If the Analytical Instrumentation Is Right, Why Do the Results Keep Changing?

Laboratories rarely question their instrumentation without good reason. Modern analytical instruments are precise, stable, and well-controlled. When results start to drift or do not make sense, the instinctive response is to look at calibration, maintenance, or operator technique.

Yet the instrumentation is doing exactly what it should.

The problem is that the results will only reveal what the sample allows.

When results are questioned, re-run, or scrutinised under QA/QC, you will likely find that was compromised earlier. During sampling and sample preparation. By the time the issue shows up at analysis, the opportunity to correct it has already passed.

When Repeating Tests Becomes Normal in Mining Laboratories

In some mining laboratories, re-running samples has gradually become routine.

It happens when results look inconsistent, when data does not align with expectations, and when QA/QC flags variability.

The impact accumulates quickly:

• Turnaround times stretch during peak periods.
• Repeat testing increases consumables and labour costs.
• QA/QC and reporting come under greater scrutiny.

If repeats are common, the issue is rarely isolated. It is systematic.

Why It Is Not Enough to Only Trust the Analytical Instrumentation

Laboratories trust their analytical instrumentation, and rightly so.

However, the instrumentation does not correct the variability but reveals it.

A sample that is not representative or has been inconsistently prepared will always produce questionable results, even when the analysis itself is technically sound.

The Sample Warning Signs Labs Tend to Miss

When accuracy is lost before analysis, the indicators are often subtle. You may notice that results vary between shifts or experience unexpected contamination during QA checks.

These signs point away from analytical failure and towards problems in sample preparation control.

Where Variability Is Introduced, Affecting Sample Preparation Accuracy

ELS site work shows the same contributors appearing repeatedly:

• Crushers operating outside target jaw gap settings.
• Crusher selection not matched to material hardness or throughput.
• The crushed product size is too coarse
• Worn bowls are producing inconsistent grind sizes.
• Poor cleaning discipline between samples.
• Throughput pressure overrides preparation control.

Each of these introduces variability before the sample ever reaches the analytical phase.

Getting Control Starts with Crushing but Does Not End There

Crushing sets the foundation for sample preparation accuracy. It does not need to be complex, but it must be intentional.

Define the Required Crushed Product Size First

Crusher selection should always begin with the required crushed product size.

The target top size determines the crusher type. Throughput and turnaround refine the choice. Budget should set the boundary, not dictate the process.

If the required product size is not defined, equipment selection becomes difficult.

Match the Crusher Type to the Requirement

Different crushers serve different purposes.

High-end jaw crushers such as the Essa JC Jaw Crushers are suited to hard, coarse feed where controlled size reduction is required, delivering up to 85 per cent passing 2 mm.

Boyd and Orbis crushers offer strong throughput with enclosed designs that reduce noise and contamination risk, achieving up to 90 per cent passing 2 mm and allowing integration with rotary sample dividers.

Standard jaw crushers remain fit for purpose in many laboratories, but require closer control of wear and gap settings to maintain consistency.

Control the Crusher Jaw Gap

Jaw gap defines the actual product size.

If the gap is too wide, material passes through under-crushed, compromising subdivision and accelerating downstream wear. If it is incorrect, jaw wear increases and mechanical stress rises.

Crusher gap must be set to achieve the target product size, not adjusted to chase throughput.

Pulverising: A Frequent Cause of Analytical Repeats

Pulverisers cannot fix upstream issues. On the contrary, they magnify them.

When excessive sample mass is loaded into the pulveriser:

• Milling times are extended.
• Bowls, rings, and pucks wear faster.
• Grind consistency drifts between batches.
• Most repeat testing issues originate here.

How to Match the Pulveriser to the Process

Pulverising must be treated as a controlled step, not a corrective one.

Start with the Crushed Product Size

Oversized particles pulverise inefficiently, accelerate wear, and increase mechanical stress.

A wide size range produces uneven grinding and poor homogeneity.

Match Bowl Size to Sample Mass

Bowl volume must suit the sample mass, not operator convenience.

Overloading a bowl restricts puck or disc movement, reduces grinding effectiveness, and could enable the pulveriser to stop. If the sample mass is too large, it must be subdivided or processed in a larger bowl.

Shortcuts here almost always lead to repeats.

Match Bowl Material to Hardness and Abrasiveness

Material hardness and abrasiveness drive wear and contamination risk.

Incorrect bowl material selection changes grind behaviour over time, introduces contamination, and reduces repeatability. Consumables must be selected as part of mining laboratory QA/QC, not treated as housekeeping items.

Control Fineness to the Analytical Requirement

Grinding finer than required does not improve accuracy.

Typical targets include approximately 80 per cent passing 75 µm for digestion-based methods, and 90 per cent or more passing 75 µm for XRF. Over-grinding increases segregation risk without analytical benefit.

A Simple Check That Prevents Sample Particle Size Headaches

The finger test is a quick, practical bench check for grind adequacy under production pressure.

The Finger Test in Practice

Rub a small amount of pulverised sample between your fingers and palm.

A smooth, powder-like feel indicates an acceptable grind for most methods. A gritty or granular feel indicates that the sample is too coarse.

If particles can be felt, the grind is not finished.

This test catches under-grinding early, supports consistent subdivision, and prevents unnecessary re-runs. It does not replace PSD testing, but it works when time is tight.

Why Particle Size Control Protects Mining Laboratory QA/QC

Many analytical methods are particle-size dependent.

When particle size is uncontrolled, segregation increases, splitter performance degrades, and results vary from the same original sample. This is why laboratories repeat tests even when the analysis is correct.

Key takeaway: Consistency, not fineness, protects sample preparation accuracy.

Consumables Are Part of Process Control, Not Housekeeping

Consumables wear gradually. When they drift, results drift with them.

Poor cleaning, especially of pulverising bowls, inconsistent pulverising times, and worn bowls, rings, and pucks all introduce variability.

As Dave Harvey, Technical Director, explains: “Improper cleaning of pulverising bowls is the most common issue we see leading to repeats. Residual material left behind causes carryover contamination that only shows up during QA checks.”

He also notes that when results vary between shifts, shortcuts during pulverising are usually involved.

“A sample not taken correctly is not worth assaying, as it is unrepresentative.”

When Repeats Point to a Systemic Sample Preparation Problem

Repeats usually indicate that something in sample preparation is not under control. At that point, an independent review is often the fastest way to restore confidence in results.

A Practical Next Step to Restore Confidence in Results

If your laboratory is experiencing inconsistent results or repeat testing, look upstream.

A focused sample preparation audit by ELS can identify where variability is being introduced before it reaches analysis.

Accuracy does not start at the analytical instrument.

Contact us to recommend the right crusher and pulveriser setup for your lab.