Reducing thickness of aluminum bar stock

Tcollins9 · Jun 1, 2017

Hi,

I am having standard dog bone tension specimens manufactured. I am supplying 0.25" thick 6061 Aluminum bar stock to a machine shop and they are going to reduce the thickness to 0.125" and make two holes in the material. Two machine shops I spoke with said you have to reduce the thickness by milling from both sides. They claim that if you don't mill from both sides you are going to cause the material to "pucker" or "camber". Another machine shop claims that you can just mill one side of the material. I want to see if there is any basis for the claim that you have to mill from both sides. The machine shop proposing to mill from just one side is significantly cheaper ($1,000 vs. $2500 for 150 specimens)

Thanks in advance!!

Techguy · Jun 1, 2017

Can you give us some information on the size of the parts besides thickness? How flat do you want them to be and what is your thickness tolerance?

Finegrain · Jun 1, 2017

One thing to consider is what material are you intending on testing -- real-world 1/8" thick Al, or the 1/8" core of 1/4" Al? There may be significant differences due to the properties of the outside material being different than the inside.

Next, if the sourcing shop takes the outside material off of one side only, then you have a mix of inside and outside material. OTOH, if the shop takes the outside material off of both sides, then you definitely are testing the inside material.

Also, is the bar extruded, or cold-rolled? There will be differences there too.

As to flatness, I doubt it makes any difference given what these samples are going to be undergoing -- they'll straighten out PDQ

.

Regards.

Mike

Milland · Jun 1, 2017

I've made many flat and cylindrical tensile test coupons, and what I strived for was uniform test sections, including from the center of larger stock, standard very smooth surface finish, minimal tooling marks in the test and transition zones, etc. Even edge breaks must be uniform for consistent results.

If you're trying for best accuracy and uniformity of results, you need to specify what your surface finish requirements and accuracy requirements are, and if this is standard "off the shelf" rolled 6061 then you should have the tensile area cut from both sides to eliminate any harder skin from rolling. But if you're using a generic Al, why not use 1/8" sheet?

If you can tell us a little more about what the coupons are being tested for and if anything is special about the material we can give better advice.

Tcollins9 · Jun 1, 2017

Thanks for all of the advice so far, I really appreciate your feedback. Some have asked for more information so I will do my best to clarify. We purchased 2" wide by 72" long rolled 6061 aluminum bar stock from four vendors (McMaster Grainger, etc.). We are going to paint all of the specimens with a heat resistant paint and test the material in uniaxial tension at high temperatures. The reason for the paint is we are using a camera to measure strains.

Here is a link to the material: GRAINGER APPROVED Blank,Aluminum,661,1'/'4 x 2 x 12 In - 1NYK6'|'SB-661-

-2-12 - Grainger

We are certainly trying for uniformity of results. Could someone recommend surface finish requirements?

The reason we are using 0.25" sheet is we have different specimens that are 0.25" thick and we'd like to cut both specimens from the same material.

Also, the specimens are 9" long and 2" wide and reduce to 0.5" in the middle (2 inch gage length) with a 0.75" radius

Milland · Jun 1, 2017

OK, if this is a university study then you'll control for metallurgy of each sample, but if this is a commercial venture then you've got to be aware that one material from a generic supplier will not necessarily be the same when it's ordered again, or even the same as the bar sitting next to it in the rack. So if you need to understand material properties that closely you really have to deal with a mill and have standards in place that they'll adhere too.

Beyond that, when dealing with low grade, low care stock (as these bin materials will be) you'll have significant variations in surface condition due to banging, scratching, dents, etc. So for best results you should be either using stock thicknesses (1/4", 1/8") even though they'll be different lots, or start with 3/8" thick stock, so you can use the same lot with a prepared surface for both your 1/4" and 1/8" tensile samples.

So I guess I'm still confused about what your goals are, and what you'll be finding out with your testing...

Tcollins9 · Jun 1, 2017

Yes this is for a university study. You have raised a lot of good points and I'm going to raise these questions when I speak to my PI because I really don't know what we are going for.

Milland · Jun 1, 2017

Tcollins9 said:
Yes this is for a university study. You have raised a lot of good points and I'm going to raise these questions when I speak to my PI because I really don't know what we are going for.

Heh. Years of working with MIT's Materials Science department made me good at asking questions...

Tcollins9 · Jun 1, 2017

I just have a couple more questions. How does the surface preparation affect the tensile strength of the aluminum? Also, is the inside material really that different than the outside material? If you have any literature I could refer to I would really appreciate it. If not then a simple explanation will do for now.

Thanks so much for your thoughtful replies

Milland · Jun 1, 2017

I can't give canonical reasons for why samples are made as they are, but there are references for tensile sample preparation that you can search for on the web. Generally the more uniform and consistent the test area is made, the better (more consistent) the test results will be. For flat materials this will include surface finish, type of tooling marks (longitudinal or cross-wise), flatness, sharp or broken edges, post-prep (bead or shot blasting, anodize, etc.). And of course, basic accuracy of the gage area.

Tensile strength for most aluminums is influenced by cold working, which is more present at the surface of rolled materials. Extruded materials can be more uniform, but a lot depends on subsequent heat treat, and even placement in ovens and process uniformity. Just from anecdotal comments you'll find on PM, there's some thought that generic materials are now more inconsistent in composition and heat treat, especially with foreign sourced metals.

If you're using optical methods for length change recording that should take a variable out, but one thing I remember from early work with a student was watching her apply the (knife edged) extensometer to the gauge diameter I'd polished carefully after long preparation (an aluminum/Si matrix material). The knife edges should have gone into the larger stock above and below the gage area, so I had to explain that and re-do the sample...

Dave G. · Jun 1, 2017

As per Post #10 "...there are references for tensile sample preparation that you can search for on the web." Start with a read of: 1) E8 Test Methods for Tension Testing of Metallic Materials; and 2) ASTM B557 - 15, Standard Test Methods for Tension Testing Wrought and Cast Aluminum- and Magnesium-Alloy Products. These specs will detail if you can use the reduced section sample as you are proposing.

6061 is a precipitation hardened alloy, there should not be any cold work in the bar - the heat treating will have removed strain from cold working. You have not told us the temper you will be testing or the test temperatures. Both would be helpful. Ideally you should have the Mill Test Reports that details the chemistry and heat treatment history of the sample you are testing. It's better to have them and not need them than do all your testing only to find they were required and then need to start over.

As stated above, there can be differences in aluminum stock from surface to core. The changes are due to segregation during solidification (distribution differences in the silicon). The core can have measurable differences in strength. How you can prepare a reduced section tensile bar will be spec'd out in the ASTM specs.

Milland · Jun 1, 2017

Dave, thanks for the additional information. I'm curious about your comment on silicon distribution being the primary reason for strength differences from surface to core of stock, would this account for the significant "potato chipping" that can occur when machining one side of thin rolled Al stock? To me, this seems much more likely to be the remnant of compressive stresses, but I've not got data to back that up.

Bobw · Jun 1, 2017

My 2 cents, for whatever that is worth now a days...

Making a pile of test specimens, at a university.. These are for playing with.. Lab time
etc...

Whenever I've had to make test specimens, they HAVE to come out of the same material as the
parts they are associated with came out of.

Tcollins9 · Jun 1, 2017

Thanks for the input Dave. Just FYI our testing is at room temp, 100, 200, and 300 Celsius. I will have a look at the ASTM specs

litlerob1 · Jun 1, 2017

Bobw said:
My 2 cents, for whatever that is worth now a days...

Making a pile of test specimens, at a university.. These are for playing with.. Lab time
etc...

Whenever I've had to make test specimens, they HAVE to come out of the same material as the
parts they are associated with came out of.

Whoa Bob, I like mine to be made out of ®Delrin, regardless of the process being analyzed. Man I could take this one to a whole new level. It's funny writing that word.

R

Bobw · Jun 1, 2017

litlerob1 said:
Whoa Bob, I like mine to be made out of ®Delrin, regardless of the process being analyzed. Man I could take this one to a whole new level. It's funny writing that word.

The same material, as in the same batch of material.. 3x4" Ti, and have to make qty4 1" diameter
test specimens, plus the parts from the same material.

And YES everything should be made from Delrin, it would make my life a hell of a lot easier.. And if it
can't be make from Delrin, it should be made from 2024 or 7075, and if that's still not good enough,
it should be 17-4 at an H1150. And if that isn't good enough give me something with lead in it.. And
NO 304L is not leaded. Sulfur doesn't hurt my feelings any either (303)..

I could never teach a materials class, we wouldn't have had any space shuttles left.

Dave G. · Jun 1, 2017

Milland said:
Dave, thanks for the additional information. I'm curious about your comment on silicon distribution being the primary reason for strength differences from surface to core of stock, would this account for the significant "potato chipping" that can occur when machining one side of thin rolled Al stock? To me, this seems much more likely to be the remnant of compressive stresses, but I've not got data to back that up.

Depends on the grade. Some are work hardened and will absolutely move if you remove the surface, but this one is heat treated. For 6061 the HT is solution heat treat at just below 1000 F and quench to keep the alloys in solution. 6061 aluminum is SOFT after quenching. Then you age at the right temp (250 to 400F)for the strength/temper you are trying to achieve. That's why I don't think there can be cold working involved. If there's cold work it would be the minimum needed to level the sheet.

The work hardened grades (3000, 5000) will be the ones with surface compression and are going to move when you remove the surface.

Milland · Jun 1, 2017

Dave G. said:
Depends on the grade. Some are work hardened and will absolutely move if you remove the surface, but this one is heat treated. For 6061 the HT is solution heat treat at just below 1000 F and quench to keep the alloys in solution. 6061 aluminum is SOFT after quenching. Then you age at the right temp (250 to 400F)for the strength/temper you are trying to achieve. That's why I don't think there can be cold working involved. If there's cold work it would be the minimum needed to level the sheet.

The work hardened grades (3000, 5000) will be the ones with surface compression and are going to move when you remove the surface.

Interesting. I know I've seen movement with 6061 sheets, not just 3xxx and 5xxx. I did a (very quick) search on the topic but didn't come across anything really helpful, I may did deeper into this to see if I can back up my impressions.

Dave G. · Jun 1, 2017

Tcollins9 said:
Thanks for the input Dave. Just FYI our testing is at room temp, 100, 200, and 300 Celsius. I will have a look at the ASTM specs

You'll need to know the temper of your 6061 bars. The bar should be marked "6061-T6" or something like that. From the temper you can guess the aging temperature. The affect of the test temperature will be the greatest when the test temp is greater than the aging temp. You also need to know the time at temp (during tensile testing) because the strength will be changing continuously when the test temp is above the prior aging temp.

CalG · Jun 1, 2017

Dave G. said:
You'll need to know the temper of your 6061 bars. The bar should be marked "6061-T6" or something like that. From the temper you can guess the aging temperature. The affect of the test temperature will be the greatest when the test temp is greater than the aging temp. You also need to know the time at temp (during tensile testing) because the strength will be changing continuously when the test temp is above the prior aging temp.

Dave

You know that T6 is an age hardening spec. It is inevitable. And needs no Certs except for the time the material has been warehoused.

Reducing thickness of aluminum bar stock

Plastic

Hot Rolled

Diamond

Diamond

Plastic

Diamond

Plastic

Diamond

Plastic

Diamond

Cast Iron

Diamond

Diamond

Plastic

Diamond

Diamond

Cast Iron

Diamond

Cast Iron

Diamond

Similar threads