Don Terrill’s Speed-Talk

I've been running through several iterations trying to obtain the ideal valvetrain geometry but am finding that there are physical limitations.
Using photographs I've taken of the rocker arm as it cycles through, I can adjust the pushrod length to bring the rocker height perfectly in line with the valve tip. That is, at mid lift, the rocker arm is exactly halfway through it's motion. For all intents and purposes this is the ideal with loss of information from the cam and minimal sweep across the valve tip. Of course I'm after the narrowest possible contact patch however at what point does a compromise have to be made and forsake perfect geometry for reasonable contact positionally speaking?
We all know the factors that affect pushrod length. Block decking, milled heads, gasket thickness, valve stem length (over stock), lifter seat height, rocker arm style, etc but some items end up fixed such as the distance from the rocker stud to valve centre line.
I've attached a few pictures of the valve tip witness marks.
I'm looking for opinions.
I'll also post up (later) the pics form the rocker arm cycle with the mid-lift method data points overlayed.
For the record the specs for the engine in question (although I'm involved with several at the moment):
Gen 1 SBC, RHS Pro Torker iron heads w/.100" taller valves, Comp full roller rocker (1.6), Comp 15853 (short travel) hydraulic roller lifter w/2.66" seat height, .570" lift.

Here are a few of the marked up drawings.
This first series is with the "stock" 7.3" pushrod for a retro-roller. This worked well in the uncut block, stock Vortec headed retro-roller 353 with Comps 853 (2.49" seat height) lifters.
The red lines represent the valve vertical centre line (x axis) and a line drawn at 90 degrees from the rocker pivot (y axis). The white line represents the data points which are the rocker centre line and the centre of the roller tip.
Ideally the white line should overlap the red line at mid lift.

Is the pattern offset toward the intake manifold or the exhaust manifold?

Here's another with a 7.45" pushrod. The contact is narrow in comparison to the 7.3" pushrod.
The geometry is excellent with mid lift angularity about as good as you can get it.
It's the concern I have with the contact off centre the way it is that is the question.

[quote="Schurkey"]Is the pattern offset toward the intake manifold or the exhaust manifold?[/quote]

Sorry about that.
All the contact pictures show the exhaust side as the bottom of the picture. That is, the contact mark is offset towards the exhaust.

Not sure if this is right but its how I do it and is pretty simple. I put a dial indicator on the retainer, ( I machined a retainer with a flat top) Make sure the indicator is in the same axis as the valve stem. Then check max lift. Adjust pushrod length to get the highest max lift.
The mid lift method never worked out for me... as I came up with a wear pattern like you did wayyy off to one side of the tip.

A lot of it depends on what rocker arms you are using and how they were designed. I have used the actual "mid-lift" rockers and I have used others. In every case it seems that you get the best geometry when you minimize not only rocker tip sweep, but pushrod sweep as well. If you are moving the pushrod all over the place it can induce harmonics in the valvetrain you don't need. There are times (almost always) when it's the best compromise of valve sweep, pushrod sweep, and contact pattern that you end up with. You can really spend a lot of time trying to find the best compromise as well.

It isn't that important to have the pattern centered on the valve, if you are pushing down on the valve straight, you can be off-center and still not loading the guides.

[quote="Brian W"]Not sure if this is right but its how I do it and is pretty simple. I put a dial indicator on the retainer, ( I machined a retainer with a flat top) Make sure the indicator is in the same axis as the valve stem. Then check max lift. Adjust pushrod length to get the highest max lift.
The mid lift method never worked out for me... as I came up with a wear pattern like you did wayyy off to one side of the tip.[/quote]

Never heard about that method.
I can see the results regarding lift when adjusting pushrods however as I understand it, it's the relationship between the motion of the rocker and it's positon relative to the cam that's important. Getting these angles spot on gives the narrowest contact possible.
Where it falls on the valve tip seems another matter altogether. The former being adjustable while the latter, is not. Unless of course you forego the angular relationship and settle on a centred pattern [u]regardless[/u] of it's width. I tried the last approach and it's valve guide death. Quickly too.

[quote="900HP"] There are times (almost always) when it's the best compromise of valve sweep, pushrod sweep, and contact pattern that you end up with. You can really spend a lot of time trying to find the best compromise as well..[/quote]

Isn't that the truth.



[quote="900HP"]
It isn't that important to have the pattern centered on the valve, if you are pushing down on the valve straight, you can be off-center and still not loading the guides.[/quote]

That's the thing. I agree with you 100% on the smallest sweep resulting in the least force in directions other than straight down. What I'd like to see is some engineering data to support that. My testing and subsequent failure seem to support the narrow contact over centred at any rate.

Here are four contact patterns.
From 7.3" to 7.45". What I'm seeing here is that the distance to the exhaust side edge of the valve seems to be consistent (more or less) from pushrod to pushrod. What changes is the width (as is expected) and it appears to "grow" towards the intake side of the valve.
In other words, as the pushrod becomes longer, the sweep becomes less and doesn't apppear to get any closer to the edge.

As a contrast, here's what NOT to do.
I chose to keep the pattern centred and let the width fall where it may. Nicely centred pattern but very wide.

Excuse the poor transfer of the pattern. That was performed using checking springs however the location, if not the intensity, is very much the same.

Valve tip pattern is not the "be all end all" of valve rocker geometry; especially with a roller tip involved. Going solely from the wipe pattern can even result in detrimental harmonics or excessive wear in other places at times.

Rocker arm geometry being near correct WILL be the best compromise of all the vector angles and components involved within the valve train.

[quote="Walter R. Malik"] Going solely from the wipe pattern can even result in detrimental harmonics or excessive wear in other places at times.[/quote]

Which brings us to the title of this thread. I thought marking up pictures would allow for a better look at the angles involved although that could have been done by measuring. With the engine in the car, it's easier to get a camera in there than my head.

[quote="Walter R. Malik"]
Rocker arm geometry being near correct WILL be the best compromise of all the vector angles and components involved within the valve train.[/quote]

I can see that's going to be the case.

Now the questions are:
Which of the above patterns looks to be the best?
Is pushrod selection based on which length nails down the mid-lift qualities best resulting in the narrowest sweep?

As long as say 70% or more of the roller is in contact with the valve at mid-sweep, I always go for minimum sweep. If the contact zone is badly off, I look for a different rocker that positions the roller closer to center. As far as focusing on max lift, I figure I can get all the lift I want and more via lobe height and rocker ratio selection.

BTW, I recall a huge thread on this ~ 3 years back...