Xinysun Instrument (Shanghai) Co., Ltd.

Solutions

Breadcrumb: Home > Solutions > Speed Measurement

Bench Tachometry Above 10,000 rpm: Why Reflective Tape Fails and How to Mark the Shaft

Updated:2026-11-09Views:

The part that usually fails on a bench rig is the marker, not the sensor. Reflective tape curls at the edge, flies off, or loses its retroreflective layer once speeds pass ten thousand rpm, and the missing pulses raise no alarm. This article takes the bench rule "above ten thousand rpm, use a paint marker" apart: it uses centrifugal acceleration to derive the crossover speed for the radius you actually measure at, then covers color choice, line width, radial tick marks, surface cleaning and the single-pass stroke, plus how standoff distance and incidence angle have to be narrowed once the tape is replaced.

A pre-run checklist is appended at the end. The centrifugal figures are estimated from patch size, density and measurement radius to judge magnitude; they do not replace measurement. Where a metrology conclusion is involved, present it together with the calibration certificate and the test conditions.

The part that tends to fail on a bench rig is not the sensor. It is the patch of tape.

At low speed it is convenient: peel one off, stick it on, the green light blinks steadily and the reading arrives. Pull the speed past ten thousand rpm and the situation changes. The edge lifts, then the whole patch leaves, or the retroreflective layer is polished dull by grit and dust. None of these failure modes raises an alarm. The oscilloscope simply shows a few missing pulses, a stretch of data quietly disappears, and the operator still believes the speed is steady.

Hence the saying that circulates around test benches: above ten thousand rpm, do not use tape, use a paint marker. Broadly true, but copied blindly it causes trouble. What needs calculating is not the rpm but the centrifugal acceleration.

1. Ten thousand rpm is a rule of thumb, and it has to be calculated for your own rotor

Start with a blunt statement: if the speed runs into five figures and the measurement point sits on the outer part of a blade, do not gamble on tape, switch to a paint marker. If the measurement point falls on the motor output shaft or the root of the hub, where the diameter is a few tens of millimeters, tape can still be tried around ten thousand rpm, and that is about the limit.

Below that, stop deciding by rpm. Ten thousand rpm is not a physical threshold. The centrifugal acceleration on the tape is a = ω²r, proportional to the square of speed and to the radius of the measurement point. Double the speed and the load quadruples; double the radius and the load doubles too. At the same ten thousand rpm, a measurement diameter of 50 mm and one of 300 mm differ by a factor of six.

So a small-diameter rotor can hold on longer, while a large-diameter one should be switched well below ten thousand rpm. Running the numbers with the real measurement radius on your bench is far more useful than memorizing the phrase "ten thousand rpm".

It is worth explaining where the rule of thumb comes from. By the criterion below, ten thousand rpm holds up on the premise that the measurement radius is modest: with the marker on the motor output shaft, the coupling or the hub root, the diameter is typically 50 to 100 mm. The site article drone propeller speed measurement setup also recommends placing the reflective marker on the shaft or at the hub root, with the stated reason of avoiding obstruction by the blades; there is a second layer to it, namely that a small radius keeps the centrifugal load small. Put the marker on the blade tip at a hundred-odd millimeters and the crossover speed has to come down; you cannot still wait for ten thousand rpm.

2. Tape is thrown off, not blown off

Take a common 10×10 mm reflective patch, about 0.2 mm thick, with density estimated at 1.2 g/cm³. The volume is 20 mm³ and the mass about 0.024 g. Light enough to ignore.

What cannot be ignored is the acceleration. Take a blade with a 200 mm measurement diameter, radius 100 mm. At ten thousand rpm the angular velocity is about 1047 rad/s, and substituting into a = ω²r gives about 109600 m/s², roughly 11000 times gravity. Multiplied by that 0.024 g, the centrifugal force on the patch is about 2.6 N, equivalent to pressing 270 g onto one square centimeter, directed outward.

At thirty thousand rpm the angular velocity triples, the acceleration rises ninefold to about 986000 m/s², and the same patch carries roughly 24 N, that is 2.4 kg pressing on that square centimeter. The holding power of an ordinary adhesive layer does not reach that magnitude.

One detail deserves attention: this force is not a uniform shear. The outer edge of the patch is lifted, aerodynamic disturbance tears inward along the lifted edge, and at some speed the whole patch releases. That is why the failure seen in the field is often a sudden loss of signal halfway through a speed ramp. A gradual degradation of signal is less common. Adhesive creep, an oil film on the surface, or insufficient pressure when applying the patch all pull that critical speed down.

Different measurement diameters at the same ten thousand rpm

Measurement diameterAccelerationGravity multipleCrossover speed
502750028001.0-1.3
1005480056000.73-0.95
1508230084000.60-0.77
200109600110000.52-0.67
300164500167000.42-0.55

Measurement diameter is in mm, acceleration in m/s², the gravity multiple in g, and the crossover speed in units of ten thousand rpm. The table estimates the load at ten thousand rpm from a = ω²r with g taken as 9.8 m/s², and the crossover speed corresponds to the criterion of three thousand to five thousand times gravity. Values are rounded to the nearest hundred and are meant for judging magnitude, not for setting a safe speed.

A practical criterion can be stated like this: once the centrifugal acceleration passes three thousand to five thousand times gravity, tape is no longer considered. Converted, a 200 mm measurement diameter should be switched from about six to seven thousand rpm, and 300 mm even earlier. The table also shows that the ten thousand rpm line actually matches the 50 to 100 mm band, the case where the marker sits on the shaft or the hub.

For those who would rather not calculate, there is a rough method: a is proportional to speed squared times diameter, so the allowable speed varies roughly inversely with the square root of diameter. Quadruple the diameter and the allowable speed falls by half. That rough estimate is sufficient for a decision at the bench, though a formal plan still deserves the numbers plugged in.

3. Tape rarely fails all at once

One mode is edge lift. The viscoelastic adhesive creeps under sustained centrifugal load, and after tens of minutes of running the patch shifts slightly outward and the edge releases. The signal is still there, just lower in amplitude with softer edges. This mode is easy to miss, because the reading still looks reasonable.

Another is a complete release. The detached patch may stick to the inner wall of a duct or damage the blades behind it. This has to be treated as an equipment safety issue; more than a few bench incidents begin with that patch of tape.

A third is surface wear. Grit, oil and repeated wiping polish the retroreflective layer dull, and the patch is still visibly there while the returned light is too weak for the receiver to catch.

All three modes come down to the same thing in the data: missing pulses. And missing pulses raise no alarm, they just shorten the record. Anyone doing order analysis and torsional vibration testing has to watch one more layer: creep makes the marker position drift slowly, and the marker position is the phase reference. When it drifts, phase drifts with it, and the error mixes into the torsional vibration result, where it is hard afterward to tell whether the rotor really twisted or the tape moved.

4. Switching to a paint marker: the drawing rules are firm

Make one thing clear: the marker material changes, the marker itself does not disappear. A reflective probe still needs one light-dark transition per revolution, and that has not changed. The site article laser tachometer selection Q&A states that measurement cannot be stable without a reflective marker, meaning the measurement point should not be left bare. The line drawn by an oil-based marker or a paint pen is a marker as well.

The question therefore shifts from "will it fly off" to "is it drawn correctly". The following steps go in order.

1
Clean and degrease

Wipe the measured surface clean with anhydrous ethanol and let it evaporate before drawing. With an oil film present the line looks applied but wipes off. This is the same preparation as before applying reflective tape; do not skip it.

2
Choose the color against the background

Use a white marker on dark carbon fiber and dark paint, a black marker on bright, chrome-plated and light surfaces. Colored pens such as red or blue do not give enough light-dark contrast against the background, so leave them out. Pick the pen so the mark separates from the background in black and white, not so it looks good.

3
Control the width

Keep the width under one tenth of the circumference. At a 200 mm measurement diameter the circumference is about 628 mm, so the upper limit is around 60 mm, while 5 to 10 mm is enough in practice. An over-wide line raises the pulse duty cycle, adjacent pulses merge at high speed, and the counter starts undercounting.

4
Draw a short radial line

Draw a line perpendicular to the direction of rotation, that is, along the radius. The spot crosses it briefly, the rising and falling edges are steep, and the timestamp accuracy is higher. Drawing a long stripe along the circumference is a common mistake; the return light smears into one block and every edge blurs.

5
Keep the position inward

A smaller radius means a smaller centrifugal load, and the ink also survives wiping better. At the same time avoid the deformation zone at the blade root and areas shadowed by the duct; do not move inward so far that the spot cannot reach the mark.

6
Prime mirror surfaces

On chrome plating, polished aluminum and mirror coatings the laser returns specularly and the marker signal is drowned. Apply a patch of matte black locally, let it cure, then draw a white line on the black base.

The pen itself matters too. Use an oil-based marker or a paint pen; a whiteboard marker will not do, since its ink layer can be wiped off and will be carried away by airflow and wiping at speed. A paint pen adheres strongly and resists oil better. Do not trace over the same spot repeatedly: a thick, layered coating has a glossy edge, tends to overexpose the receiver, and may peel off in one piece. Let the ink cure before ramping up.

All of these rules address the same goal: give the return light a clean transition as the spot crosses the mark. Done correctly, ten thousand rpm and above still yields one clean pulse per revolution.

5. A new marker means tighter distance and angle

Changing the marker has a cost. The ink layer reflects considerably less than a dedicated reflective patch, the return light is weaker, the usable standoff shortens and the tolerance on incidence angle narrows.

The 7 m in the data sheet is a nominal upper limit; do not mount at that figure when using a paint marker. The site selection guidance already suggests designing the mounting position within seventy percent of the upper limit, and that margin is adequate with reflective tape; with ink it has to come in further, with the incidence angle set to about 10 to 15 degrees, and then the pulse amplitude on the oscilloscope decides. The allowable tilt is a separate matter: the laser version reaches 60 degrees, but the usable distance falls quickly as the angle grows, so do not push toward that limit with a weak-return marker.

If the waveform amplitude is low but still usable, check three things: whether the lens is oily, whether the mark is really at the center of the spot, and whether the background is too bright. Adjust these three before considering a new mounting position; do not replace the sensor straight away. If the reading jumps by multiples or extra pulses appear, following the sequence in tachometer troubleshooting through the optical path and wiring saves time.

6. A pre-run walk-through

  1. Compute the centrifugal acceleration from a = ω²r and divide by 9.8 to get the multiple of gravity. Use the radius where the mark actually sits, not the outer edge of the rotor.
  2. Above three thousand to five thousand times gravity, switch to a paint marker without further discussion of tape.
  3. Check the surface: matte black primer before anything else on mirror, chrome and polished parts.
  4. Set color and width: white pen on dark surfaces, black pen on bright ones, width within one tenth of the circumference, in practice 5 to 10 mm.
  5. Clean, let dry, draw one short radial line, no over-tracing.
  6. Watch the waveform on the oscilloscope: sufficient amplitude, steep edges, no extra multiple pulses.
  7. Set mounting distance and incidence angle from the measured waveform, not from the standoff limit in the data sheet.
  8. Run one full speed sweep and confirm no pulses are lost across the whole range before starting formal acquisition.

One boundary needs stating: a paint marker merely fixes the marker link. It cannot cure faults in the optical path or the signal chain. If the reading still jumps by multiples after the change, check for secondary reflections and the DAQ channel instead of working on the marker further. The troubleshooting order for such problems is described in detail in the wiring and output guide.

Frequently asked questions

Does reflective tape always work below ten thousand rpm?

Not necessarily. The criterion is centrifugal acceleration, not rpm. A rotor with a 300 mm measurement diameter already reaches three thousand times gravity at four to five thousand rpm; a small shaft with a 50 mm measurement diameter sees just twenty-eight hundred times gravity at ten thousand rpm. Work out a = ω²r before deciding which marker to use. Low-speed work has its own difficulties: an over-wide patch smears the return light into one block, and that is covered in the site service and troubleshooting article.

Is a line drawn with a paint marker a reflective marker? Does that contradict "you cannot measure without tape"?

It does not. A reflective probe needs one light-dark transition per revolution, and tape, a paint pen, adhesive tape or even a small drop of paint can provide it. That answer in the site selection Q&A addresses the case where the measurement point carries no mark at all; it does not say that T5 reflective tape is the sole accepted option. With ink the mark is still there, just returning less light, so the mounting distance and angle have to come in.

White or black marker: how do I choose?

By the brightness of the background. Use a white pen on dark carbon fiber and dark paint, a black pen on bright and light parts. Colored pens such as red or blue cannot open enough contrast on either background and tend to lose pulses. If unsure, test the drawn line with the signal indicator on the back of the probe; a regular blink is what counts.

Does a mirror-finished shaft always need a matte black base?

A mirror surface reflects the laser straight back, the spot is full of highlights, the light-dark transition of the mark is drowned, and the resulting pulses are either absent or erratic. Applying matte black locally and then a white line on it is the simpler approach for this case. Degrease before applying and let it cure afterward.

Does a wider line give a stronger signal?

Somewhat, at the cost of a larger pulse duty cycle. Adjacent pulses merge at high speed, the counter begins to undercount, and the reading actually falls. Cap the width at one tenth of the circumference; 5 to 10 mm is enough in practice. If the signal really is weak, shorten the mounting distance and set the angle to 10 to 15 degrees; that works better than widening the mark.

After switching to a paint marker, is the 7 m range still usable?

7 m is the nominal upper limit in the data sheet; do not take it as a mounting position. Ink returns less light, the site selection guidance designs within seventy percent of the limit, and with ink you come in further still; how far depends on the pulse amplitude on the oscilloscope. The same applies to the 60 degree tilt limit of the laser version: that is an allowable value, not a recommended working angle.

The tape edge has lifted; can I just apply a new patch?

That treats the symptom. A lifted edge shows the centrifugal load has exceeded the holding power of the adhesive, and a new patch merely pushes the critical speed a little further out; it will lift again on the next ramp. Compute the centrifugal acceleration once, and switch to a paint marker above three thousand to five thousand times gravity. If the failure comes from water washing, steam or cutting fluid, that is a different cause and a water-resistant patch is the fix; the approach is described in ROS/ROLS service and troubleshooting.

Phase keeps drifting in order analysis; can the marker be involved?

Possibly. Adhesive creep makes the marker move slowly outward under sustained centrifugal load, and the marker position is the phase reference, so phase moves with it. Such an error mixes into the torsional vibration result and is hard to separate afterward. For order analysis, keep the marker inward, photograph the position before each test, or simply switch to ink; any of these suppresses this term.

Product specifications, wiring definitions and related technical material can be found in the published documentation for the LUYOR ROLS-5PW laser tachometer. The centrifugal figures here are estimated for a 10×10×0.2 mm patch with a density of 1.2 g/cm³, to judge magnitude; adhesive grade, surface roughness and the aerodynamic environment all shift the actual critical speed, so bench work should go by the oscilloscope waveform and the results of a speed sweep. Where a metrology conclusion is involved, present it together with the calibration certificate and the test conditions.

Related products and reading: LUYOR ROLS-5PW laser tachometer, drone propeller speed measurement: ROLS-5PW with synchronous bench acquisition, how to choose a laser tachometer, laser tachometer wiring and outputs explained, ROS-W/ROS-P and ROLS-W/ROLS-P service and troubleshooting, reading a stroboscopic speed measurement correctly, tachometer sensor section. Xinyisheng Instruments (Shanghai) supplies the full LUYOR tachometer range and reflective marker consumables, and can help determine the marking method and mounting position from the measurement radius, speed range and bench environment.

Related sections and models

This article belongs to the speed and strobe measurement section.

Other sections on the same product line include tachometer sensors (continuous monitoring into a PLC or DAQ card, with an electrical output) and industrial test instruments (the entry point for UV inspection, speed measurement, stroboscopic observation and surface inspection).

Ranges, accuracy and accessory lists for each model are given on the product pages.

  • Official WeChat QR code

    Email: info@xinysun.com

  • Tel: 131-2093-6518

  • Address: 1525 Minqiang Road, Bldg 26, Songjiang, Shanghai

Official WeChat QR code Follow

Official WeChat QR code

Contact

131-2093-6518

Top