Turbine Lub oil Analysis

Large circulating systems require frequent monitoring. Water can cause a serious problems in a relativly short time. I would recomend a visual sample at least weekly and possibly daily. Take it in a clear bottle and allow it to settle and cool for an hour or 2 then look at it to see if it is hazy or has any bubbles and sediment on the bottom of the bottle. Turn the bottle over and observe for falling particles and bubbles. Bubbles that fall are water. If you have a hotplate you might learn to do a crackle test for water. If it the sample is clear and bright and you do not detect any free water it may be returned to the sump. Monthly a sample should be sent to a lab for analysis to include particle count, water, spectrography, and TAN. Each 3-6 months also have the RBOT and FTIR done.
If you have a centrifuge or other oil conditioner it should be operated daily which will remove sludge and water and make your online filters last much longer.

Thanks Buckhntr. Do you guys carry out wear debris analysis and do you have a set frequency?

Instead of using RPVOT (previously RBOT)you may wish to consider a RULER test. It is a much quicker and usually cheaper option. I agree with buckhntr that testing should be carried out every 4 - 6 weeks. And a visual sample is a highly underrated and extremely effective way of monitering your systems. Unfortunately if there are visible signs of change then it is often too late to prevent damage. The crackle test and blotter spot test are a great way to check your system for signs of change.

My recomendation for a visual sample is to be used as an early warning thus the increased frequency. Most companies are not going to go thru the expense of daily lab samples but a quick look for water by the operators on a daily basis is a reasonable way to get an early warning on a cooler or steam seal leak.
Wear debris analysis should be accomplished whenever the lab detects particles in your lubricants. Since turbines operate in a hydrodynamic lubrication regime the amount of wear metal should be very small and usually most readily detected through spectrographic analysis.

irishal,

Testing frequency suggested by BuckHntr and MC might be appropriate for steam and gas turbine. However, if you operate hydro turbines, than such frequency is overkill. Due to a very “gentle” operational conditions with hydro units (low pressure, low flow, low temp), it is quite appropriate to test moisture, TAN, viscosity, particle count, and spectroscopy on a biannual basis. These tests are pretty inexpensive but very telling (informative). RPVOT, rust test and/or Cu-corrosion, foaming and water separation tests are appropriate to do on the annual basis (these tests are more spendy but will give you a good overall picture of the oil). If you often experience ingression of water, then you can do this test more often (even one each month because it is really cheap).

The OEM of the turbine should have provided recommendation as to what specific tests to conduct on you oil. The oil supplier should be consulted on the frequencies.