PSN-L Email List Message

Subject: Material Creep
From: The Gladow Family glad@..........
Date: Sun, 12 Sep 1999 13:02:58 -0500


To:  Barry Lotz & other interested parties

Hey, I hadn't expected to get this deeply involved in the details, but
since you asked about "relaxation vs. time under stress" in the second
paragraph of your September 10th message:

1.  Basically, material creep is a primary reason that my initial
message included the statement "Do NOT be tempted by the ease of
plastics"--- since even the best engineering plastic will exhibit a
creep rate which is typically a couple orders of magnitude worse than
metals.

2.  As you have discovered, however, even metals--- under continuous
loads producing relatively high stress--- will exhibit some creep.  This
is not unique to flexures; you already pointed out that it can happen to
music wire, and most of us have seen cases where coil springs have
"relaxed" when under load for long periods.  Even knife-edge pivots have
been shown to have some (small) amount of "cold flow".

3.  Therefore, there should be little, if any, concern about creep in
flexures for "horizontal-boom" seismic devices, where the stress level
is low.  I am, however, a fan of using comparatively thicker
spring-metal flexures, say in the .004 to .008 inch range for spring
steels, for your typical size and weight instruments--- but this is
primarily because of the ruggedness which is provided.  For those, like
Sean-Thomas, who want to maintain gravity as the primary centering
force, you can readily compensate for flexure thickness by increasing
the free-length (i.e., distance between clamps) per my "Calculations"
paragraph of previous message.

4.  But, back to your case of the "VBB sensor using an ~.018" strip bent
in ~180 degrees".  The stress level is obviously rather high, so what
can be done?---

5.  As you have noted, creep is time dependent--- gradually decreasing
over a period of months.  Therefore, one helpful "trick" used by many
instrument manufacturers is to speed up the "aging".  A typical method
of doing this is the cycle the metal (a minimum of 100 times, with 1000
being better) to a stress level which is approximately 50% higher than
the load which will be seen in service.  Obviously, you don't want to
exceed the yield strength or proportional limit!  Application of heat---
perhaps 250 degrees F, but again well below any temperature used for
heat treatment (or normalizing) of that particular metal--- during the
cycling will also assist in the stress aging.

6.  Yes, the selection of the spring metal is also important in
controlling creep.  However, the metallurgists and other experts tend to
use a lot of caveats--- items such as differing times and temperatures
during tempering, amount of cold-working due to rolling or drawing to
final size, etc.  Therefore, I haven't seen any good comparisons in
tabulated data.  (Anyone else?)  One general rule of thumb seems to be
to avoid those spring metals which are hardened entirely, or primarily,
by strain hardening ("cold working") rather than heat ("precipitation
hardening" and "transformation hardening").  This would indicate the
avoidance of phosphor and silicon bronzes, 18-8 stainless, and the 300
series of stainless steels if your application demands minimum creep.

	---Dean E. Gladow---

_____________________________________________________________________

Public Seismic Network Mailing List (PSN-L)


[ Top ] [ Back ] [ Home Page ]

Larry Cochrane <cochrane@..............>