NHML Resources - Selecting Steel & Heat Treatment: A Design Engineer's Checklist

Selecting the wrong steel and heat treatment could result in premature part failure and an unhappy customer. Here are four examples of occasions when you really need to know what steel was in the original part and exactly how it was heat treated. If the original material specifications are not available then you will need laboratory assistance. Discuss your design expectations and constraints with your material advisors. Consider these Examples:

  1. The higher strength alloy resulted in reduced fatigue life.

    The original spindle was about 1 ½" diameter and was made of AISI/SAE 4130 steel alloy. The original heat treatment had been chosen so as to harden the surface to Rockwell C 50 which was then drawn back to HRC 45. The core was HRC 32. In service, the shaft was subjected to a mild side load while rotating. Ultimately, the shaft wore out and had to be replaced.

    Recall that most steels expand during hardening. In conventional heat treating, the surface hardens first and goes through expansion. The core expands later, and with the surface already hard, the core expansion then throws the surface into tension. This resultant tensile stress on the surface makes for terrible high cycle fatigue life. The original steel selection and heat treatment had been selected so that the core hardened to less than 50% martensite, thereby allowing the core's soft phases to absorb much of the cores expansion and considerably reduce the surface tension.

    The rebuilder used 4140 for the replacement shaft, believing it to be a higher strength alloy. It was hardened and then drawn back to the original shaft's HRC 45 surface hardness. Because of 4140's higher hardenability, the core of the shaft came up to nearly full hardness.

    The resulting core expansion put excessive residual tensile stress on the surface. The fatigue life was considerably reduced and the 4140 replacement shaft failed prematurely due to high cycle fatigue.

    Conclusions: Call NHML if you have to identify the original steel alloy or before making an alloy substitution and need our recommendations for heat treating.

  2. A change in the tempering temperature embrittled the steel

    The application involved occasional shock loading. The original part was made of AISI/SAE 8630 and it wore out prematurely. A check of its hardness indicated that it had been tempered at about 850 °F. The rebuilder offered to improve the wear resistance by increasing the hardness through tempering at a lower temperature. 600 °F was selected for the new tempering temperature. The replacement part soon fractured.

    Analysis, performed by NHML, determined that the replacement part failed by low cycle fatigue. The 600 °F tempering put it into the temper embrittlement range which severely reduced low cycle life, thereby leading to premature failure.

  3. Free machining grade reduced the fatigue life.

    A complex component in a paper mill mixer was made of 316 stainless steel. The rebuilder offered a price reduction if he could use the free machining grade, since in making the part, there was a significant amount of chip removal. The customer permitted the change.

    The part failed prematurely. Failure analysis at NHML showed high cycle fatigue failure due to the free machining additive in 316F stainless steel.

  4. Rolled fillets increased fatigue life.

    Hardness testing by the rebuilder indicated that a crankshaft had received a standard quench and temper to Brinell 229-269 in accordance with the original specifications for the crankshaft. Even so, the replacement crank failed prematurely due to high cycle fatigue. Unknown to the rebuilder, failure analysis by NHML revealed that the fillets in the original crankshaft had been rolled and were very much stronger and had residual compressive stress.

Distortion Control

Prehardened bar can be the rebuilder's best friend. Keep in mind that the larger the bar, the greater the differences in residual stress between the core and the outside. Unless the residual stress is relieved, the part may "walk" between rough machining, or may distort in service.

The rebuilder can do a gross check for residual stress. Cut a thin disk off of the bar and then make a bandsaw cut from the edge to the center of the disk. If the kerf opens or pinches the blade, then there is a significant amount of residual stress. The same test should be used on cold finished stock.

Ultrasonic stress relief can be a useful tool but it doesn't always work in all cases. Always check the part for residual stress after using ultrasonic stress relief.

Sample Identification

For non-magnetic stainless steels, about two grams of chips or little chunks are consumed. Most other metals can be fully analyzed with half a gram or less. Call NHML's chemists if you are unsure of the amount required.

In order to get an average composition, with castings or solder samples, plan on sampling from the outside (first to freeze) to the inside (last to freeze). If there is a machinability problem, consider taking both an average sample and a sample from the area being machined.

With wrought metal, significant differences in composition, from inside to outside or end to end, are pretty rare. Any sample will be acceptable.

If you choose to take drill chips then it's important to take every precaution to avoid carbon contamination from cutting oil or coatings on the drill. Use an uncoated drill bit, clean it ultrasonically and remove any adhering chip buildups. When you drill, throw away the first few chips. Sampling the non-magnetic stainless steels for carbon analysis requires the utmost in cleanliness.

Please do not use sandwich bags to send non-magnetic stainless steel samples since many of the plastic films have a coating of oil. NHML will send non-contaminating plastic bottles or you can pick up clean containers from a local pharmacist. Baby food jars may be used but be sure that the covers do not have a wax lining which would contaminate a carbon analysis.

Identifying an alloy temper can sometimes be done through a hardness test but also requires tensile testing. Call NHML's Metallurgist or Metallographer to talk over your specific problem.

Lots of problems, including machinability questions, require metallography in addition to chemical analysis and physical properties. Feel free to talk it over with either our Metallurgist or Metallographer.

Welding Tips For Builders

Knowledge of a steel's chemical composition or alloy is essential for effective weld repair. Most structural steels can be welded but the techniques and pre and post weld treatments can make the weld repairs time consuming and expensive.

A general rule involved the carbon equivalent (CE) of the steel. Steels with a carbon equivalent less than 0.47% can be welded without special precautions. A carbon equivalent above 0.47% requires at least a preheat in order to prevent hydrogen cracking.

The carbon equivalent can be calculated by the following formula:

CE = C% + + + +

A more sophisticated formula for carbon equivalent is shown in Figure A. The figure takes into account both the carbon equivalent and the actual carbon content in the steel in predicting the susceptibility of a steel to hydrogen induced cracking.

Zone I steels, even with extensive alloying, have a carbon content so low that even rapid cooling will not result in cracking. Steels in Zone II have a modest alloying (hardenability) and may be subject to cracking. Preheating will minimize their tendency to crack.

Zone III steels combine elevated carbon content with alloying. Preheating low hydrogen welding practices and post weld thermal treatments may all be required to avoid cracking.

See our Industry Definitions for further insight.

 

<< return to Resources

News

Nuts & Bolts, V23

All News >>

Related Materials

Validation Compliance

Valid XHTML 1.0 Transitional Valid CSS!

who's online