Thirteen rows, and this time the table does the work
Most listings on this site give a parameter block that says nothing. This one gives thirteen rows with real engineering content: repeatability with a statistical qualifier, three separate trigger forces, overtravel in both degrees and millimetres, an IP rating, and two temperature windows. That is a serious data sheet, and the rest of this article can be written because of it.
What follows is not a complaint about missing data. It is a reading of a good table that still leaves four things unresolved: the model name, the instrument class, the radio, and a lost minus sign.
The thirteen rows as published
| Row as published | Value |
|---|---|
| Principal application | Measuring instrument for milling and EDM machines (insulated probe) |
| Gripper shank diameter | 20 mm |
| Weight | 120 g |
| Recommended styli | Ceramic, lengths 30 mm to 100 mm |
| Switch-on / switch-off | Trigger open, no trigger close |
| Sense directions | ±X, ±Y, +Z |
| Unidirectional repeatability | 2.00 µm 2σ |
| Contact life time | 8 million |
| Stylus trigger force | 0.45 N XY low, 0.85 N XY high, 5.75 N Z |
| Stylus overtravel | XY plane ±12.5°, +Z plane 6.2 mm |
| Environment | IP67 |
| Storage temperature | 25 °C to +70 °C (−13 °F to +158 °F) |
| Operating temperature | +5 °C to +55 °C (+41 °F to +131 °F) |
Two rows deserve to be read closely before anything else. Switch-on method: trigger open, no trigger close is a statement about signal behaviour, and it is the only hint that this device has an electrical output at all. Unidirectional repeatability: 2.00 µm 2σ is the most disciplined number on the page, because it names its own statistical basis.
The body says MMP 25 and the title says EMP25
The dimension drawing shows the engraved marking on the barrel: MMP 25, with a serial number below it. The title says nothing about either, the description calls it EMP25 twice, and the photograph of the barrel is the only authority on the page.
This matters more than it looks. A model number is how a buyer looks up a data sheet, orders a replacement, buys a second stylus, or compares two suppliers. If the number on the body and the number in the listing disagree by two letters, every downstream search starts in the wrong place. It is a five-minute question in an enquiry and a five-hour problem after delivery.
"3D Taster" names a different instrument
The phrase 3d taster appears in the title and twice in the description. In machine shops a 3D taster is a specific thing: a mechanical edge finder with a dial gauge on top, where a stylus deflects and the dial needle moves off zero. You read the needle. It has no battery, no electronics, and no output signal. The Haimer 3D Sensor is the best known example, and copies of it are everywhere.
What is in these photographs is not that. It is a cylindrical stainless body with a 20 mm shank, a colour band around its middle, a ceramic stylus and a ball on the end. There is no dial. The feature text calls it a 3D photoelectric probe, which is correct, and the parts list names an indicator lamp belt and a battery cover, which settles it.
So the title advertises a mechanical dial instrument and the product is an electronic one. A buyer searching for a taster will find this page and may buy the wrong class of tool; a buyer who wants an electronic finder may skip the page because the title sounds mechanical.
What the photographs show instead
The exploded and assembled views together give a complete account of the hardware, which the parameter block does not:
- Body 114 mm long overall, 53 mm of shank and 55 mm of head, which sums to 108 mm; the remaining 6 mm is the tail section carrying the battery cover
- 20 mm shank, 25 mm maximum body diameter
- Cylindrical ceramic stylus, 4 mm rod, on an M4 threaded interface
- Stylus 50 mm overall with 35 mm effective, so 15 mm sits inside the mount
- 6 mm ball, one and a half times the rod diameter
- A colour band that the parts list identifies as the indicator lamp belt
- A battery cover at the tail, and four M3 set screws around the head for adjustment
- Total mass 120 g, from the table
No cable leaves the body in any photograph. That single observation is what makes the next section worth writing.
Three accuracy numbers on three layers
The page contains three different figures for how precisely the ball finds a surface, and they are not in conflict. They describe three different things.
| Where the number appears | As printed | In micrometres | What it describes |
|---|---|---|---|
| Feature text | 0.001 mm | 1.0 | Best case, described as reachable |
| Parameter table | 2.00 µm 2σ | 2.0 | Unidirectional repeatability, statistical |
| Adjustment drawing | 0.01 mm | 10.0 | TIR after four M3 set screws |
Read as a chain, the numbers say: the instrument can resolve about a micron in the best case, it repeats to two microns as a statistical statement, and once it is bolted into a spindle and trued with four set screws the ball runs out by up to ten microns.
That ordering is not a defect. It is how touch probes work. What it means for a buyer is that quoting 0.001 mm as the accuracy of the measurement is quoting the smallest number on the page, and the smallest number is the one that matters least.
0.001 mm against 2.00 µm: a best case against two sigma
The feature text says the repetition accuracy can reach 0.001MM. The table says 2.00 µm 2σ. One micron against two microns is a factor of two, and the difference is not sloppiness; it is a different kind of claim.
Two sigma is a distribution statement. It means the stated band contains about 95.45% of readings, not all of them.
| Band | Share of readings inside | Reading |
|---|---|---|
| 1σ | 68.27% | About two readings in three |
| 2σ | 95.45% | About nineteen in twenty |
| 3σ | 99.73% | About three hundred and seventy in three hundred and seventy-one |
So of two hundred touches, roughly nine will fall outside two microns. That is the honest way to describe a repeatability figure, and this table does it, which is rare and worth crediting. The feature text's can reach is a best case with no distribution attached, and a buyer should treat the two-sigma row as the specification and the feature line as marketing.
Compare this with the GP-800 page already covered here, which printed 0.001 mm as both accuracy and repeatability and then contradicted itself in a drawing. This listing is better behaved even though its headline number is twice as large.
Runout at setup is five times the repeatability
The adjustment drawing is specific: rotate the spindle, test the ball for runout, and adjust four M3 set screws until runout is 0.01 mm. Then lock the screws.
Ten microns of runout against two microns of repeatability is a ratio of five. The instrument will faithfully repeat a position that is itself offset by up to ten microns from the spindle axis. Every measurement inherits that offset, and because it is a constant geometric error it does not average out over repeated touches.
Two consequences follow. First, the truing step is not optional maintenance; it is the difference between a ten micron instrument and a two micron one. Second, it has to be redone after a crash, after a tool change that seats differently, and any time the reading stops agreeing with a known feature. Four M3 screws and a dial indicator are the entry ticket.
Trigger force: 5.75 N against 0.45 N
Three forces are published, and the ratio between them is the most useful number on the page.
| Direction | Trigger force | Moment at the ball, 35 mm lever |
|---|---|---|
| XY, low setting | 0.45 N | 0.0158 N·m |
| XY, high setting | 0.85 N | 0.0297 N·m |
| Z | 5.75 N | 0.2013 N·m |
The Z force is 12.8 times the XY low force. On a 35 mm stylus that is a bending moment of 0.20 N·m at the ball, against 0.016 N·m sideways. Five and three-quarter newtons is about the weight of a small apple, and it is applied through a 6 mm ceramic ball. On soft aluminium that will leave a witness mark; on thin sheet it will deflect the work before it triggers.
The two XY settings are the useful part of this table. Low force is for soft and thin material, high force for rough castings and interrupted surfaces where a light touch would trigger on a burr. How the two are selected is not shown, which is the kind of gap that turns a feature into a phone call.
Overtravel in degrees, and what it means at the ball
The table gives XY overtravel as ±12.5° and Z as 6.2 mm. Degrees are the correct unit for a pivoting stylus, but they are not the number an operator feels. At 35 mm effective length, 12.5 degrees is 7.58 mm of ball travel each side, 15.15 mm total.
Compare the Z figure: 6.2 mm. So the stylus survives more than twice as much sideways abuse as vertical abuse, which makes sense for a device that pivots. Against the GP-800, this is the opposite trade: that probe allowed ±20° sideways but only ±3 mm vertically. Neither is better; they are different compromises, and the one you want depends on whether your operators crash sideways or plunge.
It is wireless, and the table never says so
The parts list names a battery cover and an indicator lamp belt. The photographs show no cable. The device is therefore battery powered, and it communicates with something or with nobody.
Now look at what the table does not contain: no frequency, no protocol, no range, no battery type, no battery life, no receiver. The GP-800 page had all of these except battery life, plus a wiring diagram naming specific input addresses. This page has none of it.
For a buyer this is the largest single gap in an otherwise good data sheet. A wireless probe that cannot be connected is a very expensive indicator light, and whether it can be connected is not answered here.
No receiver and no wiring diagram: what "trigger" means here
The usage drawing resolves the ambiguity, and not in the direction the parameter table implies. It shows the probe above a workpiece with an arrow down, and the instruction is to approach gradually until the red light of the edge finder is stable.
That is a manual procedure. The operator watches a lamp and reads a position off the machine's own DRO. No receiver appears in any of the eight images, and no output connector is documented. The trigger open row and the sense-direction row describe an electrical behaviour that the page never shows you how to use.
The title's other phrase, manual touch probe, is therefore the accurate one, and it sits in the same title as the inaccurate one. For a shop that wants to press a button and have the control record the position, this is not that device. For a shop that wants a lamp instead of a dial, it is exactly that device, and a good one.
The related CF-38 trigger edge finder is the other end of that spectrum and worth reading alongside this page.
Eight million contacts
The table gives a contact life of 8 million. That number is worth converting, because as printed it is unreadable.
- At six touches per part: 1,333,333 parts
- At 200 touches a day: 40,000 days, or about 110 years
- At 500 touches a day: 16,000 days, or about 44 years
Nothing in a shop reaches eight million contacts. The number should be read as a statement about design intent, not a maintenance interval. The parts that will actually wear out are the stylus, which is ceramic and brittle, and the battery, which has no published life at all.
The storage temperature lost its minus sign
The row reads 25 °C to +70 °C (−13 °F to +158 °F). The Fahrenheit half is correct and the Celsius half is not: minus 13 Fahrenheit is minus 25 Celsius.
As printed the span is 45 kelvin. As intended it is 95 kelvin. The published figure understates the storage window by 50 K, which is the difference between a probe that can sit in an unheated store through a northern winter and one that cannot.
The operating window, +5 to +55 °C, converts correctly at both ends. So this is a single dropped character rather than a misunderstanding, and it is the kind of thing that only shows up when someone checks both unit systems against each other.
IP67 and the temperature window
IP67 is a genuine two-part claim: six means dust tight, seven means protected against temporary immersion in water under defined conditions. It is the right rating for a probe that will meet flood coolant, and it is stated once, cleanly, without contradiction.
What IP67 does not cover is the joint between the body and the battery cover, which is the one place a seal has to be opened by the user. It also says nothing about resistance to coolant chemistry, only to water. Both are worth confirming for a machine running neat cutting oil rather than soluble coolant.
Against the GP-800 already covered
The two probes sit on the same site twice now, and the comparison is more useful than either page alone, because they trade in opposite directions.
| GP-800, already covered | MMP25, this page | |
|---|---|---|
| Shank | 8 mm | 20 mm |
| Body diameter | 42 mm | 25 mm |
| Overall length | 114 mm | 114 mm |
| Repeatability | 0.001 mm claimed | 2.00 µm 2σ |
| Runout at setup | under 0.005 mm | 0.01 mm |
| IP rating | IP56 and IP67 both printed | IP67 |
| Radio | 2.4 GHz stated, receiver shown | Battery cover, no receiver, no radio stated |
| Wiring | Full I/O diagram | None |
| Stylus | 3 mm ball, 30 mm | Ceramic 4 mm rod, 6 mm ball, 35 mm effective |
| Overtravel | XY ±20°, Z ±3 mm | XY ±12.5°, Z 6.2 mm |
| Mass | Not stated | 120 g |
Both bodies are 114 mm long. That is unlikely to be a coincidence; it reads like a common envelope that toolholder and ATC clearances are designed around. The 20 mm shank against 8 mm is the more important difference: this probe wants a proper collet chuck, and it will not go into a small machine that the GP-800 fits.
Eleven questions for the quotation
| Question | Why it decides the purchase |
|---|---|
| Is the model EMP25 or MMP25? | The body is engraved MMP 25; the title and description say EMP25 |
| Does it output a signal, and on what interface? | The table says "trigger open" but gives no connector, voltage or protocol |
| Is a receiver included, and on what frequency? | There is a battery cover, no receiver in any photograph, and no radio specification |
| Battery type, and how many hours of use | Nothing published; this is the one consumable in the instrument |
| How is the XY low and high force selected? | Two figures are given; the method is not shown |
| Which stylus length ships in the box | 30 mm to 100 mm is a range, and the drawing shows 50 mm overall |
| Ball material and grade | The drawing says ceramic D6; ruby is the usual choice and reads differently |
| Can the stylus be bought separately, at what length | It is a wear part on a threaded M4 interface |
| Is a calibration sphere supplied | A bracket with a sphere appears in one photograph and is never described |
| Storage temperature, confirmed as minus 25 | The row prints 25 C but the Fahrenheit conversion says minus 25 |
| Repeatability test method and sample size | 2.00 µm at 2 sigma is meaningful only with the method stated |
Frequently asked questions
Is this an EMP25 or an MMP25?
The barrel in the dimension drawing is engraved MMP 25. The title and description say EMP25. Ask which one the delivered unit carries, because the model number is how you order a stylus or look up a data sheet later.
Is it a 3D taster?
No. A 3D taster is a mechanical edge finder with a dial gauge; you read a needle and it has no battery. This is a photoelectric probe with an indicator lamp belt and a battery cover. The title's use of taster describes a different instrument class.
What does 2.00 um at 2 sigma mean?
It means the stated two-micron band contains about 95.45% of readings, not all of them. In two hundred touches roughly nine fall outside. That is a proper statistical statement, and it is more disciplined than the 0.001 mm in the feature text, which is a best case with no distribution attached.
Why is the runout 0.01 mm when repeatability is 0.002 mm?
Because they measure different things. Repeatability is how consistently the internal mechanism triggers; runout is how far the ball sits off the spindle axis after you true it with four M3 screws. Ten microns of offset is inherited by every reading and does not average out.
Does it connect to the CNC control?
The page does not show how. There is no receiver in any photograph, no connector, no voltage, no protocol and no wiring diagram. The usage drawing tells the operator to approach until the red lamp is stable, which is a manual procedure. Treat it as a lamp-type finder until the supplier documents an output.
What is the 5.75 N Z trigger force like in practice?
It is 12.8 times the XY low force and produces a 0.20 N.m moment at the ball on a 35 mm stylus. On soft aluminium it will mark the surface, and on thin sheet it may deflect the work before triggering. The low and high XY settings exist for exactly this reason; how to select between them is not shown.
What wears out first?
Not the mechanism, which is rated for 8 million contacts, or roughly a century at 200 touches a day. The ceramic stylus is brittle, and the battery has no published life at all. Ask for both as spares, and ask which stylus length ships in the box, because the recommended range is 30 mm to 100 mm.
Key specifications
Frequently asked questions
- Is this an EMP25 or an MMP25?
- The barrel in the dimension drawing is engraved MMP 25. The title and description say EMP25. Ask which one the delivered unit carries, because the model number is how you order a stylus or look up a data sheet later.
- Is it a 3D taster?
- No. A 3D taster is a mechanical edge finder with a dial gauge; you read a needle and it has no battery. This is a photoelectric probe with an indicator lamp belt and a battery cover. The title's use of taster describes a different instrument class.
- What does 2.00 um at 2 sigma mean?
- It means the stated two-micron band contains about 95.45% of readings, not all of them. In two hundred touches roughly nine fall outside. That is a proper statistical statement, and it is more disciplined than the 0.001 mm in the feature text, which is a best case with no distribution attached.
- Why is the runout 0.01 mm when repeatability is 0.002 mm?
- Because they measure different things. Repeatability is how consistently the internal mechanism triggers; runout is how far the ball sits off the spindle axis after you true it with four M3 screws. Ten microns of offset is inherited by every reading and does not average out.
- Does it connect to the CNC control?
- The page does not show how. There is no receiver in any photograph, no connector, no voltage, no protocol and no wiring diagram. The usage drawing tells the operator to approach until the red lamp is stable, which is a manual procedure. Treat it as a lamp-type finder until the supplier documents an output.
- What is the 5.75 N Z trigger force like in practice?
- It is 12.8 times the XY low force and produces a 0.20 N.m moment at the ball on a 35 mm stylus. On soft aluminium it will mark the surface, and on thin sheet it may deflect the work before triggering. The low and high XY settings exist for exactly this reason; how to select between them is not shown.
- What wears out first?
- Not the mechanism, which is rated for 8 million contacts, or roughly a century at 200 touches a day. The ceramic stylus is brittle, and the battery has no published life at all. Ask for both as spares, and ask which stylus length ships in the box, because the recommended range is 30 mm to 100 mm.