It usually happens on night shift. A drill has been in the spindle for three hours, the operator is not sure how many parts it has actually cut, and the next workpiece is a castings blank worth several hundred dollars. He faces the oldest decision in machining: change the Tool now and throw away usable life, or run one more part and risk a broken edge, a scrapped workpiece and possibly a damaged fixture.
Most shops resolve this with instinct. And instinct is expensive in both directions. Conservative operators retire tools early, quietly discarding a large share of the cutting edge they already paid for. Aggressive operators run tools to failure, and pay for it in scrap, rework and unplanned spindle downtime. The root cause is not carelessness — it is that nobody actually knows the count. Tool life count RFID exists to close exactly that gap.
The first misconception to clear up is that tool life is a single quantity. In practice it is a budget that can be denominated in at least three different units:
Part count — the natural unit for hole-making tools. A drill rated for 500 holes is half-consumed after 250.
Cutting minutes — better for wear-type tools such as end mills and turning inserts, where engagement time matters more than discrete parts.
Weighted wear units — a synthetic counter that multiplies time by a severity factor, so that roughing in hardened steel consumes the budget faster than a finishing pass in aluminium.
A mature system picks one unit as primary, runs a second as a guard, and triggers a warning at whichever threshold is reached first. This "hybrid" approach is what prevents the classic failure mode where a tool is well inside its part count but has spent an unusually long time cutting.
You can maintain all of this in a database, and many shops try. The problem is that cutting tools do not behave like database records. They move between machines, between shifts, into regrind, into a cabinet, into a different cell. Every physical handover is a chance for the record and the object to drift apart.
An RFID data carrier mounted in the tool holder solves this by making the tool its own ledger. The standard form factor for this is defined in ISO 69873, which specifies the dimensions and mounting location for a data carrier embedded in the retention knob or taper of a CNC tool. Typical devices are tiny — on the order of 8–10 mm in diameter and a few millimetres thick — and are built to survive coolant, oil, chips, cleaning agents and the thermal and vibrational load of a machining centre. Memory is usually in the hundreds of bytes to a couple of kilobytes, which sounds small but is ample for a structured life record. Write endurance for EEPROM-based tool tags is commonly specified in the hundreds of thousands to millions of write cycles, which matters because a counter that is rewritten every cycle can otherwise become the wear item.
Because the tag is read contactlessly by a read/write head at the tool magazine, the presetting station or the cabinet, the count can be updated at the moment of truth rather than at the end of a shift.
There is more than one way to get consumption data onto that tag, and the right choice depends on how modern your fleet is.
1. CNC-integrated. Modern controls expose tool-life data over interfaces such as OPC UA, MTConnect or FOCAS. The system reads actual cutting time, load and tool-offset changes straight from the control and writes the updated remaining-life figure back to the tag. This is the most accurate option and the least work for the operator.
2. Work-order driven counting. For older machines with no usable interface, the middleware counts instead: when a work order reports a completed part, the system debits each tool on that operation by its per-part consumption coefficient. No sensors, no protocol licences, and it works across mixed-vintage equipment. The trade-off is that it models consumption rather than measuring it.
3. Sensor-based condition monitoring. Spindle power, acoustic emission and vibration sensors infer wear directly from the physics of the cut. This is the most sensitive approach and the most expensive, and its output is still best Stored against the tool's own identity — which is where the tag comes back in.
In practice, most plants run a mix: sensor-based monitoring on the handful of operations where a failure is catastrophic, CNC integration where available, and work-order counting everywhere else. RFID is the common identity layer that lets all three write into the same life record.
Counting is only useful if the number travels. The full loop looks like this:
Presetting: the presetter measures diameter, radius and length and writes them to the tag. When the tool is loaded, the machine reads them in — eliminating the manual offset entry that is a notorious source of crashes.
Loading: a read head at the magazine verifies the tool ID against the NC program. Wrong tool or zero remaining life is an immediate interlock, not a discovered-after-the-fact scrap event.
Machining: remaining life is debited per the chosen logic and written back on unload.
Return and regrind: the cabinet knows how many regrinds the tool has had and what its current geometry is; a reground tool re-enters service with a fresh, correctly-reduced life budget.
Retirement: the tag's history — parts cut, hours, regrounds, machines used — becomes the evidence base for supplier selection and for tweaking the life standard itself.
That last point is the one most often missed. The first life standard you load into the system is a guess. The value of counting is that after a few months you can replace the guess with a distribution, and then set the threshold where the risk actually belongs.
Take a single tool with a purchase price of $120, a regrind cost of $35, and a conservative operator-managed life that ends after 400 parts. If the data shows the tool reliably holds tolerance to 550 parts, you have recovered roughly 27% of the per-part tooling cost on that operation — and, just as importantly, you have removed the judgment call from the night shift. Multiply by the number of tools in circulation and by the avoided scrap events, and the payback on tags and read heads is typically measured in months rather than years, especially on high-value PCD or carbide tooling. The arithmetic is simple; the discipline is what most shops lack.
Writing too often. Updating the counter on every single cycle burns write endurance and adds no value. Debit in batches or on unload.
Ignoring metal. A tool shank is steel. Use a tag designed for mounting in metal, and respect the specified read distance — these are deliberately short-range devices, which is what prevents cross-reads between adjacent pockets.
Treating the tag as the database. Store the detailed history in MES; keep on the tag only what the machine needs to act on if the network is down.
Setting thresholds once and forgetting them. Life standards should be re-derived from actual data quarterly, or the system becomes a more expensive version of the old guess.
Skipping the operators. If the counter adds steps to a tool change, it will be bypassed. The read must happen automatically, as a by-product of loading the tool.
Tool life count RFID is not really a tagging project. It is a decision-quality project: it converts "I think it's about done" into "this tool has 83 parts of budget left," and it makes that number travel with the tool through every machine, cabinet and regrind cycle. The shops that get the most out of it are not the ones with the most sensors — they are the ones willing to let the data override the instinct, and then keep revising the standard as the data accumulates.
Contact: Adam
Phone: +86 18205991243
E-mail: sale1@rfid-life.com
Add: No.987,Innovation Park,Huli District,Xiamen,China
Adam