How Oxygen Concentrators Produce Oxygen

How an oxygen concentrator takes ordinary room air, removes most of the nitrogen, delivers oxygen-enriched gas continuously, and why normal flow does not prove normal oxygen concentration

An oxygen concentrator can create one of the easiest troubleshooting traps in respiratory equipment.

Back to Biomed Basics

What This Page Explains

This page covers:

The Simple Version

An oxygen concentrator takes room air and uses pressure swing adsorption to separate much of the nitrogen from the oxygen. Room air is compressed and directed into a container filled with molecular-sieve material, commonly a zeolite. Under pressure, the sieve material preferentially adsorbs nitrogen while allowing much of the oxygen to pass through.

The concentrator normally uses at least two sieve beds so oxygen production can continue while the material is regenerated. One bed is pressurized and producing oxygen-enriched gas while another is depressurized so the trapped nitrogen can be released back into the room. Valves repeatedly switch the beds between those states. The oxygen-enriched product gas is collected in a reservoir and then delivered through a regulator or flow-control system.

A concentrator problem can therefore originate in the compressor, air filters, sieve material, switching valves, check valves, purge path, product reservoir, pressure regulator, flow control, oxygen sensor, or cooling system. Preserve the actual oxygen concentration from an independent analyzer, outlet flow, outlet pressure, compressor pressure, cycle behavior, operating temperature, alarm state, and whether performance changes with flow setting or runtime before assuming the sieve beds are exhausted.

Start With Room Air

The machine does not start with pure oxygen.

It starts with:

Room air.

Dry atmospheric air is approximately:

The exact composition and humidity vary somewhat with conditions.

The concentrator's job is to remove enough nitrogen that oxygen becomes the dominant remaining gas.

It is therefore better to think of the machine as:

A nitrogen-removal system

rather than:

An oxygen generator in the chemical sense.

The oxygen was already in the room.

The concentrator is separating it.

The Machine Does Not Usually Produce 100% Oxygen

This is another important misconception.

A typical medical oxygen concentrator produces oxygen-enriched gas rather than perfectly pure oxygen.

Exact concentration specifications vary by model and flow setting, but many concentrators are designed to deliver gas in roughly the low-to-mid 90% oxygen range under normal rated conditions.

The remaining gas may include:

The exact specification must come from the manufacturer.

This matters when testing.

If a unit is designed to produce approximately 93% oxygen, reading 93% does not mean:

It's 7% bad.

That may be exactly correct.

Pressure Swing Adsorption

The core separation process is called:

Pressure swing adsorption, commonly abbreviated PSA.

The name describes what is happening.

Pressure is applied.

Certain molecules are adsorbed onto a material.

Then pressure is released so that material can be regenerated.

The process repeatedly:

Pressurizes

and:

Depressurizes.

That pressure swing allows the same sieve material to be reused continuously.

Adsorption Is Not the Same as Absorption

These words sound similar but describe different processes.

Absorption

A substance enters the bulk of another material.

Adsorption

Molecules attach primarily to the surface of a material.

Molecular sieve operation depends mainly on:

Adsorption.

That surface behavior is what allows the sieve material to temporarily hold nitrogen.

Molecular Sieve Material

The separation material in many oxygen concentrators is a type of:

Zeolite.

Zeolites have microscopic pore structures and surface properties that allow them to adsorb certain gases more strongly than others.

Under the pressure and conditions used in an oxygen concentrator:

Nitrogen is preferentially adsorbed.

Oxygen passes through in greater proportion.

The result leaving the bed becomes:

Oxygen enriched.

The sieve is not acting like a simple screen where nitrogen molecules are physically too large to pass through.

The separation depends on:

That distinction matters.

The Compressor Makes the Separation Possible

The concentrator needs pressure.

An internal compressor draws room air into the machine and raises its pressure.

Without adequate compression:

The molecular sieve cannot work the way it is designed to.

A weak compressor can therefore produce a very interesting symptom.

Air still moves.

The machine still sounds like it is running.

The flowmeter may still show output.

But the pressure inside the sieve beds may never become high enough for effective nitrogen adsorption.

The user sees:

Low oxygen.

The compressor can be the real cause.

Intake Filtration

Before room air reaches the compressor, it normally passes through filters.

These may include:

depending on the model.

The filters protect:

from contamination.

A clogged filter creates a restriction.

The compressor now has to work harder to move the same amount of air.

That can cause:

A dirty filter can therefore look like a much more expensive internal failure.

Airflow Into the Machine Is Part of the Process

A concentrator needs enough room air to make the product gas.

If the unit is placed:

its performance can suffer.

This becomes especially important because the compressor produces heat.

Poor ventilation can create:

Higher internal temperature.

That can change:

So a warm room and blocked airflow can become an oxygen-concentration complaint.

Two-Bed Operation

Most continuous-flow concentrators use at least two molecular-sieve beds.

Think of them as:

Bed A

and:

Bed B.

The beds alternate jobs.

While one is producing oxygen-enriched gas:

The other is regenerating.

Then they switch.

This is what allows continuous output even though each individual sieve bed periodically has to release the nitrogen it captured.

Bed A Pressurizes

Compressed air is directed into Bed A.

Pressure rises.

Nitrogen begins adsorbing onto the sieve material.

Oxygen-rich gas exits the product side of the bed.

Some of that product gas goes toward the patient output.

Some may also be used elsewhere in the separation cycle.

Bed B Regenerates

At the same time, Bed B is released toward a much lower pressure.

When pressure falls:

The zeolite's ability to retain the adsorbed nitrogen decreases.

Nitrogen is released.

That nitrogen-rich waste gas is vented back into the room.

This is why concentrators produce:

An exhaust or purge sound.

Then the Beds Switch

After a controlled interval:

Bed A begins regenerating.

Bed B becomes the production bed.

The valve system redirects compressed air.

The machine repeats this process continuously.

A typical concentrator may cycle every few seconds, though timing varies significantly by design.

The rhythmic change in sound is often normal.

The Switching Valves Are Critical

PSA depends on gas moving through the correct bed at the correct time.

That requires valves.

Depending on the design, the machine may use:

The control system opens and closes these paths in a timed sequence.

A valve that fails completely can stop oxygen production.

A valve that only leaks slightly can be more difficult.

The concentrator may still produce oxygen.

But concentration falls.

Why a Small Valve Leak Matters

Suppose Bed A is supposed to be pressurized.

But a valve allows some pressure to leak toward the exhaust.

Bed pressure is lower than expected.

Nitrogen adsorption decreases.

Now more nitrogen passes through the bed.

The product gas becomes less concentrated.

The compressor may still sound normal.

The flow may still be normal.

The actual problem is:

The pressure sequence is wrong.

Bed Pressure Is Useful Evidence

On serviceable concentrators, manufacturers may provide:

Bed or compressor pressure can help distinguish several faults.

If pressure is low everywhere:

Consider:

If pressure is produced but oxygen concentration remains poor:

Consider:

The exact interpretation depends on the architecture.

Equalization

Some concentrator designs briefly connect the two sieve beds during the transition between cycles.

This is called:

Pressure equalization.

Instead of wasting all the pressure from the bed that just finished producing, some compressed gas can flow into the lower-pressure bed.

This can improve efficiency.

It also means that an equalization valve or pathway becomes part of the separation process.

A fault there may not stop operation completely.

It can reduce performance.

Purge Flow

A small amount of oxygen-rich gas may be sent backward through the regenerating sieve bed.

This is called:

Purge.

The purpose is to help remove nitrogen from the sieve material.

That prepares the bed for the next production cycle.

Too little purge:

Nitrogen may remain.

The next cycle starts with partially loaded sieve material.

Oxygen concentration falls.

Too much purge:

Useful product gas is wasted.

Efficiency falls.

The purge path is therefore small but important.

Check Valves

Check valves ensure gas moves in the intended direction.

For example, oxygen-enriched gas from one sieve bed may need to enter the product reservoir without flowing backward during the other part of the cycle.

A leaking check valve can allow:

Again, the machine may still run.

The quality of the gas can degrade.

Product Reservoir

The gas coming directly from the sieve beds is not perfectly steady.

One bed is producing.

Then the other.

Pressure pulses occur.

The patient should not receive strong alternating surges every few seconds.

So concentrators commonly use:

A product reservoir or accumulator.

This tank stores oxygen-enriched gas and smooths the output.

The reservoir helps create:

Product-Tank Leak

A leak in the reservoir or its fittings can reduce:

The compressor may run normally.

The sieve beds may separate gas correctly.

But the final delivery system loses product gas before it reaches the patient outlet.

Pressure Regulator

The product reservoir may operate at a pressure higher than what is delivered to the patient.

A pressure regulator reduces that pressure to a controlled outlet level.

This stabilizes the delivery system and supports predictable flow.

A failed regulator can therefore create:

without necessarily changing oxygen concentration inside the reservoir.

Flow Control

Many concentrators use a variable flowmeter or calibrated flow-control system.

The operator selects:

1 L/min.

2 L/min.

5 L/min.

and so on.

The concentrator has to maintain oxygen concentration across its rated flow range.

This is where another important relationship appears:

Higher flow places greater demand on the separation system.

Why Oxygen Concentration Can Fall at High Flow

The concentrator has a finite capacity.

At low flow:

The system has more time and product-gas reserve relative to demand.

At maximum flow:

The compressor, sieve beds, valves, and reservoir are working much closer to their design limit.

A marginal unit may therefore show:

2 L/min → 94%

3 L/min → 93%

5 L/min → 84%

That pattern is extremely useful.

The machine is not completely unable to concentrate oxygen.

It cannot maintain concentration under full demand.

Rated Flow Matters

Do not test a concentrator at a flow beyond its rated maximum and then call the concentration low.

The concentration specification applies under defined operating conditions.

The OEM procedure matters.

Flowmeter Ball Position

Traditional flowmeters may use a floating ball.

Correct reading depends on:

Where the manufacturer specifies the ball should be read.

Often that may be:

Center of the ball

but designs vary.

Read the scale correctly.

Flowmeter Is Not Always the Actual Flow

A rotameter indicates flow based on:

It can become inaccurate.

So if output flow matters:

Verify it with a calibrated flow analyzer according to the service procedure.

The front-panel flow indication is not an independent proof.

Backpressure Changes Flow

The outlet may normally deliver into:

These create resistance.

A concentrator should maintain appropriate flow within its rated outlet-pressure capability.

If the downstream path becomes highly restricted:

Flow can fall.

The flowmeter or internal compensation behavior depends on design.

Long Tubing

Long lengths of narrow tubing create resistance.

This can reduce delivered flow, especially at higher flow settings.

The concentrator outlet may be correct.

The flow at the patient end may be lower.

That is a downstream-delivery problem.

Humidifier Bottle

Some users connect a bubble humidifier.

The gas has to push through water.

That creates backpressure.

A blocked humidifier diffuser can create even more resistance.

If flow is normal without the humidifier but poor with it:

The concentrator may not be the failure.

Oxygen Concentration Must Be Measured Independently

One of the strongest tests is simple:

Connect a calibrated oxygen analyzer to the outlet using the specified setup.

Allow the concentrator to warm up and stabilize.

Measure concentration.

Do not guess from:

Oxygen is colorless and odorless.

Your senses cannot tell the difference between:

85%

and:

95%.

Warm-Up Matters

Immediately after startup:

The sieve beds and product reservoir may still contain ordinary air.

The system needs time to purge nitrogen and establish stable PSA operation.

Oxygen concentration may therefore rise gradually.

A test taken thirty seconds after startup can incorrectly fail a healthy unit.

Follow the manufacturer's stabilization time.

But Slow Concentration Rise Is Also Evidence

Suppose the OEM expects stable concentration after several minutes.

A unit takes twenty minutes.

That may suggest:

The trend matters.

Oxygen Sensor

Many concentrators monitor their own oxygen concentration.

Sensor technologies can vary.

Possible designs include:

The sensor provides the control system with an estimate of product-gas quality.

Electrochemical Oxygen Sensor

Some sensors generate an electrical signal related to oxygen partial pressure.

They can age over time.

Aging can produce:

The concentrator can therefore produce correct oxygen while falsely reporting:

Low oxygen.

Ultrasonic Oxygen Measurement

Some systems determine oxygen concentration using differences in acoustic properties between gases.

Because oxygen and nitrogen have different physical properties, ultrasound propagation can help estimate gas composition.

These sensors may also incorporate:

measurements depending on design.

Again:

Sensor architecture varies.

Internal Sensor vs External Analyzer

If the concentrator alarms low O₂:

Do not assume the gas is actually low.

Check it.

External analyzer says 94%

Internal system says 82%.

Now the oxygen-sensing path becomes suspicious.

Both say 82%

Now the gas-production system deserves attention.

That distinction is extremely powerful.

Low-Oxygen Alarm

Many concentrators alarm if product oxygen falls below defined thresholds.

Exact alarm points and delays vary.

The system may provide:

depending on concentration.

The alarm protects against silent deterioration.

Without it, the unit could continue delivering gas at normal-looking flow while oxygen concentration falls substantially.

Why Concentration Can Deteriorate Slowly

Sieve material can gradually lose effective performance.

Possible contributing factors include:

The compressor may also gradually lose capacity.

This produces the classic:

Still works, but not as well as it used to.

Moisture Is Bad for Molecular Sieve

Zeolite strongly interacts with water.

Excess moisture can occupy adsorption sites and interfere with nitrogen separation.

The machine therefore depends on keeping the air path appropriately conditioned and the sieve material protected.

A concentrator should not ingest:

through its air system.

Compressor Contamination

Oil or particulate contamination can also damage the separation system.

Many medical concentrators use compressors designed to avoid introducing oil into the gas stream.

The exact compressor design varies.

Compressor Failure Is Not Always Binary

A compressor can:

and still be weak.

Possible causes include:

depending on compressor type.

The useful measurements are:

not simply:

Motor spinning.

Compressor Current

Electrical current draw can provide clues.

High current may indicate:

Low current with poor pressure may indicate:

The compressor is not building a proper load.

Interpret according to OEM specifications.

Cooling Is Part of Oxygen Production

Compressing gas creates heat.

The compressor itself also produces heat from electrical and mechanical losses.

A concentrator therefore needs:

depending on design.

High temperature can reduce compressor efficiency and shorten component life.

Cooling Fan Failure

A concentrator may work perfectly when cold.

After thirty minutes:

That pattern is extremely useful.

Works Cold, Fails Hot

Always think:

before assuming sieve material is simply exhausted.

Exhaust Gas

The nitrogen removed from room air has to go somewhere.

It is vented back into the room.

The exhaust may contain:

More nitrogen

and less oxygen

than room air.

For ordinary use this is part of normal design, but ventilation requirements must still follow the manufacturer and facility guidance.

Exhaust Muffler

The depressurizing sieve beds can make a noticeable:

Pssh—pssh

sound.

A muffler reduces noise.

If the muffler becomes restricted:

Bed regeneration can be affected.

The machine cannot vent nitrogen efficiently.

That can reduce oxygen concentration.

A simple exhaust restriction can therefore look like bad sieve beds.

Cycle Sound Can Be Diagnostic

The switching cycle often creates a repeating acoustic pattern.

A unit that normally sounds like:

Pssh...

Pssh...

Pssh...

and suddenly becomes:

Continuous hiss

may have a valve leak.

A missing purge sound may indicate:

No bed switching.

Sound is not proof.

But changes from historical behavior can be useful clues.

Timing Matters

The control board determines how long each sieve bed remains in:

Exact timing is designed around:

A timing or control fault can reduce separation even when the mechanical parts are healthy.

Valve-Coil Failure

A solenoid coil can fail open.

Now the valve never actuates.

The machine may remain stuck with one bed pressurized or the wrong path open.

Depending on architecture:

Oxygen concentration can fall dramatically.

Valve Mechanism Can Fail Even If Coil Works

A solenoid may click.

But the internal valve may be:

As always:

Click does not prove gas flow.

Sieve Bed Degradation

When concentrators cannot maintain oxygen concentration and:

the sieve material becomes more suspicious.

Sieve beds can lose performance over long service life or after contamination.

Do Not Replace Sieve Beds First

They are an obvious candidate whenever oxygen is low.

But many other faults can create the exact same symptom.

Before condemning the beds, verify:

That is much stronger troubleshooting.

Paired Sieve Beds Can Fail Differently

If one bed performs worse than the other:

The oxygen concentration may fluctuate with the switching cycle.

An analyzer may show periodic concentration variation.

That can suggest an asymmetrical failure involving:

rather than equal degradation of both beds.

Product-Gas Pulsation

Some pressure pulsation may exist before the reservoir.

The patient output should be adequately smoothed.

If output visibly pulses:

Investigate:

depending on design.

Outlet Pressure

Concentrators generally deliver at relatively modest pressure compared with high-pressure oxygen cylinders or pipeline systems.

The exact outlet pressure varies.

This matters when users try to connect equipment requiring:

More pressure than the concentrator is designed to provide.

Normal oxygen concentration does not prove:

Adequate pressure for every accessory.

Concentrator vs Cylinder

A cylinder stores oxygen under high pressure.

A concentrator continuously produces oxygen-enriched gas from room air.

That creates different failure modes.

A cylinder can run out.

A concentrator can lose separation efficiency.

A cylinder's purity does not depend on a compressor.

A concentrator's production absolutely does.

Concentrator vs Hospital Pipeline

Hospital oxygen piping normally provides high-purity medical oxygen from a central source.

A portable/home concentrator is not simply a mini version of that pipeline.

Its:

can be very different.

Alarm Systems

Concentrators may monitor several conditions.

Possible alarms include:

depending on model.

The alarm does not always identify the failed component.

For example:

Low pressure

could arise from:

It identifies a condition.

Not necessarily the root cause.

Power-Failure Alarm

Some concentrators have a battery or capacitor-backed audible alarm that sounds if AC power disappears.

The machine may stop producing oxygen immediately.

The alarm's job is to warn the user.

During PM, this function should be verified when required.

Starting Under Low Line Voltage

Compressors require substantial starting current.

Poor power conditions can cause:

In ordinary hospital conditions this may be uncommon, but electrical supply is still part of the system.

Start Capacitor

Some compressor designs use:

A failing capacitor can produce:

The compressor itself may not need replacement.

Circuit Breaker

Repeated breaker tripping indicates a fault condition such as:

Do not simply reset repeatedly.

Flow Below Setting

Suppose the unit is set to:

5 L/min.

External analyzer measures:

3.5 L/min.

Oxygen concentration is:

94%.

This suggests the separation process may be healthy.

The delivery path is not.

Investigate:

Concentration Low but Flow Correct

External flow:

5 L/min.

External O₂:

82%.

Now look toward:

The two measurements have separated the problem.

Both Flow and Concentration Low

Now a shared upstream problem becomes more interesting.

Possible examples:

Flow Good at 2 L/min, Bad at 5 L/min

Capacity problem.

The machine can operate under low demand.

It cannot meet full rated output.

That is a much better description than:

Sometimes weak.

Concentration Good at Startup, Falls Later

Thermal or dynamic failure becomes especially interesting.

Possible areas include:

Concentration Low at Startup, Improves Normally

That may simply be warm-up and purge behavior.

Timing matters.

Concentration Never Rises

Now ask:

Is the PSA cycle actually occurring?

Check:

High Pressure Alarm

Possible causes may include:

depending on sensor location.

The machine can create too much pressure in one region even while patient flow is low.

Low Pressure Alarm

Possible causes include:

Again:

Pressure sensor location matters.

Internal Tubing

Flexible tubing hardens and ages.

Possible failures:

A small internal leak can significantly reduce PSA performance.

Because the machine is constantly switching pressure, leaks may be audible only during certain portions of the cycle.

Leak Detection

Use OEM-approved methods.

Do not spray random flammable leak-detection solutions inside oxygen-enriched equipment.

Oxygen service requires material compatibility and cleanliness.

Oxygen-Enriched Atmospheres Change Fire Risk

Oxygen itself is not a fuel.

But higher oxygen concentration can make combustible materials:

This is why contamination with:

is especially important.

Use only approved service materials.

Preventive Maintenance

A useful concentrator PM may include:

depending on model.

Concentration Should Be Checked at the Required Flow

Testing only at:

1 L/min

can miss a unit that fails at maximum rated flow.

Test Multiple Conditions When Required

A strong test strategy may compare:

This exposes capacity-related degradation.

External Oxygen Analyzer

The analyzer itself must be:

A bad analyzer can condemn a healthy concentrator.

Verify the reference before adjusting the device.

Humidity Can Affect Some Oxygen Sensors

Sensor technology matters.

Follow analyzer instructions regarding:

Flow Analyzer

A calibrated flow analyzer provides stronger evidence than:

The unit's rotameter.

Pressure Analyzer

Can verify outlet pressure.

Together

O₂ concentration + flow + pressure give a much better picture of performance.

The Best Troubleshooting Measurements Work Together

Consider three numbers:

Oxygen concentration

Flow

Pressure

They describe different things.

Concentration

Did separation work?

Flow

How much product gas is being delivered?

Pressure

Does the machine have enough pneumatic headroom to deliver that flow?

Those three measurements immediately narrow many failures.

Real-World Example: Normal Flow, Low Oxygen

Unit is set to 5 L/min.

External flow analyzer confirms 5 L/min.

Oxygen analyzer reads 81%.

Bed pressure is substantially below specification.

Compressor sounds normal but cannot create rated pressure.

The problem is not the flowmeter and the sieve beds are not yet proven bad.

Real-World Example: Low Oxygen Only at Maximum Flow

At 2 L/min:

94%.

At 3 L/min:

93%.

At 5 L/min:

85%.

The concentrator can separate nitrogen but no longer has enough production capacity for full rated demand.

Compressor output, internal leakage, and sieve efficiency now deserve comparison against specification.

Real-World Example: Low-O₂ Alarm but External Analyzer Passes

Internal display reports low oxygen.

Alarm activates.

External calibrated analyzer measures 94% steadily.

Flow and pressure are normal.

The oxygen-sensing path becomes much more suspicious than the PSA system.

Real-World Example: Works Cold, Fails After Thirty Minutes

Unit produces 94% oxygen shortly after warm-up.

After thirty minutes:

Concentration falls to 86%.

Internal temperature is high.

Cooling fan is not operating.

The PSA hardware may still be fundamentally healthy; thermal conditions are degrading its performance.

Real-World Example: Flow Drops With Humidifier Attached

Concentrator produces rated flow and oxygen concentration with no accessory connected.

After connecting the patient's humidifier bottle:

Flow falls substantially.

Diffuser in the humidifier is restricted.

The concentrator is not the failed component.

Real-World Example: One Bed Is Not Cycling

Compressor builds normal pressure.

Valve sound is asymmetric.

Pressure data shows one sieve bed repeatedly pressurizes while the other does not.

A failed switching valve has disrupted the PSA sequence.

Replacing both sieve beds would not address the cause.

Real-World Example: Concentration Slowly Declines Over Months

Historical PM results:

94%.

Then:

92%.

Then:

89%.

Compressor pressure remains within specification.

Valve timing and flow remain normal.

No leak is found.

Sieve performance now becomes much more credible as the degrading component.

That trend is stronger evidence than one isolated low reading.

Common Mistakes

Thinking the concentrator creates oxygen from nothing. It separates oxygen already present in room air from much of the nitrogen.

Assuming airflow means oxygen output is good. Normal flow can contain abnormally low oxygen concentration.

Condemning sieve beds before measuring compressor pressure. Weak compression can produce the same low-O₂ symptom.

Testing only at a low flow setting. Marginal units may fail only near rated maximum demand.

Trusting the internal oxygen sensor without independent verification. The sensor can fail while product gas remains correct.

Calling every rhythmic hiss a leak. PSA concentrators intentionally vent nitrogen during bed regeneration.

Ignoring filters and cooling. Air restriction and high temperature can significantly reduce performance.

Judging compressor health by sound. A compressor can run smoothly while producing inadequate pressure.

Ignoring downstream accessories. Long tubing and blocked humidifiers can reduce delivered flow while the concentrator itself is healthy.

Replacing parts before allowing proper warm-up. Oxygen concentration requires time to stabilize after startup.

A Useful Oxygen-Concentrator Framework

Think:

Room Air

↓

Intake Filter

↓

Compressor

↓

Pressurized Air

↓

Switching Valve

↓

Molecular Sieve Bed

↓

Nitrogen Adsorbed

↓

Oxygen-Enriched Product Gas

↓

Check Valve / Product Reservoir

↓

Pressure Regulation

↓

Flow Control

↓

Patient Outlet

At the same time:

Other Sieve Bed

↓

Depressurization

↓

Nitrogen Exhaust

↓

Purge / Regeneration

And monitoring the process:

Pressure + Oxygen Sensor + Temperature + Alarms

Another Useful Troubleshooting Split

Ask four separate questions:

Can the machine build pressure?

Can it remove nitrogen?

Can it deliver the required flow?

Does the machine's own sensor agree with an independent oxygen analyzer?

Those questions separate:

much faster than starting with:

Low oxygen means bad sieve beds.

What Did You Actually Prove?

If gas is coming out of the concentrator:

You proved:

The machine is moving some amount of gas through the outlet.

You did not prove:

If the flowmeter indicates:

5 L/min,

you proved:

The device indicates approximately 5 L/min according to its own flow-control system.

You did not necessarily prove:

An independent flow analyzer would measure 5 L/min.

If an external oxygen analyzer reads:

94%,

you have stronger evidence that:

The product gas at that test point contains approximately 94% oxygen under those operating conditions.

You still have not proven:

The unit can maintain that concentration at every rated flow or after extended operation.

If oxygen concentration, flow, pressure, alarms, and required pneumatic performance all meet specification after stabilization and at the required test conditions:

You have much stronger evidence that the complete oxygen-concentration and delivery system is functioning correctly.

Final Thoughts for Biomeds

An oxygen concentrator makes much more sense when you stop thinking of it as:

A little oxygen tank that plugs into the wall.

There is no tank of pure oxygen hidden inside.

The machine starts with ordinary air.

The compressor raises the pressure.

The molecular sieve temporarily traps nitrogen.

Oxygen-rich gas passes through.

The beds switch.

The nitrogen is vented.

The sieve regenerates.

A reservoir smooths the output.

A regulator controls pressure.

A flow system controls delivery.

An oxygen sensor watches the result.

And that entire process repeats every few seconds for as long as the machine is running.

That means the most important troubleshooting distinction is:

Gas movement is not gas composition.

A concentrator can have:

Normal flow and bad oxygen.

Good oxygen and bad flow.

Good output with a bad internal sensor.

Good performance cold and bad performance hot.

Good output at 2 L/min and poor output at 5 L/min.

Those patterns are not noise.

They tell you which part of the system is losing control.

So when someone says:

It blows air fine.

that is only the beginning.

The real questions are:

How much gas is it delivering?

At what pressure?

What percentage of that gas is oxygen?

Can it maintain all three at rated demand?

And when the front panel says:

O₂ Normal,

the question stays the same:

What did you actually prove?

— Jake

Important Note

Oxygen concentrator compressor designs, pressure-swing-adsorption timing, sieve materials, valve arrangements, purge flow, reservoir pressures, oxygen sensors, alarm thresholds, outlet pressures, rated flows, oxygen-concentration specifications, maintenance intervals, and service procedures vary significantly by manufacturer and model. Oxygen-enriched gas increases fire risk and requires compatible, clean service materials. Follow current OEM documentation, approved oxygen/flow/pressure analyzers, facility respiratory-equipment procedures, electrical-safety requirements, oxygen-service cleanliness requirements, and authorized service scope when evaluating oxygen concentrator performance.

Related Biomed Basics

Back to Biomed Basics