"A few of the machines at our plant have had recurring issues with varnish. How often should you test for varnish potential? Are there any guidelines?"
Varnish can be devastating to certain machines that are prone to its formation. Varnish has too often been the cause of costly downtime and unplanned outages. Testing for varnish potential in a lubricating oil allows you to keep track of the stages of varnish formation so it can be mitigated early.
The rate at which varnish potential testing is performed will depend on several factors, including the machine's clearances and overall geometrical complexity, the age of the lubricant and/or machine, the previous history of varnish formation, the machine's overall criticality, and the associated safety concerns.
Consequently, the frequency for varnish potential testing will not be static but instead will fluctuate based on the many factors. For example, if the machine is early in its service life, you should test more frequently, as varnish has been known to become more evident at this stage primarily as a result of caution based on the lack of historical information. A new machine is a wildcard in terms of condition monitoring results.
On the other hand, a vast amount of historical data collected over an extended period of time can provide a better understanding of the likelihood of varnish potential. This is considered the bathtub curve, which is applicable to many aspects of oil analysis.
Regarding the age of the fluid, there is a greater chance of degradation at the end of the lubricant's life. Therefore, testing more frequently toward the end of the lubricant's lifespan is recommended.
Ultimately, this is a classic case of the cost-benefit tradeoff. Certain tests, whether they are part of the routine schedule or not, will be justified by the potential cost avoidance of recognizing early indicators of varnish potential. This is where machine criticality and any safety concerns can play a significant role, along with the cost of repairs and downtime.
The optimal testing frequency will be a balance between the two extremes of this inherent tradeoff. Testing too often (such as daily or weekly) can lead to varnish avoidance but high annual test costs, while testing too infrequently (annually or by exception) will result in a higher chance of costly downtime and machine repair. On which side of the equation do you want to err?