6063 Aluminum Micostructure Question 2

[ahughes71]

I was wandering if anyone could shed some light on an issue that I've been looking into here at work. I have 2 aluminum extrusions (6063 alloy, different profiles / dies but of similar wall thicknesses and mass) and was able to heat treat one of the profiles to a yield 24.7ksi while the 2nd sample only reached a yield of 10.4ksi. Both samples were heat treated at the same time in a small laboratory oven.

Chemical analysis of the two samples result in similar results. The only variation between the samples was Chromium (0.02% vs 0.01%), Iron (0.24% vs 0.21%), and Magnesium (0.48% vs 0.45%). Note that the first number for each element is for the sample with the low yield stress.

Looking at the micro structure at a 500X magnification with no etching yield no noticeable variation in the precipitate size.

Etching the two sample with Etchant 2 per the ASM Handbook again yielded similar results (grains were visible in both samples and averaged 0.005" for the 24.7ksi sample and .0055" for the 10.7ksi sample at 100X).

Etching with Keller's Reagent is where the difference showed up. Again viewing the results at 100X, the sample that achieved 24.7ksi had a well defined cell structure while the 10.7ksi sample had almost no observable cell structure.

I'm attempting to run down the root cause for the difference in yield strength and was curious if the difference between the two samples when etched with Keller's Reagent is significant. And, if it is what does it actually mean?

Alan

 

[CoryPad]

In the unetched state, those are constituent particles not precipitates. You need a transmission electron microscope to see precipitates.

Yield strength comes from the precipitates and from grain size. It is possible that there is a large difference in grain size, and that is what you are seeing when you etch with Keller's. Can you share the images?

 

[Compositepro]

Did you actually place a thermocouple on your parts in the oven, or do a temperature survey on the oven? Most small lab ovens have no blower, and many people who use them do not understand how long it takes for heavier objects to actually warm to oven temp. and reach a stable temperature.

 

[EdStainless]

Do you have samples of the original billets to look at?
There may be a large difference there.
And it wouldn't hurt to HT samples from the billets to see if they are capable of making the properties.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[ahughes71]

Sorry it's taken me a couple of days to get back to this. My spam filters struck again.

@CoryPad: I've attached unetched images along with both the Etchant 12 (I incorrectly said Etchant 2 in my original post) and Keller's Reagent images.

@Compositepro: I have done a temp survey on the oven in the past and it was extremely consistent (it does have a blower); however, I did not actually thermocouple the parts in question.

@DdStainless: I agree the billet could be the issue (along with billet temp and die temp); however, by the time I received the parts and found out there was an issue the billet was history.

 

[ahughes71]

Here are the rest of the images:

 

[ahughes71]

Just noticed I never said this. Sample A is the sample that had the low yield strength.

 

[Maui]

Were both samples etched in the same solution at the same time? I would suggest re-polishing and re-etching sample A. Make certain that the surface has been thoroughly cleaned before etching. Do not let the sample sit for any significant length of time after the final polishing and cleaning steps are completed prior to etching.

Maui

 

[Compositepro]

It really does look like an etching an etching issue. The grain structure is dimly visible under an inadequately etched surface.