Reverse engineering of extruded products

[Znjmech]

Hi All

I had an an interview with a company that extrudes some profiles of aluminum. The company was going to improve its work by doing reverse engineering on the products produced in the past, using a FEM software to redesign some new products . I was wondering if someone could tell me what could the details of such work be? since the die design for aluminum extrusion is done mostly by experience. so how would a FEM software help on this issue? I assume that anyway the finite element mrìethod would not precisely replicate the friction and contact

 

[drawoh]

SFTmech,

Are you proposing to use FEM to improve the product, or to improve the extrusion process?

--
JHG

 

[Znjmech]

Not me, that will be the decision of the company that has recently interviewed me. as I explained : improvement of the products, or making other similar ones, I don't know what would be the second case , do you have some idea ? shall you share to discuss?

 

[MintJulep]

Consider posting your question in forum1010 and asking for help in making it comprehensible. Then repost here.

 

[Znjmech]

@Mint : who are you referring to ? if you have seen some trivial grammar errors that you could simply mention here, It doesn't seem a nice recommendation

 

[MintJulep]

The errors are not trivial, hence my suggestion.

 

[Znjmech]

@drawoh : I should add that I have had experience with FEM softwares, and I have simulated some, but about extrusion, I know that If I have to simulate some straightforward reduction, I won't get many results, and instead, If I do with a lot of details, I mean the feeder etc, since I haven't seen similar activities, I was wondering which parameters would be of interest and modifiable using FEM which is not possible to handle via experience

 

[MintJulep]

Possibly English is not your first language.

The relationship between a sentence's Subject, Verb and Object are different in other languages, and this is frequently a source of confusion in the English of non-native speakers.

Another frequent problem is the excessive use of pronouns (he, she, they, it) when the subject of the sentence has not been properly established.

The language forum exists to help with this. Make use of it.

There seems to be a good and interesting question here, but it is not clear.

 

[Znjmech]

Ok , I surrender, I posted in the mentioned group

 

[CWB1]

IME I would be amazed if any semi-modern shop was relying purely upon "experience" for designing process or product. I have no experience with creating aluminum extrusion, however forging shops and steel rolling mills I've worked with both have used specialized software based upon common FEA/FEM. Realistically, few outside of that niche industry will have experience with the necessary simulation tools so no worries about your own abilities, let them teach you what you need to know to do the job properly.

 

[GregLocock]

Hmm. Having dealt with an aluminium extrusion shop I was pleasantly surprised by the amazing tolerances they could hold (basicallty we had a semi-cylindrical shell, 2mm wall thickness 50mm OD. The two halves snapped together.) That was all designed by hand and experience. At the time (mid 90s) I don't believe they'd have used FEA, and since the tooling was only $2000 I can't believe there were any iterations to get to a good design.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Znjmech]

CWB, I can't believe that, even the small shops? I wish you could share more details. But thank you anyway. There's one point that I'll be the only engineer and the others are technicians, and they haven't been using such software. I'm gonna start to do that, if they hire me

Greg, so you're saying that a FEM isn't required? Maybe it's a piece that they know how to design its tooling over the time?

 

[GregLocock]

I'm not saying it couldn't be used usefully, for instance you might find you could reduce tool wear even if the section that was produced was no better than a hand analysis method. The equivalent in injection moulding, Moldflow, is nigh on indispensible. There again injection moulding tools are far more expensive than extrusion dies.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Znjmech]

I was mostly wondering, whether all of the work would be focused on bearing length and shape of the die, or also the feeders would be included? Since I have not a clear idea how the material flows in the feeder to get mixed and become homogeneous before entering the die, I can't predict if the simulation would be feasible

 

[Jboggs]

Here's a different take: I don't think that were planning on using FEM for the die design at all. I think they want to use FEM to learn more about the strength, stiffness, and inherent weaknesses of their extruded products, which in turn could lead to "tweaks" in the die designs to improve the performance of their products.

 

[CWB1]

SFT,
You'd likely be amazed by the technology most small shops have today despite what they can do without. Production is driven by cost and quality, and the cost of FEA or other CAE tools is pretty cheap compared to the potential savings when it leads to a few product or production improvements. Sure, most analysis done today via CAE was once done via hand calcs and experience gained through iterative testing, it was also relatively inefficient as to people, time, and material. Not to say many businesses don't get by without modern technology but as the manufacturing world becomes increasingly competitive its getting pretty difficult to do so. I've seen far more shops go under than survive because they failed to modernize.

 

[ThomasH]

Hi
I do not fully understand the question. But if I assume that the work means analyzing extruded aluminum profiles to study how they perform and investigate possible optimization. That type of work probably means modelling the profiles and analyzing them for different load cases. For that type of work I would consider FEM is an exellent tool. Regarding friction and contact I would say that it can be simulated with the appropriate software and a competent user.

On the other hand, if the question relates to analyzing the extrusion process itself, that is not my area at all. I know that the metal forming of steel plates to different shapes (like car body's) can be modelled but regarding extrusion of aluminum, sorry. But I would not be surprised if it is done.

But I would not call what I described "reverse engineering" so I may have missunderstood completely.

Thomas

Edit: When I read some later posts my impression is that the work probably relates to the extrusion prosess. So just ignore my comment, sorry.