Back to all posts
Control Systems7 min read·October 25, 2026

What Should an HMI Show When a Machine Will Not Start?

TB
ThynkBlox Automation Team
Automation Engineering

When a machine will not start, the HMI should show which start conditions are unmet, whether the controller received and accepted the start command, and where the sequence is waiting. Put those answers beside the start control, with a clear next action and a route to the relevant equipment detail. Use words and symbols alongside colour, and distinguish a confirmed blocking condition from missing or stale information.

What should the operator see first?

Design the machine overview to answer a short chain of questions:

  • Which machine or section am I controlling?
  • Who has control: this screen, a local station or another system?
  • Is a start request allowed?
  • What happened to the last request?
  • What is preventing progress now?

Use a plain summary such as “Start blocked: discharge conveyor unavailable.” Under it, show the current mode and sequence state, followed by the unmet conditions.

Avoid making “Not ready” the final answer. It describes the outcome without explaining it. Likewise, a disabled start button needs a visible reason; the operator should not have to press it to discover why it is unavailable.

Keep the summary concise. Put supporting detail one navigation step away.

How should an HMI display start permissives?

For the screen design, define a start permissive as a condition that must be satisfied before the controller accepts a start request. Agree the exact meaning with the controls team because the conditions and terminology belong to the machine's operating design.

Write each condition in language that describes the required state. Prefer “Discharge conveyor ready” to a raw controller tag or an unexplained code.

Give each condition an explicit status:

ConditionStatusDetail or next action
Operating mode is automaticMetAutomatic selected
Discharge conveyor is readyNot metOpen discharge conveyor detail
Product route is confirmedNot metConfirm the intended route
Drive status is availableUnknownStatus communication unavailable

This table is illustrative; the actual conditions must come from the machine's control logic.

Show unmet conditions first, while keeping the complete list available. Include an “Unknown” state so missing data cannot look like a satisfied condition.

Separate requirements for starting from conditions that stop an already running machine. If the same condition serves both purposes, explain that relationship in the detail view. Do not use one ambiguous “Interlock” label for every restriction.

What should happen after someone presses Start?

Specify command feedback as a sequence of distinct events. A button animation should confirm only the interaction it actually represents.

Useful feedback states include:

  • Request sent: the HMI has issued the command.
  • Request received: the controller has acknowledged receipt.
  • Request rejected: the controller has declined it, with the reason shown.
  • Starting: the controller has entered the start sequence.
  • Running confirmed: the required operating feedback is present.

Only display a state when the system has evidence for it. If controller acknowledgement is unavailable, show that limitation instead of implying acceptance.

Keep the result of a rejected request visible long enough to inspect. Show the rejection reason recorded for that request separately from the current readiness state. Otherwise, a condition that changes immediately after the press could leave the operator unable to understand what happened.

Define this behaviour in the functional design specification, including command ownership, acknowledgement and communication failure.

How should the current sequence state be explained?

Name the active step and describe what it is waiting for. “Starting” alone leaves too much unanswered.

An illustrative sequence display might read:

Current step: Establish extraction

Command: Extraction fan start requested

Waiting for: Airflow confirmation

Next step: Enable material feed

This separates the instruction from the evidence needed to continue.

If the sequence includes a defined waiting period, show its purpose and progress. If the permitted wait expires, replace the progress message with the resulting fault and the approved recovery action.

Keep “Waiting,” “Paused,” “Aborted” and “Complete” distinct. Specify what each means and which actions are available. Do not label a sequence “Paused” if continuing requires a reset and a fresh start request.

For maintenance detail, retain the technical step identifier alongside the readable description. The operator-facing explanation should make sense without it.

Which equipment details belong beside the diagnosis?

Show the equipment information that explains the blockage. For a drive, that could include control location, ready status, fault status, command and running feedback.

Keep requested and observed states separate. “Run requested” describes an instruction; “Running confirmed” describes feedback. Where they disagree, make the disagreement visible.

Also identify what each observation proves. If the display relies on drive running feedback, label it accordingly rather than calling it “Product moving.” The latter needs evidence about product movement.

For analogue conditions, show the measured value, its unit and the required condition where those values are defined in the operating design. Avoid asking an operator to interpret an unexplained bar or gauge.

When data is unavailable, retain the equipment identity but mark its state as unknown. Do not leave a previous “Ready” indication looking current.

How should navigation support fault diagnosis?

Make the blocking condition a direct route to the relevant equipment or sequence detail. Preserve the machine context and provide an obvious return path.

A practical navigation pattern is:

  1. Open the machine overview and read the blocking summary.
  2. Select an unmet condition.
  3. Inspect the equipment state, command feedback and fault detail.
  4. Follow the approved operating or maintenance procedure.
  5. Return to the overview and confirm readiness again.

Keep navigation controls visually distinct from actions that command equipment. Opening a diagnostic page should not also acknowledge, reset or start anything.

Use consistent equipment names across overview screens, detail pages and alarms. For the relationship between faults, alarms and operator response, see why operators ignore alarms.

How can the display work without relying on colour?

Pair colour with readable status text and a consistent symbol. A condition should still make sense when its colour is unavailable or difficult to distinguish.

Use labels such as “Met,” “Not met” and “Unknown.” For equipment, use explicit states such as “Stopped,” “Starting” and “Running confirmed.”

Do not make position, shape or flashing carry the whole meaning either. Keep labels visible, use sufficient contrast and check the screen in the lighting and viewing conditions expected at the machine.

During review, inspect a version without colour. The blocking reason, current state and next action should remain understandable.

What common mistakes should the design avoid?

Sending every diagnosis to the alarm list. Give the overview its own readiness explanation. Agree which conditions warrant an alarm and which simply explain why a start is unavailable.

Showing only the first blocker. Display all known unmet conditions, while distinguishing the condition that rejected the last request.

Offering Reset as the default answer. Show what requires resetting, the conditions for doing so and what the reset changes.

Displaying instructions beyond the operator's authority. Link to approved procedures and identify when maintenance support is required.

Treating an HMI indication as permission to intervene. Safety-related design and recovery procedures should be developed with a qualified assessor. Diagnostic text should never instruct someone to bypass a safeguard.

How should the design be checked before handover?

Agree acceptance checks against the actual control sequence:

  • Each unmet start condition produces a specific explanation.
  • Rejected commands show the recorded reason.
  • Accepted commands remain distinct from confirmed operation.
  • Sequence waits identify the expected feedback.
  • Lost communication produces an explicit unknown state.
  • Diagnostic links open the correct equipment detail.
  • Text and symbols preserve meaning without colour.
  • Recovery instructions match approved procedures.

Include combined conditions, changing conditions and unavailable feedback. A screen that explains a single planned fault still needs checking when several restrictions exist together.

What else do plant teams ask?

Should the Start button disappear when starting is blocked?

Prefer a stable location with a visible availability state and explanation. Hiding it removes a useful point of reference.

Should every permissive appear on the overview?

Show the blocking summary and unmet conditions there. Keep the full readiness list accessible without crowding the normal operating view.

Does every machine need a sequence diagram?

Use one when it helps explain several dependent steps. For simpler equipment, a current-state label, command result and readiness list may be sufficient.

Can clearer screens resolve the underlying machine fault?

The design goal is to explain the available evidence and guide the next authorised action. Mechanical, electrical and control faults still require appropriate investigation and repair.

Next step

Want to apply this to your plant?

Every project starts with a fixed-fee assessment, and you can stop after the report.