Conveyor Belt PLC Programming: Logic, VFDs and Safety

A reliable conveyor PLC program separates five concerns: safety, operating mode, motor/VFD command, product tracking and fault recovery. The standard PLC may request motion only when its permissives are true; a separately designed safety function removes or controls hazardous motion. Start downstream before upstream, stop upstream before downstream, and verify every transition with feedback and timeouts.

The ladder or Structured Text is rarely the hardest part. The real engineering work is defining what “ready,” “running,” “blocked,” “safe” and “recovered” mean for the actual conveyor.

Reference architecture

Layer	Typical components	PLC responsibility
Safety	E-stops, pull cords, guard switches, safety relay/PLC, safe drive functions	Read diagnostic status; do not replace the safety function
Motor power	Contactor, overload, VFD, brake, motor	Issue command and verify feedback
Detection	Photoeyes, prox sensors, encoders, jam switches	Debounce, track and diagnose
Sequence	Zones, merges, diverts and downstream handshakes	Coordinate starts, stops and product flow
Operator	Pushbuttons, selector, HMI and alarms	Present state and accept authorized requests
Supervisory	SCADA, historian and maintenance data	Publish diagnostics without making local control network-dependent

Define the control contract before programming

For each conveyor drive, define:

Signal	Direction	Meaning
xRunRequest	PLC → starter/VFD	Standard control requests motion
xDriveReady	starter/VFD → PLC	Drive can accept a command
xRunningFeedback	starter/VFD → PLC	Drive or verified auxiliary contact reports running
xDriveFault	starter/VFD → PLC	Drive has an active fault
xSafetyHealthy	safety system → PLC	Diagnostic indication only; not the safety function
xDownstreamAvailable	sequence → zone	Next zone can receive product
xProductAtExit	sensor → PLC	Product occupies the discharge sensor

Then write the acceptance condition for each signal. “Motor running” might mean a VFD status bit, contactor auxiliary contact, zero-speed switch or measured encoder motion. Those are not equivalent.

Stop-dominant motor logic

The command must drop when stop, fault, lost permissive or mode change requires it. A restart should require a new deliberate request unless the functional design explicitly justifies automatic recovery.

xReady :=
    xAutoMode
    AND xSafetyHealthy
    AND xDriveReady
    AND NOT xDriveFault
    AND xDownstreamAvailable;

IF xStopPB OR NOT xReady THEN
    xRunLatched := FALSE;
ELSIF xStartPB THEN
    xRunLatched := TRUE;
END_IF;

xRunRequest := xRunLatched AND xReady;

tonRunProof(
    IN := xRunRequest AND NOT xRunningFeedback,
    PT := T#2S
);

xStartFail := tonRunProof.Q;

The two-second timeout is an example test value, not a design recommendation. Set it from drive behavior, contactor pickup, brake release and process risk.

VFD integration

A VFD interface needs more than speed reference:

• run/stop command;
• drive ready and active status;
• fault code and fault present;
• commanded and actual speed;
• communications health;
• local/remote ownership;
• safe-torque-off diagnostic status where applicable; and
• reset request with edge and authorization control.

Network or hardwired?

Interface	Strength	Limitation
Hardwired run plus analog speed	Simple failure boundary and familiar commissioning	Less diagnostic detail; analog scaling and wiring drift
Industrial network	Rich status, speed, current and diagnostic data	Switch, addressing, configuration and communications become dependencies
Hybrid	Hardwired critical command with network diagnostics	More design and test effort

Select the interface from the risk assessment, process response and maintenance model. Do not assume that a networked command is either inherently safer or less safe.

Scale a speed reference explicitly

If the PLC command is 0–100% and the drive network reference is 0–16,384:

rSpeedPct := LIMIT(0.0, rOperatorSpeedPct, 100.0);
iDriveReference := REAL_TO_INT((rSpeedPct / 100.0) * 16384.0);

Record the units and range at both ends. Test 0%, minimum operating speed, 50%, 100%, out-of-range input and communications loss.

Product detection and tracking

Debounce sensors without hiding real products

Use an on-delay only when the minimum product dwell time is known:

tonEntryDebounce(
    IN := xEntryPhotoeye,
    PT := T#30MS
);

xEntryStable := tonEntryDebounce.Q;
rtrigEntry(CLK := xEntryStable);

The 30 ms value is an example. Calculate sensor dwell from product length and maximum belt speed, then test the smallest legitimate product and the shortest gap.

Track product by events, not assumptions

For a simple conveyor, count a product on a rising edge at entry and remove it at exit:

rtrigEntry(CLK := xEntryStable);
rtrigExit(CLK := xExitStable);

IF rtrigEntry.Q THEN
    diInCount := diInCount + 1;
END_IF;

IF rtrigExit.Q THEN
    diOutCount := diOutCount + 1;
END_IF;

diProductsOnConveyor := diInCount - diOutCount;
xTrackingMismatch := diProductsOnConveyor < 0
    OR diProductsOnConveyor > cdiMaxProducts;

For sorting, store a product record in a FIFO with destination, quality result and tracking ID. Shift or advance it only on a proved conveyor movement event. Define what happens after a stop, manual jog, sensor failure and product removal.

Three-zone zero-pressure accumulation example

For zones 1–3, run a zone when it contains product and the next zone can accept it:

xZone3Available := NOT xZone3Occupied OR xDischargeReady;
xZone2Available := NOT xZone2Occupied OR xZone3Available;
xZone1Available := NOT xZone1Occupied OR xZone2Available;

xZone3RunRequest :=
    xAutoPermissive
    AND xZone3Occupied
    AND xDischargeReady;

xZone2RunRequest :=
    xAutoPermissive
    AND xZone2Occupied
    AND xZone3Available;

xZone1RunRequest :=
    xAutoPermissive
    AND xZone1Occupied
    AND xZone2Available;

This teaching example omits transfer overlap, sensor placement, product length, coast distance and drive feedback. A production program usually needs a state per zone:

Empty → Receiving → Occupied → Releasing → Empty
                       ↘ Faulted

Each transfer should have a maximum expected time and a destination proof. If Zone 2 releases but Zone 3 never proves receipt, stop upstream feed and raise a diagnostic that names the failed transfer.

Startup and shutdown sequence

For a conveyor train:

1. Prove the discharge machine is ready.
2. Start the furthest downstream conveyor.
3. Prove its running feedback.
4. Start the next upstream conveyor.
5. Continue toward the infeed.

For a normal stop:

1. Stop accepting new product.
2. Stop the furthest upstream feed.
3. Allow downstream conveyors to clear as designed.
4. Stop downstream sections after clear proof or a justified timeout.

An emergency stop is different. Its behavior comes from the risk assessment and safety design, not from a production clear-out sequence.

Conveyor safety boundary

The standard PLC sequence must not be presented as the emergency-stop function. ISO 13850:2015 specifies principles for emergency-stop functions, while electrical realization is addressed by the applicable machinery electrical standard. In the United States, OSHA 1910.147 covers hazardous-energy control during servicing; OSHA specifically uses a jammed conveyor as an example of hazardous unexpected release.

At minimum, the project safety process must address:

• accessible emergency-stop and pull-cord devices;
• guard and access-point hazards;
• stored mechanical, pneumatic and gravitational energy;
• restart prevention and manual reset behavior;
• stopping time and coast distance;
• safe drive functions where selected;
• lockout/tagout for jam clearing and servicing; and
• validation of every safety device and fault response.

Fault model and recovery

Use specific faults rather than one ConveyorFault bit:

Fault	Detection	Recovery requirement
Failed to start	Command on, no running proof before timeout	Stop affected/upstream zones; inspect drive path
Unexpected running	Running proof without command	Remove standard request; follow safe response
Transfer timeout	Source released, destination not proved	Stop feed; inspect product/sensor
Sensor blocked	Photoeye active longer than justified dwell	Diagnose jam, contamination or failed sensor
Tracking mismatch	Count/FIFO outside valid bounds	Stop sort decisions; controlled resynchronization
Network loss	Drive or remote I/O quality bad	Defined degraded state; no automatic restart
Overload/VFD trip	Protection reports fault	Correct physical cause before authorized reset

Reset should clear a latched diagnostic only after the initiating condition is gone. It should not force a sensor, bypass a guard or automatically restart the conveyor.

Commissioning test matrix

Test	Injection	Expected evidence
Normal start	Start with all permissives true	Downstream-first sequence and running proof
Start feedback failure	Block one run feedback	Timeout identifies the correct drive; upstream feed stops
Product transfer	Move one product through all zones	One entry, one exit, no count drift
Short product/gap	Run minimum product at maximum speed	Sensor filtering still detects valid events
Blocked exit	Hold final zone occupied	Upstream zones accumulate without collision
Sensor stuck on	Hold one photoeye active	Specific blocked-sensor or transfer diagnostic
Communications loss	Remove one networked drive/I/O path	Defined stop/degraded behavior and bad-quality display
Stop during transfer	Issue stop in each zone state	No unintended restart or duplicate tracking event
Power cycle	Restart with products present	State follows documented recovery/resync procedure
Safety validation	Qualified test of every safety device and fault	Separate signed safety evidence

Practice the sequence in a simulator

Open the conveyor exercises in PLC Simulation Software. Disclosure: PLCProgramming.io and PLC Simulation Software have common ownership. The product is a training simulator; it does not model the stopping time, safety performance, VFD firmware, electrical protection or mechanical hazards of a real conveyor.

Related practical guides:

• Motor start/stop ladder logic
• E-stop safety circuit and ladder logic
• Packaging machine PLC programming
• PLC troubleshooting workflow
• PLC scan cycle explained

Limitations

• Example timer and scaling values are placeholders to be replaced from project evidence.
• Product tracking examples omit encoder slip, physical removal and multiple-lane complexity.
• Safety design and validation must be performed by competent personnel under the applicable standards and law.
• A simulator result cannot certify stopping distance, guarding, electrical design or production suitability.
• Never clear a jam or service a conveyor based only on an HMI stop indication; follow the site energy-control procedure.

Frequently asked questions

Should conveyors start upstream or downstream first?

Normally start downstream first so material has somewhere to go, then prove each section before starting the next upstream conveyor. Validate the sequence for the actual process.

How do I detect a conveyor jam?

Use a defined transfer expectation: the source sensor releases a product and the destination sensor must detect it within a justified time. Combine this with drive current, speed or zero-motion evidence where appropriate.

Can the PLC emergency-stop the conveyor?

The standard PLC code in this article is not an emergency-stop function. Implement emergency stopping through the risk-assessed safety architecture and validate the complete input, logic, output and stopping behavior.

What should happen after a power failure?

Default to no unintended automatic restart. Reconcile product positions, controller state, drive state and safety reset before accepting a new start request.

How do I prevent product collisions?

Use occupied/available zone states, destination proof and transfer timeouts. Include product length, sensor placement, belt coast and maximum speed in the design.

Primary sources

• ISO 13850:2015 — Emergency stop function, principles for design
• OSHA 1910.147 — Control of hazardous energy
• OSHA: Control of hazardous energy overview
• OSHA 1918.64 — Powered conveyors

For warehouse-scale routing, load identity and transactional handshakes, continue with the automated storage and retrieval system PLC-controls guide.