Straight grate or grate-kiln: build the comparison around the ore
A testwork-led approach to thermal routes, preheated strength and energy boundaries.
Follow the material through the equipment
| Route | Material path | Important interface |
|---|---|---|
| Straight grate | Pellets travel as a bed on grate cars through thermal zones; the zone arrangement is design-specific. | Bed loading, sealing, gas distribution and the thermal history through bed depth. |
| Grate-kiln | Drying and preheating on a travelling grate precede firing in a rotary kiln and subsequent cooling. | Strength before grate-to-kiln transfer and behaviour during kiln tumbling, as well as the connected gas circuit. |
These arrangements create different mechanical and thermal duties. They do not establish a universal winner for energy, quality or operating cost. Evaluate actual proposals against the same ore and product requirements.
Evidence: KOBELCO Pelletizing Process, Technology Review 29 (2010), pp. 61–65 ↗ describes the configurations and testwork sequence; it is a supplier-authored technical account.
Preheated strength is a process requirement of its own
For a grate-kiln route, pellets must tolerate the transfer into the kiln and subsequent motion before firing is complete. Testing only final cold strength misses this intermediate requirement. Include preheated strength and degradation in the programme and relate the test to the transfer and heating sequence being designed.
In a straight-grate bed, pellets experience different positions and thermal histories. Surface temperature or one furnace setpoint cannot establish that all material received the required treatment. Sampling and testwork should investigate variation, not only the plant-average result.
Mineralogy changes the heat balance
Magnetite oxidation to hematite releases heat. Its contribution depends on how much magnetite is present and how far oxidation proceeds. The location and timing of this heat release matter to the thermal profile; treating it as an interchangeable external fuel input would obscure the mechanism.
An interrupted pot-furnace study at LKAB investigated different oxygen levels in an approximately 100 kg magnetite pellet bed. It measured temperatures and pellet properties at different bed heights. The results support the importance of oxygen availability and local thermal history; they are not an operating recommendation for oxygen enrichment in another furnace.
Evidence: Eriksson et al., ISIJ International 61 (2021), 1439–1449 ↗ — pot-furnace experiments, not a full-plant fuel guarantee.
Do not rule out a grate-kiln route solely because the feed is hematite. KOBELCO documents hematite applications. The defensible question is whether representative testwork and relevant operating references support the proposed ore blend, product and scale.
Request a testwork chain, not just a fired pellet sample
| Testwork stage | Preserve in the record | Engineering decision |
|---|---|---|
| Preparation and balling | Ore blend, size distribution, surface area, moisture, binder and flux recipe | Can the planned feed consistently form a suitable green product? |
| Drying and preheating | Heating rate, gas conditions, time, cracking and preheated strength | Will pellets survive the next thermal and mechanical step? |
| Firing and cooling | Representative thermal programme and position-dependent results | Is the quality envelope achievable across the relevant conditions? |
| Product qualification | Chemistry, physical test distributions and relevant reduction tests | Does the receiving BF or DR operation accept the product? |
A pilot result narrows uncertainty; it does not remove scale-up questions. Record which heat-transfer, material-handling and control features are represented in the test and which depend on engineering scale-up. Repeat on credible feed extremes rather than relying only on the easiest ore blend.
Put energy claims on the same denominator
Separate purchased fuel, electricity and internal reaction heat. State fuel heating-value basis, auxiliary-system scope, production period and whether output means gross fired pellets or accepted product. Compare steady performance tests separately from annual figures that include startup, outages and off-specification production.
Illustrative arithmetic — no supplier performance claim
Plant A reports 800 GJ of purchased fuel for 1,000 t gross output, or 0.80 GJ/t gross. If only 950 t meets the agreed acceptance definition, the same fuel is 800 ÷ 950 = 0.842 GJ/t accepted. A competing figure stated per accepted tonne must be compared on that basis.
Keep fan and other electricity visible even when the commercial comparison is expressed in total energy cost. Explain whether recovered heat is credited within the plant or exported across the boundary. Avoid counting the same recovery twice.
Make a route decision with explicit unresolved risks
Compare demonstrated feed compatibility, product quality, capacity and availability assumptions, fuel supply, emissions-control scope, maintainability and lifecycle cost. Identify which items have measured evidence and which remain supplier projections. A cost ranking should not conceal an unresolved mandatory product requirement.
Include access to grate cars, seals, refractory, burners, fans and critical wear components as appropriate to the proposed design. Review outage duration and spares with the operating team. A process that performs well under a test programme still needs a maintainable installation.
Use this guide to frame testwork and engineering reviews. Actual settings and acceptance limits belong to the approved plant-specific procedures.