Performance testing

Implicit in the use of ISO 9001 is the need for agreement be­tween the supplier and customer as to the product specification and performance. Essentially the customer is buying a given airflow against a pressure. He will have calculated these ac­cording to known formulae or experience. Much data is avail­able on the pressure loss through duct elements and termina­tions. Without knowledge of the interaction of fan and system, however, problems can arise.

Aerodynamic testing

In the early days of fan manufacture, each individual company would have its own methods of test. These might vary greatly according to the whims of individual engineers. Whilst some, but not all, would be grounded in acceptable theory and prac­tice, the results achieved could not be easily compared. Defini­tions and measurement of pressure were a particular difficulty — as witness the terms used such as total pressure, effective pressure, resistance head, total static pressure. How these could be correlated with calculated system pressures was a mystery.

Into this morass stepped the then Institution of Heating and Ventilating Engineers who set up a Fan Standardisation Com­mittee. Around 1924 it made recommendations as to standard­ised methods of testing a fan’s performance. It is believed that this was the first attempt anywhere to settle what is still a chal­lenge to the specialist in flow measurement.

The conclusions were made available to the British Standards Institution (BSI) from which grew BS 848:1939, authorised by the Electrical and Mechanical Industry Committees as an ex­tension of work originally asked for by the India Store Depot. It was part of a series of test Codes covering various fan types. Others produced around the same time were BS 367:1932 for Ceiling Fans, BS 38 :1930 for Desk Fans and BS 707:1936 for Mines Fans.

By 1952 these Codes were generally found to be unsatisfac­tory. The Fan Manufacturers’Association (FMA), noting the ur­gency and unwilling to wait for a new British Standard, produced FMA Code 3:1952. This introduced new methods primarily for the testing of axial flow fans which were becoming popular, and extending the range of pressures which could be measured by taking into account the compressibility of air.

This work, in turn, was made available to BSI who updated its own Standard to produce BS 848:1963 Part 1 which largely fol­lowed the FMA Code but was brought into line with BS 1042 Code for Flow Measurement. The subject of tolerances was also introduced.

Concurrent with all the above, similar activity was being carried out in other major industrialised countries. Test codes were pro­duced in the USA, France, Germany and Italy which were char­acterised by certain significant differences. Further information is given in Chapter 4 from which it will be noted that international agreement has been achieved after more than 30 years of ef­fort.

Customers are strongly advised to specify that all performance data shall be obtained from tests carried out in accordance with ISO 5801. Only if this is the case can valid comparisons be made between competing nations’ products. The tolerances applied to this data are detailed in the shortly to be published ISO 13349.

Sound testing

BS 848 Part 2 Fan Noise Testing was published in 1966 as a first attempt in the UK to satisfy the need for meaningful results which could be used by acousticians.

An earlier attempt had been made in the USA with the NAFM test Code of 1942 (NAFM was a forerunner of AMCA, the Air Movement & Control Association International). Whilst this en­abled comparisons to be made between competing products, the information produced was not in a form which could be used by acousticians in the design of systems.

This need had to await the production of BS 848 Part 2:1985, which gave a comprehensive series of tests designed to give in­formation of open inlet /open outlet noise, noise break-out and in-duct noise levels.

It must be recognised that there are at least 12 noise levels as­sociated with a fan for a particular duty. Both the manufacturer and the customer must agree on which particular level is appro­priate for a particular installation.

Further information on Fan Noise is given in Chapter 14 from which it will be noted that six new international noise Standards have been produced which update the methods described in BS 848 Part 2:1985 and additionally incorporate French, Ger­man and American methods.

Again customers are strongly advised to specify that all sound data shall be carried out in accordance with ISO 13347 Parts 1 to 4, ISO 5136 or ISO 10302 as appropriate.

Balance and vibration testing

Again, the world has changed considerably from the days when balancing of a fan impeller was carried out with a piece of chalk and a length of string. This is not to decry the efforts of the skilled craftsmen of those years. Their results were often as good as those obtained from the early balancing machines.

However, the need for a quantitative Standard was apparent which resulted in BS 848 Part 6:1989. This specified in detail how such tests were to be carried out and was largely a copy of a UK MOD Defence Standard. It did not, however, lay down standards for specific types of fan and the first attempt at this was incorporated in AMCA 204.

Again international work has largely mirrored that carried out for sound testing and we now have IS014695 which is largely a re­write of BS 848 Part 6 and ISO 14694 which is a specification for balancing and vibration. This latter is very similar to AMCA 204 but with vibration levels quoted in r. m.s. (root mean square) values as internationally agreed rather than the peak-to-peak levels favoured in North America.

Again it is recommended that these two international Standards be specified so that valid comparisons can be made. It should be noted that out-of-balance forces are not the sole reason for fan vibration. Further information is given in Chapter 15.

Run tests

For many years, after final assembly, it was customary to give every fan manufactured a run test. This would typically be for an hour or more, during which time a check would be made on bearing temperatures, vibration, impeller integrity, tightness of fastenings etc.

With the advent of batch and even mass production of small fans, especially axial flow units, the need for these tests was questioned. It was pointed out that tests at other parts of the production cycle could obviate the need for run tests whilst the adequacy of the customer’s foundations (a more likely cause of poor running) was never questioned.

The author’s belief is that, for the present, this must continue to be the subject of agreement between the manufacturer and the customer. He would however strongly recommend the inclu­sion of a run test in the specification of any fan above about 18.5 kW.

It should be noted that every fan supplied for the Sydney Har­bour Tunnel (see Chapter 1, Section 1.2.8) was subjected to a 24 hour run test; the electricity consumption, and therefore the cost, being considerable.

Posted in Fans Ventilation A Practical Guide