Two-stage compression and what a 10-100% range really tells you

What two stages change

In a single-stage centrifugal machine the whole pressure rise happens across one impeller. In a two-stage machine it is divided, and that division changes two things worth understanding before the architecture label means anything.

The first is the work each stage has to do. A smaller pressure ratio per stage generally allows each impeller to operate closer to where it is efficient, and the benefit grows as total lift grows. The second is that a two-stage arrangement creates an intermediate pressure — a place in the cycle that can be used, most commonly for economising, which is why two-stage machines and economised cycles so often appear together.

Neither of these is automatic. Two stages are not a guarantee of better performance at every condition; they are an arrangement that pays off most clearly where the lift is high enough to justify it.

Three architectures, three intents

The published architecture fields separate the families in a way that is easy to miss because they share the same first two words.

The S series is published as integrated, two stage, with its stated use being air- and water-cooled chiller platform development. The T series is published asintegrated, two stage, single-end, for application-specific chiller and thermal-system integration. The H series is published as integrated, two stage, symmetric, and its stated use is high-lift and low-temperature operating conditions.

The pattern is that these are not three sizes of the same machine. They describe different arrangements around different intents, which is why lift and application tend to move a project between families more often than capacity does.

Reading a 10-100% range

The S, T and H entries publish capacity regulation from 10 to 100 per cent. It is worth being precise about what that figure is.

It is a capability statement: the range across which the catalog publishes the machine as regulating. It is not an efficiency curve. Nothing in a regulation range says the machine is equally efficient at 15 per cent as at 85 per cent, and assuming otherwise is one of the more expensive assumptions available in chiller design.

It also says nothing about where any particular plant will run. Two projects quoting the same machine can have completely different load profiles, and the profile is the project's fact to supply.

Hours decide, not peaks

Chiller plant is normally sized for a condition it rarely meets. Design day exists to make sure the plant can cope; the rest of the year is where the energy is actually spent.

That is why a part-load profile is worth more to a selection conversation than a single peak figure. A machine that is excellent at the design point and mediocre where the plant spends three thousand hours will lose to one with the opposite shape, and neither of those outcomes is visible from a capacity number.

Wide regulation helps here, but in a specific way: it delays the point at which a machine has to cycle or hand over. What it does not do is make low-load operation free.

What this asks of the control strategy

A wide regulation range moves work into the control layer. Three questions follow from it and are better answered during selection than during commissioning.

Where is the machine expected to sit for most of its hours, and is the control strategy tuned for that region rather than for the design point? What happens at the bottom of the range — does the machine hold, cycle, or hand over to another? And in a multi-machine plant, what stops two machines with wide regulation from chasing each other around the same load?

None of these are answered by the architecture or by the regulation range. They are answered by the load profile and the sequencing strategy, which is why both belong in the brief alongside the duty point.