage of the energy consumption of the whole building (5-15%
of the total costs, depending on what other services are running in the building). KONE estimates it at 5-10% [5],
whereas Schroeder [6] estimates it at 15% of the total building electric energy consumption (assuming no air
conditioning or oil heating). Moreover, Schroeder [4] also estimates it at 1% of the total cost of the rental of the
office space. The difference between these figures shows the difficulty in applying them, and depends on the actual
application.

2.1 COMPARISON OF DIFFERENT TYPES OF LIFT DRIVES

Doolaard [7] compared the energy consumption of various types of drives. He compared the following four types of
lift drives:

• Hydraulic lifts
• Two speed lifts
• Eddy current braking systems (DC Injection braking)
• Variable frequency drives

Energy consumption of various drives (Doolaard, 1992)

350

300

250 Hydraulic

200 Two speed

150
DC
injection
100 braking
VVVF

50

0
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4
Rated Speed (m/s)

Energy (Joules/Kg)

Figure 1: Energy consumption of various types of drives (Doolaard, 1992).

As Figure 1 shows, the hydraulic system is the least efficient system. The use of a flywheel on two speed systems,
makes them less attractive when compared to solid state systems. The curves in Figure 1 have been derived by
measurements carried out on a large number of installations. In order to be able to compare different installations,
the y-axis has been normalised against the size of the car, and thus the energy consumed has been divided by the mass
of the passengers.
The other interesting point, is that energy consumption rises dramatically with the increase in rated speed.

2.2 Regenerated Energy in Lifts

In any lifting system, potential energy can be gained or lost by the system. Thus in some cases it draws energy from
the supply and in other cases it returns energy to the main supply. Lifts always need to dissipate excessive energy

2
from the system. In some cases this is returned to the main supply naturally (if a direct connection to the mains is
used as would be the case for a single speed or two speed AC motor, which will attempt to regenerate energy back
into the mains when acting as an induction generator) or through a special power electronic set-up (as in the case of
inverter driven systems with regenerative capabilities). In other cases the energy is dissipated through resistors, or just
as heat in the motor. This sub-section discusses some of the aspects of regeneration, and gives a practical example.
Had the lift system been an ideal system, with no friction and 100% efficiency, the regenerated energy would
be equal to the consumed energy. However, this regenerated power is not equal to the consumed power, for the
following reasons:

1. The system needs to overcome friction in both cases of motoring and regeneration (e.g., friction in guide
rails, air resistance...etc.).
2. The motor (when acting as a generator) still needs to overcome its internal losses (i.e., the copper losses,
the iron losses and the friction and windage losses).
3. If the gearbox is of the worm type, its efficiency in the reverse direction will be significantly lower than
in the forward direction.

2.2.1 Example of the regenerating mode

The following is a practical example to demonstrate the concept of regeneration. Although the example is unique
(employing a DC prime mover, in a motor generator set-up), the principle of energy flow could be applied in any
other system, provided it has the capability to return energy back into the main supply.
When the lift is travelling down empty or up full, the hoist motor will be acting as a motor, and thus drawing
power from the generator. The generator will be driven by the prime mover (the DC motor in this case of the motor-
generator set). The DC motor of the MG set will be drawing power from the DC supply.
However, when the lift is moving up empty or down full, the hoist motor will act as a generator (Figure 2).

Current Current

G M G
Hoist motor
MG set, mechanically coupled

Generating mode (e.g., lift moving up empty, or down full)

Figure 2: The hoist motor in generating mode.

This mode is referred to as regenerating (or generating) mode. In this mode the system will be sending power back
into the supply, as shown by the reverse direction of the current from the prime mover.
When the lift is run up empty, it regenerates around 15A back into the main supply, as shown in Figure 3.
This is the current needed to stop it from overspeeding when it is driven as a generator.

3
 Current drawn in an up empty journey
15

10

0
05:18:41 05:18:48 05:18:56 05:19:03 05:19:09 05:19:15 05:19:22
-5

-10

-15

Time
Amperes (DC)

Figure 3: Current trace for lift moving up empty (generating).

2.3 Factors Affecting Energy Consumed in Lifts

As seen from the previous sections, several factors, among many others affect the energy consumption of the lift
system:

• The type of drive: hydraulic, two speed...etc.

• The mechanical design aspects: This is discussed in much details by Stawinoga [8;9].
• The efficiency of various components: Especially the motor and the gear-box for a geared system.
• The reduction of inertia: The use of a flywheels (and all other moving masses) reduce the efficiency of
the system,
• The type of gearing (if applicable): Worm-wheel gears have generally a lower efficiency than helical
gears.
• Possibility of regeneration back into the mains: This depends on whether the system can return energy
back, and whether the metering system can cope with reverse energy.
• The running power factor (especially important on Ward-Leonard systems).
• The loading level: The level of usage, the numbers of passengers and the number of journeys.

Obviously, the tariff structure will affect the final cost of the electricity consumption.

2.4 Calculating Energy Losses In Lifts

Schroeder (1986) developed a generalised formula for calculating the daily energy consumption of a lift installation.
He based the formula on the typical trip time, which he labelled as TP. This factor TP depends on the number of
floors, the type of drive and consequently the rated speed. Table 1 shows the values calculated by Schroeder for
various installations and drives.

4
 Drive Floors above TP(seconds)
ground Range Mean
Hydraulic without less than 6 5-7 6
counterweight
Geared AC 2-speed 6 9-12 10.5

ACVV 12 7-10 8.5

(high mass)
ACVV 12 5-8 6.5
(low mass)
Gearless Motor- generator 18 4-6 5
Thyrisor 18 3-5 4

Table 1: Values of the parameter TP for various types of drives and installations [4].

Once the relevant installation has been found in the table, then the mean value of TP in the table is used.
Alternatively, for more accuracy, “the lower end of the range is used for 1:1 roping and a relatively large motor, or
the upper end of the range is used for 2:1 roping and a relatively small motor” [4].
This is then used in the following formula:

E = (R x ST x TP)/3600
...where
E is the daily energy consumed in kWh/day
R is the motor rating in kW’s
ST is the number of starts per day

The difficulty with the parameter ST is that it has to be estimated or measured, thus affecting the final accuracy of the
estimate.

The value found is then used to calculate the yearly energy consumed (e) per floor unit area (m2), as follows [4]:

E ( kWh / day ) × days / year × 0.85

e=
population × space / person

This gives the annual energy consumed in units of energy (kWh) per square metre.

2.4.1 Example
As an example let us take a building, with six lifts, each running at 4 m/s, 23 passengers, with a gearless thyristor
drive. TP from the table is 4 seconds (the mean value). If the motor size is 45 kw, the only other figure which needs
to be estimated is the number of starts per day. A simple method of estimating the number of starts per day, is to
assume two peaks of two hours each, during which the maximum value of 240 starts per hour is achieved, and assume
another eight hours of off peak activity, producing 40 starts per hour (this assumes a 12 hour day, 7:00-19:00). This
gives a total number of starts per day of:

Starts per day = 2*240 + 2*240 + 8 * 40 = 1280 starts per day.

Then the total consumption per day per lift will be:
E= (45 x 1280 x 4)/(3600)= 64 kwhr/day/lift

For all six lifts in the group,

E= 64 x 6 = 384 kwhr/day

The yearly consumption would be equal to:

5
Total energy consumption cost per year =
384 x 260 (working days per year) x £0.065/kwhr = £6490 per year.

If we assume 2000 persons in the building, and 20 m2/person, gives:

e= (384 x 260 x 0.85)/(2000 x 20)= 2.12 kwhr/m2/year.

3. ESCALATORS
When escalators are running unloaded they need some power to overcome friction in the step-band and the handrail;
and inefficiencies in the motor and the gearbox. This power is referred to as fixed losses (or overheads, in analogy to
overheads in a business or a factory which are not used to produce any tangible product or service; only to keep the
business or the factory going and to support other functions).
As passengers start boarding the escalator, the power consumed starts to rise if it is an up going escalator, or
to drop if it is a down going escalator. The power needed to move passengers is termed as variable losses or variable
power (in analogy to the variable costs of producing products in a factory, which will be sold and will create a
revenue).
So, in addition to the fixed losses in an up moving escalator, it is imparting energy to the passengers
(variable losses) and thus is consuming power. Thus, the total power consumed is equal to the summation of the
power needed to move the passengers between levels to the power needed to overcome the friction within the
escalator.
On the other hand, a down moving escalator, is recovering the potential energy stored in the passengers. The
total energy consumed by a down moving escalator is the difference between the fixed losses and the energy
recovered from the passengers. As passenger loading progressively increases on a down moving escalator, the
escalator gets to a point where it is consuming no power at all, because it is driven by enough passengers to generate
power equal to the losses needed to drive it at no load. As passenger loading further increases, the escalator starts to
feed power back into the supply (i.e., regenerate). This crossover point depends on the level of fixed losses, and the
reverse efficiency of the gearbox.
The main factors which affect the energy consumed by the escalator are:

• Rise.
• Machine type.
• Number of passengers boarding the escalator.

3.1 Fixed Losses On An Escalator

We can think of the fixed losses of an escalator as the base mark from which it start. The fixed losses are equal to the
power consumed when no passengers are travelling on the escalators.
The easiest and most straightforward method to calculate the fixed losses, is to assume that the escalator for
the last 30 minutes of it operation in the day is very lightly loaded with few passengers using it. If the average of the
power consumed over these 30 minutes is taken, this will be a good representation of the fixed losses of that
escalator, with a very small error. The error will arise from the small number of passengers who might use the
escalator in these 30 minutes. A diagram of the power consumption for a public service escalator is shown in Figure
4 (17.2 metres rise and 18,000 passengers per day). It can be seen that due to the high loading on the escalator, it is
still being used in the last 30 minutes, and this will lead to errors in the estimation of the fixed losses. It is not
advisable to use the first 30 minutes, when the escalator is started, as the machine in the first half hour will be
warming up, and thus the reading in that period might not be representative of the whole day. This can also be seen
in Figure 4, where the power drops during the first half hour.

6
 Figure 4: Power trace and passenger trace.
This method is very attractive because it is easy to use, and only needs the measurement of the power consumption in
the first half hour and the last half hour. However, it suffers from some inaccuracy because of the reasons explained
above. This method has been referred to as method A for deriving the fixed losses. The other more accurate method
(referred to as method B) is to use the passenger data as well as the power consumption data, and to plot the points on
a scatter diagram. This method is obviously more accurate, but involves more data collection and processing, and is
discussed in more detail in [1]. An example of the use of method B is shown in Figure 5.

Relationship betw een passengers and pow er (dow n escalator)

12.00
10.00
y = -0.1495x + 10.185
8.00 2
R = 0.8717
6.00
Power (kw)

4.00
2.00
0.00
-2.00
-4.00
-6.00
0 20 40 60 80 100

Passengers (in one minute)

Figure 5: Scatter diagram between power and passengers, for a 17.2 metre rise escalator used by 18,000
passengers per day.

3.2 Relationship Between Fixed Losses And Escalator Rise

The method for deriving the fixed losses (method A) has been used on a group of 7 escalator and the results plotted in
a scatter diagram in Figure 6, which shows the fixed losses as a function of rise, and gives the resultant equation.

7
 Relationship between rise and fixed losses.

12 y = 0.52x + 2.7294
2
10 R = 0.8233

8
losses
Fixed

(kw)

0
0 5 10 15 20

Rise (m)

Figure 6: Scatter diagram of fixed losses against rise.

As shown in Figure 6, the relationship between fixed losses and rise can be approximated by a straight line, with a
non-zero intercept.
In other words, the fixed losses of an escalator (in kW) are dependent on rise and are nearly equal to half the
rise in metres plus a fixed amount equal to nearly 3 kw. This 3 kw would be needed by a fictitious zero rise escalator.
We could think of this 3 kw quantity as being the power needed to run the gearbox and the motor.
From the scatter diagram, r2 is equal to 0.8233, which indicates that the rise is a strong factor in explaining
variation in fixed losses (rise accounts for 82.33% of the variation in fixed losses; there may be other factors which
explain the other 18% variation).
If method B is used, the resulting equation is [1]:

Fixed losses (kw) = 0.55 x Rise(m) + 1.95

3.3 Calculation of the total energy consumed for an escalator

Once the fixed losses of the escalator have been calculated, as shown above, then all that is needed is the average
number of passengers boarding it per day. The total estimated energy consumption per day can then be calculated as
follows:

Total losses = Fixed energy consumption +/- variable energy consumption =

Fixed energy consumption +/- (Escalator rise x 9.81x 75(kg) x Number of passengers x walking factor)

The plus and minus signs apply to up and down escalators respectively. The walking factor varies between 0.7 and 1,
and is used to account for the passenger who walk up or down the escalator, with respect to their percentage and
speed. The lower value would apply to a high percentage of walking passenger at higher speeds. The upper limit
(i.e., 1) would apply if there were no walking passengers.
If the traffic on that day is considered to be a representative sample of the whole year, then the total yearly
cost could be find by multiplying the figure by the applicable number of days in the year.

3.3.1 Example
If we take as a practical example one escalator, with a rise of 7.623 metres, running in the down direction, used by
24910 passengers per day, and assuming that the average passenger mass is 75 kg, and that the escalator is running
for 20 hours per day, 365 days per year, then the following losses can be calculated:
Fixed losses = (7.623 m * 0.55 + 1.95) kw * 20 hours per day = 122.8 kwhr/day
Variable losses =
0.7 (walking factor) * 24910 passenger/day * 75 kg/passenger * 9.81 m/s2 * 7.623 m =
9.7797546 Joules =
9.7797546 Joules/(3600000 Joules/kwhr) =
27.16 kwhr/day

8
 The walking factor is a correction factor used to account for the passengers who choose to walk on the escalator, and
represents the difference between the calculated energy and the measured energy.
This being a down escalator, then the net daily consumption will be the difference between the two values,
Net daily energy consumption = 122.8 - 27.16 = 95.64 kwhr/day
This works out as a yearly cost of:
Yearly cost of energy = 365 days/year * 95.64 kwhr/day * £0.065/kwhr = £2270 per year
This answer does not take into account the power factor of the machine and any resulting extra charge for
low power factor.

4. CONCLUSIONS
The energy consumption of lifts has been estimated as varying between 5-15% of the total energy costs of a typical
building. The choice of the type of drive system and the rated speed affect the energy consumption, where the
hydraulic system is the least efficient, and the VVVF system the most efficient. A general equation has been
presented, which depends on the type of drive, number of floors, number of starts per day and the rating of the motor
to calculate the daily energy consumed. This can then be used to calculate the yearly energy consumed per unit floor
area. Lifts can and sometimes do regenerate power back into the main supply. A practical example was used to
illustrate the general concepts involved.
Escalator energy consumption is based on calculating the fixed losses incurred in keeping the machine
moving and overcoming friction. This mainly depends on the rise of the machine. A generalised relationship has
been found between rise and fixed losses, based on a group of 7 escalators. Once the fixed losses are calculated, then
a general formula can be used, which depends on the number of passengers transported in one day. This variable
energy depends on the number of passengers and the rise of the machine, and is added to the fixed losses for an up
moving escalator, and subtracted for a down moving escalator. A correction factor has to be used which accounts for
the fact that some passengers walk up or down the escalator.

BIOGRAPHICAL NOTES
The author graduated in Electrical Engineering in 1987, and worked for two years as an electrical and electronic lift
systems design engineer. He received his M.Sc. in Remote Lift Monitoring from UMIST (Manchester, U.K.), in
1990, and his Ph.D. in Artificial Intelligence Applications in 1992. He was then appointed as Senior Electrical
Engineer for Lifts & Escalator at London Underground, and is still working for London Underground, currently as
Team leader for Electrical Design and Lift Projects. He is also a part time lecturer in electronics and electronic
systems at the South Bank University and South Thames College, in London, and contribute regularly to Elevation,
Elevatori and Lift Report.

REFERENCES AND BIBLIOGRAPHY

1. Al-Sharif, L., 1996a, “Escalator Energy Consumption, Part I: The relationship between escalator power
consumption and escalator handling capacity (the ETP method)”, London Underground Ltd., Internal Report,
January 1996.
2. Al-Sharif, L.R., 1996b, “The use of power measurement to calculate the numbers of passengers travelling on
an escalator”, Lift Report, Issue 1/96.
3. Al-Sharif, L.R., 1995, “Lift Power Consumption”, Elevation, Autumn, 1995.
4. Schroeder, J., 1986, “The energy consumption of elevators”, in Elevator Technology, Editor George Barney,
Ellis Horwood, 1986.
5. KONE, ____ , “V3F: The green power”.
6. Schroeder, J., 1987, “Elevator traction drives”, Lift Report, May/June 1987.
7. Doolaard, D.A., 1992, “Energy consumption of different types of lift drive system”, in Elevator Technology 4:
Proceedings of Elevcon ‘92, Editor George Barney, The International Association of Elevator Engineers.
8. Stawinoga, R., 1994, “Designing for reduced energy elevator costs”, in Lift Technology 1, Proceedings of
LiftTecH 94, Editor George Barney, The International Association of Elevator Engineers, 1994.
9. Stawinoga, R., 1995, Private Communication, 22 May 1995 (set of slides accompanying presentation
“Designing for reduced energy costs”, at CIBSE 19 May 1995).
10. Greenwood, P.B., 1983, “Energy efficiency report”, Brook Crompton Parkinson Motors, H1311, CP/E/2/83.
11. Kolmeder, W. & Hofbauer, K., 1990, “Energy losses and energy saving”, in Elevator Electric Drives,
Concepts and principles, control and practice, George Barney and Loher A.G., published by Ellis Horwood.

View publication stats