Question:
Any suggestions for a good reference for spring relaxation?
How constant will the force of a spring used as an electrical
contact in a new design be after 1 year of service? 5 years? 10
years? This is the question I need to answer.
I am using .020 spring steel C1050 heat treated to a hardness of
C40 to C44.
Is there an accelerated life test that I can conduct? Does
conducting such a test at elevated temperature truly simulate an
accelerated test, or does it merely degrade the spring by undoing
what the heat treating did?
Answer:
The question of spring relaxation is a very easy one to answer provided
you know the stress level and temperature. There are all sorts of charts
and formulas in these design guides to help you answer that question. I
have copies of both and they're really useful, as is my other favorite,
the o-ring design handbook put out by Parker. Great stuff.
I would be very reluctant to use temps above actual service conditions in
tests. I don't believe there is a correlation which can be easily made
which says the spring lasted 100K cycles at 200 F therefore it should last
500K cycles at room T. Much more often the cycle rate is increased, i.e.
if in application it flexes at 10 cycles per second, increase it to 20, or
30, or more. If you keep the spring cool and don't allow it to heat up
beyond what it might do in service, I beleive that would result in much
more meaningful testing than elevating the T.
If you are at a fairly low stress level, oh lets say 10% of ultimate (in
the linear portion of the stress strain curve for unhardened 1050), the
heat treatment should not matter at all. The modulus does not differ
(measurably) between hardened and unhardened 1050. The hardening is
useful by increasing the yield and giving you a better fatigue life at
higher stress levels and allows the spring to be used at higher stress
levels.
