Application of the M&H Measurement System in Mold Machining
This page is for mold shop engineers and programmers who are deciding whether to probe on the machine or keep sending work to the CMM. It covers how an application mh measurement system is set up, what it can and cannot hold, and how to tell whether it fits a given mold cavity, insert or electrode.
What an on-machine measurement system actually does
A probe, a receiver, and software that turns touch points into a decision before the part leaves the spindle.
Where the probe touches and where it stops
An M&H measurement system is a touch-probe setup that runs inside the machine tool, not on a separate granite plate. A spindle-mounted probe, an infrared or radio receiver, and software such as 3DFORM sit between the control and the part. The probe finds surfaces by contact; the software converts those points into position, size, and form data.
The appeal for mold work is simple. A cavity that is still clamped in the chuck can be measured, adjusted, and re-cut without a trip to the coordinate measuring machine. Setup errors and datum shifts show up while the part is still dialed in. That alone removes a large slice of scrap risk.
One thing to keep straight: on-machine probing measures the part in the machine's coordinate frame. It does not replace a temperature-controlled CMM for final certification. It verifies the process while the part is still cuttable. Treat it as process control, and it earns its keep.
- 1ProbeSpindle-mounted touch probe, typically ±1 μm repeatability class.
- 2ReceiverInfrared or radio link to the control, compatible across brands.
- 3SoftwareTurns touch points into datum, size, and form decisions.
When in-process gauging pays off in mold work
Not every mold job needs probing. High-mix work with short runs rarely justifies the programming effort. The payback shows up on parts with deep pockets, thin walls, or long cycle times, where a single re-fixturing error costs more than the probe cycle.
Mold inserts are a good fit. Deep ribs, tapered walls, and blending radii are hard to reach with hand tools and easy to scrap with a single offset mistake. Probing each cavity after roughing confirms stock-on before finishing, so the finisher cuts to the correct allowance.
Electrodes for EDM are another fit. A copper or graphite electrode can be probed for wear and undersize before it burns the cavity. The measurement is quick, and the decision is binary: burn or re-cut. On five-axis work, that decision loop runs in minutes.
Where it does not pay: simple prismatic plates, one-off prototype brackets, and parts that will be cut and shipped the same day. The programming and verification time exceeds the value of the check. Be honest about the part mix before buying software.
Probe use by part type
A rough guide, not a rule.
| Part type | Probe worth it? | Why |
|---|---|---|
| Deep-cavity mold insert | Yes | Stock-on check before finishing saves the cavity. |
| EDM electrode | Yes | Wear and undersize caught before burning. |
| Thin-wall 5-axis part | Usually | Deflection and wall thickness verified in setup. |
| Prismatic plate | Rarely | Datum and size are stable; CMM is faster. |
| One-off prototype bracket | No | Programming time exceeds the check value. |
| High-mix short-run batch | Case by case | Only if cycle time or scrap cost is high. |
Thermal drift, probe verification, and the numbers that matter
The biggest error source on a mold machine is not the probe. It is heat. A spindle that has run for three hours can shift the relationship between tool and part by tens of microns. M&H software compensates for thermal influence, tool-machine error, and shaft runout, but the compensation only works if the probe is verified against a known artifact.
Probe verification is a short cycle, usually a few minutes. A master ball or ring gauge is measured, the software compares the result to its known value, and any offset is applied. Skip that step and every subsequent measurement inherits the drift.
What tolerance can you hold? Our shop runs ±0.005 mm (±0.0002 in) on the machine under controlled conditions, with surface finishes from Ra 0.2–0.8 μm on fine work. Probing does not improve those numbers; it protects them by catching the error before it becomes a cut. Think of the probe as a guard rail, not a performance upgrade.
The real gain is time. A test that used to sit in a queue for a week can run in the setup itself. On a five-axis machine, a user can adjust an inclined surface in roughly 30 minutes rather than 4 hours on a three-axis setup, because the probe confirms the position before the finishing pass.
How we set it up on the floor
We run 127 high-precision CNC machines across three plants, including 16 simultaneous five-axis centers and 16 mill-turn centers. On mold and insert work, the probe routine is written into the CAM setup, not bolted on afterwards. The operator calls it up like any other tool change.
The sequence is boring on purpose. Verify the probe. Touch the datum. Check stock-on at three points per cavity. Cut. Re-probe the finished surface. If the last step fails, the part goes to rework before it leaves the machine. That order matters more than the brand of probe.
For parts up to 4,000 mm, the probe has to reach without extensions that introduce their own bending. We keep styli short and stiff, and we accept that some deep features need a CMM instead. Knowing the limit is part of the job.
All parts ship after 100% inspection, with reports on request. The on-machine data feeds that report, but the final sign-off still happens on the CMM. Two checks, two purposes.
Common questions
Does on-machine probing replace the CMM?
No. The probe measures in the machine's thermal and geometric frame, which is fine for process control but not for final certification.
We use on-machine probing to catch errors while the part is still cuttable, then confirm the finished part on the CMM before shipment.
What accuracy can an application mh measurement system hold?
Under controlled conditions, our machines hold ±0.005 mm (±0.0002 in). The probe verifies that the part is inside that band; it does not tighten it.
Probe repeatability is typically around ±1 μm, so the machine and thermal state dominate the result.
How long does probe verification take?
A master-ball or ring-gauge check usually runs in a few minutes. That is short compared to the cost of measuring with a drifted probe.
We verify at the start of a mold job and again after long runs when the spindle has been hot for hours.
Can it be used on five-axis and mill-turn machines?
Yes. The receiver mounts on the machine and the software runs alongside the control. Five-axis work benefits most because inclined surfaces are awkward to check by hand.
Mill-turn centers need a probe routine that accounts for the rotating axis, so we program it per machine.
What materials does it suit?
Probing is material-agnostic: aluminum, stainless, tool steel, copper, brass, titanium and plastics all work. What changes is the stylus and the touch force.
Soft materials like copper and graphite need a light touch to avoid marking the surface.
Do you offer it on all jobs?
No. We apply it where the part geometry or cycle time justifies the programming effort, typically mold inserts, electrodes and thin-wall five-axis parts.
For simple prismatic plates, a CMM check is faster and cheaper.
Send us a mold part and we will quote the probing routine
Upload a drawing or STEP file and we will tell you whether on-machine measurement fits the job, with a quote in 12 hours.
12-hour quote100% inspectionNDA on request
