manga_blue wrote:I'm learning more about springs than I ever wanted to know. Firstly, let me tell you that spring retainers can wear a lot when they've got a broken spring bouncing around underneath them. The retainer wear is a consequence, not a cause, of spring breaking.
I had an interesting talk with the spring expert today. Two springs can have the same size and same spring rate and one will perform well in a head and the other will destroy itself. It's a matter of resonance, he said. All springs have natural resonance frequencies. When they are operating in a resonance frequency they store energy and additional inputs at that frequency can amplify the resonance to the point of destruction. Think of a wineglass shattering when exposed to a tone of the right pitch.
The MX5 head, he said, is notorious as a dirty head in terms of range and strengths of resonance frequencies produced in it. He knows 5 or 6 instances of MX5 springs breaking as their first harmonic matches natural frequencies within the head, independent of spring rate. Springs for MX5s need to be tuned and tested until they are known to be safe for these heads. This may involve varying wire diameters, coil numbers and tempering while maintaining the same broad parameters of free length, rate, etc. It's not just a simple matter of getting right size and rate.
He thinks my failures are indicative of harmonic failure. I never will know whether or not they were tuned for an BP. It's very likely both Woz and the spring man are right. There's a resonance problem exacerbated by inappropriate cam profile, or vice versa.
I'll be going for the belts and braces solution: known springs (choice of Eibachs, Supertechs or Performance Springs) with a matched cam/lifter solution, whichever Woz and the spring man agree is best. If I have to go back to using bloody HLAs again then I'll just stencil a sign of the cross on the cam cover first.
What your spring man says about resonance and spring breakage is 100% correct. There has to be alot of thought go into design, not just spring rate. I am unaware of the MX5 engine being any worse than other engines of similar design, but it is possible. This is also one of the reasons I don't like unusual camshaft specifications, and I have seen alot of people try and "reinvent the wheel" so to speak when it comes cams for these engines. Whilst I'm all for developement on new and improved profiles for high performance applications the fact is that this configuration of engine has been around for 40+years, and for most people a tried and proven camshaft profile is all that will ever be required unless you have the time and money for developement work, testing and tuning.
Cheers
Woz
