Stress Relieving

Should this thread be indexed?


  • Total voters
    20

Kevin R. Cashen

Super Moderator
Stress relieving: “the heating of steel to a temperature below the transformation temperature, as in tempering, but is done primarily to relieve internal stress and thus prevent distortion or cracking during machining. This is sometimes called process annealing…” ASM Heat Treater’s Guide

Stress relieving is rather subtle and hard to discern when compared to other treatments that involve actual phase changes within the steel. Annealing, normalizing, hardening etc… all involve actual physical changes in the phase and internal makeup of the steel, i.e. austenite, martensite, pearlite. Under the microscope it is easy to see if these treatments have been performed by their effects on the microstructure, the effects on stress is a secondary result of the physical changes they make in the microstructure. But stress relieving is the opposite; it deals directly with relieving the subtle and unseen effects of stress.

Proper normalizing and annealing will heat and cool the metal in a very even fashion so that the natural expansion and contraction that occurs will also be even, but operations such as forging, grinding, welding, and machining will all produce heat and some level of deformation of the metal's crystalline structure that by its very nature will not be evenly applied. Heating or deforming metal in such a localized fashion will result in the surrounding unaffected material having a hindering effect on the natural expansion and contraction causing these forces to remain as internal stresses. This problem can also arise from the very large expansion and contractions that occur during transformations in other heat treating operations. There will be nothing as notable or dramatic as new phases created, or carbon precipitations formed, but there will be subtle changes in the stacking of the crystalline matrix which result in unwanted potential energy in the steel. In subsequent machining operations or heat treatments these concentrated areas of strain energy will result in distortion or uneven rates of transformation. I have personally encountered and observed the effects of this unseen strain energy in blades not only in the most common issues of distortion, but even in the finishing of blades where things like etching will recreate coarse scratches long removed in the polish from the strain energy record left un-erased in the steel. This is also why I have often disagreed with knifemakers who are convinced there is no effect to turning a blade blue on a grinder simply because they can see no outwardly visible effects. It is easy to dismiss because how does one see strain energy? One must know the exact physical effects that coincide with it to look for even under a microscope. The only way to directly observe the cause and effects of recovery is to map the crystallographic changes with X-ray diffraction, not something bladesmiths typically have access to.

The problem and the solution
So everything is how we want it as far as carbide distribution, grain size and phases present, but we still have this pesky, unseen problem that we want to deal with but normalization is just overkill, and possibly a step backwards in the process. We have a nice, well- machined, or semi-polished blade with everything as we want it, except for this invisible stress thing. We don’t want to scale it all up, decarburize the surface, or undo our carefully planned internal conditions. We need a heat treatment that is designed just to deal with that unseen stress, that heat treatment is called, most appropriately, “stress relieving.”

The other heat treatments tend to leave the impression that on heating a piece of carbon steel nothing happens until you reach 1335F (Ac1) when suddenly recrystallization occurs and a new phase is formed for carbon to move about in. But the full process of heating involves many other changes that occur in a much smaller scale before this point is reached. Besides carbon based precipitates forming from room temperature up to 1000F other not so obvious changes occur from 900F up to the actual recrystallization temperature. The process occurring in this range is known as “recovery”. Recovery will display no outward changes in mechanical or physical properties (except, oddly enough some rise in the electrical conductivity), but instead involves changes at the atomic stacking level, where it realigns dislocations of the atomic lattice to remove the physical effects of previous stress. As recovery progresses it results in the formation of very small subgrains (1 micron or less) within the previously affected grains. But in order for this homogenizing of stress to be maintained the steel must obviously be cooled in a very even fashion or stress relieving itself can contribute to the problem. The unseen mechanisms involved in proper stress relieving are no different than the other heating operations in time being equal to temperature in its effects at that temperature. Heating to a higher temperature will accomplish the stress relieving in much less time, but care must be taken to heat as evenly as possible and not overheat to the point that one will be annealing or normalizing. One may also be able to keep the piece cleaner by keeping the temperatures lower and extending the time. It is also worth noting that the more time the steel is at temperature, the more thorough will be the stress relief. Cooling may be better accomplished in the very still air of the oven itself rather than the open air that is used in normalizing to avoid precipitations that are not a problem at the lower temperatures of stress relieving. Without going into detail about the mostly unrelated topic of creep it should also be mentioned that care should be taken to suspend or support the blade in such a way that it will not be distorted by its own weight, an issue that can be a special problem in the range that recovery occurs in.


How to stress relieve
To do a proper stress relief it may be best to place the blade in the cool oven and bring it up to temperature slowly. The effective range is from 900F to 1200F and the hold time, as with most heat treatment soak recommendations, is 1 hour per inch of thickness, but if one divides this by the fractions of an inch we use for knife blades, at least a 15 minute hold time would be good with more time resulting in more thorough recovery, depending on the temperature.



Recommended reading on this topic: “Metallurgy” by B.J. Moniz



If this post, and the resulting thread, has information that you feel is useful enough to be linked to in the sticky index at the top of the page you may indicate that by voting in the pole. If you feel the information, or resulting thread, is not helpful, productive, or positive enough for all visitors to benefit from, do feel free to indicate that with a “no” vote in the pole. Both opinions will be regarded with equal value in ensuring only the best information is highlighted in this forum. I have changed the rule regarding two pages to be considered for the sticky, and will participate with answering any questions to better see if the thread is of interest.
 
Last edited:
Thanks Kevin. I think definitely part of the tool box. I read through it a bunch of times and have questions, but I think you have the answers mixed already.

Thanks again, Craig
 
Thanks Kevin. I think definitely part of the tool box. I read through it a bunch of times and have questions, but I think you have the answers mixed already.

Thanks again, Craig

Whew! Thanks for posting Craig, I was beginning to think this one was a real bomb, or people have finally had enough of my tomes:3:. If you have questions please ask away so that this thread has more than crickets chirping in it.
 
Not enough of tomes at all but I first saw this at work and didn't have time to read it through. Thanks
this is a very informative one.

"Without going into detail about the mostly unrelated topic of creep" are you suggesting spine down, on
the side, or what?
 
In a horizontal situation blades should always be heated either spine down or edge down to avoid any distortion, even in tempering. I personally prefer spine down.
 
Recovery questions

Kevin,

Lets see if I'm understanding this process correctly.

I'm interested in these very small sub-grains. If I understand things accurately-

Upon austinizing to harden these new sub-grains will diffuse carbon from the larger previously created grains there by reducing grain size in comparison to what we would have had if we had skipped the stress relief cycle. Although we haven't created any phase changes during stress relief we will see a benefit of reduced grain size at the next phase change that occurs in the hardening cycle ????

Assuming the last statement is accurate then it will always make sense for a stress relief cycle as a final step before hardening. I've been dealing with warp in a mechanical fashion and getting great results with that so I've been considering abandoning annealing and/or spheroidizing all together.

This process of recovery has me re-thinking this logic though.

Also- I'm mostly working 1095 so I'm curious if we should stick with cooling in still air to prevent carbon in the grain boundries or is this not an issue in this temp range because we're not moving carbon around ?


As always, Thanks for your active participation on the forums and being dedicated to increasing our collective understanding of metallurgy !

Take care,

-Josh
 
Back
Top