The energy saving potential when a frequency converter is used depends on the type of load being driven and the optimisation of the efficiency of the pump or the drive by the frequency converter, as well as how much time the system operates under partial load. Domestic water and wastewater systems are designed for rarely-occurring peak loads, so they are usually operated under partial load.
Centrifugal pumps and fans offer the largest potential for energy savings. They fall in the class of fluid flow machines with variable torque curves, which are subject to the following proportionality rules. The flow increases linearly with increasing speed (rpm), while the pressure increases quadratically and the power consumption increases cubically.
The decisive factor for energy savings is the cubic relationship between rpm and power consumption. A pump running at half its rated speed, for example, needs only one-eighth of the power necessary for operation at its rated speed. Even small reductions in speed thus lead to significant energy savings. For example, a 20% speed reduction yields 50% energy savings. The main benefit of using a frequency converter is that speed control does not waste power (unlike regulation with a throttle valve or damper, for example), but instead adjusts the motor power to match exactly the actual demand.
Additional energy savings can be achieved by optimising the efficiency of the pump or drive with frequency converter operation. The voltage control characteristic (V/f curve) supplies the right voltage to the motor for every frequency (and thus motor speed). In this way, the controller avoids motor losses resulting from excessive reactive current.
Article by Gregers Geilager
The savings are not as dramatic as it seems. If the pump needs to pump 10,000 litres of water it saves more energy if run at 100% speed DOL rather than a VSD. Sure the VSD is a good soft starter but it has 4-5% losses! and the payback from initial cost needs to be under 3 years to be cost neutral. The best way to save energy is to run a motor at 100% load @ 100% speed. Too many people oversize the motor. The old multistage pump solution is still the most efficient pump solution without the losses associated with a VSD costs, harmonics and cable issues
I agree the potential energy savings are not as “dramatic as it seems” and in reality a 20% speed reduction will often not yield 50% energy savings, but the energy savings are very significant, very often when using a VSD instead of some other method of setting or controlling the flow on these types of applications.
One reason the savings are not as dramatic in reality is because the cubic relationship between rpm and power consumption of a fan or centrifugal pump is only a theoretical relationship applicable to fixed systems with no static pressure (fan systems) or static head (pump systems) and constant pump efficiency. Examples of systems that come closest to this description are centrifugal pump systems with minimal lift, pumping water out of an open ended pipe or a condenser fan or a cooling tower fan. However many pump and fan systems are not like this. Many systems have a significant static pressure or static head which the fan or pump operates against all the time and/or have changing pressure/head in the system due to other components and the operation of the system. This changes the system curve which can impact the potential energy savings.
Another reason is that pump efficiency changes as the speed of the impeller changes. In addition, if the density of the fluid being pumped changes, that also affects the power consumed by the pump. This can happen in a mine processing facility for example. Similar statements apply to fan systems.
Understanding the fan or pump curve and system curve and the system as a whole, including the static pressure or static head of the system, is the best way to be able to estimate the realistic energy savings using a VSD.
I agree that in the ideal world, if a pump needs to pump 10,000 litres per minute of water it saves more energy if it operates at 100% speed rather than using a VSD, because there are typically 2%-3% losses in a VSD. However getting a pump which is designed to deliver exactly the design flow at the actual head of the as built system when operating at the fixed speed of a 2, 4 or 6 pole motor (3000rpm, 1500rpm or 1000rpm) and have a power demand exactly the same as a standard motor power (e.g. 110kW or 132kW or 160kW) is virtually impossible. A pump’s impeller can be trimmed to ensure it delivers only the required flow at a specific head (pressure) when operating at the selected motor’s 100% speed, but if the system pressure is not exactly the same, it will pump at a higher or lower flow rate once it’s installed in the system. At the same time, although theoretically it would be possible to manufacture a motor which delivers exactly the power required by the pump at this speed/flow/head design point, it’s highly unlikely that power will be exactly the same as the power of a standard motor and therefore in reality, even if the pump is matched to the actual system design operating point, it’s highly likely that it will have a power demand which is different to the design power of the motor (e.g. the pump will have a power demand of say 97kW and therefore a standard, oversized 110kW motor would be selected).
Because of the many different factors that can impact the energy consumed by a fan or centrifugal pump once it’s installed in a real system, operating at a fixed speed without an oversized motor, will often not be the simplest to engineer or most energy efficient solution.
Instead using a VSD is often considered a cost effective solution which makes it simple to set the pump or fan to its optimum operating point during commissioning and provides significant energy savings and reduced operating costs even when there is no requirement for variable flow and it’s just used as a flow setting device, rather than using a throttling valve or damper for example.
It’s not uncommon that on a new system, when all costs are considered, optimally designing a VSD into the system can be a lower cost solution, or at worse only a small premium on the initial investment, which is recovered within a few months of operation and even retrofitting a VSD to an existing system often results in a payback of the investment costs within 18 months due to the significant energy savings and resulting lower operating costs.
Great article forwarding to all contatcs,