Mixed-vendor machine monitoring rig
Three PLC brands and one machine with no controller at all, read through one gateway and normalised into a single tag model — the situation almost every Indian plant is actually in.
Disclosure. This is our own test rig, not a customer installation. The equipment is real and the data path is real; the 'production' it monitors is a demonstration load. No customer, no customer data, and no production results are represented here.
Context
Why this exists
The most common objection we hear is that a plant's machines are 'too mixed' to monitor. This rig exists to demonstrate that the mix is the normal case rather than the obstacle, and to let us show the data path end to end rather than describe it.
The constraints
- 01
Three controllers using three different protocols, with different addressing models
- 02
One machine with no controller, exposing nothing electrically except a motor
- 03
Inconsistent tag naming, scaling and units across the three controllers
- 04
A deliberately unreliable network link, to test buffering rather than assume it
The approach
- 01
Read each controller natively
Modbus TCP on one, OPC UA on the second, and a Siemens S7 connection on the third — each polled read-only at a rate appropriate to the signal rather than at one global rate.
- 02
Instrument the unmanaged machine
A current transformer on the motor feed for run/stop and load, plus a proximity sensor for piece counting. No modification to the machine.
- 03
Normalise at the edge
All four sources mapped to one tag model with consistent names, types, scaling and units, timestamped at the edge rather than on arrival.
- 04
Derive state, not just values
A state machine on the gateway classifies running, idle, stopped, changeover and fault from raw signals, so downstream systems consume state rather than re-deriving it.
- 05
Prove the buffer
The uplink is cut deliberately during operation. Data queues locally and back-fills in order on reconnect, with de-duplication verified.
Want this walked through on a call?
We will screen-share the rig, show the data path end to end, and answer the awkward questions about where the limits are.