I notice in the basic demo video that both node sets and surfaces are set up for where load will be applied to the model.
What is the role of each of node sets and surfaces - what is distinct about each, do I need to define both for every region where I’ll apply load, restrain, set up an interaction??
I’m finding the terminology and usage in PPM very different from other FEA systems that I’ve worked with.
The nomenclature is the same as in Abaqus. They are used for different purposes. Node sets are needed for boundary conditions, initial conditions and loads applied directly to nodes (such as concentrated force). Surfaces are needed for distributed loads (such as pressure) and interactions like contact. However, you don’t have to create them manually at all. They are automatically created when defining the aforementioned analysis features.
Never used Abaqus - which may explain my puzzlement. The systems I’ve worked with don’t differentiate - everything is done using surfaces, there are no ‘node sets’.
I don’t understand what you mean by
They are automatically created when defining the aforementioned analysis features.
I can’t reconcile that with the basic demo video. Unfortunately the videos I’ve seen so far just show a series of steps, there isn’t much explanation of what is going on.
When you define an analysis feature, internal sets/surfaces are created for use in the CalculiX input file. You won’t see them in the tree though. Basically, you don’t have to worry about sets and surfaces unless you want to create them manually for some reason. For example, to have everything ordered and named as you wish instead of relying on the default names - especially important when working with input files directly. Or when using the keyword editor to add unsupported CalculiX features.
So some of the steps in the tutorial video are in fact ‘optional’ ? That was slightly misleading.
I can see some value in using a surface or node set to group things together - for example multiple bolt head contact faces, that will touch a common part face.
If I’ve previously defined these as a node set (because I didn’t know better), is there an easy way to derive ‘surface’ from that, or do I have to repeat the selection of faces…? It’s OK either way - I have to learn if I’m to get this to work.
PPM has been able to mesh parts that my previous system fails with.
I don’t know which tutorial you are referring to. In mine, I mention those two alternative approaches. This is also mentioned in the user’s manual.
Another example when predefined sets/surfaces are useful is when something is difficult to access. For example, you want to create contact between the parts and have to hide some to select the faces covered by them. So you do that, create the named surfaces and then use them to define contact pairs (although there is a Search Contact Pairs tool that can do that for you in most cases but some manual control is often desired).
Yes, surfaces can be created from the existing node sets. Just change the Region type from Selection to NodeSet.
I was referring to this video - the first one from the first referenced set of videos linked form documentation page https://www.youtube.com/watch?v=nGUzVGwSUrk&list=PLaWK58rtjzvS2ZKClbldRYN8Iu_gTmsC0&index=1&pp=iAQB
Thanks for the tips - much appreciated.
Ok, I see. That one is from a version so old that it wasn’t possible to work without creating sets and surfaces manually yet and only the .unv meshes could be imported (no step files) 
So if you watch Matej’s channel, be careful about the versions and possibly choose new videos as this channel shows the progress of the development across the years.
You can find more recent basic tutorial here: https://www.youtube.com/watch?v=XMNVtqrb3is
Since PrePoMax is actively developed, some things changed since then (especially meshing) but most stuff should be up-to-date.
That is really an old one 