"For the past three years, our company has successfully been using on-site oil analysis. Currently, as many as 500 samples per quarter are analyzed in the on-site lab. Over the past week, condition-based maintenance personnel have noticed that the particle count results seem to be backward, i.e., more particles are showing up after the in-line filter than before the filter. It was suspected that the unusual particle counts were due to a labeling mix-up, confusion on the direction of oil flow through the filter, a damaged filter or background contamination.

The plan initially is to resample. If the results are the same, we will change the filters and then sample again. What other steps could help identify the root cause of the abnormal particle counts?"

Assuming that the test instrumentation is in proper working order and calibrated correctly, you would first want to look at the historical data to determine if there is a correlation to seasonal changes or if this has happened in the past. Verify correct labeling, oil flow path and that the filter has not failed.

You then should consider personnel issues, including whether there has been a new technician assigned to pull samples and if he has been properly trained and following the correct procedures for taking oil samples. Ask questions such as, "Were two different technicians involved in taking the samples?" "Was the sample taken in the correct location?" "Has a new shipment of sample bottles arrived?" "If so, were they the correct bottles?" "Were they certified super clean, just ordinary sample bottles or were they mixed?"

If the problem is occurring in a specific unit, was the sample taken during a period in the process when flow surge could have hit the filter element and dislodged particles that went through the bypass valve and migrated downstream? Is the filter sized correctly to handle the flow surge without opening the bypass? Has the flow surge been measured? Is there an in-line flow meter to determine the flow?

In addition, look at the unit’s maintenance history. Has the system been worked on recently? Has the unit developed a leak between the downstream side of the filter and the sampling port?

Don’t forget to examine the filter. Is it the same type and style as had been used previously? Could it have been changed and a different design type installed inadvertently? Is the element of the same style and design? This greatly affects its ability to withstand flow surge and hold onto particles. Manufacturers use different designs within their respective filters. If the filter has not changed, did the manufacturer make a design change and fail to notify the end user?

These would be some of the things you should look at to help determine the root cause of the abnormal downstream particle counts. Pending resolution of the issue, you may want to plug in your portable filter cart and circulate the fluid to clean it to allowable cleanliness levels.