Introduction & Context

Variance calculation for powder mixtures quantifies how uniformly a minor component—commonly table salt—is dispersed through a bulk snack matrix. A low variance (high "mixedness") guarantees every portion reaches the labeled salt content, avoiding consumer complaints and regulatory fines. The method is applied at mixer validation, batch release, and continuous-process adjustment points in snack, pharmaceutical, and detergent plants.

Methodology & Formulas

Record sample mass fractions
Let \(x_i\) be the salt mass fraction (%-w/w) in the \(i^{\text{th}}\) sample taken from the blend.
Mean salt fraction
\[ \bar{x} = \frac{1}{n}\sum_{i=1}^{n} x_i \]
Sample variance
\[ s^{2} = \frac{1}{n-1}\sum_{i=1}^{n}(x_i - \bar{x})^{2} \] Units: (%-w/w)\(^{2}\). Note: This formula uses \(n-1\) to provide an unbiased estimate of the population variance from a sample.
Sample standard deviation (root-mean-square deviation)
\[ s = \sqrt{s^{2}} \] Units: %-w/w, identical to the data scale and the specification limit.

Empirical operating limits for binary, free-flowing powders:

Parameter	Range / Criterion
\(\bar{x}\)	0.1–20 %-w/w
Minimum number of samples	\(n \geq 10\)
Acceptable homogeneity for snack foods	\(s \leq 0.3\ %-w/w\)

For a perfectly mixed binary blend, the theoretical variance σ² of the concentration of the key component is:

σ² = p(1–p) ⁄ N
Where p = mass fraction of the key component and N = number of particles in the sample.

To convert this to practical units:

Estimate N from sample mass and average particle mass.
Multiply σ by 100 to express variability as % relative standard deviation if concentrations are expressed as mass fractions.

Target a sample that contains at least 10,000 particles of the minor ingredient:

Measure d₅₀ of both components via laser diffraction.
Calculate single-particle mass assuming sphericity and true density.
Determine required sample mass = 10,000 × (single-particle mass / minor fraction).
Collect a minimum of 30 such samples to obtain a 95% confidence interval on the variance that is ±20% of the estimated value.

Excess variance commonly stems from:

Segregation during transfer or sampling—use a spinning-rifler or thief probes with entry angles ≤45°.
Cohesion or static charging of fines—add 0.1–0.5 wt% silica or perform humidity conditioning to 40–50% RH.
Multi-modal particle size distributions—reduce the top-size ratio below 3:1 or pre-granulate the finer component.
Analytical error—validate assay repeatability by running 10 replicate measurements on the same sample and subtract this analytical variance from the total.

Adopt the 10% rule often used in pharmaceutical and food powders:

Calculate the upper 95% confidence limit on the measured standard deviation s: UCL = s × √(n / χ²₀.₀₅, n-1).
Compare UCL to the specification limit L (e.g., ±5% relative of target concentration).
If UCL ≤ 0.1 L, the blend is deemed uniform; otherwise, continue blending or address segregation sources.
Document both the variance ratio (s²/σ²) and the confidence interval in the batch record for regulatory traceability.

Worked Example: Variance Calculation for Powder Mixtures

A process engineer is assessing the homogeneity (or "mixedness") of a salt and maize puff snack mix. A batch is prepared by mixing 10 kg of salt with 90 kg of maize puffs, targeting a mean salt mass fraction of 10 %-w/w. After 5 minutes of tumble mixing, 12 spatial samples, each approximately 10 g, are collected and analyzed for salt content. The engineer must calculate the sample standard deviation s to determine if the mixture meets the homogeneity specification.

Knowns (Input Parameters and Units):

Number of samples, n = 12.
Measured salt mass fractions (x_i) for each sample in %-w/w: 9.2, 10.1, 9.8, 10.3, 9.9, 10.0, 10.2, 9.7, 10.1, 9.9, 10.0, 9.8.
Target mean salt fraction, x̄_target = 10 %-w/w.
Homogeneity specification limit: s ≤ 0.3 %-w/w (based on customer requirement for ±5% relative of declared mean).
All measurements and calculations are performed at ambient conditions; no temperature or pressure conversions are required.

Step-by-Step Calculation:

Record the measured salt mass fractions x_i in %-w/w, as listed in the Knowns.
Compute the sample mean x̄ using the formula x̄ = (Σx_i) / n. From the data, the calculated mean is x̄ = 9.917 %-w/w.
Compute the sum of squared deviations from the mean: Σ(x_i – x̄)². The calculated sum is 0.897 %-w/w² (rounded to three decimal places from 0.896667).
Compute the sample variance s² using the formula s² = [Σ(x_i – x̄)²] / (n-1). Thus, s² = 0.897 / 11 = 0.0815 %-w/w² (rounded).
Compute the sample standard deviation s by taking the square root of s²: s = √0.0815 = 0.285 %-w/w.
Compare the calculated s to the specification limit of 0.3 %-w/w. Since 0.285 ≤ 0.3, the mixture meets the homogeneity target.

Final Answer:

The sample standard deviation s for the salt mixture is 0.285 %-w/w. This value is within the specification limit of 0.3 %-w/w, indicating that the mixture is sufficiently homogeneous and no further mixing is required.

"Un projet n'est jamais trop grand s'il est bien conçu." — André Citroën

"La difficulté attire l'homme de caractère, car c'est en l'étreignant qu'il se réalise." — Charles de Gaulle