Learning Forum

[Kuba]

Hello nWhen I try to perform a pre-stressed analysis of a rotating structure, I get 'unknown error'. In Solution Information there is:nn*** ERROR *** CP = 0.641 TIME= 20:04:02n Element 1 is part of rotational components CM_ROTVE and CM_ROTVE (and n possibly more). Please redefine components (CM command) and make sure n that each element is only part of one rotational component. nn *** ERROR *** CP = 0.641 TIME= 20:04:02n The step data was checked and there were errors found. n Please check output or errors filennWhat are those CM_ROTVE components and how should I redefine components with CM command to get through this? nAnd the second task, how can I find an error file if the directory shown in error doesn't exist? nPlease, help me. nRegardsn

[BenjaminStarling]

The issue is not in redefining your components, but in the scoping of your rotational velocity. As it says in the error message, there are currently two rotational velocities being applied in your model. You must ensure that these rotational velocities do not have overlap in their scoping. If you have already ensure this is the case, there may be some thing odd going on with Mechanical or APDL. If so, attach an archive of your project and I will be able to investigate further.n

[Kuba]

Thank You for reply.n I have one rotational velocity in Static Structural defined as a vector using Components to have several load steps. I want to use the results from this analysis to perform a pre-stressed Modal analysis. So, in Modal I defined the second rotational velocity, which is an identical vector of components as in Static, so I could find natural frequencies including stress stiffening. The rotational velocity has to be applided both in Static, to get stress field and in Modal, to obtain pre-stressed results of frequencies and to plot a Campbell Diagram. Is there any way to do this? n

[Kuba]

could You, please, look at the attached file and tell me what is wrong? I made a simple project just to show what kind of problem I am facing. I just want to include prestress from static structural and get Camplbell Diagram in Modal, that would include centrifugal effects.nn

[BenjaminStarling]

Hi Kuba,nPlease see Section 9.5 of the MAPDL Structural Analysis Guide. This is linked below. This page in particular emulates what you are trying to acheive.nThere is also a whole guide for Rotordynamics you should check out.nThis confused me a bit, as the static reference frame has the omega defined in the perturbation as well as the pre-stress analysis, and the rotating reference frame has the omega defined only in the pre-stress analysis. Not sure what the difference is or if it has any implications, or if it is an error or oversight. Also be aware of the blanked out F command in the static reference frame example.nThe key to your model is the use of the OMEGA vs. CMOMEGA command. In mechanical when inputting a rotational velocity, it can be scoped to a body (or bodies) or it can remain unscoped, referring to All Bodies. The latter leads to the use of the OMEGA command, which can be specified in the Static Structural and the Modal Environment.nIf you have some bodies that are stationary, and some that are rotating, you will be forced to use the CMOMEGA command and you may not be able to find a way around this. Because of this I have actually reached out to my customer rep. for further information. I am somewhat guessing it is a bit of an oversight on the use of CMOMEGA in MAPDL, but I may be wrong. There may actually be a technical reason it cannot be used.nYou will not be able to use the Campbell Diagram feature. You will have to create 3 modal analysis environments that each use a different step of the static structural as it's pre-stress. You will then have to create your own Campbell Diagram in excel (or similar). This is not ideal, but is to be expected as part of the workflow.nYou also included a PSTRES command in a command snippet. Although it seems intuitive, I do not believe this command is intended to be used with a rotordynamics analysis. I ran a couple of examples and the PSTRES command sometimes leads to some strange results in the perturbation. I have also asked for some explanation from my customer rep. regarding this, as it is not clear to me why it is not included in rotordynamics.nI have attached your test model with corrections.nn

[Kuba]

Arraythank You for a comprehensive response. I couldn't open the attachment because I'm a user of a 18.2. release. However, your description encouraged me to try some settings. When I switched to rotating reference frame, it turned out that I am able to plot the Campbell diagram. I didn't try this before, as I have read in some guide that only stationary reference frame gives this possibiity. Anyway, the procedure conducted in rotating reference frame works, but now the problem is that FW frequencies decreases with increase of rotational speed and vice versa:nnI guess I should take Your advice and create a few modal analyses, obtain modes frequencies and draw a campbell diagram by myself in excel, because, suprisingly, this approach provides expected results. nYou have mentioned the PSTRES command. The explanation of use of it gives ANSYS Advanced Analysis Techniques Guide for release 10.. In chapter 8 You can find this information: nHowever, I typed this command to get non-zero stress and deformation in stationary reference frame with included coriolis effect. nI feel quite confused about the results of our investigations on this issue and I guess I am going to try figure something out tomorrow. By now, thank You very much for involvement and wish You Merry Christmas. n

[sbo]

nI try to explained how to make analysis in SRF in following link. you shouldn't add pres-stress due centrifugal force in your analysis but you can add pres-stress due to axial loads. So, if you dont have any axial loads, just do your analysis without static analysis. n/forum/discussion/comment/102561#Comment_102561 n