mechanica - assemblies

[ptc-2404135]

I'm trying to do a fairly simple bending analysis of an assembly. The assembly consists, basically, of two tubes. One fits snuggly (but not press fit) inside the other. The outside tube is PEEK (polyetheretherketone - a sort of plastic) and the inside tube is 316 steel. The steel tube inserts to about 80% of the length of the PEEK tube, so about 20% of the PEEK tube's length is not supported by the steel... sort of cantelevered. I want to fix the end of each tube in 6dof, and then apply a bending load to the free end of the catelevered section of the PEEK tube. The PEEK tubes are failing near where the steel ends, and we want to see what sort of load produces yield stresses & where they are localized. I've attached a screenshot of the setup from Mechanica. The steel tube is outlined in red. Whenever I run this analysis, the PEEK tube bends and the steel tube remains unloaded. 10% Scaled deformation shows the PEEK tube bent and the steel tube straight - poking out through the PEEK tube, and the stresses on the PEEK tube are localized where they would be without the steel tube present. So I assume I'm not constraining this properly, but I don't know how to get it right. I've read through much of the help docs, but can't figure out what sort of constraint I need to use. As it stands there is a very small but real clearance in the assembly model between the two tubes (i.e. they do not touch). Any help/advice would be greatly appreciated!
This thread is inactive and closed by the PTC Community Management Team. If you would like to provide a reply and re-open this thread, please notify the moderator and reference the thread. You may also use "Start a topic" button to ask a new question. Please be sure to include what version of the PTC product you are using so another community member knowledgeable about your version may be able to assist.

[RickGiguere]

There is no force(load) applied to the steel, that's the first problem I see, you need to define the part as one piece consisting of two different materials. Rick

[ptc-2404135]

Thanks, but how do I do this? If I make both bodies one solid in Pro/E-std, then how do I create two sections of different materials? Or, is there a way to make mechanica treat/analyze them as one solid but still be able to address them seperately for material assignments? I thought that I read that if two solids in an assembly share a face, then Mechanica will treat them as one solid for analysis. But I couldn't get this to happen. But regardless, I don't know that I want them to be treated as one solid... the interface between the PEEK and steel is fairly low friction. Thanks again for the help...

[RickGiguere]

You have the tubs fully constrained in assy. mode and not just packaged? The load is being applied to the plastic tube in the drawing, Mechanica is not recognizing the steel, there could be a number of reasons for this, I couldn't get into it on the internet. I would have to go through some notes to tell you how to create one part out of two materials. To start, make sure everything is fully constrained in assy, it will take more than two surfaced being constrained. If that doesn't work you'll have to figure something else out.

[ptc-2404135]

"Richard Giguere" wrote:

You have the tubs fully constrained in assy. mode and not just packaged?

[ptc-2404135]

Okay, I found the help topic on "packaging". I also discovered the place where Mechanica tells you if a part in an assembly is fully or partially constrained. So, the steel tube was not fully constrained. Now it is, as is the PEEK tube. But I still get the same results when I run an analysis.

[RickGiguere]

Hello Keiran Since the fully constrained method of an assy. did work, if it was me I would redefine the assy as on part(re created it as one solid piece). There is a way to assign different materials to different areas of a part, I've done it, I can't tell you off the top of my head how I did it, but it wasn't that hard. Then the part will be one solid piece and you should get a good FEA. FEA is kinda hard and I try not to get into on the net. Knowing that you got a correct result takes a lot of experience(which I don't have). have fun regards Rick

[RickGiguere]

I knew I was going to get the e and i backwards on your name...sorry about that.

[RickGiguere]

"Richard Giguere" wrote:

I knew I was going to get the e and i backwards on your name...sorry about that.

[RickGiguere]

Here is a screen shot of a part with two materials. The help files should have something on this. Hope this helps lator rick

[AndyDeighton-di]

In Mechanica, you can split a single part into two volumes (Insert > Volume Region) and assign different materials to each volume. However, I don't think this is the best solution for your case - the assembly you have should be fine. If the two parts are of the same diameter then Mechanica will assume they are bonded. If there is a gap between the diameters, or you want to allow the parts to slip where they contact, you need to define contact regions between the parts: WF4 - Insert > Connection > Auto Detect and Create Contacts When you set up the static analysis, you will find that "Include Contacts" is selected. The analysis will take longer to run than a normal static analysis. I hope this helps.

[RickGiguere]

"Andrew Deighton" wrote:

In Mechanica, you can split a single part into two volumes (Insert > Volume Region) and assign different materials to each volume. However, I don't think this is the best solution for your case - the assembly you have should be fine. If the two parts are of the same diameter then Mechanica will assume they are bonded. If there is a gap between the diameters, or you want to allow the parts to slip where they contact, you need to define contact regions between the parts: WF4 - Insert > Connection > Auto Detect and Create Contacts When you set up the static analysis, you will find that "Include Contacts" is selected. The analysis will take longer to run than a normal static analysis. I hope this helps.

[RickGiguere]

It sounds like you are saying, if the two diameters are the same, a fully constrained assembly will work in Mechanica, but if they're not and you want them to move or have a gap you need to do the "Insert > Connection > Auto Detect and Create Contacts" before you fully constrain the assy. just curious what you mean, this new to me. Rick

[AndyDeighton-di]

Mechanica pays no attention to how things are assembled in Pro/ENGINEER - it sees the parts as seperate bodies. The bodies are joined if they have coincident surfaces and the default connection is bonded, or you can make connections - links, bolts, contacts etc.

[AndyDeighton-di]

If your assembly contained two gears with a gap between the teeth, you need to define contact regions between them to pass the load from one part to the other, the same in this case with the shafts, assuming there is a gap between them.

[RickGiguere]

"Andrew Deighton" wrote:

If your assembly contained two gears with a gap between the teeth, you need to define contact regions between them to pass the load from one part to the other, the same in this case with the shafts, assuming there is a gap between them.

[AndyDeighton-di]

It makes no difference - constrain first, contacts first....

[RickGiguere]

"Andrew Deighton" wrote:

It makes no difference - constrain first, contacts first....

[RickGiguere]

"Andrew Deighton" wrote:

It makes no difference - constrain first, contacts first....

[ptc-2404135]

Sorry for not updating this thread. Short story: with the contact connection defined between the two cylindrical surfaces, the simulation ran successfully. Long story: while trying to figure out a solution, I needed a quick & dirty result. So I set the o.d. of the inner tube equal to the i.d. of the outer tube, so that the two surfaces were flush (no gap). When you do this, Mechanica meshes the assembly as one solid, but still allows you to define different material properties for each part. This allowed the analysis to run successfully, without any further constraints needed (asside from those already described in my earlier posts). The problem with this solution (asside from that the dimensions were not real any more) is that the interface of these two surfaces is then assumed to be bonded (no slip). This was not realistic for my assembly. Anyway, with the contact connection inserted, it worked great, and slippage and gaps were allowed between the two surfaces, and loads were trasmitted between the two parts.

[RickGiguere]

"Kieran Coghlan" wrote:

Sorry for not updating this thread. Short story: with the contact connection defined between the two cylindrical surfaces, the simulation ran successfully. Long story: while trying to figure out a solution, I needed a quick & dirty result. So I set the o.d. of the inner tube equal to the i.d. of the outer tube, so that the two surfaces were flush (no gap). When you do this, Mechanica meshes the assembly as one solid, but still allows you to define different material properties for each part. This allowed the analysis to run successfully, without any further constraints needed (asside from those already described in my earlier posts). The problem with this solution (asside from that the dimensions were not real any more) is that the interface of these two surfaces is then assumed to be bonded (no slip). This was not realistic for my assembly. Anyway, with the contact connection inserted, it worked great, and slippage and gaps were allowed between the two surfaces, and loads were trasmitted between the two parts.

[ptc-2404135]

I had both parts fully constrained in assy mode. I then constrained one end of each tube in 6dof as shown in the above images. I then defined a contact connection between the o.d. surface of the inner tube and the i.d. surface of the outer tube. That's all. In the attached image, you can see that there are varying gaps between the steel (inner) tube and the PEEK (outer) tube.