Simulate Before You Weld

Interactive 3D simulations of the failure modes that destroy pyrolysis plants — reactor weld cracks, tar fouling, condenser clogging — solved with the same multi-physics framework APChemi uses on every commercial project. Drag, hover, scrub the time slider. Above 99% accurate vs physical reality.

Multi-Physics Simulations for Pyrolysis Plants — CFD, FEA, and Process Solvers Coupled

Two interactive 3D simulations below. Click ▶ Play, drag the time slider to scrub, left-drag the model to rotate, right-drag to pan, scroll to zoom. Both run the actual physics in your browser via WebGL — same library family as the simulations we deliver to every commercial project.

Sim 1 — Reactor Thermal Bending (CFD + FEA coupled)

A horizontal cylindrical reactor heated by two floor burners. The bottom side runs hotter than the top, the steel weakens at temperature, and the cylinder sags downward at the midspan under its own weight. Where the curvature is highest — at the welded joints near the endplates — the steel yields and the welds crack. The simulation predicts both the temperature field and the deformation in a single coupled solve.

Reactor Thermal Bending

Asymmetric burner heating · 60-min ramp · CFD + FEA coupled

Solvers: conjugate heat transfer · compressible vapor flow · phase-modified k-ω SST · linear elasticity · thermal expansion

t = 0.0 min
READY
T MIN300K
T MEAN300K
T MAX300K
δ MAX0.00mm physical
Asymmetric heating predicts weld-joint stress concentration — uncorrected, 1-month shutdown for repair
Temperature
700 K 600 500 400 300 K
THERMAL EXPANSION
🖱️ left-drag · rotate   right-drag · pan   scroll · zoom   hover · inspect
Initialising WebGL scene…

What is shown: Vertex-colored temperature field on the reactor surface (blue = 300 K, red = 690 K) — the bottom runs hotter because the burners are below. Two coupled deformations are visualised: (a) vertical sag at the midspan (bimetallic + gravity creep) and (b) thermal expansion in one direction — the left end is anchored, the right end slides outward as the cylinder grows (see the yellow expansion arrow). Drag the Deformation × slider (0–50×) to exaggerate both. Toggle Stress to see the high-stress zones near the welded flanges.

Sim 2 — Tar Settling & Clogging Tendency (horizontal vapor pipeline)

A horizontal vapor pipeline with a multi-modal particle population — fine carbon black (5 μm) through to coarse tar splash (400 μm). Heavy particles settle under gravity as the vapor decelerates; sticky tar adheres to the pipe bottom and grows a solid deposit layer. The simulation predicts the clogging rate and the cleaning interval — not just the particle trajectories.

Tar Settling & Clogging Tendency

Horizontal vapor pipeline · 500 mm dia · multi-modal particle population (5 μm to 400 μm)

Solvers: Lagrangian particle tracking · semi-solid drag + adhesion · gravity-corrected settling · wall sticking probability · deposit-layer growth tracker

t = 0.0 s
FILLING
PARTICLES0in view
SETTLED0.0%
DEPOSIT0.0kg/hr
WALL CLOG0%
Heavy tar + carbon-black accumulate at pipe bottom — predicted cleaning interval < 48 h at current flow
GRAVITY
VAPOUR FLOW
Particle Ø (m)
4.0e-4 2.5e-4 1.0e-4 5.0e-5 5.0e-6
🖱️ left-drag · rotate   right-drag · pan   scroll · zoom
Seeding particles…

What is shown: Each sphere is one tar/carbon particle, coloured by diameter on a log scale (blue = 5 μm fine carbon, red = 400 μm tar splash). The dark layer growing on the pipe bottom is the predicted solid deposit. Use the Gas velocity slider to see how flow rate trades off against settling — low gas velocity means more carbon drops out and more cleaning shutdowns.

See your plant simulated before you build it — preliminary scan in 10 days.

Contact Us

Why Simulate Before You Weld

Every pyrolysis plant operator we have ever worked with has the same story: the plant ran for a few months, then a weld cracked, or the condenser clogged with tar, or one side of the reactor heated faster than the other and the cylinder bent. Each event is a one-month repair shutdown, plus capacity loss, plus the insurance and warranty conversations that come with it. The cost of finding these failure modes after construction is between $200,000 and $2 million per incident — and the learning curve is just that: you learn one expensive lesson at a time.

Multi-physics simulation finds every one of these failure modes before the first weld goes in. The cost of simulation is a fraction of the cost of one repair shutdown. Both interactive visuals above are the same class of simulation we run for every commercial project.

🔥
Weld joint cracks
Asymmetric heating bends the reactor; weld cracks at the feed point. 1-month shutdown.
⚠️
Tar-clogged condenser
Heavy carbon and tar mist accumulate where vapor decelerates. Capacity drops 30–50%.
📉
Bent reactor shell
Thermal expansion + non-uniform burners = permanent shell deformation. Unrecoverable capex hit.

What We Simulate — 8 Solvers + 7 Coupling Libraries

Every project we deliver is solved on a custom multi-physics framework that integrates the following modules in a single coupled solve, rather than running them sequentially:

8 Solver modules

  • • Master multi-physics transient integrator (PIMPLE-loop base)
  • • Compressible turbulent gas-phase flow with reacting chemistry
  • • Conjugate fluid ↔ pipe wall heat transfer
  • • Vapor-to-liquid condensation (Lee / kinetic model)
  • • Liquid-to-solid tar solidification (enthalpy-porosity)
  • • Lagrangian discrete tar-particulate cloud (semi-solid drag, adhesion)
  • • Liquid wall-film formation, transport, and break-up
  • • Spray-droplet injection with cone-profile droplet–vapor coupling

7 Coupling libraries

  • • Multi-region solver bridging (combines region outputs into one solution)
  • • Moving-region coupling at transient interfaces (sliding mesh + phase-aware)
  • • Phase-dependent non-Newtonian tar rheology closure
  • • Wall-impact sticking-probability adjudicator
  • • Wall-deposit growth tracker (couples cloud + film + solid deposit)
  • • Parcel ↔ wall-film mass / momentum exchange
  • • Phase-modified turbulence closure (k-ω SST aware of local phase fraction)

What Commercial CFD Cannot Do

Ansys Fluent and Siemens STAR-CCM+ are excellent for the standard CFD physics (compressible flow, energy transport, single-phase Lagrangian tracking). For pyrolysis, four critical capabilities are missing — and the workaround is to stitch separate simulations together by hand, which loses accuracy at every interface.

Missing in commercial CFD #1

Semi-solid particulate phase

Tar in pyrolysis is intermediate between a Newtonian droplet and a rigid solid. Commercial packages force you to pick one — and lose the chunk-like adhesion behaviour that actually drives fouling.

Missing in commercial CFD #2

Non-Newtonian Eulerian + Lagrangian DEM coupling

In one solve loop. Commercial workflow: run them as separate simulations, then patch the interface by hand. Errors compound.

Missing in commercial CFD #3

Sticking-probability wall closures

Required to predict fouling rate — not just where particles fly. Commercial packages predict trajectories, not deposit growth.

Missing in commercial CFD #4

Unified multi-mode phase-change framework

Vapor → liquid → solid in one solver, hosted in the same framework as the gas flow. Commercial workflow: three separate runs.

Mesh and Compute

Mesh density and compute budget scale with the question you are asking. A preliminary screening to identify the dominant failure mode runs in 10 days on a coarse mesh. A full coupled multi-physics study with sub-millimetre wall-film resolution and per-nozzle spray refinement takes 2–3 months on 64-core MPI hardware.

Phase 1

Preliminary scan

~130,000 cells
10 working days · single dominant failure mode · fixed fee
Phase 2

Multi-physics base

~1.5M cells
6 weeks · 4 solvers coupled · scoped fee
Phase 3

Full multi-physics

~3M cells
2–3 months · all 8 solvers + 7 libraries · 64-core MPI

How Pricing Works

Engineering and simulation effort is billed against a fixed scope at hourly rates ($50/hr engineer, $150/hr senior). For full technology-license engagements where simulation underwrites a performance warranty, we charge 10–20% upfront (advance + disclosure milestones), the simulation fees as delivered, and the remaining 70–80% as success fees — paid only when your plant runs at the contracted 300–330 working days per year.

For external benchmarking: a commercial single-solver agitator simulation (one physics, off-the-shelf solver) is quoted at ~$10,000 from US providers or ₹6–8 lakh from Indian providers, plus the recurring Ansys / Simcenter license fee. The simulations on this page combine 8 solver modules and 7 coupling libraries in a single coupled solve — scope that scales accordingly.

Track Record

47+
Plants developed
20–30
Plant troubleshoot
12
Patents
99%+
Sim vs reality

Partnerships: Shell Petrochemicals (Singapore — Technip Energies collaboration) · BASF (plastic-to-naphtha, government-cofunded in India) · ISCC Plus certified for refinery-grade pyrolysis oil supply.

Frequently Asked Questions