Determining necessary oil flow volume for main and rod bearings 2

[racinvrxse]

Is there a way to determine the amount of flow volume needed if you know what your HP, RPM, journal diameter, etc?

 

[Lou Scannon]

There is. It's a fairly esoteric branch of tribology, and the main topic headings are hydrodynamic bearings or hydrodynamic lubrication.
It's been too long ago since I studied that to be any direct help, but I'd be surprised if there isn't a forum on Eng-tips that has a few experts on the subject.
In addition to the variables you listed, you will need to know the bearing clearance, oil viscosity (at max operating temperature), effective bearing width (taking into account grooves), peak normal loading, and probably the available oil feed pressure to the bearing, though that could be considered an output variable, i.e. required feed pressure.

 

[Lou Scannon]

...effective bearing width also taking into account chamfers.

 

[tbuelna]

racinvrxse,

Here's a link to an entry level engineering lecture slide deck that should get you started. For a more in depth technical reference, try this NASA document.

As hemi notes, the design and analysis of hydrodynamic journal bearings is a complex subject.

Good luck.
Terry

 

[Lou Scannon]

I believe the angular position of the oil feed hole(s) relative to the loading vs angle is also an important factor in the design/analysis.

 

[romke]

the flow needed is in fact determined by two parameters.

first the geometry of the bearing set up: you need at least a flow that is sufficient to keep the crankshaft and the bearing surfaces separated - a fluid film thus sufficiently thick that the unevenness of the surfaces is "filled with fluid" and then a little more - a filmthickness of 35 micron or more usually is sufficient. the flow you need in fact needs to be just a little more then the leakage that occurs.

a second consideration is bearing temperature. since the oil is used to carry away the heat generated in the bearing (hopefully only the result of fluid friction and not of contact between shaft and bearings) you will keep the temperature of the outflowing oil within reasonable limits. a ingoing temperature of about 100 deg C and a outgoing temperature of say 175 deg can be used as a starting point.

this second consideration usually leads to a quite a bit higher flow rate needed then would be needed for lubrication alone.

thus, the overriding requirement is the "bearing out" temperature - if that gets too high you need a higher flowrate.

as a start you could determine the flowrate as the engine originally had - if you tweak the engine you will at least need the same amount og litres/min/hp as before, and maybe more if the bearing out temperature rises beyond what can be considered "safe".

 

[racinvrxse]

WOW! I think my brain has mushed into oatmeal after looking at a tiny fraction of the nasa document! With my limited engineering knowledge and Associates in Engineering Technology degree, I'm certainly at a loss of understanding its contents. Thank you to all of you that responded!

Hemi, I'm intrigued though about your statement concerning the "angular position of the oil feed holes" of the main journal bearings (most connecting rod bearings I've seen have no feed holes). Its seems that would be a good idea...cut the holes into an angular position...I'm guessing in the same direction as the crankshaft rotates. It seems that would be a needed improvement since it would help to fill the channel more efficiently.

And romke, I guess the big question is how do I determine liters/min/hp, but after seeing the nasa document above it's probably a level of math beyond my skill.

 

[Lou Scannon]

Engine block and crankshaft geometry being what it is, I suspect there isn't much leeway for jockeying the oil feed hole positions, but nevertheless, I believe the angular location that the pressurized oil is being introduced into the journal bearing plays a minor role.
There are also different schools of thought regarding optional grooves, e.g. in the main bearing bottom half (the upper grooves are required, I believe, in order to ensure adequate supply of oil to the rod bearings via the crankshaft drillings.

 

[Tmoose]

"most connecting rod bearings I've seen have no feed holes ".

The rod bearing inserts themselves may not have holes except in those odd engines that feed oil up to the wrist pin, or spit oil at the cam or cylinder walls (and completely ignoring the era when the crank had no holes and the con rods were fitted with dippers and dived into oil filled troughs).

Chevy 265/283 - feed hole thru parting face

http://image.superchevy.com/f/34410...l_block_chevy_dormant_engine+rod_bearings.jpg

Jag XKE - oil feed

http://cdn1.autopartsnetwork.com/images/catalog/wp/full/W01331620766GLY.JPG

Today, and for some time, the oil arrives at the rod bearing via holes drilled in the crank, leading from (some of ) the main journals where pressurized oil is available.

http://image.chevyhiperformance.com...evy_buildups+rpm_international_crankshaft.jpg

reil feed timing - For a while now, the "Chevrolet power" book has advocated adding a 1/2 inch long lead-in groove in a particular direction to the oil hole in each main bearing journal for over 7500 rpm service. They prefer this to "cross drilled" mains that, when combined with a fully grooved upper main shell, would provide ~ 360 degree/full time oil flow to the rod bearings. SBC oiling systems have proven basically sound, so less robust designs need more extreme measures to get sufficient oil to the rods, including fully grooved mains that bring a significant (~70%) reduction in main bearing load capacity.
Or, cross-drilled rod journals as used by some.
1960s/1070s Alfa 2 liter

http://autovolante.tripod.com/sitebuildercontent/sitebuilderpictures/ep065.jpg

 

[Lou Scannon]

Tmoose, thanks for the input and links. You're obviously way more "hands-on" on this subject than myself.

 

[tbuelna]

racinvrxse,

I don't blame you for not wanting to read all the way through that NASA document. But even if details of the physics and fluid mechanics theory are beyond your technical level, you still should be able to grasp some of the basic ideas presented.

There are two things to consider with your main and rod bearing oil flow problem. First, what is the minimum oil flow required by the bearings. Second, what is the actual flow through the bearing.

The minimum oil flow required in recip engine main and rod bearings will usually be determined by heat rejection requirements, and thus is mostly a mass flow problem. There are several factors that go into determining the required mass flow rate. There is the local heat transfer between the hydrodynamic oil film, bearing, journal and housing. There are defined temperature limits for the bearing shell and journal materials. There is the lube oil inlet temperature and specific heat value. There are the losses generated in the bearing contact. Most importantly, there is the critical issue of flash temperature in the hydrodynamic fluid film. If the film flash temp gets too high, it can reduce viscosity enough to cause the contact to degrade into boundary conditions, which would then eventually result in scuffing/scoring failure.

The actual flow through the bearing is determined by factors such as fluid properties, pressure at the point of regulation, flow losses in the block and crank galleries, flow losses in the bearings due to groove and hole geometries, flow variation due to dynamic forces, and properties of the discharge orifice (ie. the radial and axial bearing clearance).

The analysis problem is made even more complex by the fact that the operating conditions are constantly changing. Since the oil flows must be sized for some combined worst-case set of operating conditions, the flows will usually seem a bit excessive for normal operation.

Hope that helps.
Terry

 

[BrianGar]

racinvrxse,

Im going to make the most lax engineering statement the forum has ever witnessed so get ready.

Over the last 15yrs Ive studied a LOT of oil pumps from a LOT of different engines.
Engine ranges were,
All 1.8-2.0l
8, 16, and 20 valves.
N/A, turbo, and supercharged.
Petrol and Diesel.

I must have taken apart at least 60 pumps, measured, cc'd(hot wax), worked out the rpm in relation to crank rpm and so on.

Keep in mind the engines were all pretty different, for example 2.0l T diesel, to a 2.0l 16v petrol with a redline of nearly 10k, I found that the displacement per rev of all pumps was very similar. The pressure relief valve seat pressure where fitted(to pump) was also very similar between all the pumps.

Keeping in mind, that I once upon a time fitted a spectacular ''on the fly remote adjustable pressure relief valve'' to a golf gti makes me wonder just how over-sized oil pumps actually are, as mentioned above by Terry.

I found that even on hot idle, I could damn near screw the relief spring pressure out all the way(that amount you can guess-a lot) before I got the oil pressure light to blink.

Up the revs, going hard, I screwed it out all the way until it met a safety pin I had fitted to stop it falling out.

OBVIOUSLY all this jargon does not tell you anything on how to size, or design a pump with respect to the mains size, but what it does or should tell you is that if you're trying to size a pump for 2.0l engine, look around you, but dont blame me if anything explodes.

I certainly opened my eyes anyway, and burned my hands a few times with both hot oil, and hot wax.

Brian,