StartNot sure where to start? Pick your standard in 1 minute — free, no registration. Members: sign in with email on the right.
Quality Tool

SPC Control Charts

X̄-R / I-MR / p charts with Western-Electric rules and live out-of-control flags.

Free · runs in your browser
Filled once, reused on every report — the header that makes your SPC report auditable.
Saved only in your browser — reloaded next visit, reusable for every report.
Chart type (8)
Variables data
Attributes data
One subgroup per line (values separated by spaces/commas)
Optional — adds Cp/Cpk (within, from R̄/S̄)

Serves these standards

About this tool

Statistical process control distinguishes two kinds of variation: common cause (the natural noise of a stable process) and special cause (something specific changed — a new lot, a worn tool, a different operator). Control charts make the difference visible, so you act on real signals instead of chasing noise.

Control limits are the voice of the process, computed from your own data at ±3σ — they are not specification limits and must never be drawn on the chart as such. A stable, in-control process can still be incapable, and a capable process can still go out of control; the two assessments answer different questions.

Frequently asked questions

Which control chart should I use?

Choose by data type and subgroup size: individual readings use I-MR; small subgroups of 2–9 use X̄-R; counts of defectives use p (or np with constant subgroup size); defects per unit use c or u. This tool covers the X̄-R, I-MR and p chart families with automatic limit calculation.

What are the Western Electric rules?

The classic sensitising rules flag: one point beyond 3σ; two of three consecutive points beyond 2σ on the same side; four of five consecutive points beyond 1σ on the same side; and a run of eight consecutive points on one side of the centre line. Any hit is evidence of a special cause worth investigating.

Control limits vs specification limits?

Control limits come from the process itself (±3σ of your data) and describe what the process will do; specification limits come from the customer or drawing and describe what you need. Mixing them on one chart is a classic SPC error — judge stability with control limits and conformity with specifications.

How much data before the limits are valid?

Establish limits from at least 25 subgroups (or 100+ individual points) spanning normal operating conditions. Recalculate only after a verified process change — never after every out-of-control point, or the chart will chase its own signals.

Related resources