MotorTach reads an AC induction motor's shaft speed and load from the sound it makes, using your iPhone held against the machine.
Questions, problems or field reports: info@harmonicsentry.com
The app's own Help screen, the ? at the top right of the readout, covers all of this in more detail, and is the authority if the two ever disagree.
Tap the gear to open Motors, then tap a motor's chevron to open its page and enter the two nameplate values: supply frequency (50 or 60 Hz) and rated RPM. You do this once per motor. Everything else (pole count, synchronous speed, slip) is worked out from those two and from the sound.
You can also photograph the nameplate there, and the photo then appears on every recording of that machine. Tap any plate to open it full screen and zoom in.
In a hurry? Quick reading at the top of the list takes values without naming a motor, so a first reading never waits on paperwork. Tap a motor's row to switch to it.
Hold the bottom corner of the iPhone firmly against the center of the motor frame. The microphone is in that corner, and it is acting as a contact microphone.
Keep still. A reading appears a few seconds after contact settles, and needs about five seconds of steady sound for full precision.
That is MotorTach working as intended. It abstains rather than guessing, a number appears only when the signal supports it. The usual causes, in order of likelihood:
No. Held near a running motor it will usually read anyway. Contact is the more reliable method and is what the app is tuned for, but if a machine is not safe to touch, hold the phone near it instead.
One limitation worth knowing: at a distance MotorTach cannot tell which machine it is listening to. In a room with several motors running, make contact, so that the machine you are touching is the one you are measuring.
Because a number is shown only where MotorTach can stand behind it. Every reading carries an error band, and when that band is too wide the app shows a class (LIGHT, MEDIUM or HEAVY) instead of a false precision.
Two things widen it. One is the nameplate: a rated RPM carries a tolerance, and that uncertainty grows with load, so a mid-load reading is often a class even after calibration. The other is the missing zero: without a no-load zero the app is measuring load relative to no slip at all, which reads high at light load and settles as load rises.
So on most motors a number appears at light load once you have calibrated, and at heavy load, somewhere between 70% and 76% depending on the pole count, even if you have not. On a slow motor (six poles or more) the band can be too wide for a number at any load; that is the measurement being honest about a small slip range, not a fault.
Set one on the motor's own page, using Set no-load zero. Capture it warm: a motor's slip drifts noticeably as it heats up, and a zero captured cold makes every later reading report MORE load than the motor is actually carrying.
Relative load and the load trend are reliable whether or not you have calibrated.
Slip rises as a motor warms, and load is derived from slip, so the same machine reads higher warm than cold. This is real and is a property of induction motors, not an instrument fault. One measured example: a pool pump read 62% cold and 92% after seven hours running. Compare readings taken at similar temperatures.
VFD support is beta and has not yet been confirmed against a real drive in the field.
Enter both the VFD panel frequency and the pole count. A panel value alone cannot resolve the speed. For load on a drive, also enter the rated RPM and set a no-load zero.
Two known difficulties: on a lightly loaded motor the drive's own switching carrier can land on top of the shaft signal and prevent a lock, and an incorrect pole count shows up as a speed above synchronous. If you see a speed above synchronous, check the pole count first. Treat any VFD reading as provisional and check it against the drive's own display.
It measures how impulsive the 2–10 kHz band is, whether anything up there is hitting rather than hissing. Around 3.0 is steady noise, which is what a healthy bearing gives. Green SMOOTH is under 3.8, amber TICKING to 6, red IMPULSIVE above that. Grey QUIET BAND means there was too little signal to judge, usually weak contact, and is not a pass.
Use it to notice a machine getting worse over time, not to certify one as healthy. The number rises as a defect develops, but once a race is worn rough all over the impacts blur back into a hiss and it falls again, so a late-stage bearing can read green. Keep a recording of each machine while it is known good, and compare against that.
Yes. Swipe a saved card in Recordings and choose Share. You get a single .zip containing the audio, a plain-text summary of the reading and the motor settings it was taken under, and the minute of diagnostics log leading up to it, then you pick where it goes: Mail, Messages, AirDrop, Files, anywhere.
That package is also the most useful thing you can send us about a reading that looks wrong, because the log says what the app was seeing when it decided.
PDF report export is a separate, later feature.
Line-fed AC induction motors, 50 or 60 Hz. This is the validated path.
VFD-driven motors, beta. See above.
It cannot read synchronous, brushless DC, DC, universal or stepper motors. It also cannot reach the high slip of small residential fan motors. The condenser fan in a home air conditioner typically runs 8–15% slip, outside the range MotorTach searches.
Full precision needs about five seconds of steady sound.
RPM on a line-fed motor with the mains tone detected is referenced to the power grid and repeats to within about a rpm. On a VFD, or when the app falls back to a nominal supply frequency, readings are just as steady but the absolute value can sit a rpm or two off.
Load depends on calibration, as described above. Nameplate ratings carry tolerance and slip drifts with temperature, so treat absolute percentage load as good to a class unless you have set a warm no-load zero.
Mind couplings, belts, fans and hot frames, and keep clear of live parts. If a machine is not safe to touch, hold the phone near it instead.
MotorTach is an indication, not a protection device. Do not use it as the basis for a safety decision.
If a motor will not read and you would like it looked at, the app can send a small numbers-only log (no audio, no personal information) from Help → Share diagnostics log. It records what the algorithm saw, which is usually enough to say whether the problem was coupling, the nameplate values, or the motor being out of scope.
MotorTach is published by Harmonic Sentry LLC.
