I want to compare strength of a metal before and after heat treatment, to assess to effectiveness of that treatment. (See my previous question, Aluminium softened by brazing. How to harden?).

Specifically, I want to find the yield point (the strain a material can take without being deformed, ie the range where it remains elastic).

I have two ideas that use a similar setup. Shove a length of material (constant width, height, length, shape) in a vice. Apply a force to the end. Fasten a $1 digital luggage scale to the other end. Hang a container of water or sand, adding a little each time. Deflection could be measured as the distance from that end to some reference point with a digital caliper.

Then, either:

  1. Apply an increasing force, until the force is no longer proportional to the deflection (since up till the yield point, materials bend according too hooke's law). Or,
  2. Apply a force and release. Repeat with increasing forces until the material no longer returns to its original shape.

The problem is, to achieve any real accuracy, both 1 and 2 above would involve 10's of manual iterations per assessment.

Besides buying or hiring a rheometer, are there any practical alternatives?

  • 1
    $\begingroup$ Are you trying to determine bending or tensile yield? I would avoid using vice unless it has very sharp teeth, otherwise the specimen could slip which would render your results useless. I would vote for continuously increasing force by adding/pouring water at a steady flow rate. Any chance you can perform the test on dog-bone shaped specimens? $\endgroup$ May 6, 2016 at 13:49
  • $\begingroup$ Great idea. Then cycles of adding weight and measuring force would be eliminated. Would just need a way of plotting deflection, easily done since cheap digital calipers have a serial data output port, which would just need to be logged. Quick Arduino hack, or a serial to usb dongle should work. Dog bone - why? To make the length of the section under test more controlled? $\endgroup$
    – Jodes
    May 6, 2016 at 22:36
  • $\begingroup$ From what I understand the dog-bone shape somewhat ensures where the specimen will fail, which eliminates problems when your specimen can fail in random places. $\endgroup$ May 9, 2016 at 14:20

1 Answer 1


Determining the yield point for aluminium alloys can be difficult as it is not as well defined as steel and the stress/strain curve tends to gradually peel away from the straight elastic line rather than there being a distinct cut-off.

For a low tech experiment ultimate tensile strength is easier to measure as you are more likely to observe an obvious failure rather than interpolating from a graph.

Similarly using a cantilever loading setup (i.e. bending it with a mass hung off one end) is a bit easier to arrange and should make the failure point a bit more obvious. This makes it more difficult to derive actual numbers for stress but it will certainly make direct comparison easier, especially if you have a known 'good' sample to compare it to.

Another possibility for this sort of application is to use a hardness tester, very accurate one's aren't cheap but it's possible that a cheap one may do the trick, depending on the range you are looking at.

Also if you are looking at a practical performance test for fabricated assemblies then basic load test which is reasonably close to real world conditions is as good a bet as anything. Bearing in mind also that stress raisers from any joining method may have an effect on practical strength so you may as well check if brazing is any better than bolts for your application.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.