Bearings and Filtration

Status
Not open for further replies.

MolaKule

Staff member
Joined
Jun 5, 2002
Messages
24,022
Location
Iowegia - USA
A White Paper on Bearing Design
and Filtration Requirements
By MoleKule

Keywords: Journal bearings; lubrication; contamination; filtration.

1.0 General
Bearings reduce the friction that robs machines of horsepower. Bearings are classified by type as Roller bearings, Ball bearings, Pin/Needle bearings, Journal (sleeve) bearings, and sliding bearings. Roller and ball bearings are usually found in applications such as transmissions and differentials. Pin/Needle bearings are also found in automatic transmissions and manual transmissions that use Automatic Transmission Fluids, and in 2-cycle engines. Sliding bearings in engines can be found in cam/tappet assemblies and of course in the cylinder, where the ring pack interfaces with the cylinder liner. In this paper we will focus on the design and filtration requirements for the sleeve bearings that support journals, such as crankshafts, connecting rods, and camshafts.
1.1 Sleeve Bearing Design
A number of factors affect bearing design including loads (forces), diameter of journal/bearing, length of journal/bearing, rotational speed of journal bearing, eccentricity of journal-bearing system, clearance between journal and bearing, the minimum oil film, the viscosity of the oil film, the heat produced by all frictions, the coefficient of friction of the journal-bearing system, and the volume of oil flow.

The journal does not rotate inside the bearing in a concentric fashion, rather is has a slight wobble or “eccentricity” (denoted as “e”) that generates not only a wedge of lubricant film, but also “pumps” the oil around the wedge. Eccentricity is defined as the distance between the bearing center and the journal (shaft) center. Assume the shaft is rotating counterclockwise in the sleeve and the bearing sleeve has a hole in it, and the shaft and bearing are immersed in an oil bath. The oil will be drawn into the “diametral” clearance when the shaft is rotating and will force the wedge of oil around the inside of the sleeve. In most case, we provide both oil-in and oil-out passageways, with a higher pressure on the “in” side to keep a clean and cool oil supply flowing.

Bearing friction is determined by an empirical equation (derived by test data) called the “McKee” equation as: Eqn. 1, f = 1x10^-10[473(ZN/p)D/C] + k, where k is a constant determined from another equation or chart for various L/D ratios; L is length of bearing (in.), D is diameter of bearing (in.), N is speed of journal in rpm, N’ is speed of journal in rps, Z is absolute viscosity in cP, C is diametral clearance between bearing and journal in inches. For most bearings in automotive use, k = 0.002. p is pressure = W/LD, W is bearing load in lbs. An average value for D/C is 1000. In automobile and piston aircraft engines, the loads W, are: Main – 700-1700 lbs., Crankpin – 1,400-3,400 lbs., and Wrist pin - 2,000-5,000 lbs. The ratio of e to the radial clearance is called the “attitude” and is defined as: Eqn. 2, A = 2e/C = 1- 2Ho/C. Ho is the “Minimum Film Thickness.” So how does one find the viscosity required for a bearing? One uses the “McKee” equation, along with the “Lasche” equation, for bearing heat generation and dissipation, Hd and Hg, and equates these equations to find the viscosity required!

Typical minimum film thicknesses, in hydrodynamic lubrication for these types of bearings, are on the order of 0.1 to 1um, with most operating films (in hydrodynamic lubrication) on the order of 1um to 18 um or less. Bearing clearances range from 0.001 to 0.002 inches for every 2.54 cm (1”) of journal diameter. During normal operation the journal is supported on an extremely thin film of oil, and the two parts should have no contact. As the rotational speed increases, the lubricant is drawn into the bearing by the rotating action of the journal (shaft). Consequently, the oil film becomes thicker and the increase in friction is therefore less than directly proportional to the speed. Conversely, at lower speeds, the oil film is thinner. At extremely low speeds (such as starting or “lugging”) there is likelihood that the bearing may fail or be permanently damaged unless there is a “barrier” lubricant between the shaft and bearing that can take the load and not shear under loads.

1.2 Filtration.and Bearing Wear
In Duchowski’s paper, he shows that the range of most abrasive particles that cause bearing wear are in the range of 3 um to 32 um. He suggests that the filter have a beta of B >= 200 at 6 um, “Given the film thickness which exists under standard (steady-state) operating conditions, the filter element appropriate for the application should exhibit a high removal efficiency for particles down to 10 um in size. In addition, the selected filter element should also exhibit a significant removal efficiency for particle removal down to about 3 um in size to allow for thinner film which exists under startup conditions.”

References:
1. Hall, et. Al., “Theory and Problems of Machine Design,” Schaum’s Outline series, Chapter 23, “Lubrication and Bearing Design.”
2. John K. Duchowski, Examination of Journal Bearing Filtration Requirements,” Lubrication Engineering, September, 1998, pp. 18-28.

[ September 03, 2002, 09:37 AM: Message edited by: MolaKule ]
 
You know, I understand the basics on the wedging effect of the hydrodynamic property of the oil, as the bearing rolls, it creates a wedge of oil infront of the roller, increase speeds,increases wedge, and slower=thinner, BUT, quite frankly, I have GOT to pay more ATTENTION to these bearins. I have never noticed on to be eccentric. I have ALWAYS thought that the cams were completely true in cercumfrence as was the bearings. Even wheel bearings seem to "look" completly round. I have worked with ecentric bolts and such but cams, cranks,ujoints, and such, Never have noticed that. I'll take my trusty little measureing tool and head to the machine shop and see if I can learn more on this as I just see it being the case, that there is ecentricity in bearings, cranks,and cams.
 
grin.gif
HeHe, I doubt you will see this unless you have some very thin gages, say 0.0005" or less.

One thing I think the paper points out is that a good barrier lube is needed during start-up and during low rpm, high load situations when the film is of the thickness of the surface finish.

Another thing is the filtration requirments and what particle size gives the most wear.
 
One other aspect of this thinking to consider and has served me well as a theory is that the 15um and smaller particles in a WELL additivized oil are going to be safely carried in a protective micelle for a long enough period of time to either be filtered or drained before they can do any damage.

This is the basic premise of my resistance to bypass filtration being needed and or cost effective for the smaller sump application.

[ September 05, 2002, 01:42 PM: Message edited by: Terry ]
 
I was told to run nondetergent oil in my 1952 cushman engine (single cylinder, flat head, four stroke, 18 Cubic Inch, upright, horizantal shaft) because it does not have an oil filter. Can anyone steer me in the right direction to figure out if a modern oil could be used? I was told that the detergent oils kept the particles in suspension so the filter could take them out and that in a non detergent oil the particles fall out and stay out of harms way in the sump until removed at oil change.
Any truth to this or can I just use a good synthetic?
 
Wheel bearings are specified to have 7 micron max runout on the outer ring. Any more runout may make the bearing noisy and susceptible to premature failure. The runout usually shows up as ovality, but may sometimes be tri-lobal or 5 lobes, depending on the design.
 
Status
Not open for further replies.
Back
Top