Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Choosing materials.

  1. Oct 22, 2008 #1
    A cautious hello to the forum :)

    Im seeking a tool for selecting materials for a given application.

    Im a mechanical eng. student so my real world experience is limited.

    For example:

    torque applied = 0.25Nm
    Must be weather-resistant

    Given this information, how do I narrow my search down from a wide selection?

    There are several plastics and metals that fit the bill, but how and where do I begin my search?

    Thanks in advance, Claws.
  2. jcsd
  3. Oct 24, 2008 #2


    User Avatar
    Science Advisor

    You need to start by understanding ALL if the loads on the shaft. No shaft in a real application experiences pure torsion. There is always a bending component in there as well. You also need to know how it is going to be coupled, i.e. through a key way or splines, etc...
  4. Oct 25, 2008 #3


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi Claws. I don't know of any commercial material selection tool, though I suspect some companies may have some kind of proprietary selection tool. As you must be aware, material selection is done by comparing the conditions and requirements imposed on a part such as a shaft to the properties of the available materials. Conditions and requirements might include, but not be limited to:
    1. Material stresses and strains (including fatigue)
    2. Effects of environment on material such as corrosion, radiation, H2 embrittlement and other long term affects of the local environment.
    3. Effects of temperature on material properties both high and low temperature.
    4. Specific industry requirements for material testing, chemical makeup, material processing or heat history. For example, ASME Boiler and Pressure Vessel (BPV) standards require all materials to be certified to the BPV Code.
    5. Economic considerations such as how the part is to be manufactured (ex: cast, machined, injection molded) and overall lifetime cost of the part (ex: if a 1$ part wears out after 10 days and costs $500 to replace, and stops production of $100 worth of profit, then the cost of the part is $60.10 per day).

    The information you've provided on the part isn't nearly sufficient to select a material. Go through the above list as a start and see if you can better define the conditions and requirements imposed on the part.
  5. Oct 25, 2008 #4


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    I often get caught up in materials selection because I want to find the absolute best material for the application. My progress rate slows down when I do this. So don't be like me. Pick something and try it out; that's how we learn and develop intuition. Aluminum or stainless steel (for metals) and polypropylene or Delrin (for plastics) are easily available, easily machined, and easily evaluated.

    You could probably get all four materials from McMaster-Carr for <$50 in a day or two. Try a couple of them out to evaluate your design and you may discover some constraints you haven't considered yet. (3mm diameter, 500mm length sounds more like a wire than a shaft, so you might check their wire stock as well.)
  6. Oct 25, 2008 #5
    Hi and thanks for your replies. Someone have mentioned CES education pack since Im a student. I will have a look at that.

    I know it is a very general question, thats why I asked it in the first place :)

    Since it is impossible for one person to know all the details of every material, I wanted to know how you guys start out with a broad general definition (like the one I posted) and discover materials you did not know about - and then narrow your search down.

    My problem is, that I mostly know about metals, since thats how my education is formulated. I guess you have to dig deep to find some sort of selection-tree/database.

    ps. I have much more data on the shaft, but it is a bit besides the point :)

    pps. if I was doing my own projects I would have no problem getting and trying out different materials, but school spells data to back up every decision :)
  7. Oct 25, 2008 #6


    User Avatar
    Science Advisor

    This tends to be the crux of many a design problem. IMO this is where experience plays a big part. Knowing the end use and conditions is a must know and why young designers need to work with experienced designers. However, experience will tell you right off the bat what you can get away with. If you really need a starting place, I would suggest, given the info you presented, that you look at the max shear stress and compare that to material allowables. HOWEVER, realize that this step is only the first in many steps before finally choosing a material. In my line of work, we don't get to do the trial and error method because shafts and components are not simple to machine and they take a long time to make. In other words, the material is usually one of the smaller investments.
  8. Nov 12, 2008 #7
    A shaft normally has bearings. To fit ball bearings you better use a metal shaft, rather hard.

    Now if you have sliding bearings, very few materials are adequate. One material, usually the static one, must avoid seizure, and is generally bronze, and then the shaft is chrome-plated or heat-treatment steel.

    Weather-resistant: be very cautious with stainless steel, as seizure is really critical with it, and even when paired with bronze or Cu-Al or Cu-Be, I would keep far away. This holds for hard stainless steel as well: PH17-4 is worse than Aisi304 for that, and even Aisi440C isn't safe. Same for titanium, by the way.

    Nitriding is not a good weather protection.

    So I would chrome-plate a steel shaft to make it weather resistant. The final Cr thickness is generally 50µm, it is ground to the optimum roughness required by the bearings.

    For a short usage, nickel-plating is possible. Nickel can include Ptfe that lowers the friction coefficient, and it doesn't seize as chrome does, so you may avoid bronze then.

    Then, you may choose between rather hard steel (to be machined easily) to very hard if needed (temper after machining if this helps). Remember alternate stress requires specific margins.
    Last edited: Nov 12, 2008
  9. Nov 12, 2008 #8
    I hadn't read L=500mm and D=3mm. Bad news, you can't turn such a thing, and I guess you can't grind it neither.

    I see no obvious solution.

    It must be hard steel as alone the shear stress wants it. But you can only cut your shaft from material at the right diameter.

    I think I would take a piece of precise d=3 hard steel and plate it with 5µm nickel (check how much carbon plating accepts). But remember that the diameter won't be very precise. And don't forget to exodiffuse the hydrogen after plating your hard steel, or it will become brittle.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook