Failure Analysis
A large general-purpose steam turbine was experiencing recurring sleeve bearing failures. The failure was consistently associated with the observation of black oil. The oil in the bearing housing would turn black and soon after, the turbine bearings began to vibrate or experience complete failure.
Due to the evidence discovered after failure, it was initially believed that the lubrication failures were caused by excessive water in the bearing housings, or the shaft not being concentric in the housing. Previously collected data had been taken only after failure. It was decided that in order to determine the root cause of failure, data needed to be collected during operation.
Steps in Collecting Data
- Turbine fits, bearing housing fits, and concentricity inspected.
- Oil rings replaced with OEM rings.
- Oil level marked on bearing housing.
- Alignment targets calculated and used, based on thermal growth.
- Spring adjusted to correct small amount of pipe stress.
-
Thorough check of cooling water lines to bearing housings
(found to be plugged). - VG 68 turbine-grade mineral oil was used.
Based on the data collected and steps taken, the most obvious issue was the plugged cooling water lines. The lines were cleaned but not replaced. The flow through the lines did not appear to be enough for adequate cooling, but the turbine was placed back in service.
After the turbine was in operation for several hours, the oil temperatures in the bearing housings were inspected. Using a thermocouple, the inboard oil temperature was found to be 214°F and the outboard was 244°F.
OEM lubrication recommendations per operating temperature are ISO VG 68 up to 200°F and VG 80 to 120 for temperatures greater than 200°F.
There was insufficient cooling to the bearing housing. A generally accepted viscosity at operating temperature for plain bearings is 13 cSt. At 244°F with a VG 68 oil, the operating viscosity was approximately 5 cSt. The lubricant film could not adequately support the shaft, which resulted in bearing failures.
The short-term recommendation was to use a PAO VG 100 oil to provide adequate viscosity and robust oxidation properties due to high temperatures. This reduced the bearing failures from several per year to only one.
The long-term recommendation was to install an external lube oil circulating system with a 20-gallon reservoir, filters, pumps and a cooler. This has been implemented on the turbine and current operating temperature of the oil is 160°F.
It is expected that this steam turbine will no longer be the cause of unreliability for this equipment train. Sometimes it takes a thorough failure analysis to find the root cause of failure. In this case, the cause was lube oil contamination in the form of heat.
Controlling Vibration
When machinery vibration is attributed to oil whirl, there are several temporary corrective measures that can be applied. In the article published in Machinery Lubrication entitled “Oil Whirl and Whip Instabilities - Within Journal Bearings,” the author recommended the following: change the temperature of the oil, purposely introduce a slight unbalance or misalignment to increase the loading, temporarily shift the alignment by heating or cooling support legs, scrape the sides of or groove the bearing surface to disrupt the lubricant wedge, or change the oil pressure.
For example, after start-up, a compressor was experiencing oil whirl with excessive vibration levels. Using the above-mentioned article as a reference, the coolers were cleaned and the oil temperature lowered to the bearing housing level. Unfortunately, vibration levels only slightly decreased, which did not stop the oil whirl. The next step was to increase the oil pressure by starting the spare lube oil pump and slowly allowing more pressure into the system. Again, this only slightly improved the vibration. Because the compressor could not be shutdown without high process debits, the next step was to shift the alignment of the compressor. Water was applied to the front foot of the compressor, causing a temporary misalignment. By cooling this foot, the oil whirl subsided and vibration levels dropped by a factor of eight. A long-term solution to this issue is currently being developed with the assistance of the bearing manufacturer.
When troubleshooting equipment, the initial corrective actions should be available and understood, because without a quick response time, irreversible damage or production shutdowns can occur.
Thank you Noria for the support!
Determining Oil Type
In the past, we’ve used synthetic oil polyalkylene glycol (PAG) in various gearboxes. Some of these gearboxes have since been changed to either mineral or a different type of synthetic oil. Operations posed the question, “can you easily tell if the oil in a gearbox is a PAG, because this type of oil does not mix with either mineral or PAO synthetic oils?” My first response was that a Fourier transform infrared (FTIR) test could be performed on the oil to determine if it is a PAG. However, our lubricant engineer suggested a test that could be performed in the on-site lab within hours. He recommended that we test for the specific gravity of the gear oil, which is comparatively higher than any other lubricant used in the gearbox.
With this simple and quick test, we were able to determine if the next oil change required a lubricant conversion procedure to use the mineral oil specified.
Editor’s Note:
The first three tips in this month’s Expert Advice section were contributed by Michael W. Istre, of CITGO Petroleum Corporation. Past issues of ML, such as for the oil whirl article, can be referenced online at
www.machinerylubrication.com.
Food-grade Lubricants
When a gearbox or transmission application requires a lubricant certified to National Sanitation Foundation (NSF) Standard 60, Drinking Water Treatment Chemicals, or Standard 61, Drinking Water System Components, I have found there are no NSF certified lubricating oils or greases having viscosities in the ISO viscosity grades 150 to 1,000.
An option may be to use an NSF registered food-grade lubricant. These can be found at http://www.nsf.com/ in the NSF Whitebook. To verify approval for use, contact your state regulatory authority, which is usually your local health or environmental quality departments.
The NSF has registered food-grade lubricants since 1999. For related articles, see “The Continuing Evolution of Food-grade Lubricants” by Jim Girard and “Understanding Food-grade Lubricants” by Martin Williamson.
Draft Standard NSF 116 - 2001 Draft NSF International/American National Standard for Food-grade Lubricants - Nonfood compounds used in food-processing facilities - food-grade lubricants maybe helpful, or contact: NSF International, general manager - Drinking Water Treatment and Distribution Systems, 789 N. Dixboro Road, Ann Arbor, MI, 48105-9723.
Richard Dornfeld,
Walker Process Equipment, Division of McNish Corp.
Drum Handling
To minimize water and particulate contamination from entering new drums of oil, it is preferred to keep drums indoors and stacked horizontally. However, when it is necessary to store drums outdoors and uncovered, here is a technique that may be used to remove water from the top of a drum:
Attach a paper towel to the top of the drum. In the accompanying photo, a magnetic clip was used. Allow the paper towel to hang over the edge of the drum so that the water wicks through the towel and down the side of the drum. Water will continue to travel from the top of the drum to the ground, so that breathing of water through the bungs is reduced.
David Turner,
Shell Global Solutions
