Can older CNC and VMC machines connect to an OEE dashboard?

Often yes, with a defined measurement scope

An older CNC or VMC does not automatically need to be replaced to become visible in a production dashboard. The practical question is which trustworthy signals can be acquired from that particular machine and what those signals can establish. A basic state timeline and a complete OEE calculation do not require the same data.

Start with the decision your team needs to make: compare shift activity, investigate stops, count production or calculate OEE by product. This helps distinguish a useful first monitoring scope from a promise to retrieve every controller value from every machine. Compatibility must be assessed at the model and configuration level.

Survey the controller, not just the machine brand

Record the machine builder, controller family, exact model, software version where available and existing communication options. Note any accessible part counters, run signals, alarm information or documentation. Machines carrying the same brand can have different interfaces or licensed options.

Confirm who is authorised to inspect or configure the equipment. A site survey should review the existing network and controller responsibilities with the owner or responsible engineer. Do not assume that an unused port can be connected or that a protocol option is enabled simply because it appears in a product brochure.

Compare three possible acquisition paths

First, an existing controller interface may expose the required data through a supported protocol or manufacturer interface. OPC UA and MTConnect are examples worth checking where the equipment implementation supports them; their names are not a compatibility guarantee. Confirm actual available signals, access permissions and any option or licence requirements.

Second, a suitable PLC or approved electrical signal may expose a limited machine-state or cycle indication. Third, a separately engineered sensor or acquisition arrangement may provide selected observations. These routes have different installation and validation requirements. A sensor used to infer activity should not be represented as a direct controller production count without evidence.

Understand what is still needed for OEE

Availability needs an agreed planned-production boundary and run-time definition. Performance also needs output quantity and an appropriate ideal cycle time for the job. Quality needs good quantity and a clear treatment of rejects or rework. A cycle pulse alone does not supply the full model.

Decide how job selection, counts and quality inputs will be obtained and reconciled. An operator input or business-system interface may be part of the agreed design when the controller cannot provide it. Begin with the measurements you can explain and label incomplete inputs rather than presenting an estimated OEE score as a measured result.

Normalise meaning across a mixed machine fleet

A common dashboard should use stable machine identities and shared reporting definitions. One controller’s automatic mode may not mean the same thing as another machine’s spindle-running signal. Agree the mapping with production personnel and preserve the original source information needed to review it.

Distinguish stopped equipment from missing communication. If the newest machine provides alarms and counters but an older machine provides only activity states, show those differences honestly. A common visual layout does not require pretending that every asset supplies the same information.

Prove the connection with a representative pilot

Select a machine from each important controller or acquisition group. Observe an agreed production window and compare the collected events against a supervised reference. Check short stops, job changes, counter resets, shift boundaries and time alignment with the application.

Test the agreed handling of communication interruption and reconnection. Verify whether missing intervals are flagged, whether buffering exists in the selected design and how the next report treats the gap. Document limitations before expanding; a successful demonstration on one controller does not prove the entire fleet.

Keep monitoring work within the approved installation scope

Read-only monitoring should have an explicit access boundary and agreed installation method. Any wiring, new instruments or controller changes need approval and appropriate engineering procedures. Monitoring does not replace the machine’s local controls, protective functions or maintenance requirements.

Schedule implementation around the actual access and shutdown needs of the selected method. Do not assume zero downtime or an installation time per machine without a survey. Record who maintains the connection and who revalidates it after equipment, software or network changes.

How to start a WhereFy connectivity assessment

Send the machine and controller inventory, the approximate asset count, sample reports you need and the available operating information. If model labels or manuals are available, share the relevant non-confidential details. Explain which machines are similar and which have different controllers or production methods.

WhereFy PLC–IoT integration and Smart Insight can be scoped around the confirmed acquisition path and production model. Ask for a per-machine capability list: measured states, available counters, manual inputs, known gaps and pilot acceptance tests. That is a more useful basis for a mixed-fleet project than a blanket promise to connect every old machine.

What to check before choosing a legacy-machine connection