VFD for Energy Saving & Retrofit

The retrofit business has one characteristic that separates it from new-build work: the saving has to be measured, because it is the only thing being sold. That changes the engineering. A new installation is judged against a specification; a retrofit is judged against a baseline recorded before anything was touched, and if that baseline is not taken, the saving becomes a matter of opinion exactly when it needs to be a number.

The opportunities themselves are well understood and remarkably consistent across industries. Throttled pumps, dampers on fans, relief valves on hydraulic units, compressors running unloaded half the time, cooling systems at fixed flow all year. Each is a case where an operator is deliberately destroying energy to control something, because until recently destruction was cheaper than control. The drive makes control cheap, and the size of the prize is set by how much energy the destruction was consuming.

A credible baseline is measured, not estimated from nameplate data. That means recording the actual power drawn by the machine over a representative period, including the operating pattern — how often it starts, how long it runs, how the load varies within a shift, a week, a season. A drive with a built-in energy meter or a portable power analyser can do this; the difference between the two is mostly the cost of the visit.

The measurement has to cover the whole system, not just the motor. A pump retrofit that saves 30 % at the motor but requires an additional 8 % of pump power because the control valve has been opened fully is a 22 % saving, and presenting it as 30 % is the fastest way to lose a client's confidence on the next project. Differential pressure, flow and temperature at the process are part of the baseline, not optional extras.

Finally, the baseline has to be normalised. Production volume, ambient temperature and product mix all move the energy bill, and a saving measured in a month when the plant ran at 60 % loading proves nothing. Recording the variables alongside the power — even approximately — is what allows the result to be defended when the finance team asks why the bill did not fall as much as the calculation predicted.

Most retrofit projects that fail to deliver their saving fail for the same reasons, and none of them is the drive. The control loop is left on default and hunts, so the plant runs at a higher average output than it needs. The minimum speed is left at zero, so the pump runs in a region where it is inefficient and the motor is unprotected. The valve that used to throttle is left in place and partly closed, so the drive is fighting a restriction. Or the operators, who were not consulted, run the plant the old way.

Avoiding these is a commissioning discipline rather than a technical one. Set the pressure or flow loop against the real process, with the real sensor, before leaving site. Set a minimum speed from the machine's own requirement, and a maximum from the process's. Remove or lock open the throttling device so the energy cannot be destroyed twice. And leave the operators with a written note of what the new setpoints are and why, because a plant that is being operated by people who do not believe in it will drift back to its old settings within a season.

Verification should be planned at the same time as the installation. The same instruments used for the baseline should be used for the verification, on a comparable production pattern, and the result should be presented against the baseline rather than against a theoretical figure. Where the client wants a contractual saving, this is also the mechanism that makes the contract enforceable.

Fan and pump systems dominate, and within them the savings are concentrated in the machines that run continuously against a variable demand. A cooling water circuit sized for August that runs at fixed flow all year is the archetype: the drive is not saving energy on the coldest day, it is saving energy on every day the plant is not at design load, which is most of them.

Hydraulic power units are the second family, and their saving has a different character. A constant-speed pump on a relief valve wastes energy continuously, and the waste appears as heat in the oil, which the cooler then has to remove. Converting it reduces both the electricity and the cooling load, and on a unit where oil temperature was the limiting factor it can also raise the useful duty of the machine.

The third family is the least glamorous and often the largest: compressed air. A compressor that unloads rather than stopping still consumes a large fraction of its full-load power while producing nothing, and a system with a leaky distribution network spends much of its life in exactly that state. Controlling the compressor on pressure with a drive, and fixing the leaks, usually beats every other measure on site in both payback and certainty.