Simulation & digital twin

One machine, watched three ways at once

Watch the donut, signal lamps and highlighted code move together. The displayed Structured Text explains the sequence; JavaScript advances the model. The scan counter represents simulated time at a nominal 1 ms interval, not measured PLC executions or browser timing. Use Step to inspect a model update, and Jam the valve to separate a command from its physical effect.

👀 Watch the link: when bDonutAtStation flips TRUE, line L2 lights, bGlazeValve goes green, and only then does glaze pour. Machine, signals and Structured Text never get out of step, because they are all the same task. That is what "running in simulation" really feels like. Hit ⏭ Step to advance one slice at a time, exactly like single-cycle mode in the online view.

Choose a target and define its I/O

A real TwinCAT project needs a selected target, a compatible runtime, a task configuration and an explicit source for its inputs. The comparison below illustrates those choices; its buttons do not activate hardware. A local PLC runtime does not automatically create an EtherCAT plant model. Beckhoff provides separate EtherCAT simulation tooling for that job.

You just broke a machine and it cost nothing

Go back up and hit ⚡ Jam the valve. Watch it: bGlazeComplete goes TRUE, line L3 tries to shut the valve, but it is stuck open. The donut drowns, the glaze counter climbs and the line faults out. On the real floor that would mean pink syrup over a machine, a stopped line, a very sticky afternoon and a maintenance callout. Here it is a number you reset with a button.

"Simulation" and "twin" aren't the same word

Use the buttons to compare test boundaries: software-only logic tests, physical-controller tests and models associated with real assets. Each answers different questions; adding hardware alone does not make a test complete.

Closed-loop inputs can be part of ordinary software-in-the-loop testing. That alone does not make a digital twin. A twin represents an identified real asset and needs a defined relationship to its operational data, assumptions and model accuracy.

Two that separate the pros

From this model to a TwinCAT test record

1. List the input and output variables used by your sequence and identify which will come from a test harness instead of a terminal.
2. In your isolated test project, select the intended runtime target. Record its TwinCAT build, runtime mode and configured task interval. Confirm licensing and compatibility against the documentation for that installation.
3. Choose PLC-level I/O substitution or an EtherCAT simulation setup deliberately. TE1111 is dedicated tooling; selecting a local target does not create it.
4. Test normal filling, stuck-open output, missing station feedback and restart after the fault. Capture expected versus observed state transitions and task diagnostics.
5. Carry the remaining hardware-dependent checks into commissioning: actual I/O latency, electrical faults, actuator response and the separately validated safety functions.

For the contrasting simulator workflow, see Sysmac Studio simulation. For the general sequence rather than a vendor setup, use the timer and CASE model.

Manufacturer references

• Beckhoff TE1111: what EtherCAT Simulation provides
• TE1111 quickstart: configure a simulation device and its model

These references describe Beckhoff software. The exercise above is an independent TryPLC educational model; no hardware test or manufacturer certification is claimed.