Improving Automatic Viscosity Monitoring In the Lab

The viscosity of a lubricating oil is perhaps its most important physical property. Among other factors, it determines the dynamic load that may be supported, what the lubrication film thickness will be, and what the flow rate will be under a given pressure. Designers of machinery specify a viscosity or “viscosity grade” which must be used in their equipment. The specification comes from a combination of design parameters such as bearing loads, machining tolerances and the required rate of heat removal.

Because of its importance, viscosity is one of the fundamental properties that should always be measured in any used oil program. Most commercial oil analysis labs use a variation of ASTM D445 to determine the kinematic viscosity of used oil samples at 40°C, 100°C or both. ASTM D445 requires that the oil sample be drawn in a capillary viscometer tube, which is immersed in an oil-heating bath at the prescribed test temperature. The time taken for the oil to flow under gravity back through the narrow capillary section is used to determine its kinematic viscosity (Figure 1).

While ASTM D445 is an accurate, reliable method of determining the viscosity of an oil, many commercial oil analysis labs modify this method to allow for faster sample analysis. This is of particular importance considering the method requires at least 10 minutes of heating in the oil bath, and at least five minutes to flow through the capillary tube. Many of the larger commercial labs typically test 500 or more oil samples per day.

To provide adequate throughput, these labs usually use automatic viscometers, based on ASTM D445. These instruments are designed to allow for much faster analysis times, while providing similar accuracy to the manual method outlined in ASTM D445.

Figure 1

In addition to determining the kinematic viscosity, the other viscosity-related property of lubricating oil is the viscosity index (VI). The VI is a number without a unit that characterizes the viscosity vs. temperature relationship of the lubricating oil.

An oil’s VI is normally determined from viscosity measurement at both 40°C and 100°C using an algorithm outlined in ASTM D2270. Sometimes used oil analysis labs will provide this number from a new oil specification, without actually making the two-temperature measurement and calculating the VI.

In other cases, viscosity is measured only at one temperature and then a viscosity for another temperature is calculated using the new oil VI, again based on data provided from the new oil specifications. This calculated viscosity can be inaccurate and misleading and therefore is not reliable.

The assumption is that the VI has not changed, however, certain contaminants, not to mention shear down of VI improvers, can change an oil’s VI. This makes the extrapolated procedure potentially inaccurate. In fact, measuring viscosity at two temperatures and calculating the oil’s VI can be a useful indication of certain problems and/or conditions, particularly in engine oil samples.

Accurate viscosity measurement at two temperatures is crucial to determining the actual VI. Used lubricants present some challenges in this respect due to the contaminants they contain. For example, a sample can leave behind deposits on the walls of the capillary tube that affect subsequent measurements. In other cases, water in the oil can cause erroneous measurements. When using an automated viscometer, there is also the possibility of an instrument malfunction going undetected.

The vast majority of automatic viscometers in use today measure each sample only once. The main reason is that it is not cost effective for the laboratory to run every sample twice to crosscheck results, because this doubles analysis times. By making only one measurement, the lab is in effect relying on the accuracy of the instrument.

While these instruments are accurate when conditions are ideal (for example with new oils) when measuring used oil, particles can collect on the viscometer tube walls, shifting the calibration. When a viscometer tube is out of calibration, it will deliver consistent but inaccurate results. To counteract this, most labs will run periodic quality control (QC) samples to verify calibration. When this QC check fails, all samples back to the previous QC check should be remeasured. However, what happens when a temporary condition presents itself: one that corrects after a few more samples have been run?

A large particle or filter fiber can partially plug a tube for the next several samples. However, if that particle is flushed prior to the next QC check, how can the lab know there was a problem with some of the measurements since the last QC check? Likewise, a sample may contain water or other contaminants that are not easily flushed by the cleaning solvent.

In this case, it may take several cycles to properly clean out that viscometer tube. The samples run in the meantime can have inaccurate measurements, which again would never be picked up by QC samples unless a quality check is run every couple of samples. If any other conditions occur, a situation may arise where a highly accurate laboratory viscometer can generate a viscosity measurement that is off by several percentage points.

To counteract these problems, a new type of automatic viscometer has been designed by Edmonton, Alberta-based Dare Instruments. The Trivisc, recently introduced by Dare, is one of only a few automatic viscometers specifically designed for used oil analysis. The instrument utilizes multiple sensors to derive three unique measurements on each oil sample.

The sensor’s unique, two-piece design allows the viscometer to detect slight inconsistencies caused by calibration shift, partial blockages, water or entrained gases. Comparative analysis allows a confidence factor to be generated.

When this result is greater than a set limit, the test is considered valid and the viscosity accepted as accurate. In the case of a low confidence factor, the sample may be retested or the measurement parameters may be adjusted. This built-in quality control check happens automatically on every sample tested.

To cost-effectively measure viscosity at two temperatures and calculate the VI, commercial labs need an accurate, high-speed instrument. The Trivisc instrument can provide rapid turnaround, allowing more than 90 viscosity measurements, at two different temperatures.

Adaptive, automatic cleaning algorithms ensure that minimum time and solvent is required between samples. Wide-range viscometer tubes are utilized so that tubes do not need to be changed. On new oil, relative standard deviation (RSD) repeatability ranges from 0.24 percent to 0.50 percent above normal viscosity ranges with an accuracy of 0.5 percent.

At this level of accuracy, commercial oil analysis labs now can have the rapid throughput of the highest quality automatic viscometer with comparable repeatability and reproducibility of the stringent ASTM D445 procedure.

For more information, visit Dare Instruments at http://www.dareinc.com.