Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Material Library 3

Status
Not open for further replies.

bigzeke

Mechanical
Aug 22, 2005
9
0
0
US
Our site (part of a large military contractor) has just obtained the necessary network licenses to run Abaqus and several of the MEs are trying to get up to speed, having used the Solidworks FEA for years. It seems odd that this expensive application would be delivered without a basic material library. Is there available a basic material library?
 
Replies continue below

Recommended for you

There are clear distinctions between the approach of Solidworks and something like ABAQUS, which maybe be throwing you off a bit.

Solidworks FEA is designed to be "push button" and aimed at design engineers. There was a vision that FEA could analyze CAD models simply and hence the material database sped this up further. It may have also been there for those who did know where to find material properties or be familiar with them (which is a scary thought). To be frank, this did not work well and any stress engineer can tell you the many problems associated with this approach.

Now enter ABAQUS (or Nastran, Ansys, etc.). A good FEM model and understanding behind it requires a good knowledge of material properties and how they are used. If there was a material library, I think most real stress engineers would bypass it anyway. They want complete control and could care less about built in libraries. You should already be familiar with you material system, its behavior, etc. before you be begin analysis.

So while I do understand where you are coming from, it is not something that I would find useful. I can't think of another true stress analyst who would find much use in a built in library. Everyone would rather enter in their own data, considering effects like temperature, statistical basis, plasticity, etc. This probably takes less than about 1% of the analysis time.

The other thing is that a stress report will require you reference the material property source. I don't think anyone would "believe" the built in code. In addition, how would you know that the material properties are up to date? Most stress reports will have a section that specifically discusses the material properties. This is part of the analyst's job, and not the job of the FEA software.

Brian
 
If you are looking for metal properties, MIL-HDBK-5 has a lot of info. It is free, though the current version is now MMPDS, which is not free.

Composite material properties are becoming more available, but you have to know where to look. Some historical data may be found in MIL-HDBK-17 (now CMH-17). You can search for AGATE as well.

It may be a red flag that the analyst is finding the lack of material properties to be an issue. I hope I am not offending anyone, but Solidworks FEA was never intended for a stress analyst (it was intended for a designer). By the same token, ABAQUS is intended for a stress analyst and not a designer.

Brian
 
I do represent Materality. However, this is a direct response to the question in the forum, and hence, I hope it will not be considered as a promotion.

As partners of Simulia, Matereality offers material libraries that feed seamlessly into Abaqus ( It also allows you to create your own material libraries, and has software that can automate the conversion of raw data into .inp files for Abaqus. There is a free webinar hosted by Matereality for those more interested in this topic:

Creating Abaqus material libraries and .inp files with the new CAE Modeler 4.1 software
The Abaqus CAE Modeler software allows you to take raw material properties in databases on the Matereality cloud and convert it into inputs for Abaqus CAE. Use the CAE Modeler to locate and transform simple as well as complex property data such as high strain rate data, hyperelastic properties and visco-elastic properties into Abaqus material models.

The Abaqus CAE Modeler is available as an add-in within Abaqus/CAE. Abaqus/CAE material libraries and legacy .inp data formats are supported.

FREE WEBINAR: Thursday, March 31
10:00 - 10:45am EDT
 
I have looked into the Materality application, which is somewhat expensive, but does look interesting. You don't package a basic material library for Abaqus?
 
bigzeke: Are you a purchaser of the software or an actual user?

Are you not concerned that the ME's are finding this to be an issue? I don't think a stress analyst would find this to be a problem. In fact, we have existed without it since the inception of FEM. One could make a good argument that FEM is being over utilized, with poor results, because of the approaches such as Solidworks FEA. ABAQUS is geared towards stress analysts, not designers. So I suppose the bigger/better question is should you be using ABAQUS if the skill set is not appropriate? If not up to speed, they why not familiarize the ME's with the proper methods rather than trying to supplement engineering knowledge with software (i.e. a continued extension of Solidworks FEA)?

That being said, I never heard of Matereality, but it looks quite useful, provided you are already competent (i.e. you don't need it, but its nice to have). I am surprised you find the cost expensive since a license of ABAQUS would be far greater.

Brian
 
I am an end user and for most of the structural problems we deal with Solidworks Sim produces accurate results, when properly used, which have been verified with ANSYS and ABAQUS. In an effort to use common engineering tools division-wide, the decision has been made to use ABAQUS and to use Pro-E as the modeling tool. Some facilities in the division have been using ABAQUS for some time and are well over the learning curve. I am the first in this facility to train on the application and did use the Cosmos Geostar application somewhat, so I have some experience with a more comprehensive analysis tool. As I expressed to my functional manager, it is unreasonable to expect mechanical design engineers with limited FEA experience to get up to speed with this product without extensive training and practice, which is not usually budgeted here. I believe that one engineer at this facility will have to be a more or less dedicated analyst and get very cozy with this software.
 
Understandable. Like I said, I can see why you may have seen the lack of material database odd at first, provided you are moving from Solidworks FEA (or the like). It is just a different mentality when dealing with ABAQUS and the like. The user wants complete control of the inputs. In Solidworks the user wants little control. But I really think you won't miss the material library once everyone is up to speed with the materials. I think there is decades of evidence of this. If it were that important (and there was real user need), don't you think all of these really expensive FEM solvers would address it?

Brian
 
Rob, that's a good point. The analyst should understand the reliability of the source and adjust accordingly. Vendor data is notoriously unreliable. MMPDS and mil-5 data should be reliable. Its the job of the analyst to really understand this aspect of engineering. Is a built in library going to tell you that? Or would it just mask these issues? Almost by definition the user is saying " I don't know where to get properties, just put them in for me. "

A possible risk of using a library (and push button FEM) is that it attracts unqualified users. In turn, poor results may occur. Then people stop trusting FEM entirely because of misuse. This then limits those who are truly capable. From what I can see, this is a real scenario. I guess the point is sometimes added convenience is not a good thing after all.

Brian
 
I realize the analysts on this forum are going to argue that FEA should remain the secretive science that it has been for decades, but simple linear elastic studies can be accomplished by competent design engineers in a more automated GUI. Solidworks has done a very good job of creating an intuitive interface for the solver, but has over-reached when including non-linear and dynamic analysis. There is no reason that accurate results for simple analysis cannot be within the reach of any mechanical engineer.
 
For run-of-the-mill bog-standard analyses you can find material properties easily. If you're carrying out more detailed analyses then you'll find that even basic material such as steel has complex properties, particularly at high temperatures. If you can't carry out the material testing yourself then you need good and reliable sources of data. It's better to be confident in your data source than be lazy and just use some data within the software. Once you have your data you can always use a macro to store and retrieve thet data the next time you run Abaqus with that material property.

Tara
 
bigzeke,

FEA is not "secretive science" at all. Rather the distinction lies in the analyst's skill set versus those of the designer. In that sense, FEA may be appear to be "secretive" to a designer.

From your original post, "It seems odd that this expensive application would be delivered without a basic material library."

I don't want to get into a debate about the good and bad of Solidworks FEA solutions (they have their place). Rather, the answer should be clear that analyst's don't want or need material libraries. Since ABAQUS is a software for analyst's, it should not be surprising that it does not have a library (just as ANSYS, NASTRAN, etc. do not).


Brian
 
Again, I represent Matereality, but this is direct response to postings in this thread.

What is expensive depends on how much time an engineer spends looking for data, verifying its credentials, getting it ready for simulation.

Matereality offers CAE modelers that convert raw data into Abaqus models for analysts. All datasets can be evaluated since they are tagged with data certificates. Costs scale down to single user.

There are over 460 FEA/CAE engineers who access Matereality for design-ready inputs, material models.

I will add no more to this forum, unless I see specific questions addressed to me/Matereality.
 
Status
Not open for further replies.
Back
Top