Elevator Energy Use
Did you know that you use more energy going down an elevator than going up it?
First, it is important to choose the right type of elevator. (For an unusual elevator choice, made before the existence of passenger elevators, see “The Cooper Union for the Advancement of Science and Art,” p. 44.)
Elevators come in two basic types: hydraulic and traction. Traction elevators typically use cables to lift the cab from above, with weights to counterbalance the weight of the cab and its occupants. Hydraulic elevators use a pump, which pumps fluid into a cylinder, which pushes on a piston to lift the cab.
Because hydraulic elevators have no counterweight, they use lots of energy lifting the entire weight of the cab—energy that is not recovered on the down ride. And energy use is not the only problem with hydraulic elevators. Years ago, hydraulic elevators used water, sometimes operating on municipal water system pressure alone. To avoid rust, modern elevators use hydraulic fluid, which is oil. But oil is problematic: A minor leak smells up a building, and a major leak is a serious environmental problem.
Despite these disadvantages, hydraulic elevators have one big advantage. The cylinder is supported by the soil beneath the building, rather than by the building itself. This simplifies the structure of the building. But given the disadvantages described above, I think it is high time to stop installing them. Kone, one of the large international elevator companies, has completely stopped manufacturing hydraulic elevators. ThyssenKrupp, another large company, takes the opposite approach. It calculates that, over the life of the elevator, the combined impact of energy use, manufacturing energy, and materials on the environment for a two- or three-story building is less for a hydraulic elevator than it is for a traction elevator. (They say this is true because elevators for these shorter buildings have shorter pistons, which require less energy to lift, and less material to make.) However, ThyssenKrupp does not say how much electricity any of its elevators uses, which makes it impossible to verify its calculations.
Traction elevators use much less electricity than hydraulic elevators mainly because—at least in theory—the counterweight counterbalances the weight of the cab. The motor only has to accelerate the load and overcome friction, because there is no actual lifting involved—except when the weight of the cab and the counterweight are unequal. Unfortunately, they are unequal for most rides.
Counterweights are sized to equal the weight of a cab carrying half of its maximum capacity in people or freight. This limits the maximum lifting load on the motor to half the cab’s capacity, regardless of how full the cab actually is. But because the cab is empty or mostly empty on many trips, and is usually only completely full close to the lobby, the counterweight is, on average, heavier than the cab and its load. Therefore, for an average ride, a traction elevator uses much more energy going down than up.
How Much Energy for a Ride?
I measured standby electricity, riding electricity, and speed for dozens of elevators over a period of 3 years, using several different true wattmeters. I made every effort to ensure that my measurements were accurate. I got instruments made by three different manufacturers to agree closely with each other; the instruments were also in agreement in measuring reasonably predictable electricity loads, such as lighting fixtures. I also checked for consistency between measurements taken for consecutive rides, between measurements taken for nearly identical elevators, and between measurements taken with different instruments on the same elevator. I measured slightly higher but nearly identical amounts of electricity used by slightly larger, but nearly identical, elevators, and other differences in energy use attributable to factors that I also measured, including cab lighting electricity and gearbox temperature.
I compared elevators by averaging measurements for three four-floor round trip rides. By four-floor rides I mean rides from, say, the second floor to the sixth floor, counting intermediate floors: four of them between the second floor and the sixth floor. I also measured electricity use on longer rides on taller elevators, but because those rides varied in length, I did not use them for comparison between elevators.
The traction elevators I measured used from 14 to 270 kilojoules (kJ) for a four-floor ride, and the few hydraulic elevators I measured used over 400 kJ. One joule is 1 watt-second. Therefore, because there are 3,600 seconds in an hour, there are 3,600 joules in a watt-hour, and 3,600 kJ in a kilowatt-hour. An elevator ride that uses 500 kJ of electricity at 20¢/kWh costs about 3¢ (500 / 3,600 x 20¢). A four-floor ride on a good elevator, using 20 kJ, costs 1/10 of one penny. Because the hydraulics use about 30 times as much electricity as the best traction elevators, they should be avoided for new installations.
The most energy-inefficient traction elevators are older types that use direct-current (DC) electricity. The advantage of using DC current is that it is easy to control speed: Just discard much of the electricity as heat by sending it to giant resistors. This allowed smooth rides in the years before fancy electronic speed controls existed. Utility companies stopped selling DC years ago, so these days the DC is generated by a genset in the elevator machinery room. This mechanism combines an electric motor that runs on alternating current (AC) with a DC generator.
The industry stopped making genset elevators about ten years ago, so while they are not a choice for new installations, they will be around for many years. Replacement probably doesn’t save enough electricity to pay for itself, but it is worth switching the genset off whenever possible. Suarez and Sirt, in an unpublished study, measured total electricity use for one week for two elevators: an AC/DC elevator, and an AC drive elevator. They concluded that at 19¢/kWh, replacing an AC/DC genset with an AC drive would pay for itself in 50+ years, but installing a timer would pay for itself in just six months.
Most elevator companies already install timers with a delay of perhaps five minutes. What this means is that if nobody has taken a ride for more than five minutes, the genset takes a few seconds to start up and accelerate to speed before the cab starts to move. Otherwise, there is no additional delay.
DC current is no longer necessary because the modern electronics industry makes black boxes that control the speed of AC motors. They do this by modifying alternating current to frequencies other than the standard 60 cycles per second; they usually also vary the voltage. The AC drive emerged as the real champion of electricity use in my study. AC traction elevators scored from 14 to about 160 kJ on the four-floor ride test. That is much less than hydraulic or genset elevators, but it still represents a more than 10-to-1 range of electricity use for doing essentially the same amount of work.
Three of the big multinational elevator companies—Kone, Otis, and Thyssen- Krupp—market energy-efficient elevators, all of which are a type of traction elevator called MRL: Machine Room Less. They have been available in the United States for 5 or 10 years, and for 15 years in other parts of the world. MRLs are fundamentally different from normal traction elevators in that they eliminate the need for an elevator machine room on the roof of the building.
A traditional elevator cab is supported by steel cables, which run up the shaft, over a large pulley mounted on the floor of a machine room on the roof, and then down to the counterweight, which rides up and down on the side of the shaft. The pulley needs to be large for two reasons: The steel cable can’t make a sharp bend, and the cable needs enough “wrap” to prevent it from slipping. But as anyone who has ever ridden a ten-speed bicycle understands, a larger diameter means a faster cable speed, which for elevators means that the motor needs to be connected to the pulley via a reduction gearbox. The gearbox requires oil, maintenance, and energy to move gears through oil, especially if the oil is viscous on a cold day. These problems have all been eliminated by a miracle of modern manufacturing: a flat steel-and-rubber composite belt similar to those that have replaced V belts in cars.
The belt can wrap around a pulley small enough to eliminate the gearbox, but still generate enough friction to prevent slippage. With the gearbox gone, all that is required is a pulley mounted directly on the motor, which is mounted on the wall at the top of the shaft. The shaft still protrudes above the roof, but the enclosure for it no longer needs a door to the roof, or a floor across the top of the shaft. This makes the enclosure much smaller and cheaper to build than a normal machine room.
Eliminating the machine room is important, because it is generally vented to outdoors and heated, while equipment manufacturers and some codes require it to be cooled to protect the electronic equipment. Heating and cooling a space that is essentially outdoors wastes a lot of energy. MRL elevators still have warranty requirements for cooling the room where the electronic controls are located, but because the equipment can be located indoors, the cooling loads are greatly reduced.
How Much Energy for a Ride?
Despite repeated requests, none of the three manufacturers marketing energyefficient elevators provided any information on either standby or riding electricity for their products. Nor did any of them make any elevators available for testing. Some product literature predicted 24-hour use for combined standby and expected rides, or said that the elevator could be expected to save a specific percentage of an unspecified amount of electricity.
Nevertheless, I asked building owners, and found some Otis Gen2® and some Kone Ecomatic® elevators to test.
The Kone Ecomatics® I tested used from 106 to 141 kJ—about the same as most ordinary traction elevators made within the last 10 or so years. The Otis Gen2s® were much better, at 16 to 62 kJ, but my measurements did not show them generating electricity on the way down, as Otis claims they do.
The winners were some nonproprietary elevators with an AC drive made by a company called Tricon, which used between 14 and 19 kJ. The Tricon drives were incorporated in control panels manufactured by Motion Control Engineering, Incorporated, of California, which manufactures components for nonproprietary elevators.
The continued existence of two parallel elevator industries—the large, worldwide companies, and smaller, local companies selling nonproprietary elevators—shows that each industry has enough advantages and disadvantages to keep both industries in business. Some fancy features, such as MRL, are available only from the big companies, but at the cost of having only one choice of company from which to buy replacement parts. Large companies have the advantage of standardized equipment familiar to service people who work on only their products, but as Mary Spink, a New York developer says, “I’ve been married twice, and that’s enough. I want an elevator that any company can fix.” Only time will tell which industry will retain the lead in energy use reduction.
Standby Electricity Use
A significant portion of the energy used by elevators is used by the equipment when it is idle. I found this difficult to measure, as lights, fans, and traffic controllers for multiple elevators are sometimes on separate circuits, and sometimes not.
The worst offender when it comes to standby electricity use is a transformer to convert grid electricity into a different voltage. For example, one building I measured had 220V power feeding a transformer that made 440V electricity to run the elevator. The transformer sits there turning electricity into heat at the rate of about 500 watts, 24 hours a day. With every other elevator in town running on 220 volts, there is no excuse for this situation, which fortunately is not very common. Other parts of the standby load—such as cab lights and fans— offer good opportunities for saving.
Cab fans are generally not required by code, so they can just be disconnected on existing elevators and omitted on new elevators, with perhaps a few strategically placed holes doing the job the fan used to do. If the fan stays, it should be controlled by a timer or motion sensor, which switches it off when nobody is in the cab.
Standby electricity use for lights varies surprisingly. One problem I encountered more than once occurs when the cab has a luxurious, dark woodpaneled interior. This makes it too dark, so someone drills a few holes and adds some halogen lights, which use hundreds of watts of electricity indefinitely. The obvious solution to this problem is either to use lightcolored cab finishes, or to plan ahead for sufficient fluorescent (or LED) lighting.
LED lighting is starting to come into use in elevators. Both Kone and Otis say that in many parts of the world they already use LED lighting, which turns off automatically when the elevator is not in use. Perhaps lighting should be controlled by a motion sensor and not a timer, so that the lights can stay on if someone is stuck in the elevator for an extended period of time. Kone also offers control of hallway lighting in some countries; the hallway lights go on only when an elevator approaches.
The Cooper Union for the Advancement
of Science and Art
Cooper Union, the first free college in the United States, was founded and endowed by Peter Cooper, the famous inventor and manufacturer. He was so forward thinking that in the mid-1800s he decided his college would be open to “men and women of any race, faith, or political opinion.” Not only that, but when he built his school, he
installed an elevator shaft in the expectation that someday someone would invent a safe passenger elevator.
History unfolded as he expected, with
only one hitch: He thought elevators would be round, so he put a round shaft in his building. But rectangular elevators proved to be more popular.
The building was renovated in about
1978, and since has had a round elevator in a round shaft, but nobody knows what was there before 1978. I asked one person who graduated in the early ’40s, and he said the shaft was definitely round, but he rode the elevator maybe only once, and doesn’t remember the shape of the cab. Therefore, I’m afraid I have to leave you, the reader, hanging, curious to know what happened.
The round elevator in a round shaft in The Cooper Union’s foundation building.
The Biggest Hole in a Building: The Elevator Shaft
Some U.S. codes require all elevator shafts—even MRL, without a machine room—to have 3 square feet of vent per cab opening to outdoors, to vent smoke in case of fire. Combined with elevator doors, which are very air leaky, elevator shafts move an obscene amount of air through a building—air that is heated in the winter and cooled in the summer.
While the amount of energy saved by reducing this air flow is difficult to measure, and impossible to calculate, the solution is simple: Just stop it.
There are several ways of doing this. Some codes permit the installation of a motorized damper on the shaft vent, which can remain closed until a smoke alarm tells it to open. The hole where the cables pass through the machine room floor is limited by some codes to a maximum clearance of 2 inches around the cables, but the hole is usually larger. Reducing the size of the hole will reduce air leakage. There will still be some leakage, however, especially for installations with machine rooms, which are generally also vented.
Another way to reduce air leakage is to install a vestibule around the elevator shaft on each floor. This is not unusual in office buildings. Not only does it reduce air leakage out the top of the shaft, but it also reduces air (and smoke) movement between floors.
Suggested Features for All Elevators
Here’s what to look for if you are buying a new elevator or retrofitting an existing elevator (all options except the choice between hydraulic and traction elevators are available as retrofits):
▪ Avoid hydraulic.
▪ Ask for equipment that doesn’t need to be heated or cooled.
▪ Ask for LED lights in the buttons.
For tall buildings with multiple elevators, ask for Thyssen- Krupp’s new Destination Dispatch® system. With this system, instead of pushing one of two buttons, up or down, passengers push a button for their destination floor, and are sent to specific elevators, with passengers grouped according to destination. This system is almost as good as the old lobby dispatchers, who were stationed in the lobby to tell passengers which elevator to get on, and tell the elevator operators which floors to stop on. —Except, of course, that the modern systems can’t recognize the passengers, greet them by name, smile, and nod in the direction of the appropriate elevator, which shows how far computers have come.
▪ Choose a light-colored cab with energy-efficient lighting on a motion sensor.
▪ Eliminate the fan, or put it on a timer or motion sensor.
▪ Adopt the sensible floor-numbering system used in Europe. The street floor is numbered 0; the floors below that are numbered -1, -2, and so on; and the floors above the street floor are numbered 1, 2 and so on. This can eliminate a surprisingly large number of extra stops that are caused by confusion about which button to push.
▪ Install a motorized louver on the shaft vent for smoke control, as permitted by code.
▪ Reduce the size of the hole in the machine room floor where the cables pass through the floor to 2-inch clearance around the cables. This is required by some codes for smoke control, but is usually not done.
The elevator industry has been reducing energy use for many years, and is still headed in that direction, but there is a lot more it can do. ThyssenKrupp already dispatches all of its service people in São Paulo by bicycle, a lead other companies can follow, maybe even in U.S. cities. And it is time to look at dropping the code and warranty requirements that demand elevator machinery be coddled with individual heating and cooling systems. Perhaps in the future elevator electronics will be as durable as the electronics found under the hood of cars, which survives very high and low temperatures, ice, snow, rain, salt, vibration, dead batteries, and voltage spikes from jump starts. (Hint: Drop the requirements now, because many of the heaters and air conditioners I’ve seen are broken or unplugged, and most of the rest are just sitting there wasting energy, because they can’t keep up with the load of the air rushing through the room.)
The next logical step in the direction of MRL is a shaft short enough to allow a roof to be built flush across the top of a shaft, with perhaps only a sheet-metal gooseneck protruding for ventilation. And maybe someone will experiment with lighter counterweights, to see if an increase in peak load is worth saving energy on an average ride.
Meanwhile, elevator buyers can start asking for energy use measurements so they can make informed choices and give companies an incentive to make further improvements, including perhaps rating the energy use of their products, as lightbulb companies have been doing for many years. And don’t forget to ask about loosening the requirements for machine room heating and cooling, reducing air movement through the shaft, and cutting energy use by lights and fans.
Henry Gifford is head of mechanical system design at Chris Benedict R.A., a New York City architecture firm that designs energy efficient buildings.
>> For more information:
An Excel spreadsheet showing the measurements I took, including descriptions of elevators, equipment, rated speeds, actual speeds, actual energy use, motor sizes, energy used for a ride, and energy used for standby loads, will be posted online at www.energysavingscience.com.
- FIRST PAGE
- PREVIOUS PAGE
Enter your comments in the box below:
(Please note that all comments are subject to review prior to posting.)
While we will do our best to monitor all comments and blog posts for accuracy and relevancy, Home Energy is not responsible for content posted by our readers or third parties. Home Energy reserves the right to edit or remove comments or blog posts that do not meet our community guidelines.