Pre-Regulator Tool: How Tool Presetting Development Trends Change Your Shop Floor

• 1
  What a pre-regulator tool actually does
• 2
  Manual presetting: micrometers, height gauges and a pencil
• 3
  Optical presetters and the arrival of the digital offset
• 4
  CCD measurement and what 1 μm resolution buys you
• 5
  Networked presetting and the tool data loop
• 6
  Where the error budget really goes
• 7
  Four presetting generations side by side
• 8
  When an offline presetter is worth it

A pre-regulator tool is an offline measuring station. You clamp the tool holder in a spindle taper that matches the machine, touch the cutting edge with an optical or contact probe, and the unit returns length, tip diameter and sometimes rake or relief angle. Those numbers are stored against a tool number and pushed to the control before the tool ever enters the spindle.

The value is not the measurement itself. It is that the measurement happens while the spindle is cutting something else. On a machine running 30 tools per job, touching off each one in the spindle can eat 15 to 30 minutes per setup. Presetting moves that time off the machine and onto a bench, where a single operator can prepare the next job while the current one runs.

Presetting also decouples two things that used to be welded together: the tool and the machine. Once length and diameter live in a data record, you can swap a worn tool for an identical backup without re-teaching the control. That is the real reason presetters spread from aerospace and automotive into general job shops.

The boundary matters more than the pitch. A pre-regulator tool cannot fix a holder with 20 μm of runout, and it does not measure thermal growth during a long cut. It sets a starting point. Cutting still has to be proven.

• 1
  Measures offlineLength, tip diameter, angles, taken away from the spindle.
• 2
  Saves spindle timeSetup moves to the bench while the machine keeps cutting.
• 3
  Decouples tool from machineIdentical backup tools drop in without re-teaching.
• 4
  Not a cure-allIt cannot correct holder runout or thermal drift.

The first generation was pure bench work. An operator set a height gauge or micrometer against the tool tip, read a number, and wrote it on a setup sheet. Diameter for a face mill or boring bar often came from a caliper across the tips. Then the number was typed into the NC program or dialed into the offset page by hand.

This worked, and in some shops it still works. For a 3-axis job with six tools and a tolerance of ±0.05 mm, a height gauge is fast enough and costs almost nothing. The errors that do appear are usually transcription errors, not measurement errors. A digit dropped between the bench and the control is the classic failure.

The real cost shows up at the extremes. When a positioning-tightening tool system demands tip diameter within ±2 μm, a caliper is not a measuring instrument anymore. It is a guess. Hand presetting also scales badly: ten machines and one setup sheet is fine, ten machines and forty tools per shift is not.

So the manual generation did not die because it was inaccurate. It died where the tolerance, the tool count, or the documentation requirement outgrew what a person with a gauge could hold in their head.

The second generation put a screen between the tool and the operator. A tool is clamped in a precision spindle, a projection lens or a camera throws a magnified silhouette onto a display, and crosshairs are moved to the cutting edge. The unit reads X and Z directly and stores them against a tool ID.

Two things changed. First, repeatability. A good optical presetter holds repeatability around 2 to 5 μm, which is enough for most milling and turning work at ±0.02 mm. Second, the offset stopped being a number on paper and became a record in a database. That is what allowed tool data to be sent to the machine instead of typed into it.

Optical measurement has a boundary that engineers hit quickly. Edge geometry matters. A coated carbide end mill with a small edge radius reads differently under different lighting than a freshly ground HSS drill. Shadow-based measurement sees the outline, not the material, so a burr or a chip on the edge will move the reading.

Cleaning becomes part of the process. Wipe the tool, check the taper, then measure. Shops that skip the wipe step get 10 μm of scatter and blame the machine.

CCD-based units replaced the projection screen with a digital sensor and image processing. The practical result is that measurement resolution dropped to roughly 1 μm on length and diameter, and angle measurement became routine rather than a special setup. Point measurement, radius measurement and angle measurement can be picked from a menu instead of aligned by hand.

That resolution matters most for small tools. A Ø0.5 mm drill or a micro end mill has an edge that occupies very few pixels on an optical screen. CCD with proper magnification resolves it. For a Ø50 mm face mill, the extra resolution is mostly wasted, because the process variation from the holder and the machine is larger than the measurement error.

The larger change is data. A modern presetter exports tool data in a format the control can read, so the operator confirms rather than types. On a shop running high-mix work, that removes a whole class of scrap events: the wrong length entered against the wrong tool number.

Do not read 1 μm resolution as 1 μm accuracy. Resolution is what the sensor can display. Accuracy depends on spindle taper condition, temperature, and how well the unit was calibrated. A presetter calibrated once a year and never checked against a setting master will drift.

• 1
  Best fitSmall-diameter tools, tight diameter control, frequent tool changes.
• 2
  Poor fitLarge face mills where holder runout dominates the error budget.
• 3
  WatchCalibration schedule and a setting master for verification.

The current trend is not a better sensor. It is closing the loop. Tool data moves from the presetter to a tool management database, then to the machine control, and sometimes back again when a tool is re-measured after regrinding. Each tool carries its own history.

This changes how a shop plans. If every tool has a known remaining life and a known geometry, you can schedule regrinding instead of reacting to a broken edge. You can also prove traceability, which matters in automotive and medical work where the process record is part of the deliverable. Our own shops run ISO 9001:2015, IATF 16949:2016 and ISO 13485:2016, and tool records are part of what those audits look at.

For a shop with three machines and a stable part family, the network layer is hard to justify. For a shop running 127 machines across multiple plants, it is the only way to keep tool data consistent. The dividing line is usually how many people touch the same tool data.

There is a cost that vendors understate. Someone has to own the database. Tool IDs have to be physical, not just digital. If the label falls off the holder, the loop breaks at the machine.

Engineers often ask what presetter accuracy they need for a given part tolerance. The honest answer is that the presetter is rarely the largest term. A typical error budget for a milled feature at ±0.02 mm includes machine positioning, thermal growth, holder runout, tool deflection, and presetting error, in roughly that order of magnitude.

If the presetter contributes 2 μm and the holder contributes 15 μm, buying a 1 μm presetter does not improve the part. Fix the holder first. Check taper contact, measure runout with a dial indicator, and replace anything above 10 μm on a finishing tool. That single step often does more than a hardware upgrade.

Presetting error does dominate in one case: multi-tool work where several tools cut the same surface. A face mill and a shoulder mill that must blend at the same Z height will show any length error directly on the part. That is where the tighter presetter pays for itself, and it is usually visible in the first article.

The practical rule: match presetter accuracy to about one fifth of the tightest feature tolerance, then spend the rest of the budget on holders and thermal control.