ASM Online Member Community

Expand all | Collapse all

Determining Heat Treat Soak/Hold Time

  • 1.  Determining Heat Treat Soak/Hold Time

    Houston Admin
    Posted 09-17-2020 10:27
    ​In the heat treat literature, there seems to be a paucity of information regarding soak/hold time for heat treatment operations. Some prescriptive specifications exist for things like aging of superalloys where hold times are very explicitly defined, but often (especially for steel) the requirement is simply to "hold for a time commensurate with thickness" or "hold for a minimum of 30min/in or 1hr/in". Then there are additional variables:

    •  how the temperature is monitored, whether that be off the furnace TC, a contact TC on the part, or embedding the TC in a heat sink of comparable material and section thickness.
    • what the requirement is to soak vs. to hold after the part has soaked - rules of thumb like 1hr/in reflect the summation of a certain amount of soak time and a certain amount of hold time but the requirement for hold time is very material/operation-dependent. As section sizes get large, the cumulative hold time portion of that 1hr/in (or comparable rule) increases perhaps past what it really needs to be.
    • how HT section size is determined - for simple geometries like plate it's straightforward but for complex shapes like castings and forgings, opinions differ. A largest inscribed sphere is customary in some industries but some codes will assign additional thickness if the section is, say, square vs. round.

    What is, in your opinion, the best way to reconcile all these variables? Do you feel that common rules of thumb adequately incorporate all critical variability and it "comes out in the wash", or do you adjust your hold times to accommodate certain variables? When a range of time exists, do you tend to aim high or low?

    Sean Piper
    Product / Process Metallurgist
    Ellwood Texas Forge Houston
    Houston TX

  • 2.  RE: Determining Heat Treat Soak/Hold Time

    Santa Clara Valley Admin
    Posted 09-18-2020 08:11
    You are correct that there are a variety of ways that heat treating requirements, including pyrometry, process parameters, and so forth have been established and documented.  Different organizations have different approaches to these matters depending on their needs, their own procedures, and the practices at the time.

    As an example, I serve on the Metals Committees for SAE's Aerospace Council.  Within that there are a variety of alloy-family committees, each with their own "AMS" specifications for heat treating. While there is a desire for commonality, inevitably there are deviations from committee to committee and even between specifications under the cognizance of these committees.   A lot of that depends on the thinking of the document sponsor and the committee at the time an "AMS" document is proposed and later when reviews and revisions are conducted.  

    In some ways this is like the childrens' game "Rumor" (sic) in which a phrase is whispered in one participant's ear who then whispers it into a second participant, and so on.  What comes out at the end often deviates significantly from how it started.  While the AMS committees strive to maintain greater accuracy and consistency than that, drift does occur.  Nevertheless, the final version is generally metallurgically sound, although there is often some quibbling "around the edges", particularly regarding the differences in the needs and limitations of producers and users and the experiences of the individual members of the committees.

  • 3.  RE: Determining Heat Treat Soak/Hold Time

    Posted 09-18-2020 08:11

        The ONLY way to properly determine hold times is by using contact thermocouples. The tips of the TC's are to be imbedded into the thickest section of the workpiece. Often times, due to the geometry of the workpiece, that proper location is not available especially in forgings, plates etc. That is why we have an inventory of various stainless steel blocks and cylinders of different thicknesses. All are drilled with a 3/8" hole half way deep into that thickness. These witness, or dummy blocks are then interspersed throughout the load. Our SOP at Solar is to run a minimum of THREE work thermocouples in each load. Once all three work Tc's equalize we hold for 30 minutes maximum.
       Feel free to call me at anytime at 724-734-9303. We know the Ellwood Group very well as we are right across the street from your crankshaft facility.

    Bob Hill

    Solar Atmospheres of Western PA

    Robert Hill FASM
    Solar Atmospheres of Western PA
    Hermitage PA
    (724) 982-0660

  • 4.  RE: Determining Heat Treat Soak/Hold Time

    West Michigan Admin
    Posted 09-18-2020 09:42

    Unfortunately the problem is that there is "more than one way to skin a cat." With so many different furnace types, heating types, cooling types, part types, part sizes, part thicknesses, etc. there is nearly an infinite amount of influences on whether or not a heat treatment has been performed properly. In the end, what is desired from the heat treatment is a metallurgical condition that produces a certain set of properties. And as long as you get these properties, you can deviate from the "explicit" heat treat recipe as needed. Typically, heat treatment is judged not from a temperature trace indicating hold time, but whether the parts treated meet various mechanical and or other production testing. For example, while there is a recommended ASTM heat treatment recipe to achieve a T6 temper in aluminum extrusions, what defines whether a part has met the temper is whether it passes ASTM strength minimums. A part that has not met the soak time will not meet the requirements. Further, as the temper property requirements might be conservatively minimum, there is a lot of wiggle room in the heat treatment times and temperatures which can be exploited and still meet property requirements. Knowing this, different OEMs may have extra requirements for how to achieve a T6 temper so that they can meet, for example, crush / formability requirements. And, so, the ASTM and other requirements are merely a starting point for figuring out a good heat treatment practice. The manufacturer must be able to understand his equipment and his customer's needs to decide how he needs to proceed. 

    Aerospace and automotive customers have strict requirements on the quality and consistency of the parts they buy. So they are going to be very prescriptive on exactly how you perform heat treatment and how you verify properties afterwards. AMS 2750 and CQI-9 are the starting point for such tight controls. However, architectural components don't even remotely need the same level of quality control. So, while a Honeywell, GE, or Arconic will specify heat sinks and TCs in hot/cold locations, regular uniformity surveys, as well as corrosion, creep, and tensile tests (and are willing to pay for it), a door-frame part producer only needs to monitor the furnace TC and perform periodic tensile testing to verify you aren't way off.

    In order to reconcile all the possibilities, you need to know what is the importance of the part you are producing and what level of quality you want to be known for. To meet a temper, perhaps you will go with a short time and high temperature to increase production rate. Or perhaps you will go with longer at a lower temperature because properties are better than short/fast. Do properties drop off quickly when you overage? Perhaps run shorter time to prevent risk. Are you running at the low end of the temp range? Perhaps run longer than recommended to make sure your furnace has enough time to get to soak.

    The best recommendation is to make sure your ovens and processes are the best they can be. Pick a pyrometry spec (e.g. AMS 2750) and meet it, even if you don't have to. Which means conduct regular uniformity surveys to ensure your oven is capable of hitting temperature through its whole volume. Conduct regular calibration and SATs to ensure, along with TUS, that your data monitoring equipment is giving you the correct information. I can show you TUS graphs of ovens that look great according to their control TC but are a complete basket case in actual temperature control. 

    And conduct good sampling of production parts to ensure you are really monitoring the process - don't just throw coupons on top inconvenient locations. Those coupons might not capture the fact that your production parts don't get good airflow and aren't reaching temperature.

    Trial your process at minimums and maximums to see what works best for your company (productivity vs. cost), making sure you get good product. Then lock that process in and make sure you have a good quality system in place that can guarantee its repeatability. 

    I also recommend proficiency testing for your testing equipment. It isn't that expensive and it makes sure you are doing good testing and getting good results. Plus, when questions arise about your quality, you have an independent 3rd party to reference that you are doing things right.

    Just my 2 cents (which isn't worth much anymore, I think....)


    David Betz
    Sr. Laboratory Engineer
    Hydro Aluminum Metals, USA