Comparison

Oxygen Concentrator vs PSA Generator: Which to Choose

OxySoothe Editorial

The question usually arrives after a supplier has quoted something at a fraction of the expected cost, and it is a good question. A concentrator and a PSA generator both produce oxygen from air, and a marketing page that says oxygen from air makes them look interchangeable. They are not. One is a bench-scale device with a single purpose; the other is a plant with a control system, a compressor and a maintenance contract.

What actually differs

The separation principle is broadly the same in both cases — a molecular sieve holds nitrogen while oxygen passes through — but everything around it differs in scale, and the differences compound.

  • Scale. A concentrator handles a few litres per minute for one or two occupants. A PSA generator handles tens to hundreds of cubic metres per hour for a building.
  • Where the pressure comes from. A concentrator has a small integrated compressor. A PSA plant has a separate, large, oil-free compressor, and that compressor is the plant's actual constraint on output.
  • Duty. Concentrators are intermittent devices, usually rated for a limited number of hours a day, and their compressors are not sized for continuous running.
  • Installation. A concentrator plugs into a socket and exhausts into the room. A PSA plant needs a plant room, ventilation, a buffer vessel and a distribution network.
  • Output shape. Concentrator output is continuous while it runs. A PSA plant's output depends on feed air, pressure and load factor, so partial-load performance is a specification item rather than a detail.

The consequence of those differences is that the comparison most buyers attempt — price per litre — is only meaningful within one side of a fairly sharp boundary.

The honest case for a concentrator

Concentrators are the right answer more often than a low quotation suggests, and the reasons are practical rather than sentimental.

  • The flow requirement is small and intermittent. A handful of litres per minute for a few hours a day sits squarely in the concentrator range.
  • No distribution network is wanted or possible. A single treatment room, a mobile clinic, a home-care service — a concentrator serves these without pipework.
  • Redundancy matters more than unit cost. Two or three portable units cost less than one plant and keep working when one fails.
  • The site cannot support a plant room, a compressed air system, a ventilation design and a maintenance contract.
  • Demand is uncertain and the operator wants to start small and grow into it.

The weaknesses deserve equal billing. Efficiency per litre is poor next to a plant. The noise and the heat belong in the room rather than in a plant space, which matters in a clinical environment. The units need regular filter and cannula changes that somebody has to remember to do, and they degrade if they are left running continuously.

The honest case for a PSA generator

A plant wins when demand is continuous and large enough that its efficiency advantage outweighs everything that comes attached to it.

  • Demand is continuous across the working day rather than episodic. This is the defining condition, and it is worth being strict about it — a plant sized for a peak and run at a fraction of it gives up much of the advantage it was bought for.
  • The volume is high enough to matter. Below a few cubic metres per hour, capital cost and engineering overhead dominate and the arithmetic stops working.
  • Oxygen is fed into a distribution system rather than carried to the bedside by hand.
  • The site can accommodate a plant room, dedicated power, ventilation and a maintenance arrangement.
  • Consistency across a site matters more than unit cost, because one plant means one set of operating parameters rather than many units drifting independently.

The weaknesses are real: high capital cost, a dependency on compressed air, a finite overhaul cycle on the compressor and sieve beds, and an all-or-nothing failure mode unless a buffer and a standby arrangement have been designed in.

Where the boundary actually sits

The dividing line is usually drawn at flow, and flow alone is a poor guide. A better way to frame it is to ask what happens to the oxygen once it has been made, because that is what determines whether the scale of production matters.

  • If the oxygen is delivered by cylinder or carried by a person to the bedside, the delivery cost dominates and the production method matters less. Concentrator territory.
  • If the oxygen goes into a pipe, the production method starts to matter, because the plant's output becomes the whole site's supply and a stop is a site-wide event.
  • If the requirement is a guaranteed quantity at a specified purity at a specified time, continuity and measurement become the deciding factors rather than unit cost.
  • If the demand can be met by holding stock — a buffer vessel or cylinder bank — the two options sit much closer together than the price comparison suggests, and the comparison should be made against the cost of that stock.

It is worth remembering that these are not mutually exclusive and frequently appear in the same building. A plant carries the base load for wards and theatres; a small concentrator or cylinder bank covers the tail end of demand and provides cover during a plant stop. Most well-designed sites end up with both, and the specification should say which is which.

The parameters that actually decide it

  • Peak and average flow, stated separately and in the same units.
  • Hours of use per day, and whether that use is continuous or episodic.
  • Delivery pressure required, and whether oxygen goes into a pipeline or is used at the bedside.
  • Purity requirement at the point of use, and how it will be verified.
  • The redundancy target, expressed as what must keep running if any single component fails.
  • Whether there is space, power and ventilation for a plant room, and who will maintain it.

Questions to ask either supplier

  • What is the output at the stated pressure, and what is the specific power consumption at our expected load factor rather than at full load?
  • What is the measured oxygen concentration at the outlet, how often is it measured, and against what reference?
  • What happens to the supply when this unit stops, and what covers that gap?
  • What are the consumable intervals and the annual service cost, calculated over five years?
  • Can the system be run at partial load, and what is the output and the purity at partial load?
  • Who holds the spares for this model, and where are they located relative to the site?

So to answer the question directly: choose a concentrator when demand is small, intermittent or delivered at the bedside, and where a plant room or distribution network is not wanted. Choose a PSA generator when demand is continuous, high, and fed into a distribution system. Where both remain plausible, the deciding question is rarely the technology itself — it is what should happen to the supply when one unit fails.

If you send us your peak flow, average flow, hours of operation and whether the oxygen goes into a pipeline, we will map it to a configuration and tell you which side of the line you are on, including the compromises.

Tell us your needs

We can tailor a solution for you

Oxygen Concentrator vs PSA Generator: Which to Choose