Discussion Groups

Robot Structural Analysis

Reply
Valued Mentor
tony.ridley
Posts: 653
Registered: ‎09-07-2011
Accepted Solution

Spectral analysis

680 Views, 18 Replies
03-11-2012 04:12 AM

Hi,

 

When running spectral analysis;

 

I go to Spectral --------->direction definition---------->Resolution of a force into directions--------->Newmark Combination.  I'm only checking X and Y directions, so I choose Group 1 and Group 2.  

 

Document1_Page_1.jpg

 

Robot then automatically assigns 4 CQC combinations as shown in the next picture.  

Document1_Page_2.jpg

 

 

Questions:

  1. Why is only Y direction given a negative sign.  I would expect to see negative X somewhere too.  
  2. When the spectral analysis is run (say X direction only), does the active mass act on the structure in both +X and -X?
  3. If the mass in X acts both positive and negative, then I guess it's the same for the Y direction too.  So why the need for a neg Y case?
  4. If I double click on the generated combinations, I can choose "signed".  I guess this automatically runs the combination for a variety of -ve and +ve signs for X and Y.  If I guess coreectly, then I only need two linear signed combinations and not four, no? (+- 100%x +- 30%y, +-30%x +- 100%y).  If I guess wrong please can you explain it to me.  
  5. I read from help menu that combination sign "main mode" defaults to zero and thus all signs for all modes are taken into account.  By nominating a main mode, am I still getting +ve and -ve signing?
  6. Have I missed something and am completey on the wrong track here?

Tony

Please use plain text.
Valued Mentor
tony.ridley
Posts: 653
Registered: ‎09-07-2011

Re: Spectral analysis

03-11-2012 03:38 PM in reply to: tony.ridley

Question 7:

If I manually define a combination say 1.0x - 0.3y, will I get the same result as the automatically generated case from the Newmark combination?

 

Please use plain text.
Product Support
Artur.Kosakowski
Posts: 5,111
Registered: ‎12-17-2010

Re: Spectral analysis

03-12-2012 12:46 AM in reply to: tony.ridley

Tony,

 

Robot automatically generates only half of combinations (only these with positive X). As these combinations are then intended to be combined with other load cases it is assumed that they will be added with both positive and negative load factors e.g. 1.0* case 7 and -1.0* case 7. The same is done (automatically) when you use automatic code combinations. Masses act in both directions but only one load case for a given direction is generated (only one X or Y) therefore you have to create combinations that have a negative load factors for them. By default when you use quadratic combination formulas the signs of e.g. bending moments will be only positive. To have them 'behaving' as these from 'standard' load cases you have to decide which mode is to be reference for its signing.



Artur Kosakowski
Please use plain text.
Product Support
Artur.Kosakowski
Posts: 5,111
Registered: ‎12-17-2010

Re: Spectral analysis

03-12-2012 12:51 AM in reply to: tony.ridley


Question 7:

If I manually define a combination say 1.0x - 0.3y, will I get the same result as the automatically generated case from the Newmark combination?

 



Yes. There is no difference if a combination is created manually of automatically.

 

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.



Artur Kosakowski
Please use plain text.
*Expert Elite* | International Moderator
Romanich
Posts: 584
Registered: ‎09-29-2009

Re: Spectral analysis

03-12-2012 01:12 AM in reply to: Artur.Kosakowski

Artur,

 

As these combinations are then intended to be combined with other load cases it is assumed that they will be added with both positive and negative load factors e.g. 1.0* case 7 and -1.0* case 7. The same is done (automatically) when you use automatic code combinations.

 

Robot don't use cases 6...9 (linear combination in Tony's example). Only case 4 and 5 will be added to automatic combinations with positive and negative load factors. Maybe it's bug?

 

P.S. If I use quadratic combination, I have the same results.

   
Please use plain text.
Product Support
Artur.Kosakowski
Posts: 5,111
Registered: ‎12-17-2010

Re: Spectral analysis

03-12-2012 01:40 AM in reply to: Romanich

They should be



Artur Kosakowski
Please use plain text.
Valued Mentor
tony.ridley
Posts: 653
Registered: ‎09-07-2011

Re: Spectral analysis

03-12-2012 02:13 AM in reply to: Artur.Kosakowski

To have them 'behaving' as these from 'standard' load cases you have to decide which mode is to be reference for its signing.

 

So if I choose a mode (mode 2 for Y direction in my model) in the analysis type---->combination sign, will my moment diagram behave as i would expect for a regular static analysis?

 

Thanks

Please use plain text.
Product Support
Artur.Kosakowski
Posts: 5,111
Registered: ‎12-17-2010

Re: Spectral analysis

03-12-2012 02:20 AM in reply to: tony.ridley

Correct. Usually the main criterion to select such modes is their contribution to participation mass for given direction.



Artur Kosakowski
Please use plain text.
*Expert Elite* | International Moderator
Romanich
Posts: 584
Registered: ‎09-29-2009

Re: Spectral analysis

03-12-2012 03:13 AM in reply to: Artur.Kosakowski

Thank you, Artur. I've solved this problem.

   
Please use plain text.
Valued Mentor
tony.ridley
Posts: 653
Registered: ‎09-07-2011

Re: Spectral analysis

03-12-2012 03:21 AM in reply to: Artur.Kosakowski

Usually the main criterion to select such modes is their contribution to participation mass for given direction.

 

OK.  So when I put on the results, should I choose mode 2 only, or select CQC, as I have already chosen mode 2 as the main mode / signed mode?

 

Thanks

 

 

Please use plain text.