Hey there,
It’s Robin from CFD Engine & I’ve been jotting down a few rules-of-thumb to help the less decisive among us navigate a world of CFD choices – I’m calling them my default settings.
Whilst you may not need them (you may even have your own) it would be great if you could help me to polish these ones up a bit?
Background
I struggle to make decisions. I’m decent at helping others, but if it’s a personal choice, then I’m a disaster.
It could be anything – choosing a new paint colour, some new headphones, or even picking a library book – the more choice there is, the less action I take. And don’t get me started on the terror that is a Subway sandwich 😨
It’s mostly an aversion to making the “wrong” decision & it’s completely unrelated to how high the stakes are. But it also crops up when I don’t have the skills / experience to tell the difference between the choices, or to figure out what the consequences might be.
In these situations a go-to option, a stock answer or a default setting can be invaluable. By limiting my choices I can avoid paralysis by analysis & keep things moving. Kind of like if you had a default breakfast – Boring? Perhaps. Sub-optimal? Probably. Useful? Definitely.
Default Settings
I realised that I have a few default settings that I apply to my CFD & that I use with clients – a set of rules that I need a pretty good reason to break.
So, until we establish otherwise, I say:
- Use a recent ESI/OpenCFD OpenFOAM release…
- …on Ubuntu Linux
- Use the binaries / don’t compile it yourself
- Mesh with
snappyHexMesh - Start out steady-state & RANS
- Post process with ParaView
- Pay for CAD software
- Overnight turnaround is fast enough
- Use (or at least try) AWS before buying new computers
I like stating them without any context to get your “Yeah, but…” juices flowing 😉
Chances are, if you have a “Yeah, but…” objection to any of them, then you’ve already evaluated your options & made a different choice that suits your particular circumstances – that’s good.
These are mainly about helping someone who can’t make that distinction (yet). Getting them moving in the right direction & closer to the dopamine hit you get from your first set of results – remember those days?
That said, what would you change from that list? I had a couple of others that I went back & forth on, but ultimately ditched. Is there anything that you’d add (or take away)?
Having a set of rules like this can also help when you have an established CFD process. They can help you to focus on improving your skills with the tools you already have, rather than looking for a magic solution elsewhere.
What would your version of these rules look like? Which ones would need tweaking to fit your process?
Over to you
Conventional wisdom seems to be that I should throw ideas like this at social media, see what works & then use the winners in these emails. I like doing it the other way around. Your feedback tends to be considered & constructive, which helps to refine my ideas before I inflict them on social media.
So, please drop me a note & share your thoughts &/or your default settings. It’s always interesting to get a peak into your CFD process.
My inbox is always open & your assistance is greatly appreciated 🙏
Until next week, stay safe,
