For our API 610 pump trains, I run a 120-point monthly route with an SKF Microlog and am evaluating battery wireless nodes for continuous coverage. If you’ve used both, what’s the actual ROI in reduced unplanned work or earlier bearing calls, and did you encounter data quality gaps around 1–2 kHz on 1,800 rpm machines?
We moved our 120‑point Microlog route to a hybrid: wireless on the API 610 trains for continuous and the route for phase/acceptance; net ROI was one avoided bearing/seal event in year one, payback about 9 months. The “1–2 kHz on 1,800 rpm” gap drove me nuts until we forced the nodes to capture 1–2 s timewaves at ≥12.8 kHz with a 500–5,000 Hz envelope band (and stud mounts over magnets), otherwise BPFO got smeared, . Which nodes are you trialing and can you set Fmax/lines and true sample rate manually?
Quick example: we caught a thrust bearing on an API 610 at 1,800 rpm two weeks early and rolled it into a planned pit stop — payback in about 9 months by forcing true 10 kHz sample, 6400 lines, and stud-mounts so 1–2 kHz didn’t smear. > set Fmax/lines manually? Yes — do that and also gate captures to on-curve runs (flow steady, >70%) so the envelope isn’t washed out; which nodes are you piloting?
On our OH2/BB3 trains, the wireless paid for itself inside a year by catching a cage defect early and rolling it into a planned stop. @Sam, can your platform deliver time‑synced waveforms (not just spectra)? The mid‑band gap we saw on 30 Hz machines was from low Fmax and decimation — spec nodes that store raw waveforms and allow true envelope in a 6–10 kHz band with stud mounts and anti‑aliasing; SKF’s enveloping primer is a decent guide: SKF.
Disabling node auto-decimation fixed the 1–2 kHz dip; payback about 8 months… @Sam, can you lock Fmax/envelope bands per asset?