Grid Point Singularity Table - specifically BF set?

[tomsing]

In Siemens' NX Nastran (and I assume other flavors of Nastran), the grid point singularity table outputs some info that I mostly understand, but I wanted to clarify some of it. Can't find great detail in the User's Guide or QRG.

(Apologies if this table comes through oddly spaced.)

0             G R I D   P O I N T   S I N G U L A R I T Y   T A B L E
0 POINT  TYPE   FAILED    STIFFNESS     OLD USET           NEW USET
   ID          DIRECTION    RATIO   EXCLUSIVE  UNION   EXCLUSIVE  UNION
    2      G      2       4.00E-10        BF       F         SB       S    *
    2      G      3       0.00E+00        BF       F         SB       S    *
    2      G      4       0.00E+00        BF       F         SB       S    *
    2      G      5       0.00E+00        BF       F         SB       S    *

[ul]
[li]The Point ID is the grid point ID.[/li]
[li]The Type is ... I can only remember seeing G there, which I assume is for GRID. I suppose you could have a singular DOF on an SPOINT which would presumably show up as S?[/li]
[li]The Failed Direction is, I'm assuming the global DOF which is closest to the principal stiffness direction which has stiffness ratio less than EPPRT.[/li]
[li]The Stiffness Ratio is the ratio of principal stiffness "close to" the failed direction, to max principal stiffness.[/li]
[/ul]

The Old USET and New USET tell me the set of DOFs that the failed DOF was in when it was defined, and the set that it has been moved into (if AUTOSPC constrains it). So, here I'm getting a bit confused.

[ul]
[li]The old Exclusive set is listed as BF, and I'm unclear what that refers to. I assume it's a subset of the Union F superset of unconstrained structural DOFs, but I can't find any reference to it. Seems like would be just the C/B set (I'm looking at static analysis only at the moment).[/li]
[li]The new Exclusive set is SB, the set of DOFs with SPCs applied thru the case control or AUTOSPC. This is part of the Union S superset of constraints. [/li]
[li]The asterisk indicates that AUTOSPC is constraining the failed direction, because the stiffness ratio is less than EPZERO.[/li]
[/ul]

 

[rb1957]

if you want the table to print properly, change to a fixed pitch font, like courier.

without understanding the table (in detail) I think it is saying grid 2 freedoms Y, Z, RX, RY are failing. Now, AIUI, it doesn't necessarily mean this (the problem may not be grid 2 but somewhere close to grid 2). Whatever, you got trouble ! How big is the model ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[ThomasH]

What is the problem you are trying to solve? Or are you just trying to understand grid point singularitys?

If you are working with a model that is currently singular, try calculating the modes. That can be very useful in showing how things are moving and if they are connected.

 

[tomsing]

I thought the tt tag would do a monospace font. Apparently I thought wrong. I don't see a way to control the font otherwise....

Anyway, this is just a toy model, single cantilevered bar element, I use as an example. It's deliberately set up to show AUTOSPC constraining something that shouldn't be constrained, by calculating area in m^2, and moment of inertia in mm^4, or something silly like that. So the principal stiffness ratio is way out of whack, and AUTOSPC picks it up. Then the other DOFs, it's just the normal, "this is a 2d problem so I'm not defining stiffness in this direction" thing.

I'm just trying to ensure I understand all the info in the table, for a training thing I'm putting together.

 

[rb1957]

is the manual or the developer's manual, or the helpdesk no help ?

are you looking at conditioning issues somewhere else in this training ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[tomsing]

Fixed the font with the

 tag....

The users manual doesn't go into a ton of detail about the GPST.  I've got a programmers manual from 1972, which I'm not particularly adept at deciphering.  Haven't tried Siemens help.  Honestly, this isn't particularly critical, it's just something I realized I've looked at repeatedly without really understanding fully.

Yes, we talk about EPSILON and how it's calculated, how that's basically the same as what you did in algebra class, find the solution and then plug it back in to check your work.  I have a spreadsheet example of a simple problem where I round off at like the 3rd decimal place to illustrate the point.  We also talk about pivot ratios and how those are calculated, with an example of an unconstrained rod element.  And we talk about how those almost always indicate problems that need to be fixed in your model, not things that should be addressed by fiddling with the default allowable values.

 

[rb1957]

"looked at repeatedly without really understanding fully" ... too right ! this message would tell me "something's effed up" and I'd put the miner's hat on and go for it. Sometimes it's at (or near) the grid reported, sometimes that's what is in memory when it has it's "nervous breakdown" (and not directly relevant to the problem).

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.