Half Journal Bearing?

[RodNewstrom]

Hello community ! I'm a new member, so I'm unfamiliar with the features here. I cannot, for example, include a .JPG with my question, so I have to ask in text only mode.

I have a cam-driven engine design in which the pistons incorporate a cam follower that engages with a cam track controlling piston motion. You can see a Solidworks animation of the current design at https://www.youtube.com/watch?v=ZK4Ndbdt2kk . The cam followers are currently sliding contact only and are under substantial peak load of 1,376 lbf over a contact angle of += 40 degrees from normal. Hertzian contact analysis indicates the material (Vibenite 150, 435 kpsi yield, HRC 63) is adequate for this load with sufficient margin of safety.

I'd like to convert the sliding contacts of the cam followers to rolling contacts. The lower 80 degrees (+- 40 off normal) of the roller must be bare as it is in contact with the cam track. I was watching the video at https://youtu.be/n_vc-BGNZsE?t=48 (48 seconds in), and saw what might be half journal bearings in a similar application. That got me thinking...

Is there such a thing as a "half journal bearing?" wherein the loads are constrained to +- 40 degrees off normal and only the upper half of the shaft is encompassed by the bearing? Intuitively, it seems possible... feed oil on the ****** side, lose some out the shaft ends (length/diameter = 2), create a high pressure film where the gap narrows due to load, then loose some some oil on the output side (all oil exiting the bearing is collected off the track and returned to the oil loop). The minimum engine RPM at engine startup is 200, the max is 2626, and the ratio of path length to shaft diameter is 11.5 thus yielding shaft RPM between 2,307 (idle) and 30,290 (max). Oil pressure is low as it's generated by centripetal force of the spinning rotor, but flow is high.

What do people think? Is this possible or am I nuts?

[Hudson]

There are half journal bearings in engines. They are confined to two stroke rods as far as I know. See the early Napier Nomad that had two rods on the same crankpin held together with a pair of thin rings. This was last tried on the Continental experimental diesel of the 1980's that had a couple of whirlly gig balancers on each end of the crank. (Making a flat 4 cylinder engine as long as a six). McCulloch drone engines just had a steel strap at the back of the rod to keep the rod on the crankpin in case of a miss fire.

Roller cam assemblies can get unhappy if the contact angle reaches about 30 degrees. (You mention 40). Things can jam and lock up.

What do I think? I think it looks like a frictional nightmare.

[RodNewstrom]

Roller cam assemblies can get unhappy if the contact angle reaches about 30 degrees. (You mention 40). Things can jam and lock up.

I got sloppy when posting my question. The max contact angle is 38 degrees when the piston is driven by expansion of combustion gases and 28 degrees when it's driven by the cam during compression. The engine is two stroke, so the piston of interest is always loaded at the piston end except during scavenge.

I think it looks like a frictional nightmare.

Do you mean the engine as a whole or just the cam, and does the clarification of contact angle affect your answer?

[RodNewstrom]

Below left is a screen grab of the Black Bruin S-Series Hydraulic Motor shown at https://www.youtube.com/watch?v=ru78okZN5CI . Below right is my proposed design modification. I have superimposed identical geometry on both and it appears the Black Bruin motor is using similar contact angle while operating at over twice the pressure (450 bar) as my engine ( see ** Link Removed ** ). I have identified other radial piston hydraulic motors from Bosch and Rotary Power with similar designs that claim 95% mechanical efficiency. Given this, I guess my question should be whether the follower used in these motors is a simple sliding follower or a roller in a half journal bearing. I assumed it was a roller in a half journal bearing because otherwise, I would expect them to have just machined the end of the piston to form the follower rather than add the complexity of a caged shaft.

Capture.GIF

[Hudson]

38 degrees is greater than the usual 30 pressure angle threshold so expect problems with the piston tilting and jamming in the rectangular slot.


Sealing is going to be big problem. It is not clear how you intend to cool the thing or get spark to the cylinder.

Consider the hydraulic motor. When the pistons leak, it is just oiling its self. The oil carries away heat and it doesn’t need ignition.

There’s pumping losses with a separate scavenge piston not to mention the complexity of the arrangement. You have three pistons moving to obtain the output of one conventional piston.


HCCI operation is not easy to obtain or control. Do you expect the engine to start in that mode?

A steel with a 450 KSI yield is hard to believe.

[RodNewstrom]

38 degrees is greater than the usual 30 pressure angle threshold so expect problems with the piston tilting and jamming in the rectangular slot.

There's a difference between trying to drive a piston up with a 38 degree contact angle versus the piston being driven against the cam at a 38 degree contact angle. If you draw a sketch of the arrangement and think through it, you'll see what I mean. The whole thing has been through Finite Element Analysis and works fine. Note the hydraulic motor is using a 45 degree angle both when the piston is driven by the cam and when the cam is driven by the piston.

It is not clear how you intend to cool the thing or get spark to the cylinder... There’s pumping losses with a separate scavenge piston not to mention the complexity of the arrangement. You have three pistons moving to obtain the output of one conventional piston... HCCI operation is not easy to obtain or control. Do you expect the engine to start in that mode?

There is no spark in HCCI. It's cooled using oil propelled under centripetal force through channels around the liners. Pumping loss is low compared to a traditional two stroke because the charge is never pressurized, just moved through large ports after both the intake and exhaust ports are already open with backpressure no greater than 0.065 bar. I have one piston serving only as the intake valve, and the opposed pistons yield the smallest possible surface area for a cylinder. You would need to read the attached paper to understand how HCCI is controlled (and yes, I do plan to start in that mode even after cold soak).

A steel with a 450 KSI yield is hard to believe.

I can't share proprietary details, but it's no ordinary steel. The US Air Force is evaluating the materials for military applications (see http://www.metal-am.com/vbn-componen...benite-alloys/ ).

[Hudson]

You will have obtained complete mastery of HCCI if you can start a cold engine in that mode.

Can you point to one HCCI engine that does not need a spark plug to start and warm up before HCCI is possible?

[RodNewstrom]

You will have obtained complete mastery of HCCI if you can start a cold engine in that mode.

I don't know about "complete mastery" but I do believe I have a higher probability of success than others, and my opinion is backed by peer review by a professional engine designer with experience on EcoBoost and Skyactiv as well as a college professor with extensive experience in modeling and simulation of HCCI. Both continue to advise me in the effort.

My cold start analysis assumes use of oil/block/fuel heaters for temperatures below 0F just as many diesels. Air temperature, however, is assumed to be -81F. Compression ratio is 36:1 at cold start, but peak compression temp is borderline at 796K due to heat loss, and the engine has to run at idle until normal oil temperature of 250F is attained before revving much beyond the minimum 200 RPM (remember, each cylinder set of the engine completes four strokes per revolution, so the pistons are moving at speeds on par with 800 RPM in a crank driven two-stroke).

Can you point to one HCCI engine that does not need a spark plug to start and warm up before HCCI is possible?

I have been at this for 6 years now, and have read literally every paper I can get my hands on including every report from the HCCI Consortium of manufacturers, academics, and US national labs that worked the problem hard for over a decade under DOE funding. I took everything I learned from my research then asked myself "what kind of engine does HCCI want?" and let the answer dictate the design.

I cannot show an engine that cold starts with HCCI. Heck, I can't show you a production engine that runs HCCI full time without spark assist. You won't find one either. Neither, however, will you find an HCCI radial having highly symmetric and uniform cylinder temperature, cooling, inlet, and exhaust structures. Nor will you find one that employs cams to drive pistons through compression in less time than the ignition delay of the fuel at low RPM and then pauses at peak compression for chemical kinetics to complete at high RPM. You will not find one capable of variable compression ratio from 29:1 (sea level standard day) to 36:1 (cold start and operation at altitude) either; it's simply not possible in a realistic crankshaft driven design.

Only a fool would think they can master HCCI using a traditional engine with in-line cylinders of three to five inch bore arranged in one bank or two. Probably a hundred million dollars have been spent trying. If HCCI is to be accomplished, we have to rethink the engine to specifically address what HCCI wants. That's why my engine is so unusual.

Will I succeed? It's yet to be seen, but experts who have peer reviewed the design are hopeful, and so am I.

P.S. If you haven't already, do read the paper I attached to my last comment. It explains all the design decisions.

[RodNewstrom]

38 degrees is greater than the usual 30 pressure angle threshold so expect problems with the piston tilting and jamming in the rectangular slot.

Below is the calculation for max pressure angle before locking the follower in its guide from Rothbart's Cam Design Handbook as well as SKF's range of friction coefficients for a plain bearing and a screen grab of my design with and without the liner installed.
PressAngle1.JPG

As you can see, my pistons are constrained inside the cylinder bore. The length from the top of the piston to the end of the skirt (B in the equation) is 0.426 inches. Note the pistons and liners are steel and the piston-cylinder gap is 0.005 in. The distance from the center of the roller to the point where the skirt contacts the liner (A in the equation) is 0.146 inches. Below are my calculation results.
PressAngle2.JPG

Using the worst case friction coefficient from SKF's catalog, the maximum pressure angle is 67.1 degrees. Conversely, for my current 45 degree pressure angle, the maximum friction coefficient is 1. Assuming I got the calculations right, my pressure angles shouldn't result in the mechanism jamming.

[RodNewstrom]

Oops! The prior post should read:

"The max pressure angle is 67.1 degrees. Conversely, for my current 45 degree pressure angle, the maximum friction coefficient is 0.59."

PressAngle2.JPG

[Hudson]

Rothbart's is probably the best book on cam design. On page 64 of the original 1956 edition he wrote, “Generally the safe limiting pressure angle in practice is 30 degrees.” You are of course free to ignore this. Just as you are free to ignore the opinion of someone that has designed several I.C. engines starting with new centerlines and actually ran a two stroke in HCCI mode on a dyno about 30 years ago. It is a little spooky when you turn off the ignition and the engine keeps running without missing a beat.

[RodNewstrom]

[Y]ou are free to ignore the opinion of someone...

I don't ignore any input. I evaluate every comment and make a decision whether it applies to my design or not. When I discount a comment, I explain why and provide the rationale to give the commenter the opportunity to make a counter argument. My objective is a good design. not winning a debate. I used to run a Systems Engineering department with over 100 high level engineers, and I've done more design reviews than I can count over the past 30 years. They are critically important, but require all parties set aside their ego to be successful.

"Rules of thumb" are great, but they can be misleading if one doesn't understand the context. When folks say "keep pressure angle under 30 degrees" what they're really saying IMHO is "if it's under 30 degrees you're probably OK but if you want to go further, you'll need to do some detailed analysis." On page 19 of the 2004 edition of the Cam Design Handbook, it says "However, if the follower bearings are strong, the cam-follower is rigid, and the cam-follower overhang is small, the maximum pressure angle may be increased to more than 30 degrees." The equation that gives the limits is equation 6.8 of the same book and is followed by the statement "Note that these values and the derivation of Eq. (6.8) are based on the ideal assumption that the follower is perfectly rigid. Thus, the coefficient of friction may actually reach a value of 0.25 or more depending on relative elasticity and backlash of the follower. A flexible stem may dig into the lower corner of the bearing. Therefore, the suggested guide in practice is to keep the coefficient of friction m, the follower overhang A, and the backlash as small as possible with the bearing length B as large as possible, in the range of B = 2A."

My overhang is very small, B = 2.9A, backlash is only 0.005 in, and all cam components including the piston body, skirt, and cylinder liner are made of very strong and hard Maraging 350 steel (or perhaps Vibenite 150).

It is a little spooky when you turn off the ignition and the engine keeps running without missing a beat.

I bet! I would have enjoyed being there to see the demo. I love engines!

[valeriedixon1970]

Journal or sleeve bearings make use of a pressure wedge of fluid that forms between the rotating shaft and the bearing. The portion of the shaft supported by the bearing is called the journal and is usually hardened for wear-resistance.

[Cragyon]

Journal or sleeve bearings make use of a pressure wedge of fluid that forms between the rotating shaft and the bearing. The portion of the shaft supported by the bearing is called the journal and is usually hardened for wear-resistance.

Hydrostatic or Bearing Lubrication.

https://www.engineersedge.com/calcul...tion_15314.htm