For oil-based lubricants, the presence of water is a significant concern.
It can decrease the efficiency of the lubricant and cause early wear to parts, in turn adding to maintenance costs as well as creating machine down-time while it is being repaired.
In a study for Bently Tribology Services Inc., Whitfield showed that the quantity of water in lubricants causes bearing life to decrease exponentially. [g1] For these reasons the water content in lubricants needs to be monitored. Additionally, water removal systems should also be monitored to ensure they are working correctly. Traditionally this has been accomplished using Karl Fisher (KF) titrators, and while these methods are accurate, they require harmful chemicals and constant maintenance. Additionally, the glassware required is fragile and expensive, which limits the locations these instruments can effectively be placed. To address these drawbacks Arizona Instrument LLC has made significant efforts to develop a solvent free moisture specific instrument that utilizes RH sensor technology. The Computrac® Vapor Pro® line of instruments are able to quantify water content in oil with the same precision as KF titrators while reducing waste costs, and test times. The design of the Vapor Pro® instruments is more rugged, which allows the instrument to be placed in more hazardous environments.
Turbine oil, virgin SAE 10W-30 motor oil, and used SAE 10W-30 motor oil were tested for water content using the Mitsubishi CA-06 Karl Fisher Titrator [g1] with the VA-110 vaporizer, and the Computrac® Vapor Pro® 3100L. The turbine oil was stored in a clean Nalgene bottle, the virgin motor oil was kept in the container it was purchased in, and the used motor oil was procured from a 1998 Plymouth Breeze [g2] and stored in 0.5L Nalgene. It was used as a lubricant for approximately 5000 miles.
The sample loading bubbler oven of the KF was filled slightly above half with reagent grade toluene and set to 90°C. The flow was set to 300mL/min. Once the test was started there was a one minute delay before titration was started in order to add the sample material to the bubbler. Once the titration started there was a minimum of 1 minute that the instrument titrated before the test could end. This was to ensure that lubricant was able to heat up and begin to evolve water. The ending criteria was set so that the test would end when the amount of water being titrated was 0.1µg/sec above the baseline titration level. Testing would not start if the baseline was above 0.3µg/sec. Toluene was added to the bubbler as needed due to it boiling over into the titration vessel. The titration vessel was filled with 200mL of Coulomat A in the large, exterior container. The interior cathode was filled with Coulomat C so that the levels of both Coulomat A and C were even. The titration cell was cleaned between testing each material and checked for accuracy daily, prior to testing, using a 1% methanol water standard. Acceptable values for standard check testing were 1% ± 0.05%. For the motor oil a sample size of 0.6mL of sample was added to the bubbler. A sample size of 3.0mL was used for the turbine oil.
Vapor Pro® 3100L
The Computrac® Vapor Pro® testing was conducted using the following conditions:
Test temperature: 140°C
Bottle Purge: 40 seconds
Ending criteria: Time → Rate, minimum 3 minutes, 0.30µg/sec
A clean bottle with a new septum was used for each test and 0.6mL of fluid was added to the bottle using a 1mL syringe for the motor oil samples. Because turbine oil has less moisture a sample size of 1.5mL was used. Once the sample was added to the vial the cap was screwed on and slightly tugged to ensure a tight seal was achieved.
A minimum of 9 test were conducted to calculate the average parts per million (ppm), standard deviation (SD), and relative standard deviation (RSD). From the tables above the CT 3100L shows strong correlation to the KF for all 3 materials.
From the graphs, the CT 3100L provides real time data for complete analysis, and the graphs allow users to optimize testing criteria.
For testing lubricants, relative humidity sensor instruments, such as the Computrac® 3100L, prove to provide significant advantages when compared to Karl Fisher titration. The solvent free testing reduced hazardous chemical waste use required for titration. Using the Computrac® Vapor Pro® 3100L also significantly reduced maintenance and calibration times, allowing for rapid throughput of sample analysis. The statistical data showed strong correlation between the two analyzers, and the CT 3100L reduced standard deviation and relative standard deviation measures.