Originally Posted by y8s
Also: at what point do the dual springs become necessary? The Supertech duals are about 50% more seat pressure (74lb) and about 30% higher rate overall.
I was over at Supertech in San Jose yesterday having a long talk with *****, and I bought a set for my '97 M3, whose cam was done by Elgin. Elgin gave me a crapload more lift on the intake, and OE springs are damn close to bind. (despite the not very aggressive ramps)
Basically if you make a single spring a lot stiffer than stock, the wire will see too much stress, and a dual spring setup will be a better engineering solution; and then the dual setup is designed for the springs to have a different resonant frequency so that the result is a pair of small resonances instead of a single big one. This resonance can cause a "travelling wave" to move up and down the spring and cause the valve to bounce off the seat after closing. A bigger problem with boost behind the intake valves, which will need a higher valve-closed pressure (force, really). This is what manifests as "valve float" and loss of power at high RPM.
I first checked if I could use just the outer spring for my M3 and it turns out they're softer than the cone shaped stockers. I don't know if this applies to the miata because the miata has standard cylindrical springs. Cone shaped springs allow thicker wire (stiffer) with shorter coil-bind-length. So I had to buy the whole shebang. ($$$, esp. with 50% more cylinders than a miata) The little ******* were more than half the price of my cam regrind.
In order to engineer dual springs you have to use their retainers. In order to sit both springs on the head properly you need their lower spring seat.
The springs primarily need to be stiff enough to decelerate the valves which peaks at the cam's nose. Deceleration is proportional to RPM squared, just like rod stress (not exponential
). So spring open pressure (force, really), needs to stiffen proportionally to RPM squared, but inversely proportional to valvetrain mass (valve+follower+shim+retainer+keepers+1/3rd spring). If the springs are too soft, the lifter can "take off" like a ski jumper and cause rapid wear on the cam and lifter due to the "harsh landing".
Because the valve deceleration profile is a function of the cam profile and RPM, the spring stiffness should be specified by the cam designer. Ironically more lift can allow for a less aggressive deceleration profile despite a more aggressive opening ramp; but many factory springs run very close to coil bind, to prevent dropping a shim during a mis-shift. The bind prevents the valve opening too far and a gap under the lobe forming big enough to drop a shim. This may be why miatas don't seem to drop stock shims during a mis-shift, unlike say, a Sentra. More duration also allows for a less aggressive opening ramp. However, an aggressive opening ramp with a high peak velocity allows for more "area under the curve" and makes for more airflow for a given 0.050" duration - too much duration screws up low end torque.