In the end I wonder, how much difference there is between 1/2 and 2/3. At what level of valve train stress would you have to be at for the difference between the two to be significant enough to matter? I don't know.
Valvetrain Geometry
- Thread starter Earlie A
- Start date
-
Earlie A
Well-Known Member
I have some data using a stock Speedmaster head with PRW 1.6 rocker arm. Test was done to compare the actual valve lift of an uncorrected shaft location to a corrected shaft location. I want to redo the test before posting the data to make sure I'm posting correct information. I'll try to do that soon.
Thats a very good book. Anyone interested in this stuff at even a hobbyist’s level can learn a bunch from his book.
As long as the rpm is near stock levels, the lobes are soft and the spring loads are minimal.
Once you start getting away from anything stick, like an aftermarket rocker and then increase the rpm and spring loads to go with it you will see parts damage and failures.
I have a 471 on the dyno that didn’t have the correct valve springs on it.
When I went to change the springs it was already beating the **** out of the locks and retainers. And that was at 6100 rpm. It was so pissed off I wouldn’t dare try and pull through it.
The customer did not really understand what the issue was, so he was here this time to see it first hand.
We used a spring that was bigger on the OD and had a larger wire diameter.
I think we went down on seat pressure by 10 pounds and up 10 over the nose.
Where is was struggling at 5800ish and was miserable at 6100, now it will zip right up to 6900.
Fortunately, his geometry was very close and he paid for overkill Pushrods’s or it would have killed some parts.
When you do this every day you see the effects of **** geometry.
I just got off the phone with Mike. We discussed some of this stuff, including the fallacy that the 1.5 ratio is perfect for the small block. It’s not.
If it was, then why make the magnum with 1.6 rockers? They could have left well enough alone but they didn’t.
Most of the time when an engine doesn’t make more power with a higher rocker ratio it is geometry related.
You can run the same lobe on a 1.5 Chrysler and a 1.7 big block Chevy and the Chevy doesn’t go out of control.
In most cases a higher ratio rocker will make more power unless there are issues somewhere else.
PRH
Well-Known Member
It’s amazing to me that in this day and age of nearly unlimited information that someone could assemble a pile of parts that were that unhappy together.
It’s pretty rare for me to encounter a valvetrain that just shuts down like that…….. other than stuff where the Hyd lifters are giving up(collapsing).
Maybe I’m just not getting deep enough into the back of the lobe catalog
It’s amazing to me that in this day and age of nearly unlimited information that someone could assemble a pile of parts that were that unhappy together.
It’s pretty rare for me to encounter a valvetrain that just shuts down like that…….. other than stuff where the Hyd lifters are giving up(collapsing).
Maybe I’m just not getting deep enough into the back of the lobe catalog
Well, it’s a bit of a funky build. I’ll leave it at that.
PRH
Well-Known Member
Nothing too high end here.
PRH
Well-Known Member
Who selected the combo of stuff that wasn’t happy?
Anything stand out with the cam lobes?
The owner did. Like they sometimes do, they build a very small box with extreme limits on what they can live with and this is what happens.
They want power and rpm but not the manners that comes with that.
I will say this. When a customer wants the torque and power peaks 3k apart this is what you get.
It’s down on power. It’s down on torque. Bit it won’t need a 4800 converter.
PRH
Well-Known Member
Without variable cam timing and excellent modern technology type heads…….. or a centrifugal s/c……..it’s just not going to happen.
Did you know what components had been used before you tested it, or was that after the fact knowledge?
Or, was it getting tested because problems had already materialized in the car?
I knew what it was going together.
It’s on the dyno again because the first time it had the valve train issues and I didn’t have time to do anything about it then. So he took it home and was going to try and make an event.
When that didn’t happen it came back here to see for himself what it was doing, and so I could fix it.
Thats fixed now, but the power and torque numbers aren’t bad but they are a bit skimpy. I’m not sure there is 30 hp left to be found at this point.
I’m waiting on some other stuff that should be shipping later this week to finish it. At that point it will be what it is, unless he wants to try a different grind.
Like the brohams say, "put da nitrogen on it"
Oh, thats in the works. I already told him just put a bigger jet in the hose and let it eat.
Earlie A
Well-Known Member
This ended up being a surprising test. The idea was to compare actual valve motion of a stock head (ended up being ProMaxx) to the same head with the rocker shaft relocated to a 'corrected' mid-lift position. Rocker arm was PRW stainless 1.6 ratio. See the first two pictures for the set-up. A solid (epoxy filled) lifter was used against a cam rotated such that the lifter was on base circle. Gage blocks were placed between the cam and the lifter to simulate valve lift. Actual valve lift was recorded from a digital indicator placed on top of the rocker roller tip.
Conventional thinking says that by raising the rocker shaft and moving it away from the valve a more proper geometry will result which will give the following benefits: Less roller tip sweep across the top of the valve, increased total valve lift, increased acceleration of the valve off the seat, decreased velocity of the valve over the nose. Interesting which of those proved correct.
The yellow paper shows the results of tests. The 'Valve Lift' column shows the actual valve lift at each cam lift increment. The rocker ratio is cumulative rocker ratio which is the total valve lift divided by the total cam lift, line by line. The 'Relative Velocity' column is similar, except it is the change in valve lift divided by the change in cam lift. For instance, between 0.100 and 0.150 cam lift (a change of 0.050") on the stock head, the valve moved from 0.161 to 0.241 (a change of 0.080"). 0.080 divided by 0.050 = 1.60. This number gives a 'relative velocity' number that tells how much the valve moved during that lift increment. It's more like an instantaneous velocity compared to the rocker ratio which is like an overall average velocity. That number is useful for comparing velocity of the valve off the seat or at any other location in the lift cycle.
The results are quite surprising. I did this test over and over again over the course of 3 days to make sure it was correct. I didn't understand what was going on until I added the cam, lifter and pushrod to the drawings. As can be seen from the last two drawings, relocating the rocker shaft away from the valve increased the misalignment of the pushrod and lifter from an average of 14.71 deg to 15.45 deg. This negatively affects the geometry on the pushrod side of the rocker so much that the valve motion suffers.
As can be seen there are some discrepancies between the actual measured data on the yellow paper and on the 'predicted' numbers from the CAD drawn papers. For instance, the CAD drawing predicts a valve lift of 0.670" at 0.430" of cam lift. Actual measured valve lift was 0.678". For the 'Corrected' head the CAD drawing predicted 0.668" and the actual measured value was 0.665". The point of the CAD drawings is not so much to come up with the exact same numbers as measured as it is to explain why the relocated shaft gave surprising results.
The geometry on the valve side of the rocker arm is a little less complex than the pushrod side. Fixing the geometry on one side alone can lead to unforeseen problems on the other side.
TT5.9mag
Two atmospheres are better than one
Nice work. Did you try changing to a cup adjuster and using the longest pushrod you can?
The pushrod side is at best a nasty compromise.
Edit: how old are those PRW rockers?
Mine are old enough that they are not the upgraded version.
Last edited:
Earlie A
Well-Known Member
The set-up is ball-ball pushrod and cup adjuster. I ran the initial tests with the adjuster 1 turn from the seated position. This morning I bottomed out the adjuster and increased the PR length to compensate. Results shown.
Earlie A
Well-Known Member
PRW rockers were purchased in 2025.
Changing to real springs would be a challenge and a totally different set-up. It would be tough to do with the gage blocks - how would you compress the spring to change the blocks? If a real cam was used to lift the valve then measuring actual lifter movement is more complicated. What changes would you expect to see? Only difference I would expect are those associated with load and deflection. Can't see that being much in a static situation (non-running engine).
How does the ridiculous lifter to pushrod angle play in all of this? If the lifter bores were say 45 degrees, would any of this geometry be "better" so to speak? It's something I've always asked WHY did Chrysler cast blocks like that?
Earlie A
Well-Known Member
I don't see how the SBM runs as well as it does with that misalignment, especially at high RPM. R blocks and Ritter blocks have the lifter bores at 48 degrees which is perfect. A line drawn through the cam centerline and the rocker adjuster cup is a 48 degree line. The 59 degree lifter angle was a hold-over from the 318 poly block. I guess Chrysler didn't have enough money or time to get it right.
Lol, next You get to find & try a W2 setup,...non-linear in 2 planes by quite a bit ...
Yeah, I didn’t think about your setup and how to make all that work together.
What is interesting is PRW was supposed to have changed the nominal (or static as some say it) ratio from about 1.64ish to 1.67ish but evidently from your numbers that is not the case.
With checking springs the ratio at some point should be the nominal ratio but your testing says it’s not.
I need to ask Mike about that the next time I talk to him. I know you should have the updated rockers if you purchased them in 2025 unless they were relatively old stock sitting on the shelf.
Do you mind taking a couple of pictures of your rockers? Like a side view and a shot of the underside of the rocker?
I have forgotten how to tell the updated rocker from the older rockers by the numbers on them but if I see a picture of them I might be able to tell.
-
Similar threads
