Pump Magazine On-Line






Article #35: What do we know and what do we think we know


Have you ever caught yourself wondering if you are truly clear on the subject you thought you knew for years? I asked myself this question occasionally, and often the answer is: not really. For years, working as an engineer for  major pump manufacturers, I thought I knew rather well what most would feel a simple subject - bearings lubrication. Sitting behind my desk, I could easily answer the customer question on expected bearing life, by reaching to the SKF catalog, looking up the formula for L10 calculations, and, ten minutes later, you got the answer. If the answer was 3 years, but the end user actually was getting only three month, - my suggestion was he was doing something wrong, which caused the bearing last that poorly. But what exactly was he doing wrong?


Too often, the pump factory "desk engineers" expect the end users to be as knowledgeable in the pumps as they are. If the end user, calling for help, does not know what the L10 is, or any other "fancy terms" they hear (specific speed, flow recirculation, etc.), they are often treated as inferior to their profession. Yet, these same plant engineers may know a thousand times more about the processes and techniques at the plant that the pump engineers at the OEM plant have no clue, and never will. It is unreasonable to expect the plant maintenance or reliability engineer to know as much as the pump expert behind the desk at the OEM factory as much about cavitation bubble size, blade angle, hydraulics, and so on. This is why we are supposed to have these so-called factory experts: to help when help is needed, instead of making the caller feel inferior about his or her question.


These days, my job is so much different from what I used to do at the pump manufacturer engineering department. I see, touch, smell, hear, and troubleshoot pumps at the power plants, paper mills, chemical plants, water and wastewater facilities. And I find the reality being so much different from the "book" that the standard behind-the-desk-over-the-phone answers. Take bearings grease lubrication, as an example.


In the recently published article (www.pumpingmachinery.com/pump_magazine/pump_articles/articles_summary.htm #31), I described the four designs I commonly see in the field. As trivial as the subject may seem to some, there is not that much knowledge and expertise, out there (in the field) about these design differences and why each applied. Interestingly enough, even my colleagues at the pump manufacturers as well as bearing manufacturers could not clearly explain which design is used, and for what reason. If a cross-flow lubrication is better then a thru-flow,  - for what reason? Is there practical case study that would show a pump user actually trying one design and then switching to another, with a tangible proof of quantifiable benefits? So while most would easily state the variety of designs: thru-flow, deadheaded, cross-flow, shielded (single or double), sealed, etc. - few would be able to explain the difference, the guidelines for different applications, and examples of actual installations.


Few would be able to tell conclusively the application rules of allowing grease (versus oil sump) to begin with. Does the rule relate to the shaft speed? Temperature? Bearing Size? The product of shaft diameter times the speed? What are the numeric values that would delineate such application rules? Can a 1.5x1-6 ANSI pump running at 3600 rpm use grease bearing lubrication? What about 4000 rpm? When would you switch to a single shielded bearing design, and at which side would you install it: the outboard side, or the pump side?


In taking the advise, ask the person that gives it to you, who do they know. And, if the answer is the someone else told them, - beware. I remember, as a young engineer at Ingersoll-Rand, coming for the first time to my boss (who I have a great respect to this day) office and seeing the sign at his desk: "Do Not Listen to Those Who Ain't Done It". I still remember the sign, and the message is still valid.



Dr. Lev Nelik, P.E. (NJ), APICS

Editor, Pump Magazine