Stock material flaw?

[5teveW]

Hi,

I've had a some turned parts come in the from a supplier, made from Aluminium 6082-T6. These parts provide a seal in an oil filled vessel. On assembly some oil was noticed leaking from a tiny witness line, and curiosity led to putting the part in a press and this happened... See pictures

http://files.engineering.com/getfil...-512e-4508-8830-f984dc392af8&file=QCpics1.JPG

http://files.engineering.com/getfil...-74b5-487b-95c6-36f29b01cfd0&file=QCpics2.JPG

http://files.engineering.com/getfil...-501a-40f8-8612-5828dffc1090&file=QCpics3.JPG

The profile between the two parts in slightly tapered, and slightly oval. Can anyone shed some light on this for me please? My thoughts are that this is a flaw in the stock material, where the material has not bonded during cooling, but having been drawn it is forced together. Would our supplier notice this flaw in the material during machining or inspection?

Thanks,

Steve

 

[IRstuff]

I'm not sure I understand what you're asking. Are these two parts supposed be separate?

Or is it that the fit tolerance isn't what it's supposed to be?

TTFN

FAQ731-376

 

[MikeHalloran]

I can't say that I've ever seen a material flaw quite like that.

I _have_ seen similar parts made from a piece of plate and a piece of bar to save machining time (or if the bigger bar was unavailable), but tapering the joint is unusual, and most vendors will ask before doing something like that.



Mike Halloran
Pembroke Pines, FL, USA

 

[Timelord]

Steve,

If I understand you correctly, there was a seam in your raw material that should not have been there? I would get back to the shop that made the parts and ask if it was made from extruded round. I am not that familar with 6082, what forms is it available in? A wild guess is that it was extruded along the axis and the seam is an extrusion artifact. See if the shop has any more of the material and even have them get back to their material supplier to check out if the seam runs thru the whole bar. I would say this is sure looks like flawed material. Another remote possibility is the shop pressed together two pieces to make your one part but this seems to be a lot of extra work unless the material is real difficult to get.

My 2 cents worth,

Timelord

 

[hydtools]

If I understand the pictures, the 'joint' is the flaw. The flange separated from the hub when the part was put in the press.

I have seen flaws that ran lengthwise but none that formed a ring like this. Probably the result of a circular inclusion that got extruded the length of the bar in the rolling or extruding process. I would have expected the machine shop would have seen unusual rings peeling off as they cut into the flaw. It would be worth asking them.

Ted

 

[KENAT]

A previous employer in the UK used a lot of 6082-T6, including extruded rings as I recall.

KENAT, probably the least qualified checker you'll ever meet...

 

[5teveW]

Hi Guys,

Yes, it should be one part, but under light pressure in the press the flange came away, but the fault line does not follow the hub/flange change in cross section, it's a little wider, and on other parts it is within the hub.

I think the stock material is extruded or drawn bar along the axis of the parts. And this is a possible inclusion.

Our stores has no batch identification number, so we cannot be sure which supplier they came from. I've spoken with both of them, but they are both a bit cagey and suprised.

I'd like to find more information about whether this flaw, if it is that, is noticeable in stock material or during machining. As I'm wondering if we have been supplied with questionable parts before, and we have missed it and it has made it to the customer.

Cheers

 

[CorBlimeyLimey]

Contact some extruders and get their take on it.

 

[hydtools]

An ultra-sound inspection may show the flaw. If the flaw runs the length of the bar, it may be visible on the bar end or on the face of a parting cut.

Ted

 

[m1ckvb]

I have seen similar things happen with CRS where the material was porous allowing Helium, under a sealed condition, to be drawn through the material. This is certainly a problem that is caused by improper crystaline structure resulting during its manufacture. For our case, we were able to flash a coating of copper over the material which provided a seal. This by no means fixes the problem but when sending it back to the supplier is out of the question you have to do something to make it usable.

 

[Tmoose]

I'd check some finished parts with a NDT dye penetrant like spot-chek.

From here It sure looks like they pressed/shrunk/loctited pieces together

 

[tripleZ]

I'm inclined to agree with Tmoose (if I'm seeing the pictures correctly). I'd tell you to check for a few things:

1) Size. We shrink fit die inserts into cases all the time with similar OD/ID shoulder configurations. The application is different though (it's common in my field). Take the hub diameter. If it was a shrink fit, figure a .001" per inch of hub OD. See if the ring ID of the flange correlates to this.

2) Use a 7X + power eye loupe and look for evidence of grind lines or some other remnanat machining.

3) Look for any discoloration (blueing or gold) in the ID of the flange that might indicate the flange was heated.

Having an ovular hole is not uncommon when you pull apart thinner flanges that have been shrunk fit to a hub. If they put the flange on with a press fit, you might have a difficult time seeing any machining lines if you scored the flange pressing out the hub.

 

[IRstuff]

Seems pretty oddly coincidental that the "flaw" is circular AND concentric with your desired circular features AND is pressfitable.

Just what are the odds of something like a circular flaw occuring in normal stock?

Have you measured the concentricity of the "flaw?"


If it's within machining tolerances, you've got the perfect scholarly article candidate for the "Journal of Irreproducible Results."

TTFN

FAQ731-376

 

[CoryPad]

Circular flaws are not common for billets (cast or extruded). If I understand the images correctly, it looks like the press force produced a stress that exceeded the strength of the material.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[5teveW]

CoryPad, IRstuff, I think you have misunderstood the images. It took a tiny force to mislodge the two parts, I could have done it in my hands. The fault line does NOT follow the change in cross section, is NOT concentric, is slightly oval, and tapered. There is another part where the fault line is through the middle of the hub, and another in a differnt position on the flange.

I can't see anything I would recognise as a machining mark on the broken surfaces, but there are marks I am not familiar with. I've done a lot root cause failure analysis work and seen all manner of sheared parts and fatigue cracks and have not seen surfaces like this. As you can see in the photo there is almost a grain to the surface like on wood.

I am still wondering if the supplier fitted a bar and hollow bar together somehow, but next step I'll contact some extruders and see if they can shed any light on it.

Thanks.

 

[texag]

I'll go ahead and come to the conclusion that most everyone here already has. The manufacturer tried to save some money by making the inner piece from bar and the outer flange from plate - then thermally press fitting the two together.

Not that there is necessarily anything wrong with this but you might want to get them to share the exact procedure. You may find some problems associated with how they made the fit, coeff of thermal expansion, stress from press fit, material strength, etc... Modulus of elasticity for aluminum is about 1/3 that of steel, tells you something right there.

My opinion, this is not a material issue, that would be the least probable cause for this failure. I just can't picture aluminum being a good candidate for a press fit assembly.

 

[5teveW]

I doubt he could have saved any time and money in fitting two parts together. The ovality of the surfaces, irregualrities and taper are very complex. None of the machinists I've asked can guess at how it was created by normal manufacturing techniques. (except possible casting one into the other.)

Please look closer at the pictures. The break does NOT follow the change in cross section between flange and hub. And one of the components a huge chunk fell out of the hub itself? I'll post another picture on Monday of that one.

There is something wrong with thermally pres fitting two parts together if it is not to my drawing. With all the info I have I still wouldn't dare draw a conclusion - I'm an engineer and have to use good science. I just thought somebody here may have seen it before. Next step I'll ask stock material suppliers.

 

[hydtools]

I would have a metalurgical examination done on the material. Especially the suspect surface for material that should not be present. The banding is suspicious.

Ted

 

[5teveW]

Images of the other two parts, which leads me to suspect they were not manufactured in two pieces. Also, it is possible to put them back together perfectly, which also leads to believe it is not an interference fit.

http://files.engineering.com/getfile.aspx?folder=9d0c326e-a78b-463c-98da-d4af2184f5ff&file=QC1.JPG

http://files.engineering.com/getfile.aspx?folder=f6a71819-9e32-4fc7-be0d-a0066695157a&file=QC2.JPG

http://files.engineering.com/getfile.aspx?folder=21341a82-1304-4737-a905-838e4da87ba0&file=QC3.JPG

http://files.engineering.com/getfile.aspx?folder=5512b28d-a910-4c3f-ba7d-919cc1304036&file=QC4.jpg

Ted, I will get a metalurgical examination done as well.

 

[SnTMan]

5teveW, I am far from my field of expertise (if any) here, but could these be powder metal parts that were not properly consolidated / sintered?

Regards,

Mike