The successful operation and maintenance of gates was the subject of a number of informative papers presented during the Waterpower XV conference
Waterpower XV, held in Chattanooga, Tennessee, in July 2007, offered attending delegates the chance to learn about the latest developments in the hydroelectric and dam construction industry. One important area discussed during the conference was the efficient operation of gates, with a number of papers presenting information on new products as well as methods for determining the condition of existing systems.
Safety concerns and economic risks associated with gates are often common problems for ageing dams and hydro plants. Knowing when gates are operating efficiently – and when and where maintenance is needed – is an important consideration for project owners. But exactly how will project owners know their water control gates need maintenance? In the paper ‘Gates and hoists: means and methods for determining equipment condition’, Richard Dulin and Sonya Reiser of TCB|AECOM attempt to answer this question by discussing condition assessment of gates and hoists while providing details of tools used to determine the general condition of equipment. The authors suggest that monitoring the condition of a gate hoist involves the use of sight, sound, smell and touch as well as various types of equipment.
A simple visual inspection of the components can help maintain equipment life, says the paper. Geared couplings help reduce unwanted loads on the equipment due to misalignment – but they do need lubrication. Most geared couplings have a fill port on each side of the coupling. A quick visual inspection of these open ports prior to testing can determine the presence of lubrication. If the coupling is dry, problems assigned to another location may simply be the lack of lubrication on the coupling.
Amperage readings taken on an annual basis can provide a history of the changes the hoist and gate system incur. Amperage readings alone, however, can provide a false sense of security as other problems lay undetected.
Physical touch can also aid in the preliminary inspection of open gearing. Lubrication of open gearing is extremely important. Grease needs to flow to maintain the life of the open gears. The authors suggest that one quick test prior to operation of the hoist is to press a thumb print onto the contact surface (the side with the least amount of grease). If the thumb print can be clearly seen adequate grease should be present. After running the hoist for a short period of time one should have a good feel of where potential problems exist.
During the initial test run, the following two major functional operations should be noted: the amount of time required for the hoist to come to full speed while rising; and the amount of time required for the hoist to stop once the gate stops in the downward direction.
The time for the hoist to come to a full speed while rising should be less than a second. When the gate stops moving in the downward direction, the hoist should stop within a second as well. Many potential problems can be present if the start-stops are not quick, including problems with the holding brake and control braking system.
Sound is a good indicator of potential problems during hoist monitoring operations. During normal operations, the only sound should be the quiet hum of the brake solenoid. Any noise from the gears or bearings in unacceptable.
Although sound is the initial method of determining hoist and gate condition, the sense of smell is useful in monitoring components as well. The smell of overheating components is a warning sign of problems with the brake, motor or gearbox. Any vibrations while the hoist is running should also be noted.
Through this use of four of the five senses, one can gain an understanding of potential gate and hoist problems, says the paper, with further detailed inspection to locate specific problem areas.
The authors provided a step-by-step guide of the power train and discussed various warning signs that indicate maintenance or replacement may be necessary. They started by detailing the brakes. The Crane Manufacturers Association of America (CMAA) requires, as a minimum, a control brake and a holding brake. In many hoist systems, the role of the control brake is taken by a worm gear reducer and the holding brake is located on the back side of the motor. The minimum worm gear reduction ratio to be considered as a control brake is 40:1 and depending on the amount of usage and the design of the contact angle, the worm may not hold a static load let alone stop a moving load.
If a control brake is not part of the system, two holding brakes are required by CMAA, although some argue that CMAA is not applicable to gate hoist since it is not an overhead load.
Shoe brakes can be easily inspected for wear and condition. The wheel should have no radial grooves. If grooves exist, the shoe may be worn to the point where the attachment hardware is making contact with the drum. A brake drum is normally pitted from corrosion and this is acceptable. Most drums are made of cast iron, however the drum should always be inspected for wear.
Motor disc brakes are a little more difficult to inspect. Most disc brakes have a manual release and to fully inspect the brake the release and housing should be removed for a thorough inspection. If any problems are noted, the authors suggest that the brake and motor should be sent to a qualified shop for controlled disassembly and inspection.
The motor is the heart of the hoist system and the next component in the power train. Assuming proper sizing it should be quiet, generate a small amount of heat and start quickly. Noise can come from many sources within the motor, with the most likely source of noise with a squirrel cage motor being a worn rotor bearing or a lack of lubrication. On a wound rotor motor, the brushes could be at the end of their useful life. In addition to noises, other warning signs are vibration and excessive heat generated in the bearing area.
Motors do not age well. Their insulation degrades and they can become home to insects. Their nests can block the flow of air resulting in overheating – with open motor frames particularly susceptible to this problem. Any heat on the motor could also be a sign that internal insulation and the wires to the motor and brakes may have deteriorated. It is difficult to assess the internal insulation condition without disassembling the motor, however if the visible wire insulation is in poor condition, the coil insulation will most likely be in bad shape as well. Once the motor is free from the gate loads, the motor rotor should be able to move axially with a small amount of manual force. If axial movement does not occur, the initial placement of the motor was not correct or the couplings are not functioning as they should. Either situation should be rectified.
The next component in the power train is the primary reduction. In an enclosed primary reduction system, very little can be observed from the outside. However a few tests can be performed to assess component condition. Oil samples taken just after hoist operation and tested by a qualified lab can provide early warning signs of potential problems. A qualified lab can inform the owner if the oil is normal or abnormal for the application.
Any gear reduction system should have some backlash present to ensure efficient operation of the reducer. If no backlash is present the teeth are effectively wedged into one another and will result in excessive loading of the shaft and bearings. A simple test to determine a reasonable amount of backlash is to rotate the input shaft by hand, says the paper. If one can feel and hear the gear teeth making contact, but no physical motion is observed, then limited and acceptable gear wear is present. If physical rotation is observed, there is most likely excessive wear on the contact surfaces, which may warrant further internal inspection. This approach can be taken for both parallel shaft reducers and worm gears.
Bearing condition within the gear box is also of great concern. Several simple motion tests can be done to check the condition of the bearings at the primary reduction. For an enclosed reducer, the exposed shaft should not move in the axial or radial directions. For worm gears and parallel shaft helical gear reducers, the input and output shafts experience an axial force by the nature of their design. The bearings are designed to absorb these forces and if there is movement, the bearings have worn out.
Upon leaving the primary reduction, the gearing is generally open with an independent lubrication system for the gearings and bearings. In most applications grease is used as a lubricant, applied by brush or aerosol can. The most significant problem with gate hoist gearing is the lack of lubrication on the tooth contact surface. Lubrication not only affects the efficiency of the gear reduction, it is extremely critical in minimising the effects of contact stresses.
Grease has a greater tenacity than oil, so the need to apply fresh grease as frequently as oils is not necessary, nor is there a need for 100% tooth coverage. Several factors do need to be considered when determining a lubrication schedule however, including the magnitude of the load and the number of times the teeth engage. These can have a significant effect on the required frequency of lubrication.
Bearings are an extremely important component in the power train. Bearings need lubrication as well as gears, but over greasing a roller bearing can cause as many problems as under greasing. For grease to be effective in a roller bearing it needs to move around. If too much is used, the grease will have no where to go and lubrication is not effective. In addition, over greasing can blow out the seals and all the grease could be lost. In the authors’ experience, the best long term approach is to refurbish the bearings every 5-10 years.
Journal bearings have an advantage in that over greasing is not a problem. Pushing new grease in and the old grease out will provide many years of reliable service. These types of bearing need periodic inspection every 10 years as a minimum. The best tool to determine journal bearing condition is temperature rise – if a bearing heats up it is most likely due to a lack of lubrication. If there is excessive wear or corrosion, a through inspection is warranted.
The drums, wire ropes and sheaves are next in the power train. Wire ropes are a working piece of machinery and need lubrication. There are several systems on the market that provide devices to clean and lubricate wire rope under hydraulic pressure. The lubricant used by these devices is grease. It is the authors’ opinion that even with the hydraulic pressure system, a grease lubricant is not able to penetrate into the core of the wire rope. The authors have had acceptable results from using Lubrication Engineers Inc 2001 Monolec Wire Rope Lubricant.
Even if there are equipment problems, the hoist will typically continue to operate. If there are significant problems the motor may thermal out during the process or something else will break. No matter what its horsepower rating the motor will turn as fast as it can until the heat build up causes failure. Historically most motors are oversized and may work harder than required, which can result in further damage to the downstream power train.
In addition to physical inspection and monitoring the hoist system can be analysed using high tech methods. Vibration tests can provide an excellent tool for isolating specific problem areas. Once a problem is identified, confirmation from technical monitors could be well worth the effort in determining magnitude and in tracking changes.
A Fluke Meter can report the instantaneous voltage and current of the system. This is a popular method of hoist inspection and monitoring because it is quick, easy and inexpensive. Unfortunately the Fluke Meter is ineffective in analysing the whole picture.
Power Meters can provide a wealth of information as well, if time is spent interpolating the findings. Tracking kilowatts over the years can provide information on how the equipment is degrading or improving with usage. The authors conclude that the most common cause of gate/hoist failure is the lack of lubrication on the moving components, or that the lubricant is so old that it no longer functions as designed. Prior to doing any testing, it is the responsibility of the inspection team to verify that lubrication is adequate on all components. This will protect the machinery.
So what types of equipment are available on the market today that can help ensure efficient operation? In the paper ‘Hydraulic gates and logic processors on medium and small dams’ Lisa A Cahill of Watershed Services explains that when the company began replacing sluice gates in 2001, it found that many of the products available on the market were the same as those available in 1950. As a result the company decided to develop a robust gate and actuation system that allowed operation from the shore, rather than having the hand-wheel or operator located directly above the valve.
The system developed by Watershed Services employs a knife valve rather than the traditional sluice gate. This is a robust valve that is frequently used in the paper and pulp industry. If the seals ever need to be replaced, this can be done without taking the valve out of service. The company assemble the valve from components produced in the US, avoiding longer lead times. Cahill says that the knife valve is much stronger than a sluice gate or other types of valves, so is able to meter flow at a very precise level.
The valve is operated by a hydraulic pump based on shore, and the hydraulic fluid used is vegetable oil based. The hydraulic pump, battery and accessories are housed in a weather-proof NEMA 4 enclosure on shore. The battery is typically recharged by a solar cell mounted on a pole attached to the controls cabinet.
The system can also include a programmable logic controller (PLC) which can open and close the valve autonomously based on a wide range of data determined by dam owners. Alternatively the PLC can be interfaced with an existing control system such as a SCADA system and the command to open or close the valve can be given remotely. Triggering events can include lake level, rate of rise, or inflow into the upstream end of the reservoir.
With the PLC and various sensors, the system can be programmed to release a pre-set amount of water downstream, thereby automating compliance with the minimum flow requirements that FERC and other agencies have for maintenance of the health of the stream below the dam. The PLC is capable of complex cascading logic sequences so it can be programmed to act in one way in drought conditions, a different way at normal pool and perform another set of actions at flood stage. Records are kept of every operation performed, and they can be transmitted to a PDA or reported automatically to the dam owner.
The valve control system can be used to transfer a set amount of water from reservoir to reservoir, to control a single reservoir, or to control a large portion of a watershed. For owners of several dams in succession, programming could be put into place so that, for example, when a large amount of inflow is detected in the uppermost basin, valves could open automatically in the lower basins to accommodate the inflow.
The system can also benefit dam owners with dams that have specific areas of concern, such as seepage. Observation of these areas can be automated by installing sensors that would pass information such as flow rates or turbidity to the PLC, along with defining exception parameters and actions to be taken.
In addition, the PLC can be a vital part of an Emergency Action Plan. The threshold events identified in a dam owner’s emergency action plan can be included in the program for the PLC, and actions that can be set automatically could include not only valve operation, but automated notification message via redundant methods to the dam owner, emergency responders, or local residents.
According to Cahill, the hydraulic valve actuation system represents a significant advance over the traditional hand wheeled operated sluice gate used in the past. She gave the example of one client, a water supply company, that had been dealing with a valve dating from the early 1900s. After many attempted repairs, it still did not function properly. After the new hydraulically actuated valve was installed, with a custom weldment to allow it to fit into the existing piping, and the old valve was removed, the water company is now able to reliably meter a specified quantity of water between reservoirs. In a future upgrade, the valve controller will be mated to the company’s existing SCADA system.
Spillway gates are important facilities to secure dam safety during floods, as recognised in the paper ‘Structural reliability evaluation for deteriorated spillway gates’ by Masato Nakajima and Kosuke Yamamoto from the Central Research Institute of Electric Power Industry in Japan. The paper says that although few failure accidents of dam spillway gates have been reported throughout the world – attention must be paid to the rare incidences when such events occur, such as at Wachi dam in Japan and Folsom dam in the US. The paper suggests that such incidents imply that buckling of radial gate arms could cause structural failure. However, few methods have been developed to evaluate aged and deteriorated dam spillway gates. As such the objective of the authors’ study is to develop a structural reliability evaluation for gates.
Structural reliability was originally developed in aeronautical engineering, and this theory as been developed and applied to evaluate quantitative safety of various civil structures. The structural reliability theory allows for the precise calculation of failure probabilities of target structures if failure modes of the structures are specified and information on an external force and structural resistance are obtained, says the paper.
However, it is sometimes difficult to evaluate the failure probabilities of civil engineering works because real structures consist of multi members and the load displacement relationships are non linear. It is also considerably rare to obtain the complete information on load and resistance because natural hazards such as earthquakes are potentially random or stochastic phenomena.
Therefore, a procedure was developed to evaluate structural reliability of spillway gates as follows:
• Step 1 – Failure modes of target structures under the considered load are specified on the basis of past failure examples and non linear structural analysis results.
• Step 2 – Parameters representing dynamic properties of the subject structure are selected, and the range of each parameter is set. Considering statistics data and empirical judgements by experts, relative occurrence probability for each combination of consisting parameters is evaluated.
• Step 3 – By conducting finite element analysis for the subject structure, response values of the target place are obtained.
• Step 4 – A performance function of the structure consists of load variables S and resistance variables R. A probability density function fS(s) of variables S is modelled taking into account the occurrence probability of corrosion and pin friction. Meanwhile a probability density function fR (r) of variables R is modelled considering service duration of a structure currently under study for replacement.
• Step 5 – Using the structural reliability evaluation method as FOSM, AFSOM and the first order Gauss-second moment method, a failure probability and a safety index are computed.
The proposed procedure considers deterioration parameters of dam spillway gates, and employs precise structural analysis methods. A more realistic probability distribution of response values is established by considering occurrence probability of each parameter combination as weight. Structural failure probability of the target dam spillway gate is evaluated without conducting heavy load calculations such as Monte Calro Simulation.
In the paper the authors adopted a design reference in Japan as the analytical model and computed structural reliability of the dam spillway gate on the basis of the model discussed above. They used the general-purpose structure analysis software ABAQUS for three-dimensional structure analysis by the finite element method, conducted in three steps: dead load analysis; stress analysis under design water level load; and stress analysis considering hoist load. The results showed that the failure probability Pf for the design flood level and for the design seismic intensity are both evaluated as 10-5. A comparison with civil engineering structures indicates that the obtained results are consistent with previous research results, although the authors do point out that more research is needed to improve evaluation accuracy by accumulating data on corrosion and pin friction coefficient.
The above mentioned papers were just a selection of information that could be obtained at Waterpower XV on the subject of gates. With such research being undertaken, and the numerous new products under development, it is clear that the efficient operation of gates is an important issue for those involved in the operation, and indeed maintenance of dams and hydro plants. It will be interesting to see if, in two years time at Waterpower XVI, case studies are presented on how new gate systems are making a difference to projects around the world.
For more information on the Waterpower XV event, please visit www.hcipub.com