The authoritative public forum
  for Metal Finishing since 1989
-----

Spring Failure Causes

 

Q. We have stainless steel springs within our solenoids that fail within the first year in application. These solenoids have been in production for the past 4 years are this year we are experiencing broken springs in the solenoids. I understand spring failures can occur due to surface finish flaws. An outside laboratory has the position that the granular structure of the springs meets ASTM specification.

Can a spring meet ASTM specification and still contain flaws that would cause it to break early in application? What types of specific characteristics should be sought relative to spring failure?

Any and all suggestions are welcome.

Patricia Estorga
- El Paso, Texas

---

Q. The spring wire is specified to ASTM A313 specification. In application, the springs may be exposed to fuel vapors. Again, these springs have been in service for 4 to 5 years and up to this year there has not been this problem of breakage. Any ideas?

Patricia Estorga [returning]
- El Paso, Texas

---

---

Q. I am an amateur astronomer who has just built an unusual observatory whose roof mechanism is supported by two large stainless steel springs. I now find that I need to slightly exceed the design specification of the springs, but still feel I don't have a complete answer to my query from the manufacturer. Can someone help me out with an off-the-record commonsense answer, please? My question raises an interesting general point about handling what must be a tradeoff between the extent to which the elastic limit can be slightly exceeded if one requires a short cycle life (2000 rather than 100,000 cycles).

I am a numerate retired scientist but not an engineer. A quick internet search tells me that the elastic limit for most stainless steels is about 1/3 the tensile strength: would I be correct in thinking that even if my springs are the lower grade (see below) they WON'T in fact SNAP (as implied below) at 25" or even 26", but would probably acquire a set? How quickly is this likely to build up? I envisage perhaps 2000 (not 100,000) cycles over perhaps 20 years in which the springs spend 95% of their time at 16" or 17", and 5% at 25" or 26", depending on what extension I settle for.

The technician's last two responses include the following: "The maximum allowable extension of the spring depends [on the] tensile grade [which] we don't know. Our product standard clearly states that[they] must be made out of a 302 stainless steel [but] the tensile strength can be any where between 1200N/mm^2 and 2200N/mm^2 the design software rates the upper tensile as 1610N/mm^2. As the springs are pre-batch it [is not] possible to request a material certificate as there is no traceability. Based on the lower grade the maximum allowable extension of the spring is approximately 24.5". Based on the upper tensile the maximum extension is approximately 27". The fatigue life will be 100,000 cycles for both sets of loaded position [cycling between 17" and 25", and between 18" and 26"- J.K.] for the upper tensile grade but for the lower grade the life will be less than 100,000. This could mean it might fail [snap or acquire a set? - J.K.] on the first attempt or could last the 2000, cycles required."

John Kemp
- Whitstable, UK

---

 

A. Personally, as an engineer, but not a spring engineer, I just would not do it. First off, your belief that ultimate strength is about 3X the elastic limit is probably not true for spring stainless steels where even a quick look showed design stresses at 45% of ultimate. Although some springs can acquire a set, as in a twisted Slinky or the retraction spring on a ball point pen, you can't depend on that. Springs do break and often it is much more dangerous than when other things break because the recoil releases so much energy. A snapped tie bar can let something fall, but a snapped spring can rip your head off :-)

Ted Mooney, P.E.
Striving to live Aloha
finishing.com - Pine Beach, New Jersey

---

? Is the spring calculated according to meet the cycles requirements and the Goodman diagram is OK?

Lluis Callejón Font
- BARCELONA, SPAIN
2006

A. John,
There's an excellent spring calculator at www.msdspring.com
(no affiliation to me). Use the worst case strength values to design your spring. You can also run steel cables through the center of the springs and attach them securely at each end for safety - similar to the springs on most garage doors. Then if the spring breaks it won't become a projectile, and also your load will have some support.

Alan Kirkpatrick
- Shelburne, Vermont
September 30, 2009

A. Hi Alan. Excellent idea about the safety cables!

Regards,

Ted Mooney, P.E.
Striving to live Aloha
finishing.com - Pine Beach, New Jersey

---

---

Q. Excuse my English, I am a Spanish spoking person. I have two helical steel springs that fractured at 45° and have longitudinal cracks each one along the inner line of high shear stress. Someone can explain me why this happened?

Hamilton Roque
Mechanical Engineer - Trujillo, La Libertad, Peru
July 24, 2018