When I calculate my modell i get the following error: instability of type 3 in RX direction
I have looked at the node but everything seems to be correct. I the releases box I have tried to tick off RX, but that did not help.
Solved! Go to Solution.
Solved by Rafal.Gaweda. Go to Solution.
Open calculatin Report , find there Max min values on stiffeness matrix diagonal and write here these two values.
if the difference between these values is greater than 1e11 \ 1e12 program reports instability number 3.
It is related to big differences in stiffenes of elements for example very short elements exist in model = some inaccuracies.
You can check it this way: on dummy model select everything, make intersection, then open Bars table on View tab. Find shortests bars. Then locate nodes \ places on original model and correct location of nodes.
If it is not the case. Send me model. You already know how.
Hi,
please find attached 2 files.
https://dl.dropboxusercontent.com/u/79282449/AUVENT_V2.zip
the first, AUVENT_V2, is calculated without sismic load and is free of instability.
the second, AUVENT_V2_SISMIQUE, contains sismic loads and shows instability.
Calculation report :
Elém. diagon. de la matrice de rigidité
Min/Max après décomposition: 4.067231e+002 2.066486e+015
Précision: 2
Shortest calculaltion element : 0.16 m -> I think it's not too short
- Why?
- How be sure than we can ignore this instability (after checking node displacements, shorts elements)?
- Is it possible to highlight calculation node?
thanks
benoît
the first, AUVENT_V2, is calculated without sismic load and is free of instability.
I do not think so :
Ignore it. Related to solver tolerances.
To check - use Skyline solver -> no instability.
For me, "AUVENT_V2" has no instability (with automatic solver). I can't make a snapshot since there's no message 😉
For "AUVENT_V2_SISMIQUE", only multithread method gives no instability.
-> How is it possible to justify that the instability can be ignored?
-> How is it possible to highlight the calculation node wich is going wrong?
thanks
benoît
-> How is it possible to justify that the instability can be ignored?
In this case - change solver onto Skyline ;
Create new case apply nodal load on direction given in warning message, after caclulations check displacement and rotations in this node
-> How is it possible to highlight the calculation node wich is going wrong?
In general this way starting from v 2014:
(In your model there is nothing wrong in this node - see above)
Create new case apply nodal load on direction given in warning message, after caclulations check displacement and rotations in this node
-> what must be the result? no displacement or rotation?
Benoît
Had the Same Error, But on verification it doesn't appear.
See attached Screenies
Dear Rafal,
I have the same issue here. can you check my model please?
Can't find what you're looking for? Ask the community or share your knowledge.