Monthly: May

It is advisable to constantly monitor or at least periodically check the energy parameters of the pumps. In practice, at the workplace, measuring lifting height, power and specific weight is relatively simple and can be carried out with high accuracy. However, precise performance measurement is problematic. Without it, however, it is impossible to precisely estimate the current efficiency of the pump, taking into account the deterioration of its technical condition due to wear.

During the pump's operation, it is advisable to constantly monitor or periodically check its energy parameters to ensure that it is operating at optimal efficiency. This is important not only for technical reasons, but also for economic reasons, because the energy costs to drive the pump are significant in the total cost of pumping.

Energy costs and efficiency at the actual operating point.

From the point of view of energy costs, what is important is not the maximum efficiency that the pump can achieve at the optimal point of the characteristic curve, but the efficiency at the actual operating point. The issue becomes more complex when the operating point is variable, because then the energy efficiency of the system is determined by the control method used. In the simpler case, where the operating point is approximately constant, energy losses are mainly due to two reasons:

1. Incorrect selection of the pump for the system, as a result of which it operates with an efficiency lower than the maximum.                                                                                                                 This results either from an error in the selection (e.g. incorrect estimation of flow losses or the assumption of too high excess parameters) or from a change in parameters that has occurred since the system was designed. The latter situation is often encountered in practice.
2. Reduced efficiency of a properly selected pump.                                                    The reduction may result from the delivery of an improperly made pump, with a lower efficiency than offered or from deterioration of the pump's technical condition due to wear. The wear rate depends on many factors, such as the type of pumped medium, rotational speed and material selection. Regardless of operating conditions, a noticeable deterioration in efficiency is expected after several years of operation. This effect often goes unnoticed in the case of automatic speed control. If the pump operates at a constant speed, its efficiency decreases over time due to wear and tear, and the user decides to renovate it because the pump no longer fulfills its function. However, in the case of automatic regulation, the effect of wear is compensated by increasing the rotational speed, as a result of which the pump maintains the required pressure or capacity, which, however, comes at the cost of increased energy consumption.

Of course, in practice there is often a combination of both factors mentioned above.

In each case, it is recommended to periodically verify the energy consumption of the pump operating in a given system, which can be carried out by measuring its energy parameters. As a result, based on the results, it is possible to assess the level of energy consumption of the pump unit, possible energy savings and estimate the economic effects of possible modernization. On this basis, decisions can be made regarding investments in improving the efficiency of pumping systems (e.g. through their modernization or renovation) based on a reliable technical and economic analysis. Without measurement data, such decisions must be made to some extent intuitively, which creates the risk of allocating investment outlays on projects that do not guarantee the expected results.

Possibility to measure the energy parameters of pumps at the workplace.

To assess the efficiency of the pump unit, measure:
a) efficiency,
b) lifting height,
c) power consumption,
d) specific gravity of the pumped medium.

In practice, at the workplace, measuring lifting height, power and specific weight is relatively simple and can be carried out with high accuracy. However, precise performance measurement is problematic. Performance measurement methods and the problems associated with them are described, among others, in: in [1, 2]. Regardless of the type of flowmeter used, obtaining accurate results requires measurements on a straight section of the pipeline. Typically, a straight section of 15 diameters is required, although manufacturers of some types of flow meters allow shorter sections. This is due to the fact that the algorithms used to convert the measured physical quantity into a result in the form of flow rate are based on the assumption that there is a stabilized velocity profile in the pipeline, close to a rectangular one. If there are elements in front of the flow meter that disturb the flow (bends, fittings, changes in cross-section, etc.) or generate vortices, the velocity distribution at the measurement site is different from the assumed one and the measurement is error-prone. Additionally, if a flow meter is not permanently installed in the pump system, the so-called non-invasive flowmeters, most often ultrasonic ones. The measurement is made using measuring heads placed on the outer walls of the pipeline. In such a case, additional measurement uncertainty is introduced resulting from the condition of the pipeline walls. Any deposits interfere with the ultrasonic signal and increase measurement errors. If the measurement of efficiency in a more complex pumping system is not made directly on the discharge pipeline of the pump being measured, but, for example, on a collective manifold supplied by several pumps, a doubt arises as to whether the measured efficiency strictly corresponds to the efficiency of the pump. Due to the lack of full tightness of the shut-off fittings, leaks may occur, e.g. return flows through pumps connected in parallel with the tested one, which results in a reduction of the measurement result.

Measurement uncertainties and ways to reduce them.

The accuracy of measurements carried out at the workplace depends on the configuration of the system (existence of straight sections of pipelines, condition of walls), the class of measuring instruments used and the experience of the people conducting the measurements. Even in favorable conditions, errors of several percent can be expected, mainly resulting from inaccurate performance measurement. For this reason, the standard [3] recommends that acceptance measurements of new pumps or pumps after major renovation should be carried out not at the workplace, but at a special measurement station, which should be operated by the manufacturer or the renovation contractor. To eliminate possible doubts as to the reliability of measurements carried out by the manufacturer or renovation contractor, they can be carried out under the supervision of a specialized third party company.

For the sake of clarity, it should be added that there are methods of measuring efficiency, such as the isotope method, which allow to reduce the measurement uncertainty also in the absence of straight sections of pipelines and when the condition of their walls is unknown. Due to their complexity and cost, such methods are mainly used for pump systems with the highest parameters (high efficiency and power) and are not used on a large scale.

The fact that the applicable technical standard [3] does not recommend carrying out acceptance measurements at the workplace should not lead to the conclusion that such measurements should not be carried out at all. On the contrary, for the reasons discussed above, periodic assessment of the energy efficiency of a given pump system is advisable. As stated, measurement uncertainty of at least several percent should be taken into account, but even such accuracy allows obtaining useful conclusions regarding the advisability of modernization or renovation.

Measurement uncertainty can be significantly reduced if the possibility of carrying out measurements is taken into account at the design stage of the pump system. Ideally, permanent installation of flow meters for individual pumps should be planned, taking into account the required straight pipeline sections, which allows for continuous monitoring of efficiency. If, for economic reasons, such a solution is not possible, as a minimum, at least a straight section of the pipeline of the required length for performance measurements should be designed. In such a section, a removable section between the flanges can be installed, which can be replaced with a flow meter for the measurement period. If you plan to use non-invasive flow meters, it is advisable to replace the entire straight measuring section to eliminate doubts about the condition of the walls.

If the installation designer takes into account all the recommendations of the standard [3] and if appropriate expenditure is allocated to the project, it is possible to carry out measurements at the workplace with an accuracy that meets the requirements [3]. Currently, however, pump systems designed in this way are rare.

Assessment of pump parameters without the possibility of conducting precise measurements.

If accurate measurements at the workplace are not possible (which, as stated above, is most often due to difficulties in measuring efficiency), the pump characteristic curve prepared during acceptance tests can be used to assess the pump operating point. If we want to draw accurate conclusions on such a basis, it should not be the characteristics of the pump type or the average offer characteristics, but the characteristics of a specific unit, which may differ from the type characteristics by permissible tolerances. Please remember that the H(Q) characteristic is calculated based on the measurement results using the formula:

where index 2 indicates the parameters at the outlet, and index 1 indicates the parameters at the inlet to the pump. (p means pressure, v means speed, ah means the height of the pressure measurement point above the adopted reference level).

In practice, the first term (pressure difference between the outlet and inlet divided by the specific gravity) is dominant. If we know the height difference between the pressure measurement points and the diameters of the suction and discharge pipelines, the H(Q) characteristic calculated based on formula (1) can be corrected to obtain the relationship Δp (Q), where Δp = p2 – p1.

Then, by measuring Δp = p2 – p1, we can draw conclusions about the location of the operating point, i.e. the current efficiency.

Fig. 1. Assessment of the pump operating point location based on H measurement.

For example, let's consider the cooperation of a pump with the H(Q) characteristic shown in Fig. 1 as a solid line with a system whose Hukł (Q) characteristic is shown in the dashed line. The pump was selected so that its operating point was at the intersection of its characteristics with the system characteristics, i.e. at the parameters Hn, Qn. Let us assume that after installing the pump at the workplace, during control measurements Δp = p2 – p1 was measured, which value, after correction by the last two terms in formula (1), turned out to be equal to H1, and therefore lower than Hn. If we can be sure about the pump's characteristic curve, we conclude that it operates with a capacity Q1 higher than Qn. In the case of a pump with increasing power consumption with efficiency, selected for maximum efficiency at Qn, as in Fig. 1, operation with efficiency increased to Q1 will take place with reduced efficiency and increased power consumption, and therefore with reduced energy efficiency. The probable cause of this phenomenon is that the actual flow resistance in the pump system turned out to be lower than assumed at the design stage (the dashed-dotted line running below the dashed line indicating Hukł), as a result of which the pump with a specific characteristic "moved" to a higher efficiency. To reduce the efficiency to the assumed Qn value and improve efficiency, the rotor diameter should be corrected.

In this way, practical conclusions can be drawn without measuring performance, solely on the basis of pressure measurement. They will be correct as long as the pump characteristics are accurate. This can be assumed in the case of a new pump or one that has been in operation for a relatively short time. However, after a certain period of operation, due to wear, the pump characteristics decrease (pump parameters drop). Then, if we find a decrease in Δp based on the measurements, we are no longer able to draw clear conclusions. The reduction in Δp may not only result, as in the case of a new pump, from reduced resistance in the system, but may also result from a reduction in the pump's characteristics. If, as a result of the deterioration of the technical condition, the pump characteristics have decreased as shown by the line below the factory characteristic H(Q), and the system characteristics are as assumed, then the operating point as a result of the intersection of the characteristics will be set at a lower efficiency Q2, and determined on the basis of measurement Δp the lifting height will still be H1. The decrease in the measured head may therefore result from both a decrease in system resistance and a decrease in pump parameters, as well as from the combined effect of both factors, but separating them is difficult. Some qualitative conclusions can be drawn in this regard. If the reduction in lifting height results from a reduction in system resistance, a pump with the P(Q) characteristic, as in Fig. 1, will show an increase in power consumption. However, if the pump shows a decrease in lifting height due to wear, this is most often accompanied by a decrease in power consumption. Therefore, by measuring, in addition to the increase in pressure generated by the pump, also its power consumption, we can draw conclusions about its technical condition. However, it should be emphasized that without measuring the efficiency, it is impossible to precisely determine the current efficiency of the pump at the workplace. By analyzing the compliance or possible discrepancies of the lifting height and power measurements in relation to the factory characteristics, it is only possible to qualitatively assess whether the pump deviates from its initial state, but without measuring the efficiency it is impossible to precisely assess the current level of efficiency.

It can be assumed that the system characteristics change much slower than the pump characteristics, which is justified by the fact that the pump elements wear out faster than the pipeline elements, because the pump has much higher liquid flow rates. With this assumption, it would be possible to mark a certain section of the pipeline as an approximate flow meter. After installing the pump, by measuring the pressure increase it produces, it would be possible to determine the efficiency from its factory characteristics. At the same time, by measuring the pressure drop in a specific section of the pipeline at this known capacity, it would be possible to estimate the resistance coefficient, assuming that the pressure loss is proportional to the square of the capacity Δp = a Q2. Having the resistance coefficient a estimated in this way, it would then be possible to assess the efficiency flowing through this section of the pipeline based on the pressure drop measurement, which would then be used to assess the condition of the pump. However, such a method will be unsuitable in a situation where deposits form on the pipeline walls over time, which precludes the assumption of constant flow resistance.

It should be emphasized that estimated methods of assessing the condition of the pump, similar to those described above, can provide qualitative conclusions, provided that such analyzes are carried out by people with appropriate knowledge and experience. They may prove useful for an initial assessment of the situation, but they will always be error-prone and will not be able to provide precise information about the pump's wear and tear and its current efficiency. Therefore, if there is no capacity measurement in the installation, accurate information about the technical condition of the pump can only be obtained by sending it for testing to an appropriate test stand, e.g. at the manufacturer's.

Summary.

For reasons of energy efficiency, it is advisable to constantly monitor or at least periodically measure the energy parameters of pumps. The possibility of such continuous or periodic measurements should be provided for at the design stage of the pump system. In particular, it is advisable to install the flow meter or, at least, to provide conditions for its temporary installation during the measurement period.

If there are no conditions in the pump system for conducting accurate performance measurements, the assessment of the pump's fit to the system can be made after its installation based on accurate factory characteristics.

Without measuring efficiency, it is impossible to precisely assess the current efficiency of the pump, taking into account the deterioration of its technical condition due to wear. Experienced people can make approximate estimates in this respect based on available, partial measurement results. However, such estimates will always be subject to uncertainty. The recommended practice would be to send the pump for testing at a specialized test stand if approximate assessments suggest deterioration of the pump's technical condition. Accurate measurements in such a case would provide precise arguments for sending the pump for renovation or modernization.

Dr. Eng. Grzegorz Pakula

1. M. Cichoń, Non-invasive flow measurements, or dealing with myths – part 1, Under Control, 04/2014 (30).
2. M. Cichoń, Non-invasive flow measurements, or dealing with myths – part 2, Under Control, 01/2015 (31).
3. Standard PN-EN ISO 9906, Centrifugal pumps - acceptance tests of hydraulic parameters - Accuracy classes 1,2, 3 and 2012, June XNUMX.