Issue 1/2007


Elevator Ride Quality

The Human Ride Experience

Roger E. Howkins (Ove Arup & Partners, UK)
Elevator ride comfort and elevator ride quality are two distinct and separate concepts that are very subjective but can be very objective depending if you are a passenger, manufacturer, elevator consultant or building owner. This paper will concentrate on the passenger expectations which are very subjective and can also be classified as imaginary as there are multiple sources of experiences transferred to the elevator passenger during a journey, these experiences being both variable or fixed. The variable experiences are changes in the effect of gravity on the human body caused by acceleration, deceleration, jerk and also noise and vibration generated by the elevator systems. The fixed experiences are the architectural design additions to the elevator cars which include lighting, finishes and car furniture.
Category: Issue 1/2007
Posted by: editor

All of this is classified as the Human Elevator Ride Experience.

When the first elevators were installed in the mid 1800‘s, passengers were probably not overly concerned about safety until Elisha Otis demonstrated his safety gear in 1852 which heralded the new business of the elevator manufacturer. Milestones in the new business were the use of electric elevators by Siemens in 1880 and the Eiffel Tower in 1889 which at 321 meters is still recognised as the symbol of Paris. The first generation of elevators used either water hydraulic or steam as their motive force and the travelling public were probably only too glad to move automatically up and down buildings rather than using stairs or ramps. Passenger comfort was not thought of as an important item for elevators, if you consider the motor car industry of that time the modern cars were open, with no heating and were very noisy, very different from today.
The elevator industry, and building owners remained stagnated in this position for over 100 years after Elisha Otis gave his revolutionary demonstration, in that elevator ride quality was not a major concern as it did not have an affect on sales or maintenance revenues.
It was not until the last quarter of the 20th Century did the concept of Elevator ride quality start appearing in specification documents. The Japanese Elevator industry had basic ride quality requirements in place during the early 1980‘s due to the cultural demands of the Japanese population. These Japanese concepts started to be incorporated into international building owners and elevator consultants‘ specifications with varying degrees of success. The ride quality requirements were also being graded into local Standards of quality without specifying other parameters and requirements essential for enhanced passenger perception. To some elevator suppliers, manufacturers, consultants and installers it was an easy way to increase sales income without understanding the dynamics of elevator ride quality or what really was good or bad elevator ride quality in terms of human perception.
Crucially there was no accepted method of testing for elevator ride quality, at the simplistic end testing was by placing a coin on its edge in the centre of the car floor and driving the elevator up and down, if the coin did not fall over it was good ride quality. A more sophisticated approach on the same theme was to balance the coin on the car handrail.
To overcome these shortfalls some elevator consultants‘ specifications attempted to incorporate very detailed testing procedures using highly complex testing and recording instruments which also required highly skilled technicians to operate, keep the instrumentation in calibration and understand the data and results obtained.
To enable a Worldwide method of standardized elevator testing of ride comfort an ISO Standard for: Lifts (elevators) – Measurement of lift ride quality ISO 18 738 was published in 2003. This went a long way to provide an International standardised method of testing and recording elevator ride quality, provided standardised terms and definitions, measuring instrumentation, evaluation of ride quality, procedures for measuring ride quality and reporting of results. ISO 18 738 crucially does not detail what is poor or good elevator ride quality.
Technical Specifications
The classification of elevator ride quality is a perception of the passenger, but technical specifications being produced by elevator manufacturers and elevator consultants have only brought confusion, as sometimes they are without technical or commercial basis and it is doubtful that they bring added value to a project in terms of base commercial cost, reduced maintenance prices, increased reliability or increased rentable value of a building.
From an analysis of four international elevator consultants ride comfort specifications for 1600 kg group of elevators at 4.0 m/sec in a common shaft the basic requirements were as detailed in table 1.
ISO 18 738 classifies noise, lateral quaking, acceleration and jerk as:
  • Noise – Sound – a weighted sound pressure level in decibels.
  • Lateral Quaking – A sideways acceleration measured in m-g.
  • Acceleration – A rate of acceleration measured on the z-axis velocity and expressed in metres per second squared (m/s2).
  • Jerk – The rate of change of z-axis acceleration, attribute to lift motion control and expressed in metres per second cubed (m/sec3).
From Table 1 it can be seen that the variations in the specifications for elevator ride quality are considerable, in that two consultants appear to concentrate on noise as the main driving factor for good elevator ride quality and two consultants appear to demand specific requirements in terms of jerk and acceleration.
Which is the correct method, the simple answer is neither as noise levels are perception as are acceleration and jerk, the human perception of elevator ride quality is a combination of all three acting on the passengers body.
International elevator companies become more complicated when the marketing departments classify their products into a ride comfort category as can be seen in Table 2 where there are considerable differences in lateral quaking, when considering a good ride quality elevator at 4.0 m/sec.
Could the elevator passenger really perceive the difference in the acceptable 35+ m-g from International elevator Company “C” or the acceptable 12 – 20 m-g offered by other International elevator companies? As there is no International norm for defining the quality of elevator lateral quaking, this solely becomes an issue of perception of the passenger. Unlike other forms of passenger transport the potential building owner will find it very difficult to find an elevator installation which is exactly of the same design he requires for his building as there are too many variables to give an exact replication of the ride quality expected.
Noise is complex, what is acceptable? Everyone has a different perception of acceptable noise levels in everyday applications, the user of an MP3 player may require a noise level of 105-120 dB(A) and a modern air conditioned office environment will have a design target of 50 dB(A).
In table (3) which is a comparison of maximum noise in the elevator car measured in dB(A) there is a considerable difference between elevator manufacturers in max sound levels and their later quaking requirements in m -g.
From the specific requirements of the elevator consultants’ technical requirements the maximum elevator car noise and lateral are detailed in Table 4.
When comparing the elevator consultants requirements and elevator companies performance only one elevator company can totally comply with the elevator consultants’ specifications. The major area of non compliance is the noise level, which is one of the major areas of perception of the passenger. Does this mean the three elevator consultants requirements are technically unachievable or the elevator companies designs are technically incorrect?
Perceived Elevator Ride Quality
Each and every elevator passenger is the sole judge of what constitutes accept - able elevator ride quality, they make their subjective assessment on single or multiple journeys in the same or similar elevators in a group or in a building. This assessment in simple terms is to ride the elevator and consciously or sub consciously analyse the journey, compare using memory and past experiences. Therefore if a group of individuals were asked to rate a particular lift journey the majority of responses would be different ranging from those who have an inbuilt fear of elevators and because of the physiological differences between men and women the perceived elevator ride classification will be different. Who is to say who is right and who is wrong as it is a subjective analysis?
It has been established and agreed that the basics of elevator ride quality is subdivided into noise, vibration, sensation of speed and elevator car design and all of these are very subjective due to gender, age, culture, well being and perception. A Multi -National Elevator manufacturer will have a differing view of ride quality from a local regional manufacturer and a local elevator consultant will have a different perception from a International elevator consultant. In simplistic terms the elevator ride experience must be a pleasant experience that the passenger wants to have on multiple journeys.
The passengers perception of elevator ride quality is transmitted from the moving elevator to human body by means of experiences and signals received, analysed from the cardiovascular and digestive systems, organs and nerves. The capillaries within the inner ear, muscle groups surrounding the spine and nerves in the soles of the feet are the key receptors which respond to changes in noise, vibrations, acceleration and levels of jerk. The major internal organs of a human body are responsible for detecting changes in blood pressure, so the perception of elevator ride quality depends on the quality of information received by the brain and this will depend on the age, gender and well being of the individual riding in the elevator.
Some individuals say that we live in a noisy world and it is becoming nosier than ever, for some it is an encouragement, it is a sign of entertainment, industry and a bustling society. For others a serious form of pollution, annoying, selfishness and the absence of concord and harmony. How this applies to elevators is difficult to quantify in pure sound levels, does a building owner who has a super fast elevator installation really want to experience super quiet and vibration free journey without the sense of speed and motion?
Noise is the easiest of all the elevator ride comfort factors to measure and compare with recognised norms, but also the most complex to control as multiple factors are all interacting with each other. Is the measurement taken with the elevator car stationary/moving, car fan on/off, doors opening/closing, voice announcements on/off and the ambient noise level on the lobbies? For comparison purposes Table 5 gives levels of noise norms in typical environments and suggested classifications. For elevators where the noise levels are specified below 50 dB(A) this is classified as “faint” and would be the equivalent of living in a semi rural environment. The question is, is this an appropriate design criteria for an elevator in a commercial building?
Noise is measured in units of decibels (dB). If sound is intensified by 10 dB, it appears to a persons ears as if the sound intensity has doubled, however if it is reduced by 10 dB it would appear as if the intensity has been reduced by half. If the change of noise level is increased or decreased by 3 dB(A) the change would be just perceptible, but if the noise level changes by 5 dB(A) the change is clearly noticeable. This is very important to the perceived noise level within a moving elevator car.
It would be very unusual for an elevator car to have a generated noise level which will have an effect or destroy a persons hearing. But high elevator noise levels can put stress onto other parts of the body including the heart. Noise will put undue stress onto the body by quickening the pulse rate, increased blood pressure. Noise within a travelling elevator car can be both unpleasant and potentially harmful to the passengers and this would not be classified as an imaginary, but a real concern. It is generally considered that continued exposure to a noise level in excess of 85 dB(A) is likely to degrade the hearing of a large percentage of the population.
In an elevator car the noise generated must be analysed into 3 areas:
1 Will the noise become a nuisance inside and outside the elevator car?
2 Will the noise generated be safe to those passengers in the elevator car?
3 Will the generated elevator noise affect productivity?
It is recognised that people do have perceived levels of noise when it becomes either unpleasant or uncomfortable and this will affect perception of elevator ride quality.
Apart from hearing which is the human response to audible sound there are other responses which can be massed together under the banner of perception or feeling. Within a moving elevator, vibration is caused by surfaces and a component vibrating strongly enough to turn them into a secondary sound source, this vibration is generated by an elevator moving through the shaft and will change in intensity during the acceleration, full speed and deceleration elements of each journey.
The effects of vibration on a human body are difficult to quantify as people will accept different criteria for different situations, a home is totally different to a hospital and a motor vehicle cannot be classified to be similar as an elevator. There are defined levels of induced vibration which can cause nausea to humans, damage structures or cause equipment to fail or malfunction.
The vibration induced feeling is complex, but it is known that people using transportation systems including elevators will tolerate more induced vibration than in a stationary environment. The human perception to vibration, is how the body responds to acceleration, velocity and displacement which are either taken singularly or as a combination.
In any elevator, the travelling passenger will have the perception of vibration acting on their bodies in 3 differing axes classified as:
x the horizontal axis – front to back of the elevator car
y the horizontal axis – side to side of the elevator car
z the vertical axis of the elevator car.
Horizontal vibration is also called Lateral Quaking.
Vibration is defined as a variation with time of the magnitude of acceleration, when the magnitude is alternately greater and smaller than a reference level. Acceleration is normally expressed in terms of milli-g (m-g) or one thousandth of a “g” (.001 g).
The human body is very sensitive to low frequency horizontal vibration but normally immune to the higher frequency horizontal vibrations. As most elevator passengers travel by standing up during the journey, at low levels of frequency the whole body moves but at higher frequencies the feet move but the upper body remains stationary. It is very unusual for an elevator passenger to feel unsafe because of vibration caused by horizontal vibration, on the x and y axis.
Normally elevator passenger concerns are due to incorrect installation of equipment, most probably caused by the guide rails not being fitted correctly or damaged, also the elevator car if not statically balanced will generate high levels of vibrations which will be perceived by most passengers.
Apart from horizontal vibration the elevator passenger is subjected to vertical vibration on the z axis, this is generally a high frequency vibration generated by a primary source vibration through the rotating drive machinery which is transmitted into the elevator car by the suspension ropes. The number of elevator rotating components transmitting vibration are classified as motors, sheaves, rollers, bearings and gears which all have differing levels of frequency generation which at high rotating speeds should be dynamically balanced to reduce unwanted vibration. Most of the transmitted vibrations can be isolated by the car frame assemblies, but an incorrect car design or installation procedures can result in mechanical short circuiting between the car body and its sling, inducing high levels of vibration.
The passengers perception of vibration is not normally apparent in an elevator at rest although a recording of 4 milli-g would not be unusual due to natural vibrations generated within the building and elevator shafts.
Passengers within an elevator car which is travelling at contract speed would not normally be able to perceive any noticeable difference in horizontal axis vibration between 10 and 20 milli-g or in the vertical axis between 15 and 25 milli-g. So defining exact vibration levels may not alter a passengers perception of elevator vibration or ride quality.
Most people understand that acceleration is the rate of change of velocity. The effects of acceleration are both physical as well as psychological and will have major variances between people travelling in the elevator. Elevator acceleration is perceived as good when it is constant. If the acceleration is by noticeable steps it will be perceived as objectionable.
Obviously when slowing down an elevator will produce a negative acceleration curve which should mirror the step-less acceleration curve.
The passenger perception of acceleration is that for values below 1.1 m/s2 the majority of passengers would not have any perception of a difference in overall elevator ride quality. Acceleration and motion is noticeable if between 1.3 m/s2 and 1.6 m/s2, and in the majority of elevators this would be an optimum value for acceleration to provide a good quality of elevator ride comfort. Acceleration above 1.8 m/s2 would often be perceived as objectionable.
This is the term used for the rate of change in acceleration and is expressed as metres per second cubed or m/s3. Jerk is probably the most important of the elevator variables and the most difficult to quantify when defining overall elevator ride quality. The human body responds very quickly to rates of change in acceleration rather than the actual velocity of the elevator car.
With the rate of change of acceleration, the forces upon the human body change and if severe enough the elevator passengers will start to become uncomfortable. However changing the jerk rate of an elevator is a very convenient and cost effective way to improve the overall elevator system performance, but at a cost to overall passenger perception of ride quality.
For an elevator with a jerk rate of 6.0 m/s3 this will be perceived to be totally unacceptable to a very high proportion of passengers, a rate of 2.0 m/s3 would be entirely acceptable for a high proportion of passengers. The only exception to this would be where elevators are installed in hospitals where the effects of a high jerk rate may be harmful to the patients and low jerk rates of 0.7 m/s3 could be required to reduce the possible risk to patients health.
Elevator ride comfort
An elevator ride should be a pleasant experience that the passenger wants to be repeated, they do not want it to be frightening, to get this they would go to a theme park. The ride in the elevator must be a pleasure, but some passengers will not be satisfied, some may experience discomfort, therefore it is almost certain that 100 % elevator ride comfort satisfaction will never be achieved.
A survey of actual passenger ride comfort perception was carried out on 367 people and also using an industry standard Elevator Ride Quality Analyzer (EVA-625) manufactured by Physical Measurement Technologies, Inc. The following parameters were recorded, acceleration, jerk and noise inside the car at contact speed, vibration on x-y axis and vibration of z axis. These have been compared with the passenger perception of ride quality from those passengers surveyed.
Table 6 details the maximum vibration recorded in milli-g on the x-y-z axis and ranks each lift in terms from best to worst.
Table 7 details the maximum acceleration in m/sec2 and jerk in m/sec3 recorded in each lift car and ranks each lift from best to worst.
Table 8 details the overall level of recorded noise within the elevator car taken at full speed and ranks each elevator from best to worst, the range of recorded noise is 5 dB (A).
Table 9 details the data collection from the elevators and ranks each lift in terms of data averages and gives an overall ranking of the elevator based on a comparison of points scored on data collection.
Table 10 details the 367 passengers surveyed and their perception of the overall lift ride quality ranking and it is significant that the overall passenger perception of good elevator ride quality is significantly different to that obtained by recording and analysing equipment. This suggests that just basing elevator ride quality on data gathering and specific specification requirements is not the only method of determining good elevator ride quality. Passenger perception of quality is also very important as they are the people who matter.
Elevator ride quality is a difficult topic to quantify in terms of good, average or bad as it is solely a perceived area of judgement of the individual riding in the elevator. But certainly guidelines can be established so that perceived good elevator ride quality can be specified and achieved within the confines of the elevator shaft.
ISO 18 738 defines how to measure elevator ride quality but does not venture into defining what is good, bad or acceptable elevator ride quality.
Elevator ride quality is being driven by international elevator companies and international elevator consultants without reference to passenger perception and what is achievable.
The majority of elevator passengers would be happy with elevator noise levels below 50 dB(A), acceleration levels lower than 1.6 m/sec2 and jerk less than 2.0 m/ sec3.
Technically and commercially there is no justification for super quiet or super low vibration elevators as the passenger perception would not significantly alter, and the overall elevator performance and reliability would not improve.
Earl Abraham (April 1984) Performance Criteria: Car Ride Quality, Elevator World.
Aldia J. M. Aranbura I. Pagalday J. M. (1993) Single Parmater Elevator Comfort Elevation – Procuedure of Elevcon 1993.
Dipl-Ing. Peter Pini (2002) Elevator Diagnostics and More Lift Report – July/August 2002
Safety Line Institute (1998) Human vibration.
League For Hard of Hearing (2003) Hearing Fact Sheet.
ISO 19 738 (2003) Lifts (Elevators) – Measurement of Lift Ride Quality.
ISO 11 205 (2002) Acoustics – Noise Emitted by Machinery and Equipment.
Torsten Wintergerste, Sabine Sulzer (2004) Ascending The Matterhorn in Eight Minutes: Low Noise, High End Elevator, Sulzer Technical Review 3/2004.
Australian Government – National Occupational Health and Safety Commission.
Hillel Schwatz (1995) – Noise and Silence: The Soundscape and Spirituality, University of California.
Lyle F. Yerges (1969) Sound, Noise and Vibration Control, Van Nostrand Reinhold Co.