Bearings and oil

[pete roper]

Given the somewhat rancorous nature of a couple of threads recently perhaps its time to examine this issue. Over a year ago I was about to try and explain something about the relationship but them my Mum set about dying and I sort of gave up. Perhaps it's time I revisited the issue. Note this is NOT going to be a discussion of the relative merits of oils. Simply a brief explanation of their purpose and function with bearing surfaces. I'll also try to keep it both simple and acurate. If people want to jump in and correct me or add adendums and caveats? Thats fine by me.

 

Firstly oil has two basic functions.

 

1.) To lubricate.

 

2.) To cool.

 

Lubrication essentially consists of forming a layer between two parts that are moving in relation to each other so that they don't touch.

 

Cooling is the ability to remove waste heat to prevent various parts reaching the point where they can change state, basically melt!

 

So how does this relate specifically to plain bearings as are used in our motors? For this purpose let us look specifically at the bearings on the crank, although the cam bearings are of the same type. Lets look at the loadings that are going to be imposed for example on the big end bearings of the connecting rods. These are a very good example because the forces imposed on them are relatively easy to calculate, at least on a simplistic level, and remember this IS simplistic.

 

OK. So the piston is at bottom dead centre, (BDC.) on the induction stroke. Theoretically the cylinder will be full of mixture at atmospheric pressure. 14 PSI. Now the crank rotates until the piston is at top dead centre (TDC) with both valves closed and the spark about to ignite the mixture. But the same amount of mixture is now crammed into a much smaller space. With a 10 to 1 compression ratio it's pressure will of increased tenfold, (In fact it increases more than this because as the gas is compressed it heats up but we'll overlook that for our purposes.) so now you have a pressure of 140PSI in the combustion chamber.

 

When the spark ignites the mixture it gets VERY hot, VERY quickly. The gas REALLY wants to expand, and so it does, pushing the piston down the cylinder. But during those brief moments when the mixture is burning the pressure rise in the combustion chamber is astronomical! Lets say for convenience sake that it increases the pressure by another factor of ten. So you now have a pressure of 1400PSI pushing down on the top of the piston. Calculate, (roughly.) the area of the piston. 92mm is about 3 and 1/2 inches. The area of a circle is Pi x the radius squared or about 9 and 1/2 inches. So 9 & 1/2 times 1400 = 13300 lbs pressure all bearing down on the rod and thence onto the big end bearing!

 

But hang on???? The oil that is being delivered to that bearing is only being delivered at 50-60 PSI? Surely if the bearing was going to have to take that load it would have to be HUGE? 13300 divided by 50? You'd need 266 sq inches of bearing wouldn't you?

 

Well, actually, No. Firstly there are a lot of other issues at play that mean that the pressures imposed on the bearing won't get to those levels but for our purposes today they explain the situation clearly and simply. The other reason you don't need VAST bearing surfaces is because of the relationship between the bearing and the journal it turns on and the way the oil behaves within the confines of that bearing and we'll go there next time.

 

Pete

[emry]

If Pete did not spend so much time with rubber chickens he could have programmed this. Just a graphical example of pressure increase during combustion.

Java Applet of Otto Cycle engine.

 

I can not find any reference to the actual pressure being calculated perhaps someone else can. ???Bar???

 

For those of you that are kinetic learners loosen one plug up til it is just barely held in, start engine and rev. Small tip - don't stand on the side of the loose plug.

 

Really though, don't try this... trust me...

[pete roper]

Alas I'm too stupid to pull up graphics off the net, (Honestly, I am, I know it's *easy* but I'm obviously the computer version of Piltdown Man ).

 

So lets look at why you don't have to have vast bearing surfaces. remember this is the bonzoid, simplified version for thickies like me. If you want heaps of sums go talk to an engineer!

 

It's actually very clever. as the jornal rotates in relation to the bearing, at the point where the greatest force is being applied to the oil film there is a phenomenon created called the 'Hydro-Dynamic Wedge'. At this point, as the journal tries to force it's way through the oil film as it spins a rolling wave-front of pressure is created within the film of oil that radically multipiles its effective pressure. The smaller the clearance, the greater the rise in the pressure wave of the wedge and the less likely it is that you will get the two surfaces rubbing together. (A phenomenon known as 'Boundary Lubrication').

 

"OK," you say. "But if smaller the clearance the greater protection you'll get, why not run plain bearings with microns of clearance rather than the standard amount?" which tends to be about 1 thou for every inch of journal diameter. Well the answer to that lies in the other property of the oil, (We'll ignore thermal expansion of jornal and bearing for now.). It's other property is to COOL. That is, take away heat and dump it somewhere else. Now to do that the clearance has to be great enough that you get a good throughput of oil. The oil has to do a LOT of work. The forces acting upon it, as we have seen by the 'Back of an envelope' calculations above are pretty big and they are, in the big ends, occuring many, many times a second. As the journal tries to *squash* the oil out of the Hydrodynamic Wedge between itself and the bearing it gets crushed quite effectively at the mollecular level. Liquids are, to all intents and purposes, incompressible. So to cope with this sort of force all it can do is make the mollecules rub together and as they do this they heat up. If you ain't got enough throughput of oil the oil can't transport the heat away and the bearing will overheat and change state, (Liquify!). Once the bearing has lost it's integrity and clearance the oil can no longer form the HD Wedge and the whole poxy lot goes to hell in a handbasket in about the time it takes you to utter a few 'Potty Mouthed' phrases and whip the clutch in!

 

So, like so many things in your engine you have to look at the bearings as another example of compromise. There is no *ideal* clearance for a plain bearing. What you want for some boring old sh!tter you're going to plod through traffic on on the way to work is going to be very different to what is required in a race engine for example. If you're going to 'Blueprint', (A daft term!) a motor for a commuter you'll make the bearing clearances as tight as possible as it isn't going to work terrifically hard and you'll want it to last for ever. For a race engine you'll deliberately try and get the clearances on the looser side so as to get a higher throughput of cooling oil!

 

The formation of the HD Wedge is also very dependent on both the journal and the bearing it rotates in being perfectly round. this is one of those things that people often overlook, especially when dealing with connecting rods. If an engine is being rebuilt properly after it has shagged it's big ends, (As V11's are particularly prone to due to the oil pick-up exposure problem.) then it is very important to check the roundness of the big end eyes. If they've been playing hop-scotch on the crankpin there is a very good chance they will no longer be round. Install the shells and they will be out of round and this will seriously compromise the engine's ability to form a wedge in the bearings and another failure will simply be waiting to happen.

 

In any of the bearings, especially on the crank, in your Guzzi motor, the speed that the bearing material is moving in relation to the journal is quite high. Say your journal is 1 and 1/2 inches across. The circumfirence of a circle is Pi x the circle's diameter. So in that case the circumfirence of the journal is about 4.77 inches. Now if the crank, and there fore the journal, is rotating at 6,000RPM the speed relatively betwixt journal and bearing will be 4.7 x 6000 = 28200 inches per minute or 2350 feet per minute or 39 feet per second! Try getting someone to tow you up the road on your arse behind a car at 39 feet per second without smearing your bum with some lubricant and see how long it would last! Even with a bum the size of mine I can tell you it wouldn't be long!

 

That is the reason it is VITAL to maintain both oil PRESSURE and VOLUME to plain bearings. Loose that, (As in when your oil light comes on during a hard launch.) and you're basically rubbing some soft tinny metal on some hard, nitrided steel at 39 feet per second while exerting a force of thousands of pounds per square inch on it! Any guesses as to how many times you can do this without serious damage occuring?

 

Answers on the back of a postcard to 'Pedants Annonymous' Care of 'Who-gives-a-toss-ville!'

 

Pete, trying not to be too Potty Mouthed.

[dlaing]

I can not find any reference to the actual pressure being calculated perhaps someone else can. ???Bar???

The animated graph looks like it is measuring in atmospheres.

Compressing to about 8 atmospheres and then ignition causing it to jump to about 24 atmospheres.

I have no idea whether the 24 atmospheres would occur at idle, while accelerating or while decelerating. It should vary proportionally to the amount of torque being produced, and I am pretty sure our engines will create more than 24 atmospheres of pressure. But yes, I could be wrong...

[dlaing]

Pete, trying not to be too Potty Mouthed.

I love it when you talk gear head

[mark.gilmore]

Your a Wizzard Pete.I love that shit. ooop's

[pete roper]

The animated graph looks like it is measuring in atmospheres.

Compressing to about 8 atmospheres and then ignition causing it to jump to about 24 atmospheres.

I have no idea whether the 24 atmospheres would occur at idle, while accelerating or while decelerating. It should vary proportionally to the amount of torque being produced, and I am pretty sure our engines will create more than 24 atmospheres of pressure. But yes, I could be wrong...

 

The 24 atmospheres would presumably be at maximum, or close to best, volumetric efficiency. As I said, my figures were rubbery and didn't take in things like blow-by or frictional losses and are only relevant when the angle of the rod is at, effectively, zero.

 

Close the inlet tract, (By shutting the TB butterfly/carb slide or whatever) and add in restrictions on the expulsion of the exhaust gasses, (Exhuat/muffler design.) and you can see that everything is going to change. I was trying to explain the principle, not the exactitude. There are simply too many variables to give a definitive answer. That's why we have sh!t like flow benches. It's all experimentation!

 

Pete

[stogieBill]

Wonderful explanation Pete. Thanks for taking the time.

[dlaing]

I was trying to explain the principle, not the exactitude.

Understood.

and BTW, Excellent explanation!

[Guest ratchethack]

Thanks, Pete. Every time you do the "hydrodynamic wedgie" I learn another half-dozen things I never knew I didn't know.

 

Makes me want to run heavier oil every time. You got me up to 20W-50. That's as high as I'm goin'.

[pete roper]

Makes me want to run heavier oil every time. You got me up to 20W-50. That's as high as I'm goin'.

 

 

Using a heavier oil can produce its own problems. The heavier the oil the greater its internal friction. This not only means that it will get hotter before it escapes from the bearing but its viscous drag on the bearing itself will increase. This will not only use power that could otherwise bring the horizon towards you faster, (Or use more fuel.) but there is a risk, albeit slight, that the oil's friction will overcome that between the back of the bearing shell in the end of the rod and its cap and the rod itself. What governs and limits the bearing's will to spin within the big end eye is not the little tangs on the bearings, they are just locators, it's the contact between the back of the shell and the seat it sits in. This is called the bearing 'Back Clearance' and can be examined on a shell that has seen long service by looking at the differential discolouration on the back of the shell when it is removed. What actually controls the effectiveness of the back clearance is called the 'Nip' of the bearing.

 

If the rod eye is correctly sized then when the shells are inserted and the bolts tightened, (This is without the rod on the crank.) the ends of the bearings will clamp against each other. Once both bolts are torqued correctly one of them can be loosened and the force of the shells pushing against each other will spread the cap from the rod slightly. This can then be measured with a feeler guage and the process is known as 'Measuring the nip'. If the nip is too large there is a danger of the bearing shells deforming as the bolts are tightened on the crank and grabing the journal. If the nip is too small there will be insufficient pressure to ensure a good back clearance and the shells will be prone to spinning in the eye of the rod.

 

The good news is that Guzzi rods are generally a.) Well made and b.) strong so you can get away with a lot on a Guzzi rod that would be much riskier on some other motors.

 

Pete

[Guest ratchethack]

By Yimminy, Pete, you got me again with another coupla new items on your semi-annual discourse of 'Nip'.

 

Much appreciated again.

 

Pete, I can't efficiently archive all this stuff from your posts! There's too much overlap and no continuity! If you don't write a book yourself, somebody's eventually gonna hafta spend literally months interviewing you in your shop over the mountains of old Ambo munt, and write all this stuff down. You'd have to swear not to throw shagged-out con rods at me for asking you to repeat y'erself whilst I get it on disk, but If I can work a stint in Bungendore NSW into my retirement schedule, and the royalty numbers look good to you -- after review of qualified publisher analysis and projections of course -- you might be persuaded to consider a book deal in a decade or so?

 

In the meantime, since I'm not running against local Guzzi speed legend Bill Ross and Team Crowbar, http://teamsubtlecrowbar.pitpilot.com/about/ross.htm at Bonneville Salt Flats this year, who're probably running something close to kerosene in the sump, I reckon 20W-50 would meet the Roper requirement for my climate and use to a "T".

 

FWIW, Team Crowbar didn't make it to Bonneville this year as planned. Short story is that days before the trip, the cam dowel sheared, taking out both intake valves. Among the endless list of other reasons for others so "inclined" not to pursure this kinda foolishness are the fact that they also had designed-in insufficient lifter diameter, and the "Diamond like Coating“ on the lifters, by a company called "Oerlikon-Balzers" turned out to be far worse in practice than no coating at all.

 

-- Just in case anyone here is considering Oerlikon-Balzers -- you might want to re-visit that.

 

Sorry -- didn't mean to side-spur the thread!

[pete roper]

As far as I'm concerned side-spurs are fine

 

One of the things I come across constantly is the fact that many people whoi are excellent 'Home Mechanics' have 'Blocks' of information missing. If I can insert something into these empty spaces? That's a good thing.

 

I am definitely NOT, (As I have oft stated before <_ the best mechanic in world. i not lucky enough to have been born with a thumb but when was doing my training paid attention and listened what mentors told me. that means while freely admit there are gaps knowledge understanding am definitely an engineer. if want advice heap of people much cleverer than me can approach greg field mike haven for example. sure emry would throw his hat ring asked nicely host others who also more happy help out start groaning holding head src="%7B___base_url___%7D/uploads/emoticons/default_anigrin.gif" alt=":grin:"> .

 

I do have a wealth of knowledge both general and Guzzi specific. To say otherwise would be bloody stupid. I'm more than happy to share it and help others if I can but as in all cases this requires people to be aware of their own limits. I know mine, (I can't weld for sh!t! Molten metal and I have an agreement. I don't f@ck with it and it doesn't f@ck with me ) I try not to exceed them! But understanding stuff like this makes it a WHOLE lot easier to understand what you're trying to achieve if you accept that anything you try and do is invariably going to be a test of cobbling together the very best set of compromises, not finding the 'Magic Bullet' that will make everything marvey!

 

Pete

[raz]

Thanks Pete. This reminds me of something I came across recently. This guy Motoman (an old wierd Internet site with some arguable but interesting views on tuning) claims that the assembled rod shells are (can be?) very sligthly elliptic, to account for big end inner diameter stretching on high RPM. Is that true at all? You wrote the opposite but that could be a simplification I guess. Or maybe his statement is for high RPM japs, not ditch pumps like ours.

 

http://mototuneusa.com/circular_logic.htm (half-way down and then some)

[pete roper]

Thanks Pete. This reminds me of something I came across recently. This guy Motoman (an old wierd Internet site with some arguable but interesting views on tuning) claims that the assembled rod shells are (can be?) very sligthly elliptic, to account for big end inner diameter stretching on high RPM. Is that true at all? You wrote the opposite but that could be a simplification I guess. Or maybe his statement is for high RPM japs, not ditch pumps like ours.

 

http://mototuneusa.com/circular_logic.htm (half-way down and then some)

 

It may be the case in very highly tuned race motors but think about the the practicality of machining an *accurate* ellipse into a rod? yes, rods will stretch and deform, generally though unless they are taken beyond their design parameters the amount of deformation will of been taken into account in the design and manufacture of the rod. Remember, we are talking about 'Production' vehicles here. Not some incredibly expensive, 'Hand Tooled' perfectionists device. It has to be easily repairable by people with access to *ordinary* tooling. But I do imagine that you are correct in the fact that the clearances used in your average rod WILL take into account deformation of the eyes. Hopefully to a point where interference with the bearing's ability to wedge won't be effected.

 

Pete