Is AUTOSPC the same as “Automatic constraints”?

[hootrpootr]

The title seems pretty self explanatory, but the automatic constraints check box is located in “solution type” —> “solution parameters”.

And if I wanted to enable AUTOSPC through the .bdf, would that just be going to the PARAM section and adding a line for “PARAM AUTOSPC YES”?

I can’t seem to find an answer online. Thanks!

 

[dmapguru]

Yes, the automatic constraints box in Patran equates to the AUTOSPC capability in MSC Nastran.

For the linear solution sequences in the range SOL 101-200, the default for this is YES, so you don't need to add anything into the input file. If you want to define this, PARAM,AUTOSPC,YES would work, but it is deprecated for a good few versions now, and the case control AUTOSPC command should be used instead; something like:

AUTOSPC=YES

The case control command wraps a bunch of other parameters related to AUTOSPC into a single command, so the options on the case control AUTOSPC command let you do things like suppress the singularity table, or only print non-zero stiffness ratios, etc.

For SOL 400, AUTOSPC is ignored for nonlinear steps; singularity processing occurs in the solver. For any linear perturbation steps, you must use the command

AUTOSPC(RESIDUAL)=YES

for singularity processing of the linearised problem.

DG

 

[hootrpootr]

Thanks dmapguru! That makes perfect sense.

You seem to have a deep understanding of Nastran, so I was wondering- could you possibly help answer another question I have?

Not sure if this violates posting by asking here, but do you understand what the use of a .pch file is? If it contains information about the stiffness matrices of specific nodes, can I import that data back into a patran database?

Thanks!

 

[dmapguru]

The short answer is "not usually". The pch file does not typically contain information about the stiffness matrices of specific nodes (GRID points). Although I know MSC Nastran very well, I am no expert in Patran, so I think I am right when I say Patran cannot read the pch file.

You asked "do you understand what the use of a .pch file is?". Yes, but first some history.

When Nastran was first conceived for NASA and subsequently enhanced into the commercial product MSC Nastran, computers at that time did not have graphics capabilities. In the late 1960's and early 70's, input to computers was prepared on cardboard cards and output appeared via dot matrix printers. The cardboard cards were called punch cards, because they had rectangular holes punched into them, each card describing a single entry in the deck of cards which make up what we now call the input file. To this day, we still refer to entries in the input file as cards and the complete input file as a deck. The cardboard cards were read by a punch card reader, but with time and abuse they could become bent or torn. As well as a punch card reader, there was a device for creating the punch cards (the puncher) which had a keyboard so you could manually create cards (hours of fun). This device could also accept input from a computer, so you could easily request an echo of some or all of your deck to be directed to the punch card puncher and obtain crisp new cards of your input. Google "Punched card input/output" on Wikipedia for the full glory.

At this time, online data (as opposed to the offline punch cards) was not stored in files as we know them, it was stored either on magnetic tape or for the high end systems, cylindrical drums. As time went by, these devices were replaced by files in filesystems, so the punch (pch) file is the modern rendition of the information that was designed to be received by a punch card puncher from an echo from the Nastran job.

People have asked "where is the format of the punch file". Well it didn't need a separate description because it was unambiguously defined by the input data, which is a documented format. With the passage of time, the pch file has taken on all sorts of functions and capabilities none of which is in the original design intent. For example, it is possible to obtain results data in the pch file (look at the DISP(PUNCH)= case control command), XY plotting data can be written therein (see the XYPUNCH plotting command). It is also used to output one of the flavours of external superelements, i.e. EXTSEOUT with the DMIGPCH option. In this case, MSC Nastran will write the boundary matrices of the external superelement (including the stiffness if you asked for it) to the pch file in DMIG bulk data format. OTM data for data recovery may also appear in the pch file.

Now I have a question. What are you trying to do when you ask if you can "import that data back into a patran database"? Patran has no way to abstract or visualise stiffness or any pure matrix data.

DG

 

[dmapguru]

Strike that question. I just saw your previous post and I will answer it there.

DG