search

idaholab / moose

Multiphysics Object Oriented Simulation Environment
https://www.mooseframework.org
GNU Lesser General Public License v2.1
1.77k stars 1.05k forks source link

Tutorial Step09 Broken? #11070

Closed permcody closed 6 years ago

permcody commented 6 years ago

Rationale

We need the tutorials to work during training.

Description

Displacement fields in the mechanics example are always zero. One issue we already know about is that the temperature isn't coupled into tensor_mechanics by default, but even after coupling temperature in, there is still no displacement values. What's going on?

While we're at it, I think the restriction of having the axis of rotation being "Y" for RZCoordinates has been relaxed. If that is indeed the case, this example should be updated so that it is consistent with the first 8 steps of the tutorial (step 10 will also need to be updated).

Impact

Embarrassment in training - it's already occurred so the damage is done :)

permcody commented 6 years ago

Also, perhaps this has been broken for awhile. We don't Exodiff these tutorial steps. We only check that they "run".

Tag @dschwen or @aeslaughter or @andrsd.

andrsd commented 6 years ago

While we're at it, I think the restriction of having the axis of rotation being "Y" for RZCoordinates has been relaxed. If that is indeed the case, this example should be updated so that it is consistent with the first 8 steps of the tutorial (step 10 will also need to be updated).

I thought the point of using a different axis in step 9 and 10 was to show how users would go about it and that it is possible. But may be I remember something incorrectly? I can understand that it can be confusing to change this in last 2 steps of the tutorial...

amjokisaari commented 6 years ago

Happy to help on this, but where exactly are you talking about? Mike Tonks and I took a different approach for teaching TM at TMS 2018...

andrsd commented 6 years ago

See #11075 for more details why step 9 and 10 are aligned with y-axis

amjokisaari commented 6 years ago

Looks like you guys are on it, but I think we (I?) need a broader discussion offline about the training materials. Things seem kind of scattered and not up-to-date but it could just be that I'm out of the loop. Tagging @dschwen @tonkmr

permcody commented 6 years ago

The materials didn't seem out-of-date to me, but they are a little scattered. The design of the course is to interleave discussions about math and framework features between the steps of the turorial. We pull live changes of code and we try to keep them fixed up. I did fix several minor things and have a list of more changes that need to be made when I get back to Idaho. However, I'm more than happy to discuss changes, new directions with the training material.

tonkmr commented 6 years ago

Which tutorial are you talking about here? @permcody are you talking about the Darcy Flow tutorial used for the general MOOSE training?

amjokisaari commented 6 years ago

I changed some slides up for TensorMechanics so that 1) it would incorporate the new Master action more easily and 2) be of the appropriate length for our 1-day MOOSE/phase field training. I wonder if we should maintain two sets of training material, one for the longer trainings and one for the phase field short course. Definitely would like to discuss more when @permcody is back.

tonkmr commented 6 years ago

@amjokisaari, I believe that Cody is talking about the general MOOSE training material that they now do around an example problem on Darcy Flow: http://www.mooseframework.org/wiki/MooseTutorials/DarcyThermoMechanical/

How we want to manage slides for the mesoscale trainings is a separate issue, I think, but also one that we need to deal with.

amjokisaari commented 6 years ago

@tonkmr, ah I see now, that was my confusion. I have much to learn, I retract previous comments re: TM!

Agreed we need to address the mesoscale trainings additionally.