Bearings...how they work.

[Shannow]

When you start modelling bearings at Universisty, the standard mathematical tricks are employed...first one is that the bearing is first modeled as infinitely long.

If you imagine an oil molecule leaving the bearing feed hole, it has two directions of motion, one is around the bearing, and the other is towards the side, the ends of the bearing where it drains out...it's a spiral motion.

With an infinitely long bearing, the sideways movement is nil, so you are only concerned about the one plane hydrodynamics. The pressure gradient across the face is inconsequential.

But with a short bearing, like we see in engines, there's a lot of "end" for the oil to leak out...the bearing can't support as much load, and needs more oil supply.

That's where design tables come into play. These are correction curves for

First one is MOFT

Sommerfeld number is again the bottom axis.

This demonstrates the effect of short bearings on MOFT.

Draw a vertical line at say 0.2 on the lower axis, and you can see the difference in MOFT for the various bearing geometries. MOFT for an infinitely long bearing is about 6 times that for a bearing that's 1/4 as long as it's diameter.

Again, consider the parameters that give you increased MOFT.

Also, as per previous comment that the X axis for the Stribeck curve is the second part of the Sommerfeld number, you can see from these curves that the actual slope on the hydrodynamic side of lubrication on the Stribeck curve is geometry sensitive when applied to specific examples.

 

[Shannow]

Second Curve is the side leakage curve.

You can see that as described, the short bearings, L/D of 1/4 use a lot more oil than the infinitely long example.

I will have a dig here, 'though.

The premise of dropping oil viscosity to close the pump relief is demonstrated here, in this graph, and the effects are in the one previously.

With all other things equal, pick a point on either the l/d = 1/4 or 1/2 (what's typical ? 20mm/50mm) and assuming there's no change to speed, load, and obviously geometry.

Your intent is to increase the bearing volume flow rate to close your bypass, so you have to reduce the viscosity to achieve more flow, moving the curve to the left.

Now, superimpose that same change back on the MOFT line for that bearing and see what your changes have done to your MOFT.

 

[Shannow]

Regarding my previous statement that bearings will only take what they need, this is one of my favourite discoveries.

bearing leakage and viscosity.

The experimenters took an engine where there was easy access to the oil gallery on number 4 main bearing, and provided their own oil supply to it.

Then they pressurised the supply to 60psi, and ran the engine, measuring the time taken for a voulme of oil to go through the bearing.

Here's a screen dump

Couple of interesting takaways.
* the volume flow measures was a cup, 250ml.
* the time to use that flow was measured in minutes...a running engine, with 60psi of pressure on that gallery and it's fractions of a cup per minute.
* They noticed the temporary shear effect of non Newtonian (multigrade) oils, where the apparent viscosity in the bearing is lower than the kinematics would suggest.

 

[JaseD]

You realise that this is a waste of time because it's not solving the RP against the rest of the world contest.

Seriously though, these technical threads are fantastic even if some of it's over my head, there's plenty to learn from them, great stuff, many thanks Shannow.

Best wishes,
Jason.

 

[Yah-Tah-Hey]

What turbine/generator sets do you have?

 

[fdcg27]

If I understand what you've posted correctly, then oil pressure isn't really all that important, since the bearings will take what they need and only what they need as long as oil is supplied in adequate quantity to replace what leaks out to the sides.
Am I thinking along the right lines?
OTOH, if pressure is a measure of the resistance to flow, then higher pressure to the main bearings will accomplish nothing beyond wasted work?
Does this also explain why engines seem to do just fine on a wide range of oil grades?
As long as enough oil is supplied to keep the bearings full, a thinner oil will allow the same wedge to develop in a running engine with higher side leakage?

 

[Shannow]

fdcg27,
yep, you are reading correctly.

The idea isn't to jam a sump full of oil every x seconds through them, it's to keep them full...that's all the pressure does (for bearing...squirters rely on pressure to provide flow).

A thicker oil will have lower side leakage, so will draw less from the galleries, and as a result will show a higher oil pressure.

CATERHAM's oil pressure viscosity statements are correct in that regard...it's the high shear rate viscosity that he is "measuring" at the revs he uses.

 

[Shannow]

Originally Posted By: HosteenJorje
What turbine/generator sets do you have?


Japanese GE licenced machines at one site, and Parsons (GEC designed) Erith at the now closed site.

 

[Oil Changer]

Great synopsis. This is why I like OVERKILL's posts. He takes the tech-heavy posts from Shannow and re-writes them for the common man. That is what you did. It shows you have a good understanding of the material.

Originally Posted By: fdcg27
If I understand what you've posted correctly, then oil pressure isn't really all that important, since the bearings will take what they need and only what they need as long as oil is supplied in adequate quantity to replace what leaks out to the sides.
Am I thinking along the right lines?
OTOH, if pressure is a measure of the resistance to flow, then higher pressure to the main bearings will accomplish nothing beyond wasted work?
Does this also explain why engines seem to do just fine on a wide range of oil grades?
As long as enough oil is supplied to keep the bearings full, a thinner oil will allow the same wedge to develop in a running engine with higher side leakage?

 

[userfriendly]

Shannow mentioned that bearing side leakage is
viscosity/time and very little to do with pressure.

Pressure therefor has little affect on oil flow through
a (plain) bearing.

Lowering the lubricant's viscosity by heating or choosing
a lighter grade, may require a larger oil pump during low
RPM, but only sufficient in size to supply oil to the gallery
feeding the bearing.
The bearing's self pumping action takes it from there.

Bearing side leakage has a time factor and most automotive
engines have a positive displacement pump.
That means X amount of oil is pumped per engine revolution.

With no other calculation than time, which is RPM,
it is easy to assume that the leakage rate would be halved
when the RPM is doubled.
Oil demand per RPM in plain bearings would be halved
when RPM is doubled.
That is the reason pressure rises
with RPM in engines that have fixed
displacement oil pumps.

Other X factors affecting bearing side leakage
would be load, perhaps shear rate (friction),
lubricant heating during it's life-cycle in the bearing,
the VI of the oil at the known lubricant temperatures,
whether the lubricant is Newtonian or not, and likely
many other inputs such as acceleration and
centrifugal force acting on the oil
and it's supply route.

 

[AdamC]

Ok, so the reasons why high volume oil pumps are often used in a high performance rebuild are;
- Looser bearing clearances
- A hedge against higher oil temperatures (which reduces viscosity)

Anything else? Since flowrate through the bearings will be unchanged there's no cooling benefit, right? (except for piston squirters i guess).

 

[userfriendly]

The larger "high performance" pumps are used
because the engine builder does not know what
the future holds for the engines once they
leave the shop.
The most common application is the small and
big block Chevs which have a factory oil
pump option.
The non Bow-Tie blocks flex and move around
under high loads and RPM.
Crankshaft harmonics can also change clearances.

Fuel mixture detonation spikes cylinder pressure
at low rpm when bearing side leakage is at it's
greatest per revolution, and oil demand per
revolution is at it's highest.

The engine may leave the shop and assume duty in
a heavy delivery truck with a standard transmission.
Lugging at low rpm all day.
Again, high load at low rpm.

The engine may have NOS or other performance enhances
added at a later date.

The new owner may read oil forums and think that 0W20
is the answer for everything and reduce engine oil
viscosity to keep the oil pressure at an imaginary
maximum for an imaginary benefit.

The engine may idle in traffic on a hot day or become
overheated at some point in it's life.

Remember, we are dealing with unsophisticated engines
operating in uncharted territory.

What is the best practice in these cases?

Large clearances, redundantly over-sized oil pumps
and thick engine oil.

 

[Shannow]

http://www.tech.plymouth.ac.uk/sme/desnotes/LubeCalcEG1.htm

works through an example, and has some good commentary re surface finish, and the use of the tables.

 

[Shannow]

used_Oil

From the 10psi per 1,000 RPM rule of thumb, here's some others.

http://www.superchevy.com/how-to/engines-drivetrain/4380-bearing-clearance-info/

Note, that they are generalities, don't have speed and viscosity, but are applicable to the ubiquitous small block.

 

[il_signore97]

Thank you Shannow for this very interesting and highly informative thread... A refreshing change from the current norm at BITOG!!!

Much appreciated sir

 

[il_signore97]

Originally Posted By: fdcg27
As long as enough oil is supplied to keep the bearings full, a thinner oil will allow the same wedge to develop in a running engine with higher side leakage?



Yes, but a thinner oil would also cause the MOFT in the bearing to be lower as well, no?

 

[Shannow]

Originally Posted By: il_signore97
Thank you Shannow for this very interesting and highly informative thread... A refreshing change from the current norm at BITOG!!!

Much appreciated sir

Thanks

 

[Clevy]

Shannow. You always post interesting stuff and I feel better informed reading them.


Now let's get into some real interesting stuff.


I've heard there are pyramids down under.

Hehehe

 

[147_Grain]

Originally Posted By: Clevy
Shannow. You always post interesting stuff and I feel better informed reading them.

+1

....and thanks for pointing us in the right direction / helping us keep our bearings straight!

 

[Merkava_4]

I think you guys are only pretending to know what he's talking about.