cardo0

Welcome back Pat. Yes I see a mark on the rocker arm. Crane cams has instructions on checking rocker arm clearances. An unbent paper clip works for me but you need to make sure its correct thickness (diameter).

Sky65

Patrick. I ran into the same problem recently on my Profiler heads. I decided to go to 7/16 studs and since I would need new rockers to fit the studs I aslo decided to go to 1.6. I bought some Scorpions bought rollers. They have good reviews. I used the 1/2 lift method to calculate PR length. What ended up with was very little polylock to slot clearance at full lift. Then I tried the blue plastic PR tool. It was way off. Next I did the valve tip contact method and came out again with very little slot clearance but a real nice contact pattern. What you found with the rocker studs being canted towards the valves, I think as well, it is the problem. By the way the Profiler website says the valves are .100 longer than stock. So I am now on the hunt for some full roller rockers with long slots so I can put mine back together.








Tom
Oh yeah. I think you can get a little narrower contact patch than you have if you fool with the PR length some more.

GearheadJoe

Hi Patrick:

Glad to hear that you are back on this. Regarding the SolidWorks model that I've been working on with my friend Gil, there have been some delays on Gil's end due to circumstances outside his control, but we have been moving forward slowly and are scheduled to review the "final" results next week.

I can tell you that the preliminary results show the "Straub" method for selecting pushrod length yields a much longer pushrod than the "centered contact pattern" method or the "Miller" method we have discussed.

The result is that with the valve in the full-open position, the Straub method leads to a more steeply angled rocker arm and a higher likelihood of the (stamped steel) rocker slot bottoming out on the stud. I'm quite certain that this is what caused you broken rocker studs.

In general, switching to full roller rockers will allow more rocker angle before the rocker bottoms out on the stud. So, simply switching to full roller rockers will likely eliminate the problem with stud contact.

However, a larger question in my mind is whether the "Straub" method is technically correct from an engineering perspective. My current impression is that it is not correct. The reference line that the Straub method draws through the roller tip axis should actually pass through the contact point between the roller and the valve stem. This is what Jim Miller advocates in the article you sent me.

The nice thing about the SolidWorks model is that we can make detailed comparisons among the various methods for selecting pushrod length, and carefully examine which method is "best" in terms of making the valve accurately follow the cam lobe.

GearheadJoe

Regarding alternate roller rockers, I have been very impressed with the Comp Cams "Ultra Pro Magnum" steel rocker. Its weight is comparable to most aluminum rockers, but it is stronger than most aluminum rockers. And, the fact that it is made of steel helps to reduce the bulk of material in the area near the stud.

I have samples of both the Comp Cams rocker and the Scorpion rocker out in the garage. I'll try to take some comparison photos and measurements tomorrow. There may be more room inside the Comp Cams rocker.

You mention using the "half-lift method" to select push rod length, so I'm interested to know what reference line you used on the rocker.

As I described briefly in Post #43, arguments can be made for either the "Straub" reference line or the "Miller" reference line. The attached photo shows a still shot from the Straub video, with the reference line drawn on the rocker in black.

I have added a red line that shows the Miller reference line. Miller says the reference line should pass through the contact point between the roller and the valve tip.

When using the half-lift method, it makes a BIG difference which reference line you use. The Straub reference line tilts the rocker at a much steeper angle at maximum lift.

Patrick has estimated that the rocker stud in the Pro Filer head is moved back about .070" from the stock position. This causes the Straub method to create an even steeper rocker angle at maximum lift.

At present I think the Miller reference line might be the one that is more technically correct in terms of accurately translating lobe lift into valve lift. Since it has the added advantage of providing a bit more room in the rocker area that concerns you, it might be possible to keep your Scorpion rockers and simply change to the Miller reference line (provided you can get comfortable with the rationale for the Miller reference line).

cardo0

Joe, I don't know anything about this Miller guy but I mentioned that true center line to to the rocker tip back in post #9. Your giving me short man disease.

Well I thought I should mention that anyone can get the grinding tool out and open up the rocker arm for clearance. Smokey Y. (RIP) mentions this though I'm sure there are limits to metal removal. IMHO aluminum roller rockers are overbuilt and the real concern is flex rather than breakage - unless of course it's smacking the stud.

Sky65

Thanks Joe. That would be great if you could take some measurements for me. I am using 1.6 ratio.

I used the Straub method. But as I work on this I think the only true 1/2 lift method is the contact pattern. True 1/2 lift will give the narrowest contact on the valve tip. How narrow and where on the tip of the valve contact is will be determined by the rocker ratio, valve length, and rocker stud placement and angle to the valve. The so called 1/2 lift methods use a 1/2 lift calculation in their process but I don't believe it has any real mathematically provable result. They do not take into account all variables I listed above. JMHO

Hope to hear from you soon on rocker measurements. I really want to get my engine back together.


Tom
I had to come back and add something. I do think the 1/2 lift methods do value. On a fresh build with no PR length reference they will get you close. And maybe even right on. But to me the final check is contact pattern. That tells the real story. Again. JMO

GearheadJoe

Wow, you did bring this up back in post #9! Sorry this didn't register in my head at the time.

Back then I wasn't particularly focused on this topic. I have always used the "centered contact pattern" method, but when the 90 degree method came up, my initial thought was that both methods probably lead to about the same result.

However, after Patrick "did everything right" and broke some studs, I became interested in trying to help him figure out what happened.

I studied the Straub video, and read an article by Jim Miller that Patrick sent me (a copy of the Jim Miller article is attached here). Figure 3 in Jim Miller's article points out the difference in the two reference lines.

The more I looked at this, the more I realized how remarkably complicated the analysis of rocker geometry is. There are several non-obvious factors that can influence the outcome when selecting pushrod length. I concluded that the best way to accurately compare the effects of the different variables was to build a SolidWorks model and then evaluate the effect of each variable and/or combinations of more than one variable.

My friend Gil is a Solidworks expert, and he agreed to to develop a SolidWorks model as a side project (i.e., I am not paying him to do this work).

The model is 99% done and we have spent some time comparing three methods for selecting pushrod length (centered pattern, 90 degree with Straub reference line, and 90 degree with Miller reference line). We also evaluated the effect of moving the rocker stud .070" back from the valve, as Patrick's measurements indicate was done on his Pro Filer heads.

The preliminary results are very interesting (at least to a valve train geek), and I will post them here when Gil and I have double-checked everything. Unfortunately, Gil has other things going on right now that are delaying our next meeting.

At present, I think the Jim Miller reference line is the correct one to use, but I look forward to getting feedback from others once I can post some results.

cardo0

What the Chevy Power Catalog emphasis is the rocker arm is 90 degrees to the valve stem at one-half the valve lift. They (Chevy PC) don't elaborate why other than a major impact on performance and reliability.
My assumption is to minimize side loading and wearing on the guide.

The difficult part is to get the lifter (or lobe follower device if something else is used) to 1/2 lobe lift. And of course if exh lift is different than intk then you have different measurements for each - let alone mixing 1.6 and 1.5 ratio rockers.

GearheadJoe

Hi Tom:

The attached two photos don't directly show whether the Comp Cams rocker has more range than the Scorpion for your particular application, but if you print out the photos and draw some lines with a straight edge, I think you can get a pretty good idea of how they compare.

What's not visible in the photos is how deep the pushrod socket is. Using a set of calipers for a rough measurement, it appears that the pushrod socket is about .140" deep on the Comp Cams rocker and about .170" deep on the Scorpion rocker.

In this comparison, both rockers are 1.6 ratio for a 7/16" stud. I placed them on a 7/16" rod and made two photos. One photo shows the rockers rotated as far as they will go in the direction of the valve closing, and the other photo shows how far both rockers will go in the direction of the valve opening.

Regarding the debate about the best method for selecting pushrod length, I found the Jim Miller article (attached to post #47) to be very thought provoking. I always assumed the goal was to center the contact pattern in order to minimize side loading of the valve guide.

Jim Miller says that the actual goal (at least for power production) is to center the arc of the roller tip travel so that it stays as close as possible to the maximum radius achieved at the 90 degree position. Placing that maximum-radius at the half-lift position of the valve travel accomplishes the Jim Miller goal.

I'm still in the process of trying to understand all the details of this very complex geometry problem, but right now I lean toward Jim Miller's view.

BTW, if you try to translate the Jim Miller method into a version of the contact pattern method, I think it can be said that the goal should be to achieve the narrowest contact pattern, even if that occurs off-center on the valve tip.

Also attached here is a very interesting rocker comparison article from Hod Rod magazine in 2013. This article is what influenced me to use the Comp Cams Ultra Pro Magnum rocker. As the comparison table shows, this is a very strong rocker whose weight compares well to aluminum rockers. I like the fact that I don't have to worry about aluminum fatigue, although that's probably not much of a risk with my street 383.

Sky65

Thanks Joe. The pics help allot. I will do as you suggest. Print the pics, draw center lines and compare the two. Just from eyeballing what you sent it looks like the Comp rocker has a longer slot on push rod end. Just where I need it. I'll take a look at the article as well.

Tom