Inliners International Forums General Discussion Hi Performance Geometry isn't fun

And I'm not talking about that evil class in high school. I was discussing my setup with Jeff Nelson and he had concerns about rocker geometry. Realizing I didn't know altogether that much about it, I started researching it on the internet. Holy Cow! Every possible conceivable configuration is touted by someone to make the best power, cause the least wear, break the least parts, give you back rubs, and end world hunger! How do you guys set your valve train for your preferred geometry?

The mid point of lift should be at the middle of angle between the valve and the push rod. That would put the least stress on the components. Roller tips on the rocker arms would present the least wear at the valve stem.

This is a really complicated topic.

1st: "mid-lift" is where the valve stem and a line through the rocker pad and rocker pivot (shaft or ball) intersect at 90° @ 50% lift. This produces the largest valve curtain area vs. time combination of all variants, but this doesn't always produce more power. It also has the lowest side-thrust vector, but that doesn't mean the least wear, since the highest spring pressure occurs at the worst angle. It's only designed into some roller rocker arms, since Miller holds the patent. The other rollers cannot be shimmed, spaced etc. to get mid-lift - it's not an adjustment.

2nd: mid-lift is not built into any "pallet" (non-roller) rocker arm (such as almost all OEM). You can modify things, raise parts and adjust the valve side lever to get this angle (but only on one side), and it produces a substantial error on the pushrod side.
The original rockers are most frequently designed to place the pivot below the stem top at 1/2 lift, which reduces the scrubbing force for best wear.

What do you have, and what's the problem?

I am researching before I order parts. Rocker arms and pushrods are the only things I need before I assemble. My plan was to run Harland Sharp 1.8 big block chevy rockers, but not if I'm going to have to make a bunch of compromises that would hurt power or cause incredible wear or broken parts. I am waiting to get pushrods until after the engine is assembled. I just don't know what to set the rocker to for measuring the required pushrod length. Just trying to know something going in.

Comp Cams and Isky as well as other cam companies have detailed info on their websites for checking and determining the best rocker geometry. Just browse them, and make sure you get a checking pushrod to get the right pushrod length before you order the ones you plan on using. To explain all of this is very detailed and complex, but actually doing it is very easy and you will have no trouble following their guidance in doing it. Ideally, if you placed and centered a nickel on top of a quater, and the quarter represented the valve tip, the nickel would be the area you would want the rocker tip to stay in as it travels thru its start to finish motion. Making sure it doesn't stray to close to the edge of the quarter, or even off of it. Good luck, keep us posted.

CNC-Dude, that's just the thing. Comp wants the roller to start on the inside of the valve, and end up at the outside. Several others want it centered on the valve stem. Jim Miller just wants a narrow pattern. Titan wants it to roll from outside in! And that's just the valve side. Most of them don't even discuss the pushrod side. I'm not worried about the actual procedure, I just don't know what to shoot for.

First you need to see where you are to start with using the 1.8 rockers and a stock length pushrod. Also, just watch the contact pattern on the valve tip as you are rotating the engine to see if you are even in the ballpark with the stock pushrod. Make a mental note of where the rocker tip is at the start, midlift and full lift points on the valve tip making sure it stays on the valve tip. Changing to a bigger ratio rocker will likely make the engine need a longer pushrod, if everything was correct to begin with before the change. If you have a checking pushrod also, you can shorten or lengthen it to see how it changes the rocker tip contact area on the valve tip. The optimum is as has been mentioned, where the rocker tip is in a close proximity of the valve tip center at mid lift. But first, you will have to see where the rocker wants to begin, either at the inside or outside of the valve tip and go from there.

Those aren't different instructions. Those are different built-in rocker geometry (the angles between the ball center and each arm). Miller says "smallest path" because he has the mid-lift patent, which produces the smallest possible scrub.
Normally, an engine with restricted breathing and shrouded valves (chamber wall is too close) uses mid-lift unless you have some reason not to. If HS does not incorporate this, you will only be able to get mid-lift on the valve side (pushrod geo is less important). For mid-lift adjustment and pushrod measurement, the rocker tip begins closest to the rocker stud with valve closed, proceeds away from the stud and is ideally centered on the stem (not critical) at 25% lift, continues away and reaches 50% at the farthest point away (rocker is now 90° angle to the stem) and reverses. On the return trip, everything repeats backwards: 75% lift at stem center, full lift at the same (closest) point to the stud.
A BBC rocker is about 1.65" long (valve side), the scrub path at mid-lift for a .600" cam is about .027". If the geo is wrong, this distance will be both larger and asymmetrical.

Thanks for those numbers, panic. That also jives with what I got from Harland Sharp. The specs they gave me are 1.59 for an inline six rocker and 1.649 for a BBC. I think I'm going to order one of each and see which is closer to the center of the valve stem when set to be 90* at mid-lift. It turns out my cam has .606 lift with a 1.8 or .589 with a 1.75, so now I know what to shoot for pattern-wise. I'm kind of hoping that the long length of our pushrods makes the geometry on that side less critical. This probably isn't all that important for an engine that will only make 425ish on motor with boost. I just figure the more attention I pay to all the little things, the less nitrous I have to run later, or the less weight I have to pull out of the car. And, learning is its own reward.

which is closer to the center of the valve stem

But - that's not a function of geometry, only of the physical dimensions of the parts. Geo is far more important than having the rocker centered (which is entirely a safety and wear concern, not breathing or power), as long as it doesn't run close to the edge at any point.
The long pushrod reduces the side-thrust and wasted motion somewhat, but has no effect on the geometry (the triangle formed by the tappet cup, and inner and outer adjuster positions is too small to do anything). The geo on the short side is only less critical because there is no "scrub"; a geo error here (which you will get with a HS rocker adjusted to produce mid-lift on the long side) will reduce valve area in the same way.
What I have not figured out (and not ever) is:
1. how to determine what geo the manufacturer is shooting for (discarding their advertising)
2. what adjustment provides the best balance of curtain/area on both sides of the fulcrum.

I finally got my engine assembled to the point that I could check pushrod length. The first thing I learned is the stock ones are like 3/4" too short. With the rocker sitting as low as it will go on the stud, there's like a 1/4" gap, and I haven't lifted the rocker to set geometry or compressed the lifter at all yet. So tomorrow I build the checking pushrod. My intakes and exhaust valves sit at slightly different heights, so I'm going to make 2 checking pushrods and order different lengths for each, since it won't cost any more that way. I just went ahead and ordered the 1.8 BBC rockers. The pattern is going to end up on the outside third of the valve tip, but isn't close to running off anywhere. Panic, I've decided with respect to #1 above that they don't even know and for #2 - I've got nitrous, I can overcome.

Too bad there will be 2 different pushrod lengths. I set my heads up so that the same length valve (intake/exh) get the same installed height.
As far as the 1/4" gap, can I see a pic of the rocker stands, specificly the height of them? I wonder if they are not milled down far enough. Tom

Sorry, I misread your original description - these are stud-mounted, not shaft conversion, so the adjuster height on the stud is a variable (but not the same effect as rocker shaft or stand height) and pushrod length is definitely relevant.
Is the stud parallel to the valve stem axis, vertical, or other?
If it's not parallel to the stem, any height adjustment also moves the roller across the tip (tight = closer to the stud).
Your roller tip is already going to be about .059" too far across the stem tip just because of the BBC lever length vs. original (1.649 - 1.590 = .059), and any attempt to center this is just going to make an error somewhere else.
The short lever (pushrod side) has to be about .916" for 1.8:1 ratio, .054" closer to the stud than stock (the stock 1.7:1 lever is .970"). I'm not sure how much pushrod clearance you have.

Panic, I have a louis tool to lengthen the pushrod holes if required. The studs are pretty much parallel to the valves, so no help for roller tip placement.

Someone a while back had asked what the valve angle was on a 6. Well according to my $10 angle finder, I got 9 degrees-ish. Interestingly, the top angles away from the pushrod, not towards it, opposite of a SBC. This means the air has to turn past 90 to go through the valve. Think how much difference going from 23 to 18 makes on a SBC. No wonder the hybrid heads add so much power. I truly didn't realize how poor the six head design was until today. It's really a good argument for drug testing in the workplace IMHO.......

Tom, I feel the same way about the pushrods, but I've changed so many things from stock that stuff like this is inevitable. The cam base circle is different, deck heights have been changed, The head has been milled somewhat, I don't believe the valves are stock length anymore either. I would have ordered two extra pushrods anyhow, now they'll just be labeled "I" and "E" is all.

The valve angle is in 9-10 degree range, depending on how level the head was before it was resurfaced. Don't sell the siamese head short, there isn't many stock production heads that can be made to support over 600 HP naturally aspirated, except a few Big Block heads.

9-10 degrees valve angle,,,, that is why we have a mile of room piston to valve clearance.


MBHD

The amount of clearance as the valve opens (assuming no other variables - and there are big diffs between engines) is a function of valve angle.
The tightest clearance (fastest approach to the dome, needs the greatest absolute vertical safety clearance) is a vertical valve (such as GMC, Buick L8, BMC Mini) since the lift is the actual distance.
As the valve stem axis is tilted away from vertical, the distance is reduced, and is easily calculated by multiplying the lift by the cosine of the stem angle.
If the engine has a high stem angle, such as the Chrysler hemi at 26.5°, the distance is substantially reduced: .200" lift (a common figure for lift during overlap) × cosine 26.5 or .895 = .179", the actual distance the valve moves down toward the piston. An even more severe angle such as the H-D Sportster at 45° moves even less, only 70.1% of .200", or .141".
Obviously, a valve tilted sideways (such as "T" head Marmon, Stutz) does not approach the piston at all: the cosine of 90° is 0.
A 9° stem angle has a vertical component of 98.8% of the lift, or .198" for .200" lift.
However, the valve's motion out of the vertical plane (the difference between the actual lift and the cosine-derived vertical distance) isn't lost, it must still be provided with a clear path but it's across the dome rather than down.
The sideways scrub across the dome is the width of the clearance notch in addition to the actual valve head diameter (or whatever part of it needs more room). Since it also deepens as it approaches the far end, the relief notch shape is 3-dimensional and weird, but it's what you need if you have a big compression dome and must keep the chamber volume small for CR.
This calculation is even smaller for any angle less than 45°. The scrub for a 9° stem angle is only 15.6%, or .031" for .200" lift. Even the Chrysler only moves .089" for .200" lift.

The amount of clearance needed varies almost inversely proportionate to the stroke length: short strokes get close at very small lifts.
it's also affected by the rod ratio: a short rod moves away from TDC quickly in both directions, and needs slightly less clearance.

I am right there with 56er on a street 292 build, except I am using hydraulic lifters with Comps roller tip rockers, and assume 56er is using solid lifters? Not to hijack, but I am measuring for pushrods too. If the hydraulic lifters have been soaked in oil, will they stay "up" during engine rotation by hand and give me an accurate indication of valve stem contact, or will they collapse when load is applied? It looks like a 11.70 pushrod length gives me a great contact area centered in the valve stem tip. I have heard of taken the lifter apart and disabling it, thus making it solid for the purposes of these measurements- but then don't you have to add additional length (.020?) for pre-load of the hydraulics?; thats something I have no experience in. advice?
thanks, Tad

You will need to use checking springs on the valves for doing this, a hydraulic lifter will collapse if using a regular valve spring. You could also modify or use a solid lifter to do this with the regular valve spring.

x2 Heater 63.

I've been told 3 different ways as the 'proper' way to determine proper push rod length. I'd really appreciate finding the right way to do it right the 1st time instead of the trial and error method.

56er
I think someone (Panic) found mastery of geometry either fun or a necessity at some point!

I am using rhoads lifters, which you end up setting like solids. Heater, if I were you I would get an extra lifter the same brand as the ones you are running, take it apart and make it solid. you just grind a bolt or something to length and stick it inside so the plunger is held up against the snap ring. No knowledge about the extra length, but a half turn on a 7/16" stud is .025" and on a 3/8" stud is about .021", which is how much most people preload a hydraulic lifter, so your numbers pass the idiot check if that makes you feel any better.

My length ended up being 12.014". Also, the pushrods bound on the rocker fulcrum, so they all had to be clearanced. The pushrods hit the inside of the slot, which couldn't be lengthened with my louis tool due to logistics, so I had to drill out the pushrod slots to 1/2" and install my guideplates. However, the guideplates are for 3/8" so I had to take everything apart and clearance everything for 3/8" pushrods. It was just a tad frustrating.

X2 - it won't hold any spring pressure.
I suspect a hydraulic won't remain "inflated" even if the engine is cranking on the starter even if the oil pressure is good, because the time in seconds for the lifter to bleed down is really long compared to even idling.

That's the most accurate method, since the collapse distance of a hydraulic will differ slightly between manufacturers.
If you have to remove any metal from the rocker itself, obviously take as little as possible but blend the cut into the body along the edge in both directions to "fade out".
BTW: the H-S are extrusions to keep the price down, and could be profile milled to save some weight if you know where, especially on the roller side. Don't reduce the height, but the width can be sculptured a bit with little bas-reliefs.
I think of adjuster motion in terms of hex flats on the head or wrench, 7/16-20 NF is 8.3 thousandths per flat, 3/8-24 is 7.

CNC-dude,

I chuckle every time I see your tag line,

"There's no such thing as too much cam....only not enough engine!

That's because I ask myself, "6re6,How big of an engine do you need for a cam with an advertised duration of 720 / 720?"

I thought that was kinda' humorous myself. It does sorta' make you stop and think about it for a minute doesn't it.