Retrofit machine monitoring
Retrofit machine monitoring for any machine, no PLC required
Put a sensor on the outside of any machine, from a 1985 press to last year’s mill, and it reports uptime, downtime and parts like a new cell. No wiring, no PLC, no change to the machine.
Trusted by industry leaders
Most machines were never built to report. They still can.
Your floor was bought over decades, from different builders, with different controls or none. Connecting each one through its controller takes a programmer, a protocol and a project per machine. A sensor on the outside takes minutes, and every machine reports the same way.
How it works
It reads the machine, not the controller
A running machine vibrates differently from an idle one, and a finished part breaks a beam. SenseAi reads vibration and sound through its magnetic base, SenseAi Embedded reads at the motor, and BeamTracker counts every part that crosses its laser.
The device decides running or stopped every second. The cloud turns it into uptime, downtime, part counts and health, live on every screen.
Install
On the machine in seconds. Calibrated in minutes.
A SenseAi attaches magnetically in seconds and calibrates in under five minutes, and so does SenseAi Embedded. Run the machine, let it idle, and running and stopped are set for that machine, not an average one.
BeamTracker bolts on with the supplied mount in 10 to 20 minutes. Nothing is wired into the controls, so the install is reversible: take the sensor off and the machine is as it was.
Network
No IT ticket. The router brings its own network.
The IoTFlows 4G/LTE router ships configured with its SIM fitted. Power it, and every device within 100 ft (30 m) joins on its own and reaches the cloud over cellular, off your plant network.
Prefer plant Wi-Fi? The devices need 2.4 GHz access and outbound HTTPS on port 443, set up from the phone app.
Results
A 1987 press reports like a 2026 mill
Once a sensor is on, the oldest machine on the floor sits on the same Assets page as the newest, with the same uptime, downtime, utilization and part counts.
Classify its stops, set its goals and alert on it the same way. Start with downtime tracking or production monitoring, or see an old press counted stroke by stroke on stamping press monitoring.
Fits the machines you already run. Whatever their age.
If it has a motor or a part path, it can report. Each machine type has a set-up guide.
SenseAiCNC mills, lathes and grinders
On the frame near the spindle. Cycle times that change with the part are no problem.
Set up a CNC machine
SenseAiPresses and press brakes
On the frame near the ram, counting 10 to 120 strokes a minute.
Set up a press
SenseAiInjection molding
On the injection unit or near the hydraulic power unit, every cavity counted.
Set up a molding machine
BeamTrackerPackaging and assembly lines
Across the part path after the last reject point. No electronics on the line needed.
Set up a packaging line
SenseAi EmbeddedLasers, waterjets and robot cells
Mounted on the cell and sealed against coolant and spray, reading every cycle it runs.
See SenseAi Embedded
SenseAi EmbeddedWashdown and outdoor positions
Sealed to IP67 and mounted on the motor or spindle itself.
See SenseAi Embedded
Enterprise
Want PLC data as well? Add it when you’re ready.
The IoTFlows edge device reads your PLCs, robots and analog sensors beside the sensors you already run. Alarms and process values feed the same rules and work orders. Our engineers typically have it live in one week to a month.
- MTConnect
- OPC UA
- Modbus TCP
- EtherNet/IP
- Profinet
- MQTT
Customer story
Sorinex had its machines reporting in under a week
Sorinex builds strength equipment for professional teams, colleges and military units in Lexington, South Carolina.
- < 1 week
- from the box to monitored machines
- 60%
- faster issue resolution
- 50%
- more efficiency on critical machines
Challenge
Critical machines ran with no live view of how they performed. Finding why one underperformed or stopped took hours of investigation.
Solution
Sensors went on the critical production machines in under five days, with live dashboards and every stop sorted by cause.
Products
SenseAi, Maintain and the Production Scheduler.
“The speed of implementation was remarkable. We went from limited visibility to complete operational transparency in less than a week.”
Storm Power Components retrofitted chemical baths, CNC machines, lasers and presses and took bath timing errors to zero. Read the Storm Power story. Datwyler runs SenseAi on 20+ injection molding machines.
Questions people ask
A sensor for every kind of machine and a way onto the internet that needs no IT work. SenseAi and SenseAi Embedded attach magnetically in seconds and calibrate in under five minutes; BeamTracker bolts on with the supplied mount in 10 to 20 minutes. Running or stopped is decided on the device, second by second, and the cloud computes uptime, downtime, part counts and health. Devices connect out to the IoTFlows cloud over HTTPS on port 443, on 2.4 GHz plant Wi-Fi or through the IoTFlows 4G/LTE router (IFCR-150), which devices within 100 ft (30 m) join on their own. Nothing is wired into the controls, so the install is reversible. Firmware updates over the air, and each device's page shows its firmware, signal strength and connection state. It fits CNC mills and lathes, presses and press brakes, injection molding machines, packaging and assembly lines, laser cutters, waterjets and robotic cells.
Yes. SenseAi and SenseAi Embedded read vibration from the outside of the machine (SenseAi adds sound, SenseAi Embedded adds temperature); BeamTracker counts parts crossing a laser beam. The sensor decides running or stopped on the device and the cloud computes uptime, downtime and part counts. Nothing connects to the controls.
Yes. Age is irrelevant to a sensor that reads the machine from the outside. A 30-year-old press, a manual lathe or a bagging line with no electronics report the same way a new cell does: a magnetic base on a flat metal surface near the spindle or main motor, or a laser across the part path.
Not with the IoTFlows 4G/LTE router, which gives the sensors their own network and reaches the cloud over cellular; it ships configured with its SIM fitted. On plant Wi-Fi the devices need 2.4 GHz access and outbound HTTPS on port 443.
No. There is no wiring, no PLC integration and no modification to the equipment, so the installation is reversible: take the sensor off and the machine is as it was.
Use SenseAi Embedded. Its IP67 housing is sealed, so it mounts on the motor or spindle itself in washdown and outdoor positions, and reading at the source adds advanced health analytics.
One. A sensor is attached to exactly one asset, the record that represents one machine, and one sensor per machine is the unit IoTFlows counts, bills and reports on. A multi-spindle machine sometimes carries more than one.
Yes, for Enterprise customers, typically plants with 50 or more machines. The IoTFlows edge device reads PLCs and controllers over MTConnect, OPC UA, Modbus TCP, EtherNet/IP, Profinet, MQTT or any other protocol, plus robots and analog and digital sensors, beside the sensors you already run. Controller alarms, program names and process values then feed the same event rules and work orders as sensor data. A project typically takes one week to a month, depending on the machines. The custom integrations page has the detail.
Documentation
Want the detail? These guides walk through each feature step by step.
- How sensors, the cloud and the dashboard fit togetherThe four layers a reading passes through.
- Install and power up SenseAiA flat metal surface near the spindle or main motor, the magnetic base, power.
- Install SenseAi EmbeddedSealed, on the motor, with M12 power.
- Install and align BeamTrackerAcross the flow of finished parts, resolved to the millimetre at up to 8 m.
- Calibrate SenseAi and SenseAi EmbeddedRunning and stopped thresholds against what the machine does.
- Connect a device to Wi-FiProvision a device onto plant Wi-Fi from the mobile app.
- Connect a device through the 4G/LTE routerBring a sensor online without touching the facility network.
- Frequently asked questionsWhich machines fit, what the network needs, what leaves the plant.
Get your oldest machines reporting
Start with 10 to 25 machines for four weeks, with no commitment to buy. Pick the ones nobody thought could be connected.