When we purchase a machine we accept that it will not last forever. The problem isn

However Condition Monitoring is only part of the solution since once you have identified a faulty machine and gained some insight into its seriousness the next and crucial step is to do something about it and this is usually referred to as Condition Based Maintenance (CBM). It is CBM that gives the benefits but these can only be gained by implementing CM with its associated costs (see Table I).

{mosimage}
 
A wide range of Condition Monitoring techniques are available, each of which has its own strengths and weaknesses and it is fair to say that no single technology can provide information on all machine faults. Even within one area of CM technology the range of instrumentation available can be bewildering. Therefore, for the newcomer wishing to get started on CM it can be a daunting prospect and two commonly encountered approaches are as follows:

Some companies approach the task in a proactive and systematic way whereby each machine is analysed in terms of possible failure modes, likelihood and consequences. The capabilities of appropriate CM techniques are then matched to each of the failure modes and a cost benefit analysis for each can then be performed. In an ideal world every company would follow such a procedure but in the real world few have the luxury of financial and manpower resources to pursue it. As a result, some companies leave CM on the back burner from one year to the next.

At the other extreme some companies get bounced into CM in response to an unexpected and serious machine failure which has had a major adverse impact on production / operations. This often results in a panic to get CM implemented so that “It never happens again”. Needless to say hasty decisions made by newcomers to CM, often with blank cheque books, are a recipe for later regrets. In the absence of knowledge and under extreme pressure human nature is to buy the most expensive as it must be the best ! 

However there is another approach to implementing CM between the two extremes of all or nothing. Why not start with those machines where past experience tells you that the consequences are most serious such as :

· where there is no spare capacity to meet production demands
· machinery for which spare parts are on long delivery times
· old machinery for which spare parts are no longer available
· machinery that has awkward access making repairs difficult
· machinery whose failure would have knock-on consequences

In short start with those machines where people will jump up & down if it suddenly fails ! You don’t need a consultant to tell you which these are.

Next its important to consider what information is needed from CM. In an ideal world you will want to identify at an early stage :

· which machines have a problem
· how serious the problem is
· how long before failure (or secondary damage) occurs
· what the fault is

No matter what CM technique(s) you choose its going to require skill, experience and sophisticated instrumentation to diagnose what the fault is. In the first instance the time to failure can be estimated from the trend of readings as a function time (i.e. from a trend plot) which does not require great skill. The difficulty of recognising which machines have a problem and ranking the relative severity of the problems is different for different CM techniques (as is the sensitivity of the different CM techniques).

For example an increase in temperature is very easy to recognise and trend but it is insensitive to the early stages of a problem and very difficult to interpret in terms of the problem, giving rise to it. Similar comments apply to an overall vibration reading. In contrast, frequency domain vibration analysis starts with the difficult task of diagnosis and follows on from this to identify whether or not there is a fault and if so how rapidly it is developing.

As an alternative to getting bogged down in assessing the relative merits and capabilities of the various CM techniques to provide all the answers why not instead simply ask “what information can CM techniques give me easily and quickly ?”. This is a question to put to the relevant vendors and holds out the prospect of getting a quick return from CBM for a minimum of investment in CM.

To illustrate this approach a number of examples are given in Table II based on the use of the MHC Memo Pro manufactured by Holroyd Instruments. This is an Acoustic Emission based portable CM instrument (see photo). First a brief word about the use of this type of instrument. Each measurement takes in the region of 15 seconds, requires no set-up, previous history or knowledge of machine design details (such as bearing type, number of balls, race diameters etc..) and the same interpretation is applied across all of the machine types; if Distress® is greater than 10 there is a problem.

Maintenance personnel periodically monitored 500 points at the site where the measurements shown in Table II were taken. Actions were only taken if Distress® was greater than 10 and Table II is concerned only with those items falling into this category for one particular month.

{mosimage}

In fact Table II provides a perfect illustration of the simplest form of CBM in action; one off measurements on hundreds of machines instantly identify those few machine points where a problem has arisen and straightforward maintenance procedures are carried out to effect an immediate and measurable improvement. (Note : It is clear that some problem still exists after re-greasing of the Flue Gas Compressor and the Big Bag Conveyor Gearbox Output since Distress® only fell to 13 and this is usually indicative of a degree of surface damage beyond that which can be cushioned by improved lubrication.)

Of course with a little further investment in time the regularly taken readings could be trended in order to reveal the rate of deterioration and it is even possible to do a frequency domain diagnosis from the AE signals but interpretation of this requires a much greater level of skill. However the important point to note is that simple CM instruments exist which can very easily provide immediately useful information, which can be acted upon to provide tangible benefits.

In fact starting simple with CM has a number of additional benefits to those of CBM listed earlier :

· immediate payback
· confidence in CM and CBM is nurtured
· existing maintenance personnel are empowered
· efficiency of the maintenance function is enhanced
· justification of expanding CM is easier
· financial savings continue to be evident as CM expands

At whatever level you decide to get into CM, it is important to bear in mind that the act of purchasing CM instrumentation is the easy part. Regularly using the instruments properly, interpreting their outputs correctly and having the confidence and organisation in place to act upon their findings in a timely way are the keys to a successful CBM strategy. To help achieve this goal it is most important to avoid buying CM instruments that are incompatible with the time and/or skills you will have available.

by Trevor Holroyd

Photo supplied courtesy of Elyo services Ltd