The reason I ask is that I find it hard to believe that a manufacturer such as Honda would design a spring which fails at the rate it does
I have a feeling that the mechanical design of the spring itself is probably sound enough, but the post-production heat treat may not. For those of you who mightn't know, heat treating steel or pretty much any other metal or alloy will give the material different mechanical properties (strength, hardness, elongation etc etc) depending on the heat treat. The heat treatment is usually done after manufacture or machining so as not to introduce any residual stresses back into the part. If we knew the exact make-up of the spring it would be simple to create a heat treatment profile suitable for the spring which might improve it's elastic limits and stop it from failing.
) that thermochemical treatment might be crucial to spring reliability, but if it is ends of spring that got broken - then it means spring should have localised treatment, rather than constant through all spring.
ha bleedin ha! 
That's kind of what I was getting at in my previous post - if this is where it's failing most or at the other bent end that slots into the CCT housing, then it's pretty clear that the springs are treated prior to forming, rather than after, or that the forming process has simply weakened that area so that mechanically it falls way beneath it's design limits. The 'corner' of that cross section is probably the weakest part of the spring (except for the other bent section at the other end). Anyone else notice the spring failing in this area??
