Quick answer
If clinker free-lime results change with sampling location, do not average the numbers blindly and do not assume the laboratory is wrong. Samples taken at the kiln outlet, within the cooler, after the clinker crusher or at a downstream conveyor can represent different material populations and different process delays. Particle-size segregation, incomplete mixing, cooling history, sample representativeness and real kiln variability can all create a location-dependent result. The correct control point is the one whose sampling method is repeatable, representative and explicitly linked to the process decision being made.
Problem / job
Free-CaO measurements from two clinker sampling points disagree enough to confuse kiln control, quality reporting or troubleshooting.
Evidence
Capture exact sampling point, process transit time, sample timing, clinker temperature, size distribution, cooler condition, kiln stability, preparation method and duplicate analytical results.
Decision
Separate real process gradients and segregation from sampling bias, time-lag mismatch and laboratory repeatability before selecting the control sample or changing kiln firing.
As-found evidence capture
Document the exact location and mechanism of every sample. A clinker sample collected close to the kiln discharge is not automatically equivalent to a sample collected after the cooler or after the crusher. Record whether the sampler cuts across the complete material stream or takes material from one side, whether the sample includes the full clinker size range, the time at which it was collected, and the estimated process delay between sampling locations.
At the same time, record kiln feed rate, fuel condition, kiln torque/current trend, burning-zone observations available to the plant, cooler grate or bed condition, cooler fan behavior, crusher status and any recent upset. Preserve the clinker sample and its size distribution when practical. If large nodules, fines or visibly different clinker are segregated during collection, photograph the condition before crushing or laboratory preparation.
Field checklist before changing kiln control
- Label each sample by exact physical location and collection time.
- Estimate material transit time between the two sampling points.
- Photograph or record clinker size distribution before crushing when practical.
- Run duplicate preparation and analysis on one homogenized reference sample.
- Compare sampler reach and whether the complete stream or only one side is intercepted.
- Record kiln and cooler stability during the paired sampling window.
- Keep the control-point role fixed until the bias mechanism is understood.
Eight-step diagnostic workflow
1. Define what each sampling point actually represents
Write down the material path from kiln discharge to each sample point. A kiln-outlet sample is valuable for fast feedback on burning, while a downstream sample may better represent the mixed product ultimately entering storage. Those are different quality-control jobs. The first step is therefore to decide whether the plant is comparing two control points for the same purpose or two intentionally different process views.
2. Align samples by material transit time
Do not compare samples solely by clock time. Estimate how long the clinker takes to travel from the upstream sampling point to the downstream sampling point under the actual operating condition. Match samples that represent approximately the same clinker production interval. During process changes, a time-lag mismatch can look like a systematic location bias.
3. Check sampling representativeness
Inspect sampler reach, cutter motion, spoon or screw condition, blockages and wear. Verify that the sample is not consistently taken from one segregated portion of the stream. A representative clinker sample must capture the range of material passing the point rather than only fines, only coarse nodules or material near one wall. If one location uses a full-stream or cross-stream sampler and the other relies on a small side grab, the disagreement may originate before the laboratory receives the sample.
4. Preserve clinker condition during sample handling
Record how hot samples are cooled, how long they are stored before analysis and whether moisture or contamination can enter the sample. Use the approved plant procedure for cooling and handling. Different handling routes can change the condition of the sample before free-lime measurement and can also affect crushing and homogenization. Standardize the route before using the result to judge the kiln.
5. Standardize crushing, splitting and analytical preparation
Process paired samples through the same laboratory preparation route. Use proper splitting after crushing and avoid selecting only easy-to-grind fractions. Run duplicate preparations from the same homogenized sample where possible. If duplicate preparation scatters widely, resolve the preparation or analytical problem before concluding that the process location causes the difference.
6. Compare location bias by clinker size fraction
Check whether the difference is carried mainly by coarse nodules, fines or a visibly distinct fraction. Real kiln and cooler behavior can produce clinker populations with different burning histories. If a sampler over-represents one size fraction, its free-lime result may be consistently different even though both lab measurements are analytically correct. The useful evidence is whether the location difference remains after the samples are made comparably representative.
7. Correlate the residual with kiln and cooler behavior
Plot the difference between locations against kiln upset periods, fuel changes, raw-meal chemistry variation, cooler bed disturbances and crusher operation. A location difference that increases only during process instability suggests real heterogeneity or changing transit/mixing behavior. A constant bias under stable operation points more strongly toward sampling or preparation differences.
8. Assign one control point and document the role of the others
Once the plant proves the sampling and analysis chain, define which location is the official control point for kiln feedback and which locations are diagnostic or product-verification points. Do not alternate control locations without compensating for their different process meaning. The operating team needs one traceable reference for trends, alarms and historical comparison.
Evidence-to-decision table
| Observed evidence | Most useful next check | Decision direction |
|---|---|---|
| Downstream result changes later than kiln-outlet result | Process transit time | Align material windows before comparing |
| One sampler contains mostly fines or coarse nodules | Sampler geometry and size distribution | Correct sample representativeness |
| Duplicates from one prepared sample disagree | Laboratory preparation and method repeatability | Resolve lab variation before process action |
| Location difference grows during kiln upset | Kiln/cooler trend and clinker heterogeneity | Treat as possible real process gradient |
| Difference is stable during steady production | Sampler bias and handling route | Audit systematic sampling/preparation bias |
| Results converge after representative splitting | Original sample composition by size fraction | Sampling segregation was dominant |
Verification and return-to-control criteria
The comparison is acceptable when paired samples are time-aligned, both sampling methods are proven representative for their intended job, laboratory duplicates are repeatable, and the residual between locations has a documented and stable interpretation. The official kiln-control sample should then remain consistent over time so operators can distinguish a real change in burning from a change in sample location or handling.
Use the plant’s approved clinker sampling and free-lime method for quantitative limits. Sampling frequency, allowed analytical variation and control targets depend on the installed system and quality plan, so this guide does not impose universal numbers.
Common mistakes and repeat-failure traps
- Comparing samples by clock time only: downstream clinker may have been produced significantly earlier.
- Using a side grab as a reference for a full-stream sampler: the two methods may not represent the same particle population.
- Ignoring size segregation: coarse and fine clinker fractions can carry different process histories.
- Changing kiln fuel from one discrepant sample: first prove the sample and analytical chain.
- Mixing control-point roles: fast kiln feedback and final product verification are not necessarily the same sampling job.
- Changing sample cooling or preparation between locations: inconsistent handling makes a process comparison inconclusive.
Related Infinity guides
Build a stronger cement technical library
The Infinity for Cement Organization technical package contains 900+ technical files for cement professionals. It is a one-time $249.99 purchase, with instant download immediately after payment and lifetime updates forever at no additional cost.