I just finished my E-Tec swap. Edelbrock says .100" longer than stock pushrod for hydraulic roller cams and they give a part number which is a 7.300" pushrod. Well of course I get those ones without actually checking the length. The rockers look like they are sitting in the open position when the valves are all the way closed. I have been having a float/lifter pump up problem too. When I took off a rocker arm to check the valve-tip wear pattern I noticed that it was a VERY thin stripe at the edge of the stem(towards the header side). I then took out one pushrod and colored the stem black with a permanent marker and re-installed the rocker along with a STOCK OE pushrod for roller cams(7.195"). Not only did the rocker look properly positioned, when I removed it the wear pattern was near perfect with no preload just at 0 lash. It was a thicker stripe that started a hair on the intake side and rolled over to the exhaust side at full lift. I have 1.6 ratio Comp Pro Magnum rockers. I checked the geometry with NO push rod and I notcied that; the shorter the pushrod length, the closer the wear pattern is to the intake side.

Should I just buy a good set of OE length roller pushrods(7.200") or should I try 7.150?

Also, is it common to need shorter pushrods when switching to 1.6 ratio rockers?

 

Here is what I mean

 

Well I did a search and found that RB83L69 said the wear pattern that I have is right, but the comp cams book that I have right here in my hands says the other way is right. It says that the wear pattern should be right across the middle.

I also found a guy with a VERY similar set up to mine and he also found that stock pushrods were the right length.

 

I've always heard center of valve stem as well

 

The stock ones look like the right pattern,I might say try the 7.150 ones but the stock ones look like they will work fine

 

There are varying theories on the valve train geometry question.
Some say center the rocker on the tip of the valve and others say it's right when the wear pattern is narrowest.
As in your own case, the narrowest wear pattern can be off center.
The whole concept of the narrower wear pattern approach is that the the rocker tip travels the least distance across the surface of the valve tip when it's perpendicular.
By perpendicular I mean, the rocker fulcrum being on a line that's perpendicular to the valve stem, at the level of the valve tip, when the rocker is at mid lift.
That is when the lever (rocker arm) is most efficient.
You can prove it to yourself with a dial indicator. Mount a dial indicator so it will measure your valve lift. Use a solid lifter, and an adjustable pushrod, and measure how much more lift there is at the valve using the "perpendicular/mid lift" system than the "centered on the valve tip" system.
I tried it, and I'm convinced. I think I got .012" greater lift with the mid lift system.

edit:
The guy who explained the mid lift method to me, told me that the method described in Comp Cams' catalog is WRONG

 

Forget anything you've heard about having the pattern centered on the tip of the valvestem, for it has absolutely nothing to do with having proper valvetrain geometry. It amazes me all the "reputable" cam companies that still insist that this the proper way to determine proper geometry, which is totally laughable.

I, like many others use the half lift method. When done correctly (like Streetiron85 has described), you will have the smallest swept pattern across your valvestem tip, which may or may not be near the center. I would just like to add that when you're determining that the rocker arm is perpendicular to the valvestem, that you go by the center of the rocker's fulcrum and center of the rocker's tip. A lot of people wrongfully reference to where the rocker tip actually contacts the valve.

On my BBC, I have heads that have been "rolled", extensively decked, and 1/4" longer exhaust valves installed. The tip pattern is about .030" from the outboard edge on all of the exhaust valves, but valvetrain geometry is perfect. The reason why is that the rocker stud centerline and the valve centerline are not parallel - they're angle towards each other. As I raised the rocker arm up the rocker stud (mainly to compensate for the longer valve), the contact pattern shifted towards the outboard edge of the valve tip. Even though this is quite close to the edge, it is perfectly acceptable.

 

Thank you everyone for the replies, I got it figured now.

 

No Kidding!!
Companies design cam lobes which supposedly maximize the "area under the curve" and then turn around and give you a method for setting your rocker geometry that un-does everything that supposedly makes their lobe design superior.
Also, there's a noticable difference in the power output between a correctly adjusted valvetrain and one that's not.

 

Centered on the valve tip? Not necessarily. The distance from the rocker stud to the valve tip can affect where the contact pattern happens. And this is a distance that usually isn't controlled to a fine decimal point on the head, and can vary from one rocker manufacturer to another. Sort of a red herring on any mass produced head with Brand X rockers.

Smallest amount of travel? Again, not necessarily. Here's why....

Many people beleive that you want to use the "1/2 way to valve lift" method. I USED TO use it myself. This will give the smallest contact pattern. This involves making sure the rocker tip starts slightly towards the intake side, reaches the furthest travel towards the exhaust side at 1/2 of max valve lift and then travels back towards where it started when it gets to max lift.

I said I USED TO use this method, and successfully so for years.

I changed my opinion based on conversations with those who know a lot more than me about valvetrains. I now use the "2/3 of max lift" method. Meaning the rocker reaches it's maximum travel towards the exhaust side at 2/3 of max lift (not 1/2) and then travels a very small amount back towards the intake in the remaining 1/3 of the rocker's travel. This means SHORTER pushrods than the "1/2" method. And you'll usually find yourself using stock length sticks or something close to it.

Why do it this way?

Because having lots of rocker tip travel near max lift, where the valve spring is exerting MUCH more force is a bad thing. The 2/3 method will have more TOTAL travel across the valve tip but almost NO travel in that last tenth of an inch where the valve spring is almost fully compressed and exerting a lot of force.

The 1/2 method will have less TOTAL travel across the valve tip but much more travel in that last tenth of an inch where the valve spring is really exerting a lot of force.

That's the long way around the barn to say I almost always start with stock length sticks, regardless of manufacturer's recommendations, and I watch what the rocker tip does across the valve tip through the lift curve. I will go contrary to the above opinions and say that the stock legth sticks are giving you a better pattern and better rocker geometry for your application.

Please note that if you had asked me this same question 2 years ago I would have agreed with above opinions and told you to stick with the +.100" sticks.

FYI- using the "2/3" method I described above will typically result in the rockers looking more "leaned back" similar to a stock orientation whereas the "1/2" method will look more like they're "nosed over" even when fully closed.

 

The 2/3 lift method is the other method that is gaining popularity, for the reasons Damon has quite thoroughly and accurately explained. The difference in pushrod length between the two methods isn't very different (according to my calculations, about .065", with my BBC set-up). Either method (based upon which theory you go by) will work satisfactorily. The main point I'm trying to make here is that you can't determine "proper" valvetrain geometry simply by checking for a centered pattern on the valvestem tip.