Roller Cams

[Michael H.]

Re: Roller Cams

Michael,

Don't forget the lube oil, which forms the thin film boundary layer, preventing most contact between lifters, lobes, rings, bearings, journals, cylinder walls, etc, etc. Although the components themselves do not contact each other (except in certain rare and cyclic occurrences, which is when wear takes place), it is the shear forces set up in the boundary layer, which create heat at a rate sufficient to raise the temp of the lube, per unit of time. This consumes energy at a specific rate.

The lube oil in the crankcase "heats up" or, rises in temperature for two reasons, the first of which, is because it shares a part in carrying away heat due to the combustion process. The second reason, is due to the shearing, on a molecular level. This action is similar to the transmission fluid heating within the torque converter of an automatic transmission (fluid shear coupling). It is also similar to the fluid heating which occurs in a vane type pump which is idling for extended periods.

For example, for a crankcase containing 6 quarts mineral oil, losing 1 HP due to shear:

6qts (1/500 lb/gm) (.8 gm/cc) (1/.00105668 cc/qt) = 9.08 lbs.

The specific heat of mineral oil is 0.4 BTU/lb. deg. F.
1 HP = 42.407 BTU/min.

42.407 BTU/min = 0.4 BTU/lb-deg F. (9.08 lb) X deg F.
where X is the temp rise per minute, of the 6 quarts of oil
solving for X yields 11.67 degrees/min.

An oil pump does not create oil pressure, it pushes oil from one place to another. It is a positive displacement pump that moves oil as it turns. Oil is incompressible so once it leaves the pump it continues to flow until it encounters resistance in the filter, oil galleries and bearings. It's the resistance to flow that builds pressure in the oil system. Trying to force oil through a small opening creates more resistance and pressure than allowing it to pass freely through a large opening. As pressure builds in the oil system, it exerts pressure. A spring-loaded "pressure relief valve" built into the oil pump (or near the pump) opens when pressure exceeds a certain limit (typically 50 to 60 psi) and either reroutes oil back into the pump's inlet or the oil pan. At idle, most oil pumps do not produce enough flow to force open the relief valve. Oil pumps that are camshaft driven turn only at half engine speed so output isn't great at idle and low rpm. Even pumps that are crankshaft driven and turn at engine speed (or double engine speed in a few instances) don't pump enough oil to overcome the relief valve spring. The relief valve generally only comes into play at higher rpm when the pump's output pushes more oil into the system than it can handle. Then the relief valve opens to vent oil and limit maximum oil pressure until the engine returns to idle or a lower rpm.

A high volume pump, having larger lobes, requires more power to drive it at all times. High pressure pumps, being equal sized to standard pressure pumps, only require additional energy when running at speeds above that at which a standard relief spring would bypass........

Vehicle manufacturers have traditionally recommend a minimum of 10 psi of oil pressure for every 1,000 rpm of engine speed. Using these numbers, most stock engines don't need any more than 50 to 60 psi of oil pressure. With tighter bearing clearances, pressure goes up requiring less flow from the pump and less parasitic horsepower loss to drive the oil pump. In racing applications, the old school of thought was more oil pressure was needed to keep the engine lubed. That's true if bearing clearances are loosened up. But most engine builders today tighten clearances so less oil flow is needed to maintain adequate oil pressure. This approach increases the horsepower output because less power is needed to drive the pump at high rpm, but sets up higher shear within the boundary layer, which leads to more losses due to shear.

Who ever wrote this had a few good points but there's a little more to it than that.
First of all, the amount of energy consumed in frictional losses at the base of the lifters isn't anywhere near the amount of energy consumed by the oil pump so the theory that it's a trade off isn't even close to being accurate.

He mentions that oil goes into "sheer" at the lifter to cam lobe contact point. That's not exactly correct. Oil can/does go into "sheer" IF the viscosity is greater than necessary for a given load/speed. (often the exact cause of bearing failure in engines operated at high load/RPM before the oil is at/near operating temp)
Sheer is when the oil fillm is ripped apart at the friction surface.

A lot of folks still think that there's a lot of friction at the cam lobe/lifter but, as mentioned previously, if that were true, cams and lifters surely wouldn't last more than a few miles.
The writer does refer to the fact that there isn't actually metal to metal contact between the lifter and cam other than on rare occasions.

Here's the problem that most people don't consider when talking about frictional losses at the cam/lifter. With a flat tappet design, the load applied to the lifter is nearly straight up along the axis of the lifter bore. The contact area between the lobe and lifter moves across the face of the lifter as it is being raised in it's bore. Thus, very little frictional losses at the bore.
However, with a roller style lifter, the side loading applied at the lifter exceeds, by a great amount, the frictional losses of a flat tappet design at the cam/lifter contact area. This is because the contact point of lobe/roller is a lot closer to the C/L of the lifter body and the lobe is trying to push the lifter up AND to the side at the same time.
If you don't believe me, ask Clem how many roller lifters were destroyed because the sides of the lifter body were scuffed.

Either way, the frictional losses of either flat tapped or roller is nowhere near as great as that of the oil pump.
In previous discussions on this topic, I suggested that, to get some idea of the energy required to turn an oil pump at operating RPM, even at idle RPM, plug in your 3/8" electric drill and drive the pump in an engine on the stand. You quickly learn that it requires a LOT of energy to turn that pump, even at that slow speed. That's why most builders use a 1/2"drill.
Now imagine the amount of energy required to turn the pump at 2500 engine RPM. If you don't think this is a factor in fuel mileage/emissions, you need to think about it smore.

The boys in engineering didn't spent all that time/cash on this project just because they thought it would be cool to use roller lifters.

I still have a ton of GM engineering paperwork/documents on this subject and if I have time, (and can find it) I'll try to dig it out and post some of the info. Some pretty impressive numbers.

[Clem Z.]

Re: Roller Cams

the big problem with after market roller cam setups is the seat pressure is 300#+ and close to 1000# over the nose which cause the problem with the side loading that michael was referring to. with this type of spring you need to go to larger diameter lifters to help spread the load. chevy lifters at .842 diameter are the smallest of any engine and the bigger lifters are close to 1.00 diameter

[Joe C.]

Re: Roller Cams

the big problem with after market roller cam setups is the seat pressure is 300#+ and close to 1000# over the nose which cause the problem with the side loading that michael was referring to. with this type of spring you need to go to larger diameter lifters to help spread the load. chevy lifters at .842 diameter are the smallest of any engine and the bigger lifters are close to 1.00 diameter

Clem,

Those are EXTREME spring force specs!!!

Most roller cams for street/strip use require valve springs with much lower rates.

Here is an example of a fairly "big" solid roller that I did a couple simulations for. Lunati 501B5LUN:

http://www.lunaticams.com/CamSpecCard.aspx?partNumber=501b5

106 (ground @ 102/110............4 deg advanced)
253/261 (@ .050)
.645/.645

This cam requires springs:
73899, or 73367

http://www.lunaticams.com/ProductGroup.aspx?id=189&cid=23&pid=2

You can see that forces are in the 200/500 lb range

[Joe C.]

Re: Roller Cams

Who ever wrote this had a few good points but there's a little more to it than that.
First of all, the amount of energy consumed in frictional losses at the base of the lifters isn't anywhere near the amount of energy consumed by the oil pump so the theory that it's a trade off isn't even close to being accurate.

He mentions that oil goes into "sheer" at the lifter to cam lobe contact point. That's not exactly correct. Oil can/does go into "sheer" IF the viscosity is greater than necessary for a given load/speed. (often the exact cause of bearing failure in engines operated at high load/RPM before the oil is at/near operating temp)
Sheer is when the oil fillm is ripped apart at the friction surface.

A lot of folks still think that there's a lot of friction at the cam lobe/lifter but, as mentioned previously, if that were true, cams and lifters surely wouldn't last more than a few miles.
The writer does refer to the fact that there isn't actually metal to metal contact between the lifter and cam other than on rare occasions.

Here's the problem that most people don't consider when talking about frictional losses at the cam/lifter. With a flat tappet design, the load applied to the lifter is nearly straight up along the axis of the lifter bore. The contact area between the lobe and lifter moves across the face of the lifter as it is being raised in it's bore. Thus, very little frictional losses at the bore.
However, with a roller style lifter, the side loading applied at the lifter exceeds, by a great amount, the frictional losses of a flat tappet design at the cam/lifter contact area. This is because the contact point of lobe/roller is a lot closer to the C/L of the lifter body and the lobe is trying to push the lifter up AND to the side at the same time.
If you don't believe me, ask Clem how many roller lifters were destroyed because the sides of the lifter body were scuffed.

Either way, the frictional losses of either flat tapped or roller is nowhere near as great as that of the oil pump.
In previous discussions on this topic, I suggested that, to get some idea of the energy required to turn an oil pump at operating RPM, even at idle RPM, plug in your 3/8" electric drill and drive the pump in an engine on the stand. You quickly learn that it requires a LOT of energy to turn that pump, even at that slow speed. That's why most builders use a 1/2"drill.
Now imagine the amount of energy required to turn the pump at 2500 engine RPM. If you don't think this is a factor in fuel mileage/emissions, you need to think about it smore.

The boys in engineering didn't spent all that time/cash on this project just because they thought it would be cool to use roller lifters.

I still have a ton of GM engineering paperwork/documents on this subject and if I have time, (and can find it) I'll try to dig it out and post some of the info. Some pretty impressive numbers.

Michael:

I wrote the above.
I cannot quantify, exactly, how much difference is attributable to the roller lifters and full-roller rockers. I CAN say, however, that when the L98 went to roller lifters for 1987, while retaining the same stamped steel rockers as 1986, the horsepower rating was increased by 10 SAE net, and torque increased by 15 ft-lbs.

[Clem Z.]

Re: Roller Cams

Clem,

Those are EXTREME spring force specs!!!

Most roller cams for street/strip use require valve springs with much lower rates.

Here is an example of a fairly "big" solid roller that I did a couple simulations for. Lunati 501B5LUN:

http://www.lunaticams.com/CamSpecCard.aspx?partNumber=501b5

106 (ground @ 102/110............4 deg advanced)
253/261 (@ .050)
.645/.645

This cam requires springs:
73899, or 73367

http://www.lunaticams.com/ProductGroup.aspx?id=189&cid=23&pid=2

You can see that forces are in the 200/500 lb range

i was referring to race roller cams that turn 10,000 RPM

[Joe L.]

Re: Roller Cams

All----


At the present time, ALL GM push-rod type PRODUCTION engines utilize roller hydraulic lifters. All push-rod type SERVICE engines, GMPP crate engines, industrial engines, and marine engines also use hydraulic roller cams EXCEPT the following:

---3.0 L inline 4 cylinder engines used for marine and industrial applications only;

---2.8L/3.1L/3.4L 60 degree SERVICE V-6 engines for older applications. These engines are, essentially, a "budget" sort of deal designed to compete with rebuilt engines and, consequently, are "bare-bones";

---5.7L Gen I V-8 GM #10067353 and 12568758 GMSPO SERVICE engines. These are intended for the same market as above;

---5.7L Gen I V-8 GM #12499529 and 12499711 GMPP SERVICE engines. These are, essentially, "entry level" performance engines designed for folks on a budget that want a performance increase. (NOTE: these engines have performance characteristics virtually identical to 1973-81 Corvettes with L-82)

---ZZ572/720R GM #12498792 SERVICE engine. This engine uses a MECHANICAL ROLLER cam and lifters.

While I don't keep up with what other manufacturers are doing, from what I do know all pushrod PRODUCTION engines used by Ford and Chrysler are also hydraulic roller type and have been for quite a few years.

[Joe C.]

Fully Rollerized Engines

The main thing to notice here, is the difference in friction in the two cases.........with conventional sleeve bearings, and with roller bearings.

http://engineeringtv.com/blogs/etv/archive/2007/05/31/episode-42-rollerized-engines.aspx

This same principle applies to the friction (whether it be "rolling" or "sliding") which exists at all of the interfaces within the valvetrain.......cam/follower, rocker/rocker ball, rocker tip/valve stem.

An added benefit of using a roller cam, is that much less oil drainback is required within the lifter valley to lube the tappets. Standpipes may be employed to limit oil splash, and smaller capacity oil pumps can be used, further limiting parasitic losses.

[Michael H.]

Re: Fully Rollerized Engines

The main thing to notice here, is the difference in friction in the two cases.........with conventional sleeve bearings, and with roller bearings.

http://engineeringtv.com/blogs/etv/archive/2007/05/31/episode-42-rollerized-engines.aspx

This same principle applies to the friction (whether it be "rolling" or "sliding") which exists at all of the interfaces within the valvetrain.......cam/follower, rocker/rocker ball, rocker tip/valve stem.

An added benefit of using a roller cam, is that much less oil drainback is required within the lifter valley to lube the tappets. Standpipes may be employed to limit oil splash, and smaller capacity oil pumps can be used, further limiting parasitic losses.

What that man proved is that without an oil pump, there's a LOT less drag on the rotating crankshaft. DUH!! Kinda what Mikey was saying all through this thread.

Actually, the test was somewhat flawed to begin with. He showed how difficult it was to spin the crankshaft in conventional bearings compared to how easy it spun in roller bearings. The amount of drag displayed with the conventional bearings is BS.
If the main bearing bores in the block are straight/true and the correct bearing clearance is used, a crankshaft will spin JUST AS EASY in conventional bearings.

I really hope you guys aren't going to run out and spend a lot of cash on rollerized everything thinking you're going to see some wonderful increase in HP. It ain't going to happen.

If you're afraid of the new blend of oil, that's a different matter and I'm all for rollerized components but don't think for a minute all this "rollerized" stuff is also going to make your car faster. It's primary function is to reduce the amount of HP that the oil pump sucks up.

Ya'll can debate this till next Xmas if ya want but I think you'll eventually understand what's happening with this.

[Joe C.]

Re: Fully Rollerized Engines

What that man proved is that without an oil pump, there's a LOT less drag on the rotating crankshaft. DUH!! Kinda what Mikey was saying all through this thread.

Actually, the test was somewhat flawed to begin with. He showed how difficult it was to spin the crankshaft in conventional bearings compared to how easy it spun in roller bearings. The amount of drag displayed with the conventional bearings is BS.
If the main bearing bores in the block are straight/true and the correct bearing clearance is used, a crankshaft will spin JUST AS EASY in conventional bearings.

I really hope you guys aren't going to run out and spend a lot of cash on rollerized everything thinking you're going to see some wonderful increase in HP. It ain't going to happen.

If you're afraid of the new blend of oil, that's a different matter and I'm all for rollerized components but don't think for a minute all this "rollerized" stuff is also going to make your car faster. It's primary function is to reduce the amount of HP that the oil pump sucks up.

Ya'll can debate this till next Xmas if ya want but I think you'll eventually understand what's happening with this.

Michael,

The man was working for Timken Bearing, not "Bubba's Bearing Shack".

He was bald, bespectacled, sheepish looking and had a pocket protector bulging beneath his cardigan sweater. So, he was probably an injuneer, not yer ordinary garage floor sweeper. So, WTF does he know, Duuuuuhh?

You're correct....they purposely skewed the test by misaligning the main saddles for the sleeve bearing test, and then they align honed the mains for the roller test. I suppose that Timken's roller bearing division needs a "shot-in-the-arm" at the expense of their sleeve bearing division.

I respectfully disagree with your position. You will not be convinced.

Your argument does not hold up, on the basis of mechanics, but you would certainly be able to make a valid argument against roller bearings on the basis of cost.

You haven't responded to my remarks above, concerning the 1987 v 1986 L98 engine.

You have, and always have had plenty of "good stuff" to offer, but in this case, you are mistaken.

Joe

PS: The use of a different type of oiling system, with lower parasitic losses is an added benefit of using roller bearings, in addition to the lower friction inherent in rolling surfaces versus sliding surfaces.
Again, the same line of reasoning applies to flat tappets v roller tappets, together with full roller trunnion rocker arms versus conventional type.................rolling friction is less than sliding friction.

[Duke W.]

Re: Roller Cams

There's little difference in friction between properly lubricated plain and roller bearings at moderate surface velocities when plain bearings are in hydrodynamic mode. Plain bearings have more static and very low speed friction ("boundary lubrication" mode), which is one reason why freight railcars went to roller bearings in the fifties. (Another is that they are a lot more reliable and durable than non-pressure lubricated journal bearings, especially as train speeds increased.) It's a lot easier to start a train with roller bearings.

At teardown or on a newly assembled engine the crankshaft should spin freely if given a flick from a counterweight. If there is significant drag there is likely a clearance or bearing alignment problem.

At the other end of the spectrum roller bearings have an advantage at very high surface velocities - above what a typical automotive engine sees.

Honda motorcycle engines from the early sixties, like the 250 cc SOHC twins were all roller bearing with built-up cranks and have no oil pump of any kind. Oil was slung up to the cam and rocker boxes by the cam chain. It was a simple design. Later, as the engines got bigger with higher specific outputs, journal bearings with pressurized lubrication were adopted, and I think it began with the DOHC 450 twin.

Chevrolet OE rocker arms probably have full hydrodynamic lubrication over nearly their entire motion range. Otherwise, they would probably wear more than typically observed. Most can be reused at engine rebuild time.

Duke

[Clem Z.]

Re: Roller Cams

the oldest engine i ever saw with roller lifters was 1937 H/D motorcycle race engine that i overhauled for a friend

[Michael H.]

Re: Roller Cams

There's little difference in friction between properly lubricated plain and roller bearings at moderate surface velocities when plain bearings are in hydrodynamic mode. Plain bearings have more static and very low speed friction ("boundary lubrication" mode), which is one reason why freight railcars went to roller bearings in the fifties. (Another is that they are a lot more reliable and durable than non-pressure lubricated journal bearings, especially as train speeds increased.) It's a lot easier to start a train with roller bearings.

At teardown or on a newly assembled engine the crankshaft should spin freely if given a flick from a counterweight. If there is significant drag there is likely a clearance or bearing alignment problem.

At the other end of the spectrum roller bearings have an advantage at very high surface velocities - above what a typical automotive engine sees.

Chevrolet OE rocker arms probably have full hydrodynamic lubrication over nearly their entire motion range. Otherwise, they would probably wear more than typically observed. Most can be reused at engine rebuild time.

Duke

I agree....

As I remember, my Harley had roller bearings throughout too. I suppose that's exactly why it was considered normal to have near zero oil pressure at hot idle.

[Jim L.]

Re: Roller Cams

the oldest engine i ever saw with roller lifters was 1937 H/D motorcycle race engine that i overhauled for a friend

Probably of similar vintage is this L A R G E stationary Diesel engine rusting in peace at Furnace Creek Ranch in Death Valley:

[Joe C.]

Re: Roller Cams

There's little difference in friction between properly lubricated plain and roller bearings at moderate surface velocities when plain bearings are in hydrodynamic mode. Plain bearings have more static and very low speed friction ("boundary lubrication" mode), which is one reason why freight railcars went to roller bearings in the fifties. (Another is that they are a lot more reliable and durable than non-pressure lubricated journal bearings, especially as train speeds increased.) It's a lot easier to start a train with roller bearings.

At teardown or on a newly assembled engine the crankshaft should spin freely if given a flick from a counterweight. If there is significant drag there is likely a clearance or bearing alignment problem.

At the other end of the spectrum roller bearings have an advantage at very high surface velocities - above what a typical automotive engine sees.

Honda motorcycle engines from the early sixties, like the 250 cc SOHC twins were all roller bearing with built-up cranks and have no oil pump of any kind. Oil was slung up to the cam and rocker boxes by the cam chain. It was a simple design. Later, as the engines got bigger with higher specific outputs, journal bearings with pressurized lubrication were adopted, and I think it began with the DOHC 450 twin.

Chevrolet OE rocker arms probably have full hydrodynamic lubrication over nearly their entire motion range. Otherwise, they would probably wear more than typically observed. Most can be reused at engine rebuild time.

Duke

There's little difference in friction between properly lubricated plain and roller bearings at moderate surface velocities when plain bearings are in hydrodynamic mode.
There is a difference, and the difference is not insignificant. Again, in addition to the reduced friction, a rollerized system would require much less lubrication than a pressurized hydrodynamic system, and thus experience less parasitic losses in order to drive the oil pump.

http://wardsautoworld.com/ar/auto_timken_bearings_boost/

Read it carefully, especially paragraph 16:
"Finally, Timken says an engine with reduced friction is able to produce more power. This is an aspect that had been exploited in motor sports, where Timken has close associations in NASCAR. But Deane says the rollerized engine was subsequently banned for having an unfair advantage."


At teardown or on a newly assembled engine the crankshaft should spin freely if given a flick from a counterweight. If there is significant drag there is likely a clearance or bearing alignment problem.
I assume, here, that you are talking about a crankshaft using roller bearings. The context in which the paragraph is written would allow one to interpret the statement as applying to either roller, or sleeve bearings.

[Michael H.]

Re: Roller Cams

There is a difference, and the difference is not insignificant. Again, a rollerized system would require much less lubrication than a pressurized hydrodynamic system, and thus, experience less parasitic losses in order to drive the oil pump.

http://wardsautoworld.com/ar/auto_timken_bearings_boost/

"At teardown or on a newly assembled engine the crankshaft should spin freely if given a flick from a counterweight. If th............................"
I assume, here, that you are talking about a crankshaft using roller bearings. The context in which the paragraph is written would allow one to interpret the statement as applying to either roller, or sleeve bearings.

Hey, wait a minute! That was MY story at the beginning of this thread. So now you agree with me?