I’m James Linfield and This Is How I Mesh

I grew up on the south coast of the UK, in a small sleepy village called Storrington in West Sussex. Not too far away from the seaside town of Brighton (that’s where I say I’m from when people, quite rightly, have no idea where Storrington is!).

As a kid I always enjoyed figuring out how things worked, and I can vividly remember building a model remote-control car (The Grasshopper R/C Model Kit) from scratch with my Dad at a fairly young age. I think projects like that helped pique my interest in engineering at an early age, although I wasn’t aware of it at the time. I also loved airplanes. I can remember being completely obsessed with the Blackbird (Lockheed SR-71 to give it its proper name) and had a scale model of it in my room. I still think it looks incredible today.

Eventually this led me up to the East Midlands to study Mechanical Engineering at Loughborough University, where I had the great pleasure of being taught fluid mechanics by the legendary Henk Versteeg. Henk and Malala Sekera (another Loughborough Uni lecturer) wrote the book “An Introduction to Computational Fluid Dynamics: The Finite Volume Method” which is still my ‘go to’ CFD bible and is generally a pretty well-respected resource in the CFD community.

I always enjoyed fluid mechanics during my under-grad studies (yes, I’m one of those people who enjoys driving down the motorway in the rain, mesmerised by the suddenly visible turbulent flow structures behind cars and lorries). However, it wasn’t until my master’s degree that I really started to get the CFD bug.

My foray into CFD really began though when we had the opportunity to simulate both reacting and non-reacting cases for an axisymmetric bluff-body burner in Fluent (version 6.something), creating the computational mesh in Gambit…shudder… but I was hooked.

We were also super lucky to be able to jump in the lab to conduct cold / non-reacting PIV studies on the same burner geometry for validation purposes. It was great to experience both the computational and experimental side of fluid mechanics and begin understanding the (many) pros and cons of both approaches!

My Masters thesis explored the novel use of embedded shape memory alloys (SMA’s) into foil type structures to form adaptive composite structures, to increase the aerodynamic efficiency of projectiles. Basically, using shape memory alloys to actuate and morph aero surfaces. Think Batman’s cape (at least that is how the project was sold to me!).

After concluding my studies at Loughborough University, I started a role as a Consultant Engineer at SCITEK in Derby, an aerospace R&D consultancy. Working on a wide range of client-based consultancy projects for a variety of clients, including a well-known aero engine manufacturer based in Derby and on several EU funded research projects under the Seventh Framework Programme (FP7). My favorite EU project was probably STREAMLINE (Strategic research for innovative marine propulsion concepts). The goal was to optimise the low speed performance of high-speed water jet propulsors by applying state-of-the-art aerospace expertise to the design and optimisation of the waterjet intake duct using CFD. It was really interesting to see how cutting-edge technology from the aerospace sector could benefit the marine industry.

Over the years I got the opportunity to apply CFD to a range of complex industrial applications and absolutely loved it. In 2014 I was appointed to Lead Consultant in Numerical Modelling, responsible for the coordination and execution of all numerical modelling activities (chiefly CFD) within the company. I also led the initiative to implement OpenFOAM within the organisation and set best practices which was both interesting and challenging.

After nearly ten years, I decided that it was time for a new challenge and the itch that I had to run my own business could no longer be ignored.

In 2017, I co-founded Navier Flow Consultants to provide small to medium sized enterprises (SME’s) with access to the same state-of-the-art CFD expertise & resources that the large OEM’s have, with the aim of helping them to innovate and solve complex unfamiliar flow & thermal problems quickly. We do this by leveraging high-fidelity CFD simulation and optimisation techniques to shorten our client’s product development cycles, decrease risk, reduce overall project costs and ultimately improve design performance.

We believe that our experience from working in high technology sectors allows us to transfer that know-how and apply it to help clients in other industries (we call it borrowing brilliance). Currently, our clients span across the Aerospace, Civil, Renewables, and Water Treatment industries.

I absolutely love doing  CFD consultancy projects, we get to work with some amazing people and companies.

  • Location: Nottingham UK (Home of Robin Hood!)
  • Current position: Director & Consultant Engineer at Navier Flow Consultants Ltd
  • Current computer: Dell XPS 15 9560, 4K UHD, Intel Core i7-7700HQ, 32GB RAM, 1TB SSD, Nvidia GTX 1050 4GB GDDR5, Dual boot Windows 10 and Ubuntu 18.04.2 LTS
  • One word that best describes how you work: Obsessively!

What software or tools do you use every day?

  • ATOM: as our text editor, I love the look & feel of it. Of course, it also has all the useful functionality that you would expect from a good text editor: folder tree, language aware, variable autocomplete, etc.
  • Putty: for connecting with our in-house HPC facility.
  • ASANA: for project & time management.
  • Linux terminal: I do seem to spend a lot of time in a Linux terminal, firing off simulations and monitoring solution progress and tweaking dictionary files on the fly.
  • Pointwise: although we don’t strictly use it every day. We are intermittent Pointwise users on a project by project basis, usually when we need a super high-fidelity grid with absolute control over the viscous boundary layer mesh, e.g high speed aerodynamics.

What does your workspace look like?

Spot the Pointwise accelerator mouse matt (very useful, thank-you to the Pointwise folks!).

You can also make out my trusty calculator that I’ve had for the last 20 years or so… I’m strangely attached to it.

We are also massive fans of the SpacePilot Pro 3D mouse, it really helps speed us up when it comes to CAD & meshing. Secretly, I love it because it makes me feel like I’m in Flight of the Navigator!

What do you see are the biggest challenges facing CFD in the next 5 years?

I think CFD potentially faces a credibility issues which is likely to get worse as tools get “easier” to use. Partly due to the ever-increasing rise of “black box” tools and how they are marketed as general purpose tools that allow the user, having no background in fluid mechanics or CFD, to compute a solution at the click of a few buttons.

There is potential for this approach to lead to CFD as a discipline being cheapened somewhat. What I mean by that is, it can lead to the situation where a novice cranks the handle on a simulation, gets some results and when they don’t’ reflect what they had intended to simulate or wildly differ from reality – the solution along with the idea of using CFD gets thrown in the bin. The takeaway then is whilst CFD is “easy,” it’s not useful and can’t be trusted. Or, the other potential scenario is that they do get some viable results from the analysis. The next question is what to do with all that data, how to characterise what is happening and what to focus on to improve it. At the end of the day, CFD is a tool and if it cannot be leveraged to provide meaningful and actionable engineering data – it is a waste of time, in my opinion.

I am also concerned that the current trajectory may lead to a lack of people with good CFD domain expertise, people who know which turbulence model or wall treatment is most suited to which flow problem. Or, what parts of the problem to focus efforts on (and which not to bother with), which numerical methods to adopt and so on.

Don’t get me wrong, I do believe that there is a place for the simpler, easier-to-use CFD tools in this industry. I think they slot nicely into the early, first pass analysis phase of a project. Focused on a well-defined, well understood type of problem for which the tool is well-tuned and the appropriate settings are already encoded by experts. I’m not against designers utilising appropriate simulation tools earlier in the design process. Ultimately, I think it’s a good thing.

Something that I also think will be really interesting to watch out for over the coming years is the use of artificial intelligence in the CFD space. There is some exciting work going on at the moment, looking at data-based approaches to turbulence modelling for Large Eddy Simulation (LES) by artificial neural networks.

I think the potential for AI to also benefit shape-optimisation campaigns is huge. But will AI become useful in CFD applications, before it becomes sentient and kills all of us…?

What are you currently working on?

The work we carry out is very varied and always changing. Currently we are working on everything from ventilation studies for industrial gas turbine enclosures to ensure cooling requirements are satisfied, to air pollution studies from power generation plants. And more recently, we’ve worked on drone aerodynamics/handling and open-deck wind comfort on cruise ship super structures.

What would you say is your meshing specialty?

Probably producing hybrid, unstructured meshes for extremely large, complex geometries. Once you’ve tackled a fully detailed aero-engine combustor, most things look simple in comparison! The CAD data that we typically work with from clients can be in the region of 5 to 10 GB, so I’m not easily intimidated by geometric complexity (famous last words).

Any tips for our users?

For bottom-up type meshing approaches, it is all about three things: surface mesh, surface mesh and surface mesh! Spend the time and effort required to get a high-quality surface mesh and you’ll reap the benefits when it comes to growing boundary layer elements off it.

What project are you most proud of and why?

I am most proud of setting up my own company because I feel very lucky to be doing a job I love, working with amazing people.

What CFD solver and postprocessor do you use most often?

Although we are code agnostic at Navier and believe in using the right tools for the job in hand, we are still big OpenFOAM users. I think OpenFOAM is great because it forces you to have a good appreciation of the underlying numerical methods and models. There is no hand holding or magic behind the curtain that converges your solution when you make a silly mistake, or have an ill-posed problem.

We also regularly use SU2 for high speed compressible flows and adjoint-based optimisation studies.

For post processing, we mostly use Paraview. We find that the ability to automate post processing through python scripts is invaluable. It also allows us to spend more time interpreting results and coming up with new ideas to test, whilst spending less time trying to make sure all the camera angles match from different runs – we’ve all been there!

Are you reading any interesting technical papers we should know about?

I am currently reading two papers. The first is titled Plasma Streamwise Vortex Generators for Flow Separation Control on Trucks that comes from Linné Flow Centre in Sweden.

The other paper is from our friend Dr. Ben Evans at Swansea University and is titled On the Kutta Condition in Compressible Flow over Isolated Airfoils. We’ve had the pleasure of working extensively with Dr. Evans on a range of industrial projects in the past. We’re also hoping to sit on the advisory board of an exciting research project next year, aiming to exploit efficient evolutionary optimisation algorithms for aerodynamic design.

Do you plan on attending any conferences or workshops this year?

We’re NAFEMS members and will likely be attending their upcoming event on Simulation-Based Optimisation in London.

Recently, we’ve also been applying our aerodynamic shape optimisation techniques to the UAV sector, so the “Commercial UAV Show” in London will be an interesting event for us.

I’ve never made it to an OpenFOAM Workshop, but that is high up my list to get an up to date flavor of how the code is being used in different industries.

What do you do outside the world of CFD?

To be honest, Navier does take up a lot of my time and energy.

I’m a bit of a coffee addict though, and decided to invest in a swanky bean to cup coffee machine which gives me the ability to constantly tweak and adjust the grind size, dosing size and temperature etc. to my heart’s content. Basically, an Engineers dream! We also get our beans delivered weekly from a local roastery called 200 Degrees… I’m worried the addiction (hobby?) may have got out of hand….

Recently, I’ve started to get into road cycling. I’m not yet a MAMIL (“middle-aged man in lycra”) but give it time. The grand plan was to cycle into work every day as a good way of incorporating exercise into my daily routine, but I must confess that when its pouring with rain (which is does a lot of here in the UK) it’s incredibly tempting to just hop in the car – must do better!

What is some of the best CFD advice you’ve ever received?

  1. Always start simple and gradually build in complexity. This is especially true when exploring new or unfamiliar flow physics. I like to start off with a simple representative test case that I can use to tweak and develop quite quickly, that doesn’t require days or even weeks of run time.
  2. Avoid “black-box syndrome,” and always try to have an appreciation of what is happening numerically under the hood – it will pay dividends in the end.
  3. When interpreting CFD results always ask yourself:
    1. Why is this happening?
    2. Why does it matter?
    3. How can I improve it?
    4. And what am I comparing it to?
  4. CFD should be used as a tool to provide meaningful and actionable engineering data, not just pretty pictures (although that is also fun).
  5. Finally, “All models are wrong, but some are useful.”- George Box

If you had to pick a place to have dinner, where would you go?

That’s a tough call, probably Coco Tang in Nottingham…they serve Vietnamese food and it’s incredible. After travelling around Vietnam about five (ish) years ago, I fell in love with the cuisine and it’s nice to be able to get my next fix…without a 12-hour flight.

About Travis Carrigan

A Pointwise engineer helping other engineers solve their meshing problems.
This entry was posted in People & Places and tagged , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s