What DEM modelling actually tells you about a transfer chute
Discrete Element Method modelling gets talked about like magic. Here is what it genuinely shows you about flow, wear and blockages, what it cannot tell you, and where it earns its keep.
Key takeaways
- DEM integrates particle contact forces and nothing else. The model is only as honest as its setup.
- It genuinely predicts stream trajectory, impact conditions, stagnation zones and relative wear intensity.
- It does not give liner life in millimetres, and it cannot model wet sticky material credibly without calibrated cohesion parameters.
- Calibration against measured material testing is the difference between evidence and animation.
Discrete Element Method modelling has a reputation problem. It is sold as a way to see the future of a chute, and it is dismissed as expensive animation. Both readings miss what the method is good at.
DEM simulates bulk material as thousands or millions of individual particles, integrating the contact forces between them and the surfaces they touch. That is all it does. Whether the output is worth anything depends almost entirely on how honestly the model was set up.
What the model is actually doing
At each timestep the solver calculates contact forces between particles and between particles and boundaries, then advances positions. Contact models range from simple elastic springs to formulations that add cohesion and adhesion for wet or sticky material.
Timestep is governed by particle stiffness and size, which is why a model with realistic fines is expensive to run. Most production models use scaled particles, and that scaling is a modelling decision that has to be justified rather than hidden.
What it genuinely tells you
- Where the material stream goes: trajectory, centring on the receiving belt, and whether the stream stays coherent through the drop.
- Impact conditions at the load point: velocity and angle, which drive belt cover wear, spillage and dust generation.
- Where material stagnates: dead zones and build-up points that become the blockage you get every wet season.
- Relative wear intensity: tangential shear work against liner surfaces shows you which panels take the punishment.
- Sensitivity: how flow behaviour changes across throughput, moisture and material variation, which is often more valuable than any single result.
What it will not tell you
DEM will not give you liner life in millimetres per year. It gives relative wear intensity, which you convert to a service interval using site experience and material testing, not simulation alone.
It will not model cohesive, high-moisture material credibly unless the cohesion parameters came from real testing. An uncalibrated model of sticky ore is a cartoon, and it will happily show clean flow through a chute that plugs in service.
Calibration is the difference
The useful models start with material characterisation: wall friction against the actual liner material, internal friction and cohesive strength from shear cell testing such as the ASTM D6128 method, bulk density, and moisture range from real samples rather than a design specification.
Calibration then tunes the contact model until the simulation reproduces measured bench behaviour. Only after that does the chute geometry mean anything. If a DEM proposal does not include material testing, you are buying animation.
Where it earns its keep
- Before steel is committed, when comparing three geometries costs simulation time instead of fabrication and a shutdown.
- On a chute that blocks repeatedly, where the value is finding the root cause rather than adding another deflector.
- Ahead of a throughput increase, to check that the existing flow path survives the new duty.
- Where dust or spillage is a safety and environmental issue, since both trace back to stream control.
Three geometries compared in simulation cost hours. The same comparison in steel costs three shutdowns.
How a DEM study runs
Characterise the material
Shear cell, wall friction and moisture testing on real samples across the operating range.
Capture geometry
Survey or 3D scan the existing chute, belts and structure so the model matches what is installed.
Calibrate
Tune the contact model until simulated bench behaviour matches measured behaviour.
Simulate the cases
Run the full range of throughput, moisture and material conditions, not just the design point.
Interpret and redesign
Translate flow, impact and wear results into geometry and liner changes, then re-run.
Validate in service
Check performance after installation and feed what you learn back into the model library.
Characterise the material
Shear cell, wall friction and moisture testing on real samples across the operating range.
Capture geometry
Survey or 3D scan the existing chute, belts and structure so the model matches what is installed.
Calibrate
Tune the contact model until simulated bench behaviour matches measured behaviour.
Simulate the cases
Run the full range of throughput, moisture and material conditions, not just the design point.
Interpret and redesign
Translate flow, impact and wear results into geometry and liner changes, then re-run.
Validate in service
Check performance after installation and feed what you learn back into the model library.
Common questions
How long does a DEM study take?
A study of a single transfer typically runs in weeks, with the schedule set by material testing and geometry capture rather than simulation time. Comparing extra geometry options costs little once the model is calibrated.
What does calibration actually involve?
Measuring the material: wall friction against the actual liner, shear strength through methods such as the Jenike cell, bulk density and the real moisture range, then tuning the contact model until simulation reproduces the measured bench behaviour.
Is DEM worth it for a chute that is not blocking?
Where a throughput increase, a material change or chronic wear and spillage is in play, yes: the model tests the future duty before it arrives. For a transfer with no symptoms and no change coming, spend the money elsewhere.
Key terms
Plain-language definitions from our glossary for the concepts this article leans on.
Standards and further reading
- ASTM D6128 Standard test method for shear testing of bulk solids using the Jenike shear cell (ASTM International)
- ANSI standards for conveyors and bulk material handling (CEMA)
- Conveyor installation standards and publications (CEMA)
- ISO 12100:2010 Safety of machinery, risk assessment and risk reduction (ISO)
- Guidelines library for mining technology and interoperability (Global Mining Guidelines Group)
Related case studies and tools
- Transfer chute redesign guided by DEM modelling (case study)
- Grain flow paths and chute upgrades guided by DEM (case study)
- Conveyor Capacity Calculator (tool)
- Agriculture solutions (solution)
Working through something like this?
See how we approach these initiatives, or tell us what you are dealing with.