Valve spring slection

i have a question out of curiosity.
how is valve spring pressure requirements determined?
I know if you go through a cam grinder they will tell you the spring they recommend (and I have done this a couple times and will in the future), but I am curious how these numbers are determined.
this may be prying too deep into knowledge that has to be earned and so be it if that is the case.
if not,
I know that ramp rates, lift, rpm, valve train weight, rocker style (rail / stud), and probably a few other items factor in. This is what leads to the realization that with out trial and error this could be a very difficult number to ascertain.

most of what I have found out on the net is useless generalism by people repeating what they were told and none of them having a real answer.

btw, I am not talking about shelf cams from Comp and others. I am talking about really aggressive solid flat and solid rollers.

Its not a simple process, for a real calculation you need a lot of specs and data you haven't even mentioned.

But its also determined by what springs are readily available or as close to needed as possible.

This is a very interesting topic for sure.

One thing to think about is how much of a factor the valvetrain inertia is in valve spring selection vs the frequency components in the lobe shape vs the natural resonant frequency of the spring itself which causes it to surge. There are computer programs like Spring Wiz which analyze the lobe shape using FFT to break it down to its frequency components and then compare that to the spring that might be used. The dimensions of the spring materials are carefully measured so its natural resonant frequency can be determined. The program then can determine if any of the frequency components that are in the lobe shape can excite the spring within the projected RPM range of the engine.

Does this work? From what I've seen in my own work it tends to be more conservative than what happens in the real world and I don't really know why. I can plot a cam, use that data and put it in the program. Then I can measure a spring that I know works great with that cam yet the program seems to think that the spring would tend to lose control earlier than I've seen in the real world.

I think the main reason is simply that the valvetrain itself is much less "linear" than the program assumes. The valvetrain simply absorbs some of the high frequency energy that's contained in the lobe.

You need a spring which will control the energy imparted upon the valve train by that particular cam lobe profile with consideration of the weight of that valve train and the RPM it will run.

It is done with math but, a lot of the equations need to have inputs which are presumed because all those parameters are usually not known exactly.

one thing that i had read was that previous spring selection has been conservative in the past. this conclusion is based on Spintron testing. But that would certainly be the way fall versus too light of a spring.

the linear topic certainly makes sense. linearizing the calcs would certainly make the arithmetic easier, but this would be one instance where linearizing across the entire range of operation is not going work.
if that is what they are doing then they would have to select the most aggressive acceleration of the lob and linearize that section. however, i would think that would tend to give a spring that is heavier than may be needed, which is what you said you have noticed.

in general, this answers my question. so there appears to be programs out there to use, but the results have to me mixed with hands on knowledge and experience.

like stated, the easy answer for us mortals is to use a cam grinder and let them give the proper answers.

however, someone has to be able to make this decision and was a little curious how that decision was made.

thank you for the information

I've often wondered the same thing. Try the SAE Bookstore. There are 1,000's of books and technical papers on 1,000's of topics. finding the one is probably going to take a while but it's probably there.

This is the most modern papers I could find. Might could find others if you spend some more time in different categories searching.

https://www.sae.org/publications/technical-papers/content/930615/

https://www.sae.org/publications/technical-papers/content/820768/

It is an interesting topic to be sure, and consider that the most knowlegable people within the topic are few and far between in the general automotive world.
Consider also, that camshafts and valvetrain and springs, which are now well over 100 years old in engine applications, are still being studied, and improved upon constantly. The "science" of valvetrain is still very active and progressive in design and manufacture.
I'm sure the formula for determining all of the "common" combo's of cam/valvetrain/springs is probably well known to the designers/engineers of such products.

But as an example, say you have the common valvetrain of a flat tappet lifter, chromoly pushrod, off the shelf aluminum body roller rocker and stainless steel valve with a spec'd spring and steel retainer.

Now we change the valve and retainer to light weight titanium parts but all else stays the same. The specs for the spring would now have to change as well.

BBFTorino wrote:It is an interesting topic to be sure, and consider that the most knowlegable people within the topic are few and far between in the general automotive world.
Consider also, that camshafts and valvetrain and springs, which are now well over 100 years old in engine applications, are still being studied, and improved upon constantly. The "science" of valvetrain is still very active and progressive in design and manufacture.
I'm sure the formula for determining all of the "common" combo's of cam/valvetrain/springs is probably well known to the designers/engineers of such products.

But as an example, say you have the common valvetrain of a flat tappet lifter, chromoly pushrod, off the shelf aluminum body roller rocker and stainless steel valve with a spec'd spring and steel retainer.

Now we change the valve and retainer to light weight titanium parts but all else stays the same. The specs for the spring would now have to change as well.  

Not necessarily but it is certainly possible that a spring that was good to X rpm will now be able to provide good control for another thousand RPM.