Opinion on rocker arm geometry

[panic]

Yet to be addressed: which result is given the highest priority?
1. lowest guide and stem wear
2. maximum area under the valve open area curve
3. most power

1. is not always compatible with 2. or 3.
1. and 2. definitely not always the same, depends on the cam, vacuum, head flow, etc.
2. is produced by mid lift

[My427stang]

Thanks to all of you guys, Steelcomp especially for drawing the arc pic. I agree, its more than a little sketch, thanks for the hard work

Panic's input certainly hits the nail on the head. I know to, and I do, build my stuff to meet a use, or desired behavior vice the other way around.

But, and its a big but LOL, in this case

Call me thick, but I am having a hard time grasping the side loads could be significant with a .100 of roller tip travel compared to the same setup with a stock non-roller tip that I used to run on this FE.

Dont be afraid to cut to the chase, my feelings wont be hurt LOL

What is the "I told you so" thats going to happen if I dont get roller travel minimized and/or centered?

BTW if anyone cares, this is the little hot rod the 427 (489 cid) is going into

[MadBill]

I'm aware of at least half a dozen methods for setting up rocker geometry, some of which address panic's excellent points, but except when trying to make the most out of a rules-mandated cam, or with less than 0.500" lift, I think scrub and the resulting potential guide wear are more important.
The area under the curve and max lift are pretty well infinitely adjustable via rocker ratio and lobe profile, but I suspect that even the freest moving roller still exerts significant side load on the stem, so guide wear and power lost to friction result from excessive scrub.

[steelcomp]

After recieving some critical PM's I guess I need to clarify, which I thought was quite obvious, that my drawing is in no way to any kind of scale, merely trying to illustrate the relationship of the rocker tip to the valve tip during it's movement. I could draw something more to scale, but it would show the same results.
Having said that, is there another geometry scenario that could be more favorable to the one I've illustrated? Mind you, this drawing isn't adressing the PR end of the rocker, just the valve end. I see this scenario as transfering maximum vertical movement to the valve for the max. available lift, and also imposing the least amount of sideloading on the valve stem, regardless of the amount of lift. Over-arcing or under-arcing the rocker seems to be ineffecient in any case, unless I haven't learned something yet.
Thanks guys.

[STK 758]

I made a mistake on my sons High School ride by not checking pushrod length and aligning the rockers with guides on the old 23 degree chevy heads. We went from a flat tappet hyd to a Retro roller hyd @ .525 lift. The first thing I noticed was oil consumption and smoke at start. The second was the kid saying it was down on power. We killed the guides and the tips of the valves , not to mention the roller rocker damage. We were lucky and caught it in time. I did find a huge differance in stock vs aftermarket guide plates. Street driven @ 11.89 and 113 mph in a 69 Camoro 355 with 9.00cr and NO...NOS !

[My427stang]

STK, how far was it off when you went back and checked?

[STK 758]

It has been about a year but if I remember correctly the pushrods were about .200 off or more . The differance was the taller roller hyd lifter with stock length pusrods. After installing the correct pushrods I had to massage the bottom of the rockers to clear the moly tappered studs. I was surprised it did not break a valve tip off after looking at the damage. I missed it and got lucky. The roller tip was on the very edge of the valve and not on center as the illistration shows above.

Barry

[MadBill]

I think you drawing was clear and correct, steelcomp. The only slight quibble I can think of is that the spring load is increasing as the rocker moves through its arc, so one might look at spring force and acceleration loads throughout the valve event at the most relevent RPM range for your application and conclude that scrub at 'X' % of lift is more harmful that anywhere else, and arrange to minimize it there to the detriment of the overall sweep window.
Example: If you decided that scrub during the last 30% of lift was most harmful, you might set your rocker geometry so that the roller to pivot centerline was at right angles to the valve stem at 85% lift, thus minimizing sweep in the 70 to 100% range.

[steelcomp]

Thanks Bill. I see what you're saying about the valve spring pressure, but not about scrub. (I'm not trying to be argumentataive, here...just want to learn)
With the roler tip, it seems that the friction incurred due to spring pressure would be more at the axle of the roller, than the contact of the valve tip, since there is virtually no friction there. That being the case, the rocker wouldn't know anything about it's orientation to the valve regarding increased friction at the axle. Also, it's obvious by the illustration that there is far more lateral movement than vertical movement at the first 20%to 30% of rocker movoment. Clearly, this is exaggerated by the lack of scale on the drawing, but still exists. By under-arcing the rocker as you are suggesting, (and clearly stated that you were taking this into account) wouldn't you be losing valuable valve acceleration by effectively increasing that initial lateral movement, and decreasing the vertical? It seems there are valuable crank rotating degrees lost there that are supposed to be moving the cam, and opening the valve. Basically changing the opening profile of the cam? This is all theoretical to me, so if what you're saying is this is worth the tradeoff, and already proven, please forgive my lack of experience. This is the kind of stuff I really like to understand, as I believe it makes a huge difference.

[MadBill]

I think the difference in our visions relates to the effectiveness of a roller, steelcase. You are considering it to be essentially frictionless and I am saying it likely has significant friction. Let's consider a thought experiment (and take some short cuts with the physics):
1. Assume a fairly stout valve spring, lets say 500 lb. open force.
2. We know that at valve float RPM, valve gear inertia forces equal spring force, so simplistically, let's say inertia forces also peak at 500 lb.
3. Therefore, the maximum load on the roller tip is 1,000 lb.
4. Now, visualize a bizarre-looking wheelbarrow with a very sturdy frame designed to carry 98% of its load directly over the wheel.
5. The wheel is ½" in diameter and is resting on a smooth, hard steel surface.
6. Load 1,000 lb. into the barrow, lift it up (only 20 lb. at the handles) and start walking.
7. Does it take zero effort to move?
8. If instead it takes 20, 30, 100 lb. (considering that besides whatever friction exists between the needle rollers, the axle and the ID of the wheel, its tiny line of contact is deflecting the steel plate like a heavy truck on warm asphalt), how will this side load affect the wear of a perhaps marginally-lubricated valve stem moving 0.750" up and down 4,000 times a minute?

Regarding the under-arcing: True, it [/i]might slow the initial movement, but maybe not. The pushrod end of the arm is simultaneously arcing the other way, possibly* equally shortening its effective lever length. In any case, the overall movement will catch up as the lift event proceeds. In fact, the scenario in my previous post may* give greater lift, as the valve end of the arm will be closer to right angles with the stem at the max lift point.

*I think one of the reasons there is little agreement on how to optimize rocker geometry is that all the points in the preceding paragraph (and more!) are also affected by rocker construction details such as the height of the pushrod seat and that of the contact point of the roller relative to the pivot. If nothing else it certainly provides lots of opportunity for discussion!

[steelcomp]

All very good points, and goes to show how far in depth these things can be looked at. I also see that you're addressing more, the guide to stem friction than the roller tip itself, which I can understand, and won't disagree with.

Regarding the under-arcing: True, it [/i]might slow the initial movement, but maybe not. The pushrod end of the arm is simultaneously arcing the other way, possibly* equally shortening its effective lever length.

With a fixed, stud type rocker, and given the rocker's geometry is correct, if the vlave side is over arc'd, then shouldn't the P/R side be equally under-arc'd? Both suffering from initial excessive lateral movement? Also, regardless of the angle of the rocker, isn't a lever a lever, and the ratio stays the same, no matter what? I'm not clear on how it could change it's effective lever.

In any case, the overall movement will catch up as the lift event proceeds. In fact, the scenario in my previous post may* give greater lift, as the valve end of the arm will be closer to right angles with the stem at the max lift point.

The arm may be closer at right angles at max lift, but lost effective lift at the beginning. As far as I can see, the only scenario for max lift, is to keep the overall movement of the rocker as perpendicular to the valve as possible.

*I think one of the reasons there is little agreement on how to optimize rocker geometry is that all the points in the preceding paragraph (and more!) are also affected by rocker construction details such as the height of the pushrod seat and that of the contact point of the roller relative to the pivot. If nothing else it certainly provides lots of opportunity for discussion!

This, if for no other reason, is a good reason to go with shaft rockers, as the geometry can be changed with the movenent of the adjuster. (ie: the height of the p/r seat) I wonder if it would be worth while for a guy, just as an experiment, to pop the pressed in sockets out of a stud type rocker, add install adjusters. Just to be different. You could experimewnt with these ideas without having to shim stands every time you wanted to change something. I can see where under arcing the rocker may be advantageous, as you've explained, but I'd guess that the p/r end still should remain in the same mid-lift scenario, being perpendicular to it's lever at mid-lift. AFA I can see, this is the only way to accurately transmit cam lobe information to the rocker. Having the p/r socket adjustable could allow for such variances.
I gotta try some of this stuff.

[My427stang]

OK guys, I shimmed up .030. I think its about as far as I care to go until it starts pushing the witness mark too far out for my liking.

Tell me what you think here. (Second pic is what you saw before)

Please confirm that you like picture 1 better than picture 2. (Less centered, but less scrub, or wheel travel). The second benefit of picture 1 is that with the shims, I got some of my adjustment back too, and wont have to buy pushrods.

Picture 1 (Crane Sportsmans with .025 oil shields and .030 shim stock under stands)




Picture 2 (Crane Sportsmans with .025 oil shields only)



Thanks again, I do appreciate it greatly

[bill jones]

-when you do the roller tip check can you show us how far the roller moves out at each .100" of valve lift.
---------------------------------
-If the majority of the roll happens at low lift and then really slows down about .350"----say rolls out .055" by .350" valve lift---and then only rolls out .015" in the last .150" of lift I'd be pretty happy even tho the pattern isn't directly in the center of the valve.
----------------------------------
-I'd suggest that you check the geometry at all four corners of the engine and make sure you see the same trends---so that you aren't dealing with misdrilled shaft bolt hole C/L's compared to the valve guide C/L's
----------------------------------------------------------------------
-It looks like you need to move the entire shaft assembly .050" towards the intake.
-That could be accomplished fairly easy if you removed the heads and could set the heads up on a mill and offset mill the shaft bolt holes towards the intake---and then install helicoils---as the helicoils would allow just about that much offset.
-----------------------------------------------------------------
-Is there ANY slop in the shaft bolts that allow you to push or pull the shaft assemblies slightly towards the intake or towards the exhaust?
----------------------------------------------------------
-Have you thought about slotting the shaft holes and elongating the stand holes to get that .050" offset?

[My427stang]

-when you do the roller tip check can you show us how far the roller moves out at each .100" of valve lift.
---------------------------------
-If the majority of the roll happens at low lift and then really slows down about .350"----say rolls out .055" by .350" valve lift---and then only rolls out .015" in the last .150" of lift I'd be pretty happy even tho the pattern isn't directly in the center of the valve.
----------------------------------

I'll map it out tomorrow Bill, good thinking


-I'd suggest that you check the geometry at all four corners of the engine and make sure you see the same trends---so that you aren't dealing with misdrilled shaft bolt hole C/L's compared to the valve guide C/L's


I'll do that as well, sorta caught me, I did one head all the way down the line, but good call, sorta silly if I am chasing a misdrilled head

----------------------------------------------------------------------
-It looks like you need to move the entire shaft assembly .050" towards the intake.
-That could be accomplished fairly easy if you removed the heads and could set the heads up on a mill and offset mill the shaft bolt holes towards the intake---and then install helicoils---as the helicoils would allow just about that much offset.
-----------------------------------------------------------------
-Is there ANY slop in the shaft bolts that allow you to push or pull the shaft assemblies slightly towards the intake or towards the exhaust?
----------------------------------------------------------
-Have you thought about slotting the shaft holes and elongating the stand holes to get that .050" offset?

I'll look close, there isnt any slop, but I'll look and see if its the shaft or stand holding it tight. My guess is I cant go much, but I might be able to pull .020 or so. Beyond that I'll have to open up all the pushrod holes, because the backside of the rocker is longer too, so it eats up some room

Thanks again, I'll dig deeper and post back tomorrow. I am certainly getting closer. I do appreciate your help

[STK 758]

www.mid-lift.com

This link has helped me.