Machinery Lubrication magazine published by Noria Corporation
Issue link: https://www.e-digitaleditions.com/i/1449093
www.machinerylubrication.com | January - February 2022 | 27 ML ML a couple of tests. Since we don't necessarily know what kind of information we will be looking for in the future, it is a good idea to run a few different tests. Fourier Trans- form Infrared Spectroscopy (FTIR) and Elemental Spectroscopy are great places to start. e results from these two tests will set you up for success later on when you're trying to figure out what is going on in your equipment. Since we now have that initial sample, also known as our "baseline" sample, we can look at what we want to learn from our oil. ere are three major categories that people typically think about when it comes to oil analysis: 1. e condition of the oil. 2. e contaminants in the oil. 3. e wear taking place in the machine. Whether or not you need to know all three things from your oil is a bit of a complicated question. Let's say that you have a very small gearbox that only holds around a quart of oil. Oil analysis may or may not be warranted on this gearbox; it all depends on how critical this little gearbox is to operations. If this gearbox could shut down an entire facility by failing, then it is probably a good idea to do some analysis on it. For argument's sake, let's say that this gearbox has the power to shut the whole place down if it goes out. So, what type of oil analysis should we do on it? Well, the oil volume isn't large, so the cost of an oil change shouldn't be too bad. is means that information about the oil's condition may not be high on our priority list. How about contaminants? Certainly, solid particles would be of interest. ese are the ones that cause a lot of mechanical wear with three-body abrasion. Something as routine as a particle count would be a good idea, even with this small volume of lubricant. Machine condition? Absolutely! Machine condition would be the number one reason to perform oil analysis on this gearbox. Even if you don't care about the condition of the oil or have little thoughts as to the contami- nants getting in there; if this gearbox can shut the entire place down, you definitely want to know what kind of wear is taking place inside of it (not to mention how much wear). Now that we know the type of informa- tion we want to gain, what tests should we run? Since we want to know more about the wear of the machine components, we need to look for what those components are made from. Firstly, ferrous density is going to tell us the concentration of ferrous metals. is is an excellent resource since components are typically made with such metals. A step further is analytical ferrography. Analytical ferrography can be done in a lab with a technician looking at the actual wear metal's particles. ese technicians are highly trained and can offer insight as to what type of wear is taking place — an extremely handy tool for figuring out how healthy the machine is. Last but not least is elemental spec- troscopy. Unlike FTIR (which looks for compounds), elemental spectroscopy looks for elements (or the building blocks). is can be a fantastic tool, especially if you are familiar with the types of metals that exist within your equipment, particularly in the given components (like the bearing material, if there are babbitt materials). If I know that a particular gearbox is a worm drive and I see signs of copper, lead or bronze on my elemental analysis report, I know where those elements are coming from. is goes a long way in telling me how much wear has taken place. Now let's move on to a large hydraulic reservoir. is system isn't mission-crit- ical; no catastrophic failures will be caused if this system goes down, but the system holds a lot of high-dollar fluid in it. My focus switches a bit: I am no longer quite as interested in the wear metals as I am in the health of the oil. I mean, if it costs thousands of dollars just to change the oil in this system, I should probably try to get as much life out of that oil as possible. So, what information do we want from this system? ML ML