See the difference between Cat I and Cat IV.
The four TCAT categories are not the same test with more sliders. Below is one machine, one fault -- switch between the four categories and watch what each one is actually asked to do with it. Every plot is computed live from that one signal, not a picture.
One machine, one fault, one measurement point.
Deliberately ordinary: a motor–pump train reading well into the alarm zone. What changes below is not the machine — it is what each category is licensed and equipped to conclude from the same record. Watch the pedestal: it moves horizontally and barely at all vertically, and that is the whole case.
The measured point is the pump inboard housing, horizontal — one point, one session. The other housings were not measured, and nothing below pretends otherwise.
This page’s simulator needs JavaScript. The case in one line: a motor–pump train at 1480 rpm reads 11.8 mm/s RMS — ISO 20816-3 Zone D — with 2× shaft dominant. Cat II calls it misalignment and is wrong: the coupling is in tolerance, and the pump pedestal has a structural mode at 49.8 Hz sitting on top of 2× at 49.3 Hz, amplifying an ordinary force twelvefold. The fix is the foundation, not the machine.
Same signal. Four different questions.
Your instrument rack at this level. Greyed controls are not missing from the page — they are not carried by the category.
Cat I never names a fault. The only questions are: can this reading be trusted, and if so, what is its alarm status? Both are answered here, in full. Offering no cause is the job done properly — not a shortcoming being apologised for.
“Zone D. Raise it. I am not saying why.”
Correct, and complete for the level. Everything below is what happens after this alarm reaches someone with a spectrum.
I-4 bad-data recognitionI-5 alarm/zone readingI-6 observation, not diagnosis
Settings are already chosen (that is Cat III's job). Read the evidence, run the discriminating test, name the fault. This analyst does everything their level asks, in the right order — and gets it wrong.
First, eliminate the bearing. A dominant line low in the spectrum is not evidence against a bearing fault, so it has to be checked rather than assumed.
II-1 settings already chosen for youII-3 systematic spectrum readingII-6 Fault Matrix discriminating testII-7 envelope demodulation
Three measurements Cat II's level does not carry. The first two cost minutes and already end the misalignment call; the third explains the number.
The bump test is the fixed-speed route to a natural frequency, which is why this page relies on it — note that coast-down is greyed out in the rack above, at every level. Want to drive a bump test and an ODS yourself? Open the Live Lab.
III-1 design the acquisitionIII-4 resonance vs forcing
Cat III found the resonance. Cat IV has to prescribe a change to the machine — and get it right, because the obvious partial fix makes this one worse. Drag the slider and find out where.
IV-2 prescribe a structural modificationIV-4 prove a measurement trustworthy, or show why it is not
Every frequency on this page is derived, not drawn.
This whole case rests on six numbers: three forcing levels, the pedestal's three modal frequencies, and one damping ratio. Every millimetre per second on this page — each spectral line, the overall, the H/V ratio, the axial fraction, the bump-test peak and every point on the Cat IV stiffness curve — is those six numbers through one standard transfer function. Nothing is typed into a caption.
That constraint is what makes the case checkable rather than merely plausible, and it leaves a fingerprint you can test yourself. One damping ratio and one natural frequency produce a twelvefold amplification at 2×, a mild 1.3× lift at 1×, and an actual reduction at 3× — because 3× sits above the resonance. A case tuned to flatter its own story would have amplified all three. Check the 1× and 3× lines against the same numbers that gave you the 2×; they hold.
The build refuses to ship unless that arithmetic passes, unless the synthesized signal measures back to the same values through a real FFT, and unless the bearing rule-out Cat II performs can be shown to fire on a genuine defect — an elimination that could never detect anything would prove nothing about the analyst who ran it.
TCAT is a TIERA-issued credential aligned with the ISO 18436-2 body of knowledge. It is not an accredited ISO certification, TIERA is not an ISO certification body, and TCAT training hours do not count towards the formal training an accredited ISO 18436 certification requires.
If what you just watched felt unfamiliar rather than obvious, you are not TCAT's audience yet — and that is a fine place to be. TCAT assesses people who already work with vibration data; it does not teach the subject from zero. The honest starting point is the free TIERA 101 primers: self-paced, beginner-first, no cost, and they issue a completion badge when you finish. Go there first if that is where you actually are — nothing about TCAT is a reason to skip them, and nothing you do there is required before you come back.
Want to know which category you should sit?
Send us a line and we will help you find the right rung.