XANSYS Message: 7178 [Go back to message list]
No rating yet
Rate item:

Subject: Re: Corrupt Database
Author: Mark Rodamaker
Date: 1999-12-22 09:34:00

Whenever we run into a corrupt database we usually take a look at the
.log file( lgwr). In the majority of cases , at least one highly
non-recommended action has been taken. Most common are changing element
types (ie 42 to 45), merging keypoints and not nodes ( always do nodes then
kp's), and Booleans on meshed solid modelling items (user will never admit
to this). In these cases it would be really nice to go back and redo the
model from before the point the mistake was made. For this and many other
reasons we maintain nice clean input files. Of course, there are times when
a corrupt database occurs and no clear reason can be determined.
Disassociating the fe and solid models with modm,deta is probably the most
reliable way of getting something to run. You can create and/or delete
elements manually but remember to get your tough b.c.'s on first because you
can no longer use solid model b.c.'s. When it comes to sacrifice , I
prefer MBA's.

Mark Rodamaker
MCR Associates
Sunnyvale,California
voice: 408-736-1636

----- Original Message -----
To:
Sent: Wednesday, December 22, 1999 10:09 PM

> From: "Crawford, John"

> David,

> How about explicitly creating the element using the EN command, and then
> deleting it using EDEL? I've done this a few times in similar instances
and
> it worked for me. I think that if I had solid model geometry in the model
I
> had to do a MODM,DETACH or MODM,NOCHECK first.

> Of course, as with all problems that deal with database corruption and
hence
> don't follow the usual Laws of ANSYS, the degree of success obtained from
> the work around is often times dependent on whether it is accompanied by a
> live animal sacrifice or not.

> John Crawford
> Honeywell Engines
> Phoenix, AZ
> john.crawford@a...

> -----Original Message-----
> From: Ballenger, David (Southfield)
> To: 'ANSYS Mail List - QnA'
> Sent: 12/22/99 2:32 PM
> Subject: [xansys] Corrupt Database

> From: "Ballenger, David (Southfield)"

> Hi out there in ANSYS-land!!

> OK, here's the story.....

> I run a model, it crashes.

> I get the message (ok, I'm writing this from memory)
> ERROR, element 97053 is not currently selected but was selected for
> the first load case........

> Hmmm....interesting since this WAS the first load case.

> I tried to list the element.....ANSYS would not list it....
> I tried to delete it...ANSYS wouldn't delete it....
> I tried to NUMC, ELEM it out....no good...

> Finally, at the suggestion of my ASD, I "unselected"
> the element for the run.....It worked!

> Obviously, the database is corrupt......

> I tried CDWR & CDREAD. I got an error message when it got to
> the corrupt element, saying all of it's nodes weren't selected.

> So then I tried unselecting the "bad" element, then doing the CDWR
> and CDREAD.....it SEEMS to have worked.

> The question: Is there any other (or safe) way to "salvage" the
> database?
> Remove this "bad element? etc.....

> Merry Christmas!

> David W. Ballenger | Some men have thousands of reasons why they
> cannot
> do
> Honeywell Plastics | what they want to, when all they need is one
> reason they can.
> 248-324-4210 | - Willis R. Whitney

Posts possibly associated with message #7178AuthorDateScore
7174Corrupt DatabaseDavid Ballenger1999/12/22 
7175Re: Corrupt DatabaseBryan Baskin1999/12/22 
7176Re: Corrupt DatabaseJohn Crawford1999/12/22 
7177Re: Corrupt DatabaseSanjeev Srivastav1999/12/22 
7178Re: Corrupt DatabaseMark Rodamaker1999/12/22 
95456corrupt databaseTed Janzen 2008/10/15-10
95460Re: corrupt databaseParis Altidis2008/10/15 
95463Re: corrupt databaseTed Janzen 2008/10/15 
95464Re: corrupt databaseschellapa42_at_tntech.edu2008/10/15 
95470Re: corrupt databaseDan Bohlen2008/10/15 
95473Re: corrupt databaseDan Bohlen2008/10/15 
95477Re: corrupt databaseTed Janzen 2008/10/15 
95480Re: corrupt databaseDan Bohlen2008/10/15 
95482Re: corrupt databaseTed Janzen 2008/10/15 
95483Re: corrupt databaseTed Janzen 2008/10/15 
95484Re: corrupt databaseKrishna Raichur 2008/10/15 
95494Re: corrupt databaseDan Bohlen2008/10/165