Compressive strength of aluminum

[Rev. Cheeseman]

TL;DR Summary

Do aluminum bars or anything made from the same type of aluminum have the same exact compressive strength for each of them?

Hello,

Do all of aluminum bars (or load cells) made from the same type of aluminum (i.e high purity aluminum or different types of aluminum alloys) share the same exact compressive strength or are there variations in compressive strength even among the same type of aluminum like high purity aluminum or same type of aluminum alloy?

 

[berkeman]

What would cause the differences? Are you asking about small variations in the stress-strain curve due to small imperfections or slight differences in the alloying process?

https://en.wikipedia.org/wiki/Stress–strain_curve

 

[Frabjous]

There will be experimental variability. If you look at the standards, there are generally ranges for various parameters. Some standards are performance standards. Some are compositional standards. Some are processing standards. So the variability can vary by standard.

 

[Rev. Cheeseman]

What would cause the differences? Are you asking about small variations in the stress-strain curve due to small imperfections or slight differences in the alloying process?

https://en.wikipedia.org/wiki/Stress–strain_curve

View attachment 346914

For example, two bars or load cells are made from the same aluminum alloy but maybe one of them is not as dense as the other thus not as strong in compressive strength due to some imperfections.

 

[Rev. Cheeseman]

There will be experimental variability. If you look at the standards, there are generally ranges for various parameters. Some standards are performance standards. Some are compositional standards. Some are processing standards. So the variability can vary by standard.

Therefore, two bars made from the same aluminum alloy should have the exact same compressive strength then.

 

[Frabjous]

Therefore, two bars made from the same aluminum alloy should have the exact same compressive strength then.

No. They will be close.

 

[Rev. Cheeseman]

No. They will be close.

Is it because there is no such thing as 100% similar or something like that? Therefore, those load cells made from the same aluminum alloy have slightly different compressive strength for each of them?

 

[Frabjous]

Is it because there is no such thing as 100% similar or something like that? Therefore, those load cells made from the same aluminum alloy have slightly different compressive strength for each of them?

Yes. The standards allow a range of values.

 

[Rev. Cheeseman]

Yes. The standards allow a range of values.

Porosity will also affect the individual compressive strength too despite made from the same material. Isn't it?

 

[Frabjous]

Porosity will also affect the individual compressive strength too despite made from the same material. Isn't it?

Yes.

 

[Rev. Cheeseman]

Yes.

In what way can we achieve to get two aluminum based load cells to have exactly similar compressive strength if possible in order to avoid inaccuracies in force measurements between those two aluminum load cells? Made with the same compositional standards, performance standards and processing standards?

 

[Frabjous]

No measurement is exact. You need to specify an accuracy.

 

[Rev. Cheeseman]

No measurement is exact. You need to specify an accuracy.

Sorry, language barrier. I don't quite understand the specify an accuracy part. How to specify an accuracy especially in the context of getting exactly similar compressive strength between two bars or load cells that are made from the same type of aluminum alloy ?

 

[Frabjous]

What does “exactly similar” mean?

 

[berkeman]

Sorry, language barrier. I don't quite understand the specify an accuracy part. How to specify an accuracy especially in the context of getting exactly similar compressive strength between two bars or load cells that are made from the same type of aluminum alloy ?

If you want 0.01% accuracy at the same temperature, for example, you would probably choose to do fine-tuning in the manufacturing and calibration of those load cells. This type of process is done all the time in high-precision resistor manufacturing, for example. Can you think of what process you might use in manufacturing/calibration for these load cell bars?

 

[jack action]

Usually, with a quality product, a manufacturer would guarantee a minimum strength value for all bars. So you can expect some bars to be stronger than the minimum value.

 

[jrmichler]

In what way can we achieve to get two aluminum based load cells to have exactly similar compressive strength if possible in order to avoid inaccuracies in force measurements between those two aluminum load cells?

The compressive strength of the material in a load cell affects only the maximum load. Load cell sensitivity is directly proportional to the modulus of elasticity, and the cross sectional area. Both of those have tolerances. The load cell signal is the result of a strain measurement, which is made using strain gauges. Good search terms to learn more are modulus elasticity aluminum and strain gauge. Strain gauges also have tolerances for both resistance and gauge factor. And strain gauge signal conditions have their own tolerances.

The sum of all of these tolerances is normally greater than about 10%. That's why all load cells need to be calibrated, and strain gauge signal conditioners have controls for both zero adjust and gain.

 

[Baluncore]

The characteristics of metals and alloys may change, depending on how the profile was formed, (extruded, hot or cold rolled), and how the crystal structure or phase changes over time.

Most metals can, if needed, be brought to a stable structure by careful heat treatment, chilling, or by simply waiting for a day, a week, or leaving it outside in the weather for a year.

 

[Rev. Cheeseman]

What does “exactly similar” mean?

Lets say, two bars made from the same aluminum alloy has 100% same compressive strength at 600 mpa.

 

[Rev. Cheeseman]

If you want 0.01% accuracy at the same temperature, for example, you would probably choose to do fine-tuning in the manufacturing and calibration of those load cells. This type of process is done all the time in high-precision resistor manufacturing, for example. Can you think of what process you might use in manufacturing/calibration for these load cell bars?

Doing drop tests like dropping a 10 kg object on it?

 

[Rev. Cheeseman]

Usually, with a quality product, a manufacturer would guarantee a minimum strength value for all bars. So you can expect some bars to be stronger than the minimum value.

So, with two bars that are made with the same aluminum alloy, we can't expect the other aluminum bar to have compressive strength with the other one. This makes someone for example me hesitate to buy just one aluminum bar but instead I have to get multiple aluminum bars despite all of these aluminum bars are made from the same aluminum alloy.

 

[Rev. Cheeseman]

The compressive strength of the material in a load cell affects only the maximum load. Load cell sensitivity is directly proportional to the modulus of elasticity, and the cross sectional area. Both of those have tolerances. The load cell signal is the result of a strain measurement, which is made using strain gauges. Good search terms to learn more are modulus elasticity aluminum and strain gauge. Strain gauges also have tolerances for both resistance and gauge factor. And strain gauge signal conditions have their own tolerances.

The sum of all of these tolerances is normally greater than about 10%. That's why all load cells need to be calibrated, and strain gauge signal conditioners have controls for both zero adjust and gain.

Is calibration the same as when we try to drop things on load cell?

 

[Rev. Cheeseman]

The characteristics of metals and alloys may change, depending on how the profile was formed, (extruded, hot or cold rolled), and how the crystal structure or phase changes over time.

Most metals can, if needed, be brought to a stable structure by careful heat treatment, chilling, or by simply waiting for a day, a week, or leaving it outside in the weather for a year.

In order to get aluminum bars that share the same compressive strength, we need to made them from the same aluminum alloy that had been brought to a stable structure by careful heat treatment, chilling and leaving it outside in the weather for some time?

 

[Baluncore]

In order to get aluminum bars that share the same compressive strength, we need to made them from the same aluminum alloy that had been brought to a stable structure by careful heat treatment, chilling and leaving it outside in the weather for some time?

That will depend on you knowing what heat treatment is available, and how the alloy behaves over time.

No two samples of a material are ever 100% identical.
Two samples, that are each within ±1% of the specification, may differ by 2%.

 

[Frabjous]

Are you saying that your application will fail if one bar has a yield strength of Y and the other bar has a yield strength of 1.01 Y?
As a side note, for most applications, one does not want to get anywhere near the yield strength.

 

[jack action]

Two samples will never be alike mainly because of the randomness of the crystal structure defects where impurities, dislocations, and grain boundaries will be slightly different.

 

[jack action]

So, with two bars that are made with the same aluminum alloy, we can't expect the other aluminum bar to have compressive strength with the other one. This makes someone for example me hesitate to buy just one aluminum bar but instead I have to get multiple aluminum bars despite all of these aluminum bars are made from the same aluminum alloy.

If your design requires a compressive strength of 600 MPa for example, and a manufacturer guarantees its product has a minimum compressive strength of, say, 640 MPa, why does it matter to you that the two different samples have, for example, 645 MPa and 657 MPa? They're both the same price and they are also both greater than 600 MPa, thus qualifying for your design.

 

[Vanadium 50]

Maybe you should describe your application. I find it difficult to envision a requirement that two bars yield at the exact same value, but that they application does not require them to yield.

 

[Rev. Cheeseman]

According to this https://www.researchgate.net/figure/Material-properties-of-aluminum-alloys-6061-T6_tbl1_289584991, the strength between these objects made from the same aluminum alloy indeed show variations in strength despite made from the same material.

But as @jack action already said above, both aluminum bars have strength higher than 600 mpa which I think is the minimum compressive strength and share the same price. So I think I just go get a single aluminum bar. Thank you.

 

[Baluncore]

Buy one bar, then cut two from it.

Notice that the -T6 tells you that it has been treated to make it hard and dimensionally stable. Avoid overheating it during cutting or machining.

"T6 temper 6061 has been treated to provide the maximum precipitation hardening (and therefore maximum yield strength) for a 6061 aluminium alloy."
https://en.wikipedia.org/wiki/6061_aluminium_alloy#6061-T6

 

[jack action]

Two samples will never be alike mainly because of the randomness of the crystal structure defects where impurities, dislocations, and grain boundaries will be slightly different.

So in other words, the inner structure is not consistent inside an aluminum bar therefore the compressive strength will be inconsistent along the length of an aluminum bar?

Metal products have defects just like wood does. The fabrication process is more controlled, so the defects are smaller and more uniform, but they are still there.

(source)
​

Metals at the microscopic level:

(source)

(source)
​

Introduction to the wonderful world of metallurgy:

​

 

[Astronuc]

Is calibration the same as when we try to drop things on load cell?

Does one mean dropping a mass or load onto the specimen? A drop test is a dynamic test, i.e., relative high load/strain rate, which will affects test results. Strain rate is a factor in both tensile, compressive and shear testing; some alloys are more sensitive than others.

Re: https://www.physicsforums.com/threads/compressive-strength-of-aluminum.1063600/post-7096651

As jack action mentioned, no two specimens will necessarily test identically for the reasons described. There can be small differences in composition (local inhomogeneity compared to bulk/nominal), which can be minimized for products from the same lot. However, one will find some process (thermomechanical processing history) variability, leading to slight variations in microstructure (grain size, grain orientations, dispersed second phase particles, . . . . ), which will contribute to the variability in testing results. In the case of aluminum, storage temperature and time (ageing) of the finished product is also important.

Factors affecting testing variability include differences in load/strain rate, alignment or misalighnment of grips or loading system, variability in specimen and testing temperature, just to name a few.

 

[Joe591]

It might be useful to make both aluminium pieces from the same bar stock. If you want the properties to be nearly the same. If made from the same bar stock then the same heat number will apply to both pieces so the material properties of both pieces will be as close to identical as you could possibly make it. Then it's down to machining tolerances on the indicidual parts.

 

[Juanda]

Buy one bar, then cut two from it.

Notice that the -T6 tells you that it has been treated to make it hard and dimensionally stable. Avoid overheating it during cutting or machining.

"T6 temper 6061 has been treated to provide the maximum precipitation hardening (and therefore maximum yield strength) for a 6061 aluminium alloy."
https://en.wikipedia.org/wiki/6061_aluminium_alloy#6061-T6

If you want them to be very similar I agree with @Baluncore. That should be the easiest way to achieve it. There will still be some variation as others have mentioned but that's always the case.

Also, as others have mentioned as well, without knowing the application for these components it's harder to assess the best approach. Typically, the small variability is not an issue since the strain limit is often far away from the working conditions. If a 1% variability breaks your system then you probably need to change your design in the first place.

Lastly, as a side note, you've talked about compression and taking it at very high values. Again, I don't know the application but, just in case, I advise you to check for buckling.