AnsweredAssumed Answered

Someone Please Help Me with Time-Sensitive Simulation!!

Question asked by Erika Nuss on Oct 15, 2019
Latest reply on Oct 17, 2019 by Erika Nuss

Like • Show 1 Like1
Comment • 13

Hello,

Please someone help me with this simulation!

Objective: Overall, I need to find the 3 independent values at which failure occurs due to 3 independent bearing load forces (each with differing accelerations – lateral, vertical, and longitudinal).

Background: This assembly is a 3-piece socket that is recessed into a ceiling panel and secured by rivets. A top of a stanchion will interface with the ceiling socket and the bottom will interface with a floor grating. The stanchion is to hold cargo in place on a ship at sea. We were asked to determine the when socket will failure given 3 different, independent accelerations (lateral - .55 G, longitudinal - .28 G, and vertical – 1.52 G). All loads should be applied to strong axis of the stanchion.

Issues: I am having trouble setting up the parts properly for simulation, determining where the fixtures should be, and how to properly apply bearing loads.

Questions:

How do I setup my parts correctly so that the assembly is not under-constrained or over-constrained? (Make sure my parts are also shelled, solid, surface bodies, etc. where needed)
Where should my fixtures be placed for this application to run an accurate simulation?
Can I apply 3 independent bearing loads (given 3 separate accelerations) into 1 study and have the study give me results for each? Or do I need to create 3 separate studies?
Do I need to add a component to represent the ceiling that the socket is recessed into so that I may add connectors (such as pins) for the rivets?
Should welds be taken into consideration? If so, help with adding welds, as SW would not let me choose the edge where the weld should be.
When trying to find point of failure, do I need to keep adjusting the force value until it says it failed, or will SW automatically keep calculating until it reaches that point of failure?

I know this is a long list, but I this is a time-sensitive project and I am not familiar with SW Simulations. I am learning as I go.

Thanks in advance for the brave soul who is willing and patient enough to help me out!! Please reach out for additional SW documents/files.

-Erika

Attachments

94276-batten socket.SLDASM.zip87.9 KB

No one else has this question

Outcomes

Frederick Law Oct 15, 2019 4:35 PM

You need to include all the parts with the assembly.

Actions

Josh Brady Oct 15, 2019 4:59 PM

If this is a student post, we shouldn't be helping you with your homework.

If this is a real world job problem, I have serious doubts about being able to get the qualified support needed to do this job for free on an internet forum. If your employer has promised this to a customer, they have overpromised and they are inviting a lawsuit or someone to get hurt. This is not a knock on you personally... You just don't have the knowledge and training required to do this analysis, and your employer needs to understand that as well. Sorry, I don't either. I only know enough to know that if you don't know what you're doing FEA can get you into real trouble. Unfortunately a lot of people (probably your boss included) think Simulation is just a plug and play thing that anyone can use. This is NOT the case.

I recommend hiring a qualified professional engineer for this.

Actions

Erika Nuss @ Josh Brady on Oct 16, 2019 8:26 AM

Thank you, Josh. That is very good advice. I have already addressed these concerns. The simulation still must be done. I'm hoping someone's input to my questions may guide me a little better. I'm hoping to attend a simulation class soon to increase my understanding/knowledge of SolidWorks Simulations.

Actions

Erika Nuss @ Erika Nuss on Oct 16, 2019 8:27 AM

Also, I am not a student. I was not that type of student even when I was in school. So, I applaud you for not being interested in helping with a student's homework.

Actions

Frederick Law @ Erika Nuss on Oct 16, 2019 8:34 AM

Simulation is to confirm what you already know.

So first, you need to know how and if the part will fail.

Have you calculated stress of the part in those conditions?

Actions

Josh Brady @ Erika Nuss on Oct 16, 2019 8:54 AM

Erika Nuss wrote:

Thank you, Josh. That is very good advice. I have already addressed these concerns. The simulation still must be done. I'm hoping someone's input to my questions may guide me a little better. I'm hoping to attend a simulation class soon to increase my understanding/knowledge of SolidWorks Simulations.

It's not just a matter of understanding how to use SolidWorks Simulation software. It's about understanding structural analysis, then using the software as a tool to do the heavy lifting. It's like the difference between understanding machine design and understanding how to use SolidWorks software. You could be an absolute whiz at running SolidWorks and design up something that looks pretty awesome on the screen, but it's actually impossible to fabricate or assemble, plus it wouldn't work due to the actual laws of physics. You may end up getting something set up and make some pretty pictures with rainbow colors and lots of megapascally numbers that give management the feels, but it won't actually be worth any more than just taking a sheet of paper and writing the word "yes" on it. I certainly wish you the best of luck, but if "the simulation still must be done" but nobody at your company is willing to put in the resources to do it properly, it really doesn't matter what you do. You've been put into an impossible situation. Even if you get some random yahoos on here to help you make the pretty pictures, who is to say their advice was good? The result you get will not tell you if it's set up correctly or not. Simulation is garbage in, garbage out. If actual decisions need to be made based on these results, and you're the one responsible for those results, and you really have "addressed these concerns" but you're still being made to do this, you need to get out of there as soon as possible.

Actions

Frederick Law @ Josh Brady on Oct 16, 2019 8:57 AM

Josh Brady wrote:
Even if you get some random yahoos on here to help you make the pretty pictures

I can see JD walking away.

Actions

John Pesaturo Oct 16, 2019 8:40 AM

I won't disagree with your comment, there is a lot to be evaluated here ... While this is way over my head if anyone could be of assistance here it would be Bill ...

Good luck with it ... Hopefully someone can get you pointed in the right direction

Actions

Shawn Mahaney Oct 16, 2019 8:45 AM

Thank you for a well written post and a well-posed problem. I'd love to have clients who know enough to ask these specific questions or at least present a coherent problem statement.

I myself got into Solidworks simulation with no formal training and learned a lot the hard way, which took some years.

Everything you're asking you can get through by trial and error, running small experimental studies that focus on each problem. As a head start I can answer each question in turn...

How do I setup my parts correctly so that the assembly is not under-constrained or over-constrained? (Make sure my parts are also shelled, solid, surface bodies, etc. where needed)

One can count degrees-of-freedom carefully, but the best general advice is to set up the problem the way it really is in real life and stop the guess work. This usually involves modeling components beyond the actual article of interest, but the models do not need to be detailed.

As for shells... Solidworks is all about solids. Everything works better with them, with less mental accounting. Stay with solids, even if a coarse draft mesh is required, for as long as possible, at least until a setup is giving reasonable results.

Where should my fixtures be placed for this application to run an accurate simulation?

Wherever they really are, and preferably away from the areas of interest.

Can I apply 3 independent bearing loads (given 3 separate accelerations) into 1 study and have the study give me results for each? Or do I need to create 3 separate studies?

Three studies. Get one running well then 'copy study' and edit the inertial loading.

Do I need to add a component to represent the ceiling that the socket is recessed into so that I may add connectors (such as pins) for the rivets?

This can be done with a reference plane and the 'foundation bolt' connector. But I prefer to make a simple solid model and regular bolt connectors. It's the same effort, similar run time, and better for visual presentation and client understanding.

Should welds be taken into consideration? If so, help with adding welds, as SW would not let me choose the edge where the weld should be.

In limited cases the weld features of Solidworks can be used in simulation. I usually model the welds explicitly, as outlined in Solidworks Simulation for Real Weldments.

When trying to find point of failure, do I need to keep adjusting the force value until it says it failed, or will SW automatically keep calculating until it reaches that point of failure?

The system does not know your failure criteria. For linear static analysis of a single solid response will vary linearly. If you want to see the change a set of studies can be run at increasing load. Setup a sensor or two and you can quickly chart items like key stress or deflection values. (I rarely do this ahead of time, and too often wind up manually probing values from a dozen different runs)

Actions

Erika Nuss @ Shawn Mahaney on Oct 17, 2019 8:04 AM

Thank you very much for taking the time to answer my questions. I greatly appreciate your help.

Actions

Steven McCallion Oct 16, 2019 9:07 AM

This is fascinating. At least the request is nicely formatted though, it's a first to be honest!

Actions

Shawn Mahaney Oct 16, 2019 2:29 PM

I have a long comment which has be stuck in moderation for almost six hours...

In short: what they said.

As for the listed questions, in general be accurate and honest. (specific advice is in my long post)

The details can be learned on one's own, but it takes time, and experimentation with small models one problem at a time.

Actions

13 Replies

Frederick Law Oct 15, 2019 4:35 PM
Mark Correct
Correct Answer
You need to include all the parts with the assembly.
Like • Show 1 Like1
Actions
- Erika Nuss @ Frederick Law on Oct 16, 2019 8:15 AM
  Mark Correct
  Correct Answer
  Will do. Thanks.
  Attachments
  - 94276-socket base_rev.SLDPRT.zip95.4 KB
  - 94276-batten socket tube.SLDPRT.zip67.9 KB
  - 94276-batten socket top.SLDPRT.zip71.2 KB
  Like • Show 0 Likes0
  Actions
Josh Brady Oct 15, 2019 4:59 PM
Mark Correct
Correct Answer
If this is a student post, we shouldn't be helping you with your homework.
If this is a real world job problem, I have serious doubts about being able to get the qualified support needed to do this job for free on an internet forum. If your employer has promised this to a customer, they have overpromised and they are inviting a lawsuit or someone to get hurt. This is not a knock on you personally... You just don't have the knowledge and training required to do this analysis, and your employer needs to understand that as well. Sorry, I don't either. I only know enough to know that if you don't know what you're doing FEA can get you into real trouble. Unfortunately a lot of people (probably your boss included) think Simulation is just a plug and play thing that anyone can use. This is NOT the case.

I recommend hiring a qualified professional engineer for this.
Like • Show 6 Likes6
Actions
- Erika Nuss @ Josh Brady on Oct 16, 2019 8:26 AM
  Mark Correct
  Correct Answer
  Thank you, Josh. That is very good advice. I have already addressed these concerns. The simulation still must be done. I'm hoping someone's input to my questions may guide me a little better. I'm hoping to attend a simulation class soon to increase my understanding/knowledge of SolidWorks Simulations.
  Like • Show 4 Likes4
  Actions
  - Erika Nuss @ Erika Nuss on Oct 16, 2019 8:27 AM
    Mark Correct
    Correct Answer
    Also, I am not a student. I was not that type of student even when I was in school. So, I applaud you for not being interested in helping with a student's homework.
    Like • Show 1 Like1
    Actions
  - Frederick Law @ Erika Nuss on Oct 16, 2019 8:34 AM
    Mark Correct
    Correct Answer
    Simulation is to confirm what you already know.
    So first, you need to know how and if the part will fail.
    Have you calculated stress of the part in those conditions?
    Like • Show 0 Likes0
    Actions
  - Josh Brady @ Erika Nuss on Oct 16, 2019 8:54 AM
    Mark Correct
    Correct Answer
    Erika Nuss wrote:
    
    Thank you, Josh. That is very good advice. I have already addressed these concerns. The simulation still must be done. I'm hoping someone's input to my questions may guide me a little better. I'm hoping to attend a simulation class soon to increase my understanding/knowledge of SolidWorks Simulations.
    It's not just a matter of understanding how to use SolidWorks Simulation software. It's about understanding structural analysis, then using the software as a tool to do the heavy lifting. It's like the difference between understanding machine design and understanding how to use SolidWorks software. You could be an absolute whiz at running SolidWorks and design up something that looks pretty awesome on the screen, but it's actually impossible to fabricate or assemble, plus it wouldn't work due to the actual laws of physics. You may end up getting something set up and make some pretty pictures with rainbow colors and lots of megapascally numbers that give management the feels, but it won't actually be worth any more than just taking a sheet of paper and writing the word "yes" on it. I certainly wish you the best of luck, but if "the simulation still must be done" but nobody at your company is willing to put in the resources to do it properly, it really doesn't matter what you do. You've been put into an impossible situation. Even if you get some random yahoos on here to help you make the pretty pictures, who is to say their advice was good? The result you get will not tell you if it's set up correctly or not. Simulation is garbage in, garbage out. If actual decisions need to be made based on these results, and you're the one responsible for those results, and you really have "addressed these concerns" but you're still being made to do this, you need to get out of there as soon as possible.
    Like • Show 3 Likes3
    Actions
    - Frederick Law @ Josh Brady on Oct 16, 2019 8:57 AM
      Mark Correct
      Correct Answer
      Josh Brady wrote:
      Even if you get some random yahoos on here to help you make the pretty pictures
      I can see JD walking away.
      Like • Show 0 Likes0
      Actions
John Pesaturo Oct 16, 2019 8:40 AM
Mark Correct
Correct Answer
I won't disagree with your comment, there is a lot to be evaluated here ... While this is way over my head if anyone could be of assistance here it would be Bill ...

Good luck with it ... Hopefully someone can get you pointed in the right direction
Like • Show 0 Likes0
Actions
Shawn Mahaney Oct 16, 2019 8:45 AM
Mark Correct
Correct Answer
Thank you for a well written post and a well-posed problem. I'd love to have clients who know enough to ask these specific questions or at least present a coherent problem statement.
I myself got into Solidworks simulation with no formal training and learned a lot the hard way, which took some years.
Everything you're asking you can get through by trial and error, running small experimental studies that focus on each problem. As a head start I can answer each question in turn...

How do I setup my parts correctly so that the assembly is not under-constrained or over-constrained? (Make sure my parts are also shelled, solid, surface bodies, etc. where needed)
One can count degrees-of-freedom carefully, but the best general advice is to set up the problem the way it really is in real life and stop the guess work. This usually involves modeling components beyond the actual article of interest, but the models do not need to be detailed.
As for shells... Solidworks is all about solids. Everything works better with them, with less mental accounting. Stay with solids, even if a coarse draft mesh is required, for as long as possible, at least until a setup is giving reasonable results.

    Where should my fixtures be placed for this application to run an accurate simulation?
Wherever they really are, and preferably away from the areas of interest.

    Can I apply 3 independent bearing loads (given 3 separate accelerations) into 1 study and have the study give me results for each? Or do I need to create 3 separate studies?
Three studies. Get one running well then 'copy study' and edit the inertial loading.

    Do I need to add a component to represent the ceiling that the socket is recessed into so that I may add connectors (such as pins) for the rivets?
This can be done with a reference plane and the 'foundation bolt' connector. But I prefer to make a simple solid model and regular bolt connectors. It's the same effort, similar run time, and better for visual presentation and client understanding.

    Should welds be taken into consideration? If so, help with adding welds, as SW would not let me choose the edge where the weld should be.
In limited cases the weld features of Solidworks can be used in simulation. I usually model the welds explicitly, as outlined in Solidworks Simulation for Real Weldments.

    When trying to find point of failure, do I need to keep adjusting the force value until it says it failed, or will SW automatically keep calculating until it reaches that point of failure?
The system does not know your failure criteria. For linear static analysis of a single solid response will vary linearly. If you want to see the change a set of studies can be run at increasing load. Setup a sensor or two and you can quickly chart items like key stress or deflection values. (I rarely do this ahead of time, and too often wind up manually probing values from a dozen different runs)
Like • Show 3 Likes3
Actions
- Erika Nuss @ Shawn Mahaney on Oct 17, 2019 8:04 AM
  Mark Correct
  Correct Answer
  Thank you very much for taking the time to answer my questions. I greatly appreciate your help.
  Like • Show 0 Likes0
  Actions
Steven McCallion Oct 16, 2019 9:07 AM
Mark Correct
Correct Answer
This is fascinating. At least the request is nicely formatted though, it's a first to be honest!
Like • Show 1 Like1
Actions
Shawn Mahaney Oct 16, 2019 2:29 PM
Mark Correct
Correct Answer
I have a long comment which has be stuck in moderation for almost six hours...
In short: what they said.
As for the listed questions, in general be accurate and honest. (specific advice is in my long post)
The details can be learned on one's own, but it takes time, and experimentation with small models one problem at a time.
Like • Show 1 Like1
Actions

Someone Please Help Me with Time-Sensitive Simulation!!

Attachments

Outcomes

Attachments

Related Content

Recommended Content