Hello guys,
first: my english is not the best.
The english description is under the post and translated with deepl.
Edit: Viele Nutzer (~80) haben den Post bereits gelesen. Eine legitime Antwort/Hilfestellung wäre auch, ich vermute das wird nicht abbildbar sein, weil so und so. Dann müsste ich das Ganze anders darstellen.
Ich habe schon einige Erfahrung mit Inventor, aber leider keine mit Nastran. Ich würde gerne für meine Masterarbeit folgenden Prozess nachbilden und benötige dafür eure Hilfe.
Das Bauteil hat anfänglich eine Temperatur von ~ 20 °C wie die Umgebung. Mein Bauteil wird am Anfang mittig in einem Bereich von circa 300 mm auf 1200 °C erhitzt. Das Bauteil hat auf beiden Seiten einen nicht erhitzen Bereich von circa 300 mm. (siehe Abbildung 1)
(Picture/Abbildung 1)
Ich möchte gerne den Abkühlverlauf für die Oberfläche in der Mitte ermitteln und im Kern. Wir haben bereits Messungen durchgeführt, welchen ich gerne nachbilden würde. Nicht wundern, die Messung wurde immer wieder durch Objekte gestört. (siehe Abbildung 2)
(Picture/Abbildung 2)
Wie würdet ihr das Ganze in Nastran abbilden? Ich habe schon mehere Versuche gestartet, allerdings ohne Erfolg. (siehe Abbildung 3)
(Picture/Abbildung 3)
Vielen Dank im Voraus für eure Unterstützung!! Wenn euch Informationen fehlen meldet euch bitte. Ich wäre echt dankbar, wenn mir jemand helfen könnte, damit die Masterabeit weiter vorwärts geht.
Freundliche Grüße Robin
Edit: Many users (~80) have already read the post. A legitimate answer/help would also be, I suspect this will not be mappable because so and so. Then I would have to present the whole thing differently.
I already have some experience with Inventor, but unfortunately none with Nastran. I would like to simulate the following process for my master thesis and need your help.
The component initially has a temperature of ~ 20 °C like the environment. At the beginning, my component is heated to 1200 °C in a central area of approx. 300 mm. The component has a non-heated area of approx. 300 mm on both sides. (see figure 1)
I would like to determine the cooling curve for the surface in the middle and in the core. We have already carried out measurements, which I would like to reproduce. Don't be surprised, the measurement was repeatedly disturbed by objects. (see Figure 2)
How would you model the whole thing in Nastran? I have already made several attempts, but without success. (see figure 3)
Many thanks in advance for your support! If you are missing information please contact me. I would be really grateful if someone could help me so that the master work can continue.
Kind regards Robin
Hi Robin,
I see that you have the analysis setup and have results. What problems are you having?
=== translation ===
Hi Robin,
Ich sehe, dass du die Analyse eingerichtet hast und Ergebnisse vorliegen. Welche Probleme hast du?
================
John
Hello John,
my solution is garbage. I have a lot of warnings on my temperature/mains settings and my simulation doesn't even begin to match reality. I just want to show how fast the surface cools down compared to the core and that there can be large differences between core and surface depending on the cross section (similar to the figure). I assume that the whole thing shouldn't be that difficult, but nobody has been able to help me yet.
It would be enough for me to see how I could show the proof/simulation.
ISBN 978-3-658-19891-6 Page 216
I am using Inventor Professional 2025, Build: 162, Release: 2025
Kind regards Robin
Hi Robin,
The difficulty is getting the initial temperature distribution. The real situation is heating the model with one set of loads, and then cooling the model with a different set of loads. Inventor Nastran is missing a few features that would make it easy to do in one analysis, so you need to use a couple of analyses to transfer the temperatures. This article describes the process: How to set up a transient heat transfer model with different initial part temperatures in Nastran.
The attached sample ("heating and cooling.zip") includes all the files using numbers that I made up. (100x100 mm steel bar, modeled using 1/4 symmetry, convection coefficients are reasonable for normal applications but not 1000 degrees steel, no radiation which would be important now that I think about it, etc.) Note the modified Nastran file in the "4x4 steel bar > InCAD > FEA" folder where I replaced the uniform initial temperature in analysis 3 (4dqolp6uj.nas) with the nonuniform temperature from analysis 1 (ko6u6gsf7.fno) by transferring them to the intermediate analysis 2 (6asx8zxgn.nas).
Let us know if you have any questions. Feel free to provide your model or details about any warnings you encounter, abnormal results, and so on. (Most likely the abnormal results are related to the extreme temperature difference encountered in the model.)
John
Can't find what you're looking for? Ask the community or share your knowledge.