AXILINEDecanter Centrifuge Balancing: What Can Really Be Done On Site
The decanter is shaking, the vibration trip fires toward the end of the cycle, and the manufacturer's service quotes lead times in weeks. We arrive with a Balanset-1A vibration analyzer, use the vibration spectrum to separate out the contribution of the bowl, the scroll, the gearbox, and the cake, and give you a straight answer: is this imbalance that's fixed with weights in the standard planes, or wear that means a trip to the service shop. We're based in Vila Nova de Gaia near Porto and travel throughout Portugal.
Two coaxial rotors: why a decanter isn't like other centrifuges
Inside a decanter, two masses turn on a single axis. The bowl runs at operating speed, typically 2000 to 4000 rpm. The scroll turns at the same speed minus a differential of a few to a few dozen rpm, set by a planetary or cycloidal gearbox at the end. The bowl is supported on two main bearing housings, and the scroll additionally rides on its own bearings inside the bowl's trunnions.
For vibration, this means two closely spaced running frequencies. The bowl produces its own 1x component, the scroll its own, a fraction of a hertz lower. On the casing they combine and produce beats: the level slowly "breathes" with a period of anywhere from seconds to minutes. An ordinary vibration meter shows a number that keeps jumping around, from which you can't tell which rotor is at fault. We separate them with a high-resolution spectrum — the breakdown of vibration by frequency — and by phase, meaning the angle by which the vibration is tied to the marker on the bowl.
The second distinctive feature is geometry. The rotor is elongated, with a length-to-diameter ratio reaching three to four, and some machines run near or above the first critical frequency — the speed at which the rotor goes into resonance. Formally, this is the flexible-rotor zone under the applicable part of ISO 21940, and end weights don't always compensate for a bow in the middle. We have a separate article on critical speed; the conclusion that matters here is that on a decanter we always look at how amplitude and phase behave on run-up and coast-down before we promise a result.
Sources: ISO 21940-12:2016
Where imbalance in a decanter comes from
This machine has its own sources of imbalance, and weights don't fix half of them.
Cake on the bowl wall
Cake — dewatered solids — builds up unevenly, especially in the dry conical section, where the layer isn't held in place by the liquid ring. This is a variable imbalance: it grows toward the end of the cycle, partly clears after washing, and settles differently with every run. Weights can't compensate for it: after the next wash, the machine will shake harder.
Wear on the scroll flighting
Abrasive material wears down the flighting edges unevenly around the circumference. Hard-facing chips off, and carbide tiles tear loose and leave with the cake. Losing even one tile at the flighting's radius already produces a noticeable force at operating speed. The tell: the level jumped and stays elevated even on a clean machine.
The discharge zone and bushings
The cake discharge ports and protective bushings work as if under sandblasting, and the erosion here is asymmetrical. Add to that the liquid-discharge weirs at the opposite end, if they were reset without swapping the whole set together.
Repair and assembly
After the scroll is re-hard-faced, the mass distribution differs from the factory balance. A bowl reassembled without matching the alignment marks, or standard weights that got moved, produce the same effect. If the decanter started shaking right after a repair, this is the first place we look.
Bearings and the gearbox
The scroll's internal bearings and the main bearing housings produce peaks at frequencies that aren't multiples of the running speed. The gearbox adds gear-mesh components and sidebands. This isn't imbalance, and weights are useless here, but on the shop floor it sounds the same to the ear.
Mounts and piping
A decanter usually sits on vibration isolators. A sagged mount, a cracked frame, or a feed pipe connected rigidly all change the vibration picture without a single gram of imbalance involved. It's checked by hand and with a ten-minute measurement, which is why it's the first item on the inspection.
Sources: ISO 281:2007
Wash first, conclusions after
A measurement on a dirty machine tells you almost nothing: what you see is the sum of the rotor's imbalance and a random buildup pattern. So the order is strict. Measure as-is, then wash the bowl per your CIP procedure or by hand, then take a second measurement on the clean machine at the same speed, ideally on water or empty.
The difference between the two readings is itself a diagnosis. If about eighty percent of the level disappears, the problem is in the process, and the right answer is a washing schedule and a vibration threshold, not weights. If the level stays, we work on the mechanics. How to tell imbalance apart from shaft misalignment and bearing defects by spectrum is covered in our diagnostics articles. Here only the decanter-specific details get added: beats, the bowl's 1x separated from the scroll's 1x, and behavior on coast-down.
- Wash the bowl before we arrive and note the time of the wash.
- Make it possible to hold a stable operating speed for a few minutes, without run-up/braking cycles.
- Export the trends from your standard vibration monitoring system for the last few months, if you have one.
- Find the page in the manufacturer's manual on balancing and permitted intervention on the rotor.
- Clear access to the main bearing housings and the bowl's end hubs.
- Recall what changed before the vibration rose: feedstock, flocculant, a repair, the discharge weirs.
Sources: ISO 13373-3:2015
Four possible outcomes of a decanter visit
Based on the measurement taken on a clean machine, the scenario comes down to one of four, and we say which one out loud before touching the rotor.
| What the measurement showed | Conclusion | What we do |
|---|---|---|
| The bowl's 1x dominates, phase is stable from run to run, and the manufacturer permits masses in the standard positions | Bowl imbalance | We balance on site in one or two planes on the end hubs, with a before-and-after report |
| The level dropped after washing and comes back toward the end of the cycle | Uneven cake buildup | We don't fit weights. We help set a washing schedule and a vibration trip threshold |
| 1x persists on the clean rotor, the level jump coincided with a repair or lost hard-facing, and beats point to the scroll frequency | Scroll imbalance or wear | Can't be fixed on site. An assessment with numbers for scroll inspection and balancing at a service shop |
| Peaks at bearing frequencies, gearbox gear-mesh components, a frame or vibration-isolator resonance | Not imbalance | Diagnostics, a repair recommendation, and a check measurement afterward |
You get a measurement that separates out the causes and an assessment regardless of the outcome: even a decision not to balance rests on numbers, not opinion.
How balancing the bowl proceeds, when it's possible
- 01
Preparation on the stopped machine
We stop and lock out the drive. We inspect the end hubs and the standard balancing positions: bolt holes, rings, threaded sockets. We stick the reflective phase marker on the bowl itself, not on a pulley or the gearbox: the phase we need is the bowl's.
- 02
Sensors on the main bearing housings
We place two accelerometers on the main bearing housings, radially, usually horizontally: on vibration isolators the machine is most compliant in this direction. We aim the laser phase sensor at the marker, and add a check point at the gearbox end if needed.
- 03
Baseline measurement and separating the rotors
We record overall vibration, 1x with phase, speed, FFT spectrum, and time waveform for each bearing. We choose the spectrum resolution to resolve the bowl's 1x from the scroll's 1x, which differ by the amount of the differential. We look at the coast-down — the free run-down after the drive is switched off: wherever amplitude and phase change quickly, there's a resonance.
- 04
Trial runs and influence coefficients
We fit a weighed trial weight at a standard position in the first correction plane and run the machine, then repeat for the second. A valid change is 20-30 percent in 1x amplitude or 20-30 degrees in phase. This is how the instrument obtains the influence coefficients — exactly how your system of rotor, bearings, and frame responds to an added weight.
- 05
Correction masses
The software outputs a mass and an angle, or, in fixed-position mode, a direct breakdown across the standard holes. Fastening is bolted only, using the standard hardware, with torque checked and the thread locked. Welding on a stainless bowl is ruled out, and so is drilling without the manufacturer's written approval.
- 06
Verification, a run with product, and the report
A check run at the same points under the same conditions, with one trim correction if needed. We calculate the tolerance against grade G — the balance quality grade — under the applicable part of ISO 21940, taking the target grade from the manufacturer's manual. We then check the level on a run with product and put together the report. We save the influence coefficients: the next trim balance (a quick touch-up) will take a single run.
Sources: ISO 21940-11:2016 · Balanset-1A operation manual
When on-site work won't work: the scroll, manufacturer restrictions, a flexible rotor
The scroll is the main honest limit. Its correction planes sit inside the bowl, and the only way to reach them is a full teardown of the rotor. Once it's apart, balancing has to be done on a machine: first the scroll on its own, then the assembled rotor, with fits and runouts checked. That's service-shop work, and we say so up front, not after four trial runs.
The second limit is documentation. A number of manufacturers only permit intervention on a decanter rotor at their own service centers, with subsequent certification. We don't work around requirements like that: we work strictly within the standard positions, or we stop at diagnostics. The third limit is a flexible rotor, one whose bow shape changes across the operating range: balancing at one speed doesn't guarantee compliance at another, and the right place for a machine like that is a run-up balancing rig.
- What you get instead of balancing: levels and 1x with phase for each bearing, spectra, coast-down, and a breakdown of the contributions from the bowl, the scroll, the gearbox, and the cake.
- An assessment of the probable cause and a list of what to check during teardown: the flighting, the hard-facing, the tiles, the trunnion fits, and the scroll's internal bearings.
- A report that keeps the conversation with the service shop about numbers, not impressions. It also records the condition before the repair.
- A check measurement after the repair and, if the manufacturer permits it, a trim balance of the bowl in the standard planes.
Environment: food-grade and chemical versions, product, vibration isolators
Food-grade decanters and wastewater-treatment machines work with a product nobody wants to see on the instrument: the machine is washed before we start work, and our sensors and fasteners are washable. In sanitary zones, fitting weights must not create gaps or stagnant pockets, so only standard positions and standard fasteners are used. Chemical-duty versions add a corrosive environment, and often a hazardous-area rating: we agree the order of work with your safety team in advance. And everywhere, the rule for stainless bowls applies: no welding and no improvised holes.
A separate note on vibration isolators. A machine on compliant mounts is entitled to move noticeably at the casing, and that's not always a failure. We assess it against the applicable part of ISO 20816, with a note on the type of mounts and measurement points, but the most informative thing is the trend: a comparison with the level at which this same machine ran quietly before. That's why we ask for the standard vibration-monitoring trends first.
Sources: ISO 20816-1:2016
Price, timing, and what to send for an estimate
Vibration diagnostics with a report costs 300 EUR per unit, balancing adds from 250 EUR, and the minimum invoice per visit is 500 EUR; the calculator on our website works out the exact amount. For decanters we always ask for data in advance: it determines the very format of the visit — balancing the bowl in the standard planes, or a measurement with an assessment.
Send us the model and a photo of the nameplate, the bowl speed, the product type, photos of the end hubs and main bearing housings, the manual's page on balancing, and a brief history: what changed before the vibration rose. We'll reply the same working day with what can realistically be done on site.
The engineers who come out are the same ones who design and manufacture the Balanset instruments and do the on-site balancing themselves. We work with the Balanset-1A: two accelerometers, a laser phase sensor, a two-channel USB module, and software on a laptop. For a fleet of decanters, the same instrument can be supplied to your own team: a trim balance using saved coefficients takes a single run, and for a repair bench there's a Balanset-1A OEM version without the case.
We're based in Vila Nova de Gaia near Porto. We travel throughout Portugal, and in the north usually within one to two working days.
Sources: Balanset-1A manufacturer specification
Frequently asked questions
Can a decanter's scroll be balanced on site?
No. The scroll turns inside the bowl, its correction planes aren't accessible without a full teardown of the rotor, and once it's apart, balancing has to be done on a machine. What we can do on site is different: use the spectrum and the beats to show that the source really is the scroll, and hand over the numbers for the service shop. That saves you from a teardown based on guesswork.
The decanter's vibration slowly ripples in waves every few seconds. What is that?
Classic beats. The bowl and the scroll turn at close frequencies, the difference equal to the differential, and their running-speed components alternately add up and cancel out. Pronounced beats mean both rotors carry a noticeable imbalance, and each one needs to be dealt with separately.
After washing the bowl, the vibration went away. Is balancing still needed?
Probably not. If the clean machine stays within its usual level, there's no mechanical imbalance worth treating with weights. The working tool here is a schedule: washing on a time or vibration-threshold basis, and controlling the flocculant and feed rate. You can't balance a dirty rotor: you'd be locking in weights against a random buildup pattern.
What's the vibration standard for a decanter centrifuge?
There's no single number. The assessment is made against the applicable part and edition of ISO 20816, accounting for mount compliance, and decanters almost always sit on vibration isolators, which shifts the limits. Two practical benchmarks work better: the limits in the manufacturer's manual, and your own machine's trend. A rise to one and a half times the usual level is reason enough for a measurement, even if the figure is formally within tolerance.
The vibration trip fires toward the end of the cycle. Can the setpoint just be raised to keep running?
Raising the setpoint blindly is dangerous: at these speeds, damage develops fast. Start with a measurement that separates out the causes. If the level rises because of cake buildup, it's fixed with a washing schedule, and the setpoint won't need touching. If the running-speed component rises on a clean rotor, it's a mechanical issue, and running with the trip bypassed brings you closer to a failure.
The decanter started shaking after the scroll was re-hard-faced. Will on-site balancing help?
Almost certainly not. Re-hard-facing changes the mass distribution along the flighting, and that imbalance sits on the scroll, which isn't accessible on site. The correct route is: balance the scroll on a machine before reassembly, then check the assembled rotor. We'll take a measurement before it goes off for repair, a check measurement afterward, and, if the manufacturer permits it, a trim balance of the bowl in the standard planes.
Related content
On-site balancing of centrifugal compressor rotors and impellers, where they operate
Partly. On site, in the machine's own bearings, we balance overhung impellers on single-stage machines through an open inspection port, half-couplings, free shaft ends, drive-motor rotors, pulleys, and oil-cooler fans. Rotors of multistage compressors in horizontally split and barrel-type casings, as well as high-speed pinion shafts on integrally geared machines, are not balanced on site: the correction planes lie inside the flow path, the rotor behaves as flexible, and the residual-imbalance tolerance calls for specialised conditions. In these cases, our work on site is measurement and cause separation: imbalance, oil whirl, shaft misalignment, gear wear, surge, blade-pass pulsations. You get a report and a clear next step, not trial runs by guesswork.
On-site balancing of centrifugal fans: single-inlet and double-inlet
Yes, we balance centrifugal fans right on site, both single-inlet and double-inlet wheels. The rotor turns in its own bearing housings, and there's no need to remove the wheel or take the volute apart. Three conditions apply. The machine has to hold a stable speed at one damper setting. The correction plane — the spot on the wheel where balancing weights go — has to be reachable: an inspection hatch in the volute, a removable inlet cone, or an open inlet pocket, and on a double-inlet wheel both sides have to open. And most of the vibration has to come from the 1x component (vibration exactly at rotating speed — the signature of unbalance), not from bearings, a worn pulley, shaft misalignment, or pedestal resonance. We work out access and regime from your photos at the request stage, we measure the 1x share ourselves in the first half hour on site, and we tell you plainly if weights won't help here.
Building Your Own Balancing Machine: Supports, Bed, Drive, and Measurement System
Yes, you can build a rig on your own, and the soft-bearing (above-resonance) scheme is the most accessible way to do it. You need four things: a rigid, heavy bed, supports with a known suspension natural frequency (the frequency at which the support's moving part oscillates on its own) well below the running speed, a drive with stable speed, and a two-channel measurement system with a phase-angle sensor. The electronics get solved by buying a ready-made measurement core; everything else has to be designed and verified by you. The key point that separates a working rig from an expensive piece of hardware: acceptance testing by geometry, by dynamics, and against a reference rotor with a known trial unbalance.
Describe your equipment and the problem
We'll answer your questions, clarify the details, and let you know what's needed for an estimate and a visit.