How to Overcome the Practical Problems of ISO 8573 Oil Measurement in Compressed Air

What does ISO 8573 actually say about oil measurement?

ISO 8573-1 states that the total oil liquid, aerosol and vapour concentration figures must be combined to attribute a purity class but if you need to achieve class 1 or 2, how can this be achieved?

ISO 8573-2 gives two ways of measuring aerosols but only one that is also suitable for liquid oil – Method A. This requires the use of ISO 12500-1 class coalescent filters in series to collect oil from the airstream during the test period and the volume collected is measured to calculate the concentration in mg/m³.

ISO 8573-5 covers the sampling of oil vapour and subsequent analysis by gas chromatography to determine the vapour concentration.

The results of these two methods then need to be added together and compared with the purity class table in ISO 8573-1 to determine the actual purity class achieved.

This is complicated and slow, surely there's an easier and faster way?

Yes, there are several but none of them follow the relevant ISO 8573 methods and therefore cannot be claimed to provide ISO 8573 compliant purity class results.

Do I really need to measure the vapour concentration?

No, not if class 3, 4 or X is acceptable. If you need to achieve class 1 or 2, yes vapour must be measured if you want or need to be compliant with ISO 8573-1.

Should you use glass Draeger tubes to measure the oil content for the ISO 8573 audits?

This is much faster to get a measurement but this method is only intended as a test for the presence of oil and a guide to the concentration. ISO 8573-5 stipulates clearly that this method is not approved for the measurement of oil and cannot be used as a test method to determine ISO8573-1 purity classes. If you are being sold a service claiming ISO 8573 compliance that uses this method, be aware that this is false and may not accurately reflect the actual system condition.

Is the Draeger Impactor a suitable test method?

The Impactor detects oil aerosols only, is a good indicator for the presence of oil and approximates a concentration figure. As a preliminary test it is fine but again, this is not an approved test method for the purposes of ISO 8573 compliance.

What about some kind of sensor for measuring oil aerosols directly?

There is no proven sensor yet invented that is capable of measuring oil aerosols specifically. Some may claim this capability but there is no documented independent proof that the sensor technology can deliver on the makers claims. If you are offered such a device, ask for details of the sensor patent application, details of the pre-production testing carried out and testimonials from existing users that you can verify.

Oil aerosol sizes do fall within the measuring parameters of laser particle counters and if there was no possibility of true particles being in the airstream, the measured output would be the aerosol concentration. However, ISO 8573 stipulates that both particles and all oil forms must be measured separately and and as it is not possible to discriminate between a solid particle or an oil aerosol with a laser particle counter, this method is not feasible for 8573-1 purity class checks.

Do we actually need to measure the oil aerosols in order to achieve purity class 2 or better (<=0.1 mg/m³ total oil)?

In order to comply with the ISO 8573-1 standard, yes. However, let's consider this in the practical world. Since the means of achieving such a low concentration is the use of ISO 12500-1 compliant coalescent filters to remove the aerosols, the main concern is how effective they are in use. A site may have several in series, monitor the oil volume captured and regularly replace them according to the filter manufacturers recommendations. They could consider that in this case, the filtration is effective and therefore there is no real need to measure for aerosols and liquid oil. They still need to measure for vapour though and by monitoring the vapour concentration it will show an increase as the filters degrade and more aerosols escape.

This idea is practical but not suited to infrequent audits and subsequent gas chromatography, as the filter performance may drop faster than anticipated and produce contaminated air for some time before the next audit discovers the problem.

So if regular sampling and gas chromatography is not practical, what choices do I have?

If you accept that achieving a high purity class is desirable but the approved and compliant methods will not provide that reassurance of high quality air in practice, the solution is to use an unapproved method on a regular basis to give an ongoing measurement of oil content in the compressed air.

The most common method would be a PID sensor that detects hydrocarbon vapours. Regular audits with this device would provide accurate and repeatable measurements that can support a high purity class rating but cannot be claimed as ISO 8573-5 compliant.

The real benefit of the PID sensors is that they can provide 24/7 continuous monitoring of the compressed air, something that no compliant method can achieve. Using a PID oil vapour sensor in this way can still allow the regular ISO compliant sampling method to take place for record keeping and provide an extra safeguard for the operator to be sure the compressed air quality is always good with respect to oil contamination.

Remember that oil vapour will always be present when there is liquid or aerosol content in the airstream but it is the most difficult to filter out. Even after activated charcoal filters, there may be some vapour present even though there will be no liquid or aerosols. Measuring vapour therefore is the best indicator of effective oil filtration and a PID sensor will react quickly to any change allowing the operator to take action before contamination can become a serious problem.

Frequently Asked Questions

I have Oil-Free compressors so I don't need to measure oil do I?

Yes, you do. There are 3 sources of oil contamination in compressed air:

Oil and hydrocarbon vapours drawn in in at the inlet from the ambient atmosphere.

Oil from the compressor seals not being 100% effective (if not oil free design).

Existing oil contamination in the distribution system in the form of aerosols that have contacted pipe walls, valves and systems to form liquid oil coatings which also produce oil vapour.

Does this mean that changing to an oil-free compressor design won't improve the air quality?

Changing the compressor will remove the addition of oil from that source but will not affect the oil already in the system and if the inlet filters are not effective, the difference is likely to be very marginal for a significant capital outlay.

Does this mean that upgrading filters and treatment to get a lower oil contamination level is a waste of time?

Not exactly. Upgrading the air treatment will reduce new contamination but unless the distribution system is new or thoroughly cleaned, the existing piping and systems attached to it may be the biggest contributors to aerosol and vapour contamination. It may actually be more effective in reducing oil contamination, to replace and re-pipe the whole distribution system than upgrade to an oil free compressor or new/additional filtration system

I have an oil-free compressor and full filtration plant but the audits still measure oil. Is the measurement method incorrect?

Probably not. Whichever measurement method is used, the results should always be respected and investigated if they are not as expected. If the distribution system is contaminated downstream of the filtration plant, oil will be present at each point of use air outlet and exhausted to atmosphere at every pneumatic valve. Installing additional coalescent and carbon filters at each point of use is a massive task to ensure all connected systems are safe from oil but it is the only option if the distribution system itself is contaminated and can't be replaced.

Sensitive PID sensors will detect other hydrocarbons such as solvent vapours in addition to oil that can give rise to higher than expected readings and while in this case, the reading may not be accurate as the sensor will not be calibrated for solvents, action should be taken to resolve the issue rather than blaming a faulty sensor.

If in doubt, use another PID sensor to verify the results. Two sensors saying the same thing will not be wrong. Using an Impactor or glass tube will not provide corroboration unless the correct glass tube for the actual solvent is used.

Do we have to be fully compliant with ISO 8573?

No, not at all but many sites and companies do insist on it. This is a European standard for determining and measuring compressed air purity but this does not mean that it is compulsory. There is no legislation in place that makes achieving certain purity classes mandatory, it is simply a mechanism for ensuring equal comparison for measurements according to the methods described. ISO 8573 could be regarded as a standard to be achieved exactly or guidelines to suggest what may be acceptable for air purity, and the individual site management decide how to implement this in a practical way.

How often should breathing air be tested?

Breathing air should be tested at no more than 3 monthly intervals according to EN 12021 but even at that frequency, there is the possibility of harmful contamination between checks. Continuous monitoring is the safest solution but not always practical and a simple and quick to use tester that doesn't need external chemical indicator tubes, could provide user reassurance on a more regular basis.