AXILINEOn-site balancing of centrifuges and separators: drums, baskets, separator rotors
The centrifuge has run out of true, and the shop floor hears it through the hum long before the protection trips. We come out with a two-channel vibration analyser, separate rotor imbalance from shaft misalignment, bearings, and resonance, and balance the drum right in the machine if the manufacturer permits it. Speeds here are high, and the cost of getting it wrong is markedly higher than on a fan, so for some rotors we honestly send you to a service centre instead of balancing on site. We work in Portugal, based in Vila Nova de Gaia near Porto, with visits across the country.
Symptoms: when it's time to measure the centrifuge, not put up with it
A centrifuge forgives less than a fan. At 3000 rpm and above, a small residual imbalance turns into a force that wrecks the bearings, pushes out the shaft seals, and loosens the frame mounting within weeks. So "started humming a bit" on a machine like this should be read as an urgent signal.
There's a second feature. On a centrifuge, vibration often changes over the course of the cycle: it rises toward the end of dewatering, drops after a rinse, spikes with uneven feed. That's not an instrument glitch, it's the product behaving inside the rotor. Note at which point in the cycle the level peaks — that tells us more than one averaged number.
- Humming and knocking appeared after cleaning the drum, or replacing the screens, the screw, or the disc stack.
- Vibration builds up over the course of the cycle and drops noticeably right after the rotor is rinsed.
- On spin-up the machine passes through a zone where the housing shakes noticeably, then settles down at operating speed.
- The vibration trip has activated, and the operator restarts the machine by bypassing the lockout.
- The bearing supports run hotter than usual, and a leak has started along the shaft.
- The cake or the centrate discharges unevenly, the discharge stream "breathes".
- After reassembling the drum, the parts were fitted without matching their factory marks.
- The mm/s RMS reading at the bearing support has moved past zone B of the applicable part of ISO 20816 (that is, above the level acceptable for continued operation).
Don't try to take a reading through the cover or the guard mesh while it's running, and don't reach toward the spinning drum by hand. We fit the sensors on a stopped, locked-out machine and take the measurement from a safe distance afterward. If the vibration trip has activated twice, it's better to keep the machine stopped until it's been measured.
Which centrifuges and separators we balance
We work with industrial centrifuges and separators of the main configurations. Below are the subgroups and exactly what creates imbalance in each, because the source is different for every design.
Decanter (screw-conveyor) centrifuges
A long conical bowl with a screw conveyor inside: two rotating masses on one axis. Imbalance comes from cake buildup in the cone, wear and hardfacing on the screw flights, erosion at the discharge zone, and lost carbide tiles. The rotor is elongated and flexible, so here it's almost always two correction planes, plus a separate conversation about critical speeds — the frequencies at which the rotor passes through its own resonance.
Filtering centrifuges
A rotor with a screen or filter cloth. Imbalance comes from uneven cake thickness on the screen, locally clogged perforations, a torn cloth, worn screen segments. After a screen change the machine sometimes shakes simply because the new set didn't match the old one by mass.
Sedimentation centrifuges
A smooth bowl wall with cake that sticks in patches. The vibration level here is especially sensitive to the rinse regime. Balancing only makes sense on a rotor that's been washed beforehand, otherwise you'd be fixing the weights to a random pattern of buildup.
Centrifuge baskets and drums
Baskets on batch centrifuges and drums on centrifuges of any configuration — the classic short rotor. Imbalance comes from wall erosion, residual cake in a pocket, deformation after an impact, a loosened shaft fit, a lost or shifted factory weight. One correction plane is often enough.
Separator rotors: dairy and oil
Balancing a dairy or oil separator means working with the disc stack and the bowl on a vertical spindle. Causes of imbalance: reassembly not matched to the marks, deposits between the discs, corrosion, wear on the locking ring. Manufacturer requirements are especially strict here, and we agree any intervention on the rotor in advance.
Food-industry and chemical-industry separators
Food-industry separators run in a sanitary zone under CIP washing, so a weight must not create gaps or stagnant pockets. Chemical-industry separators add a corrosive environment and sealing requirements. In both cases we choose the mass-fixing method to fit the regulations, not the other way around.
- Rotors of treatment and dewatering equipment: sludge-dewatering centrifuges at treatment plants, sludge and drainage rotors, spin-dry drums. Here abrasive wear is almost always mixed in with the imbalance.
- The same machine's drive: pulleys, belt drive, motor rotor. Pulley imbalance and belt runout produce their own vibration, which weights on the drum can't remove.
- Discharge screws and conveyors near the centrifuge: their vibration travels through the frame and spoils the measurement if you don't know about them.
On-site balancing applies when the rotor is accessible, the speed holds steady, and the manufacturer doesn't prohibit fitting masses. If that doesn't describe your machine, tell us the model and year, and we'll draw the line straight away instead of driving out and shrugging.
What we check before balancing, and why we ask about the manufacturer
On a centrifuge the order of operations matters more than speed. We measure and inspect first, then decide whether weights are needed. The reason is simple: at high speed, balancing faulty mechanics just masks the level for a few weeks while the damage keeps progressing.
We separately check the documentation. Many centrifuge and separator manufacturers explicitly restrict intervention on the rotor: banning drilling, welding, and fitting masses outside the factory balancing elements. Sometimes balancing is only permitted at the manufacturer's own service centre, with subsequent drum certification. We don't work around requirements like that. If a restriction exists, we either work strictly within the factory correction planes or stop at diagnostics and hand you the report.
One more honest boundary: drums have an assigned service life and material-inspection requirements. Balancing doesn't override either flaw detection or the service-life limit.
- Model, speed, rotor configuration, whether there are factory balancing bolts, rings, or slots.
- The manufacturer's requirements on intervening in the rotor and on the mass-fixing method, in writing.
- Machine fasteners: tightness, a soft foot, condition of the frame and vibration isolators, clearances under the feet.
- Fit of the drum, basket, and screw on the shaft: play, runout, condition of the key and the clamping nut.
- Condition of the bearings and lubricant, because the measurement will have to be repeated after a bearing repair.
- Shaft misalignment at the drive: a noticeable 2x (vibration at twice the rotational frequency) and strong axial vibration mean shaft alignment is needed, not balancing.
- Resonance: how the 1x amplitude and phase behave during spin-up and coast-down.
- Cleanliness of the rotor: how much cake is inside and when it was last rinsed.
- Whether the machine can reach a stable speed and hold it for a few minutes.
The method for telling causes apart is covered in detail in a separate article on distinguishing imbalance from shaft misalignment and reading a spectrum. On this page we cover only what's specific to centrifuges and separators.
Product inside the rotor: why imbalance on a centrifuge shifts
On a fan, imbalance is usually constant. On a centrifuge it's variable by nature: product doesn't distribute perfectly across the bowl wall, cake sticks in patches, part of it leaves at discharge, part stays behind. So the same machine passes through several different vibration states within a single cycle.
That gives us a practical rule. What needs balancing is the component that remains with a clean rotor. If you fit weights to compensate for caked-on buildup, the machine will shake harder after the next rinse than it did before. That's why we ask you to wash the rotor before the visit and take the baseline measurement on a clean machine.
We take a second measurement on the actual working cycle with product. It's not for calculating weights, it's for you: it shows how much vibration the process itself adds, and whether the rinse regime or the feed needs adjusting.
| What you observe | What it usually means | What we do |
|---|---|---|
| The level rises smoothly over the dewatering cycle and drops after a rinse | Uneven cake buildup on the bowl wall | We balance the clean rotor, then help set a rinse regime and a vibration threshold |
| Vibration jumps around from run to run with no pattern | Product settles differently each time, or the drum or screw fit on the shaft has loosened | We check the fit and the runout, and take readings on the empty and the loaded rotor |
| A sharp peak over a narrow range during spin-up, then the machine settles down | Passing through the critical speed — a normal occurrence for a flexible rotor | We balance at the stable operating speed and assess how long it takes to pass through the zone |
| High 1x, phase is stable, and the rotor is clean | Rotor imbalance: erosion, wear, a lost weight, reassembly not matched to the marks | Balancing in the machine's own supports, one or two correction planes |
| Noticeable 2x and strong axial vibration at the drive | Shaft misalignment or a soft foot | Fasteners and shaft alignment, balancing only after that |
| Peaks at high frequencies, not a multiple of speed, a raised noise floor | Bearing defects, rubbing, a failing bearing cage | Bearing diagnostics; balancing before the repair is pointless |
If your machine has permanent vibration monitoring, give us the trend from recent months. The point where the level started rising often points straight to the event behind it: a cleaning, a screen change, a change in feedstock.
How the work proceeds on site
- 01
Conversation before the visit
We confirm the model, speed, rotor configuration, product type, access to the supports, and whether there are factory correction planes. We ask for photos of the rotor, the supports, and the nameplate, plus the manual's page on balancing. That way we arrive with a plan already in place and the right fixing hardware.
- 02
Inspection and preparation
The machine is stopped and locked out. We check fastener tightness, a soft foot, the fit of the drum and screw, shaft runout, condition of the vibration isolators. We clean flat spots on the bearing housings for the sensors and stick the reflective tape mark on the shaft or the drive pulley.
- 03
Baseline measurement
We fit two accelerometers on the bearing supports, usually radially, and aim the laser phase sensor at the mark. We record overall vibration (the combined level across all frequencies) and the 1x running-speed component with phase — the angular reference of the vibration to the shaft mark — speed, FFT spectrum, and time-domain signal at each support. On vertical separators we add a point at the upper spindle support.
- 04
Decision: balance or not
We calculate the 1x share of the overall level, look at the spectrum, and check for resonance during coast-down. If 1x accounts for only a small part of the level, we don't spend the shift on trial runs — we move to diagnostics and tell you plainly what needs repairing.
- 05
Trial weight and influence coefficients
We fit a weighed mass on the agreed correction plane, at a known radius, and run it. We count a 1x amplitude change of 20–30% or a phase change of 20–30° as valid. If the change is smaller, we increase the mass and repeat. That's how the instrument gets the response of your specific rotor-supports-foundation system.
- 06
Fitting the correction masses
The software outputs a mass and angle, or a fixed-position number if fixing is only possible at factory points. The fixing method is agreed in advance: a bolted weight, a factory balancing ring, metal removal in a permitted zone. We don't use welding on stainless-steel or sanitary rotors.
- 07
Verification run
We repeat the measurement at the same mode, at the same points, in the same direction. If tolerance isn't met on the first pass, we add a small mass to what's already fitted. The machine usually gets there within one or two trims.
- 08
Checking on the cycle with product, and the report
We run the working cycle and watch how the level behaves under load. We write up the report with before-and-after numbers and save the influence coefficients — the machine's measured response to the weight — so the next trim balancing (fine-tuning with a small weight) after a cleaning takes one run instead of three.
All eight steps happen in a single visit, provided the mechanics are sound and access is available. A bearing defect or a loosened fit found along the way changes the plan: repair first, then a repeat measurement and balancing.
One or two correction planes for drums and baskets
Rotor shape and speed set the number of planes. The rule of thumb is simple: the ratio of rotor length to diameter at the weight-mounting zone. A short basket behaves like a disc, and one mass is usually enough. An elongated drum almost certainly produces couple imbalance — a pair of unbalanced masses at opposite ends of the rotor — and one mass won't reduce vibration at both supports at once.
In practice, across this family of machines, here's how it breaks down.
- Baskets and short drums with length under roughly half the diameter: usually one correction plane, a static picture.
- Medium-length bowls on sedimentation and filtering centrifuges: we use two planes if two factory fixing points are accessible. Otherwise we balance in one and honestly record the residual level at the second support.
- Decanter and screw centrifuges: two planes practically always, the rotor is elongated and flexible.
- Separator rotors on a vertical spindle: the design fixes the planes rigidly, there's little freedom of choice here, we work at the points the manufacturer provided.
Flexible rotors that run above their first critical speed are a separate topic. For these, the rigid two-plane approach can give a good result at operating speed and a poor one during spin-up. Cases like this are assessed under the applicable part of ISO 21940 for flexible rotors, and sometimes the right answer is balancing at a service centre on a spin-up balancing stand. The general logic for choosing the number of planes, and the L/D rule, are covered in detail in a separate article.
Sources: ISO 21940-12:2016 · ISO 21940-11:2016
When on-site balancing won't work or won't help
We prefer to say this before the visit, not after. Below are the cases where weights on the rotor are either prohibited or don't solve the problem.
- The manufacturer prohibits intervention on the rotor, or permits it only at their own service centre. Typical for sealed bowls on disc-stack separators.
- No access to the correction plane: the drum is enclosed, there are no factory mass points, and opening it up means a full teardown. In that case it's cheaper to remove the rotor and balance it on a machine.
- The rotor runs above its first critical speed and changes bending shape across the operating range. Balancing at one speed doesn't guarantee compliance at another.
- The machine won't hold a stable speed: the cycle is built on spin-up and braking with no plateau. Without a steady mode, you can't get a repeatable phase.
- The rotor's geometry is already compromised: a crack, bulging, or through-wall erosion, broken disc-stack fixings. Balancing here would mask a dangerous condition — the rotor needs to be taken out of service.
- The fit of the drum, basket, or screw on the shaft has loosened. The 1x level will change after every run, and the influence coefficients will come out unstable.
- Sanitary or corrosion regulations don't allow any available mass-fixing method in the product zone.
- The rotor runs hot and develops thermal bow. Balancing on a cold machine won't hold, and a hot one needs a separate method.
- The main vibration isn't imbalance: bearings, rubbing, cavitation on the feed, shaft misalignment, resonance in the frame or the pipework.
In any of these cases you still get a result from us: measured numbers, a spectrum, a conclusion on the likely cause, and a recommendation on where to go next. We have a separate article covering cases where balancing doesn't help, and centrifuges show up in it more often than average.
Sources: ISO 13373-3:2015 · ISO 281:2007
What you get
- A report with before-and-after numbers: overall vibration and the 1x running-speed component in mm/s RMS at each bearing support, phase, speed.
- FFT spectra and time-domain signal before and after the work, taken at the same points and the same mode.
- Mass, radius, angular position or position number of each fitted weight, and the fixing method.
- An assessment of the machine's condition by zone under the applicable part of ISO 20816, citing the edition, measurement points, frequency band, and support type. For contractual acceptance these conditions are recorded separately.
- A residual-imbalance tolerance calculation by G quality grade under the applicable part of ISO 21940, if your team or the manufacturer needs it.
- Saved influence coefficients for this machine. The next trim after a cleaning or a screen change will take one run.
- A plain-language conclusion: whether it was imbalance or not, what we found mechanically, and what's worth doing before the next shutdown.
We don't promise a specific final number in advance. The result depends on the condition of the rotor and the supports, on the applicable criteria, and on the manufacturer's restrictions. What we do promise is measured before-and-after numbers and a straight answer on whether compliance is achievable on this machine without a repair.
Sources: ISO 20816-1:2016 · ISO 21940-11:2016 · Balanset-1A operation manual
Price and how to book
vibration diagnostics with a report costs EUR 300 per unit, balancing adds from EUR 250, and the minimum invoice for a visit is EUR 500. The calculator on our website works out the exact amount for your machine, the number of rotors, and the distance. That way you see the figure before the visit, with no surprises on the invoice.
For a useful answer, send us the centrifuge or separator model, operating speed, product type, what changed before the vibration appeared, and photos of the rotor with the supports. If you have the manufacturer's requirements on intervening in the rotor, attach them right away: that's the single most common factor deciding whether we work on site or not.
- Who comes out: the engineers who design and manufacture the Balanset instruments and use them on-site themselves.
- What we work with: the Balanset-1A, a portable two-channel vibration analyser and balancer. Two accelerometers on the bearing supports, a laser phase sensor via reflective tape mark, a USB module, and software on a laptop.
- What the instrument delivers on site: overall vibration and 1x with phase, speed, FFT spectrum, time-domain signal, one- and two-plane balancing by the influence coefficient method, fixed-position mode and drilling calculation, trim balancing, G-class tolerance calculation, archive and reports.
- Coverage: based in Vila Nova de Gaia near Porto, visits across Portugal.
If you'd rather handle jobs like this in-house, the same instrument can be bought and put into service with your maintenance team. On centrifuges it pays for itself quickly: after every drum cleaning the level shifts, and a trim balancing off the saved coefficients takes one run.
Sources: Balanset-1A manufacturer specification
Frequently asked questions
Can a centrifuge be balanced without removing the drum?
In most cases, yes. We work on the machine as installed, in its own supports: two or three runs at operating speed, weights fitted on whichever factory correction plane is accessible. Removal is needed when there's no access to the correction plane, the rotor is flexible and needs balancing across its operating range on a stand, or the manufacturer only permits intervention at their own service centre. We tell you this before the visit, based on the model and photos.
The manufacturer prohibits fitting weights to the drum. What do we do then?
We don't work around requirements like that. In that case the visit turns into vibration diagnostics: we record overall vibration and 1x at each support, phase, spectrum, and coast-down behaviour, and check fasteners, fits, and signs of resonance. You get a report with numbers and a conclusion on the likely cause. With that document you can have a concrete conversation with the manufacturer or a service centre, instead of a vague "the machine shakes".
How do you fix weights on a stainless-steel drum if welding isn't allowed?
We don't use welding on stainless-steel or sanitary rotors. A weld changes the metal's structure and creates a corrosion zone and a gap where product collects. Instead we use the factory balancing elements the manufacturer provided: bolted weights, balancing rings and slots. Where the documentation allows it, we correct by removing metal in a permitted zone. We agree the fixing method before the work, not as we go.
Do you measure on an empty centrifuge or one loaded with product?
Both, and they're different tasks. We calculate the weights from a measurement on a clean, washed rotor, because caked-on cake is a variable quantity and can't be compensated for with weights. We take the measurement with product afterward, to show you how much vibration the process itself adds. If the difference is large, the question isn't about balancing anymore — it's about the rinse regime, feed uniformity, and the condition of the discharge zone.
Why does the separator shake on spin-up but settle down at operating speed?
Most likely the rotor is passing through a critical speed, and near it the amplitude rises while the 1x phase swings around quickly. For flexible rotors that's normal, the important thing is just that the machine passes through the zone quickly and doesn't linger in it. Balancing needs to be done at the stable operating speed, outside that zone. If the spin-up peak has become noticeably stronger than before, that's already a reason to check the rotor and the supports.
Down to what mm/s values do you bring centrifuge vibration?
The target value isn't set at random. In the software we set a target residual 1x, we assess the machine's overall condition by overall vibration and the zones of the applicable part of ISO 20816, and where needed we assess balancing quality via the G class under the applicable part of ISO 21940. These are three different tolerances, easy to mix up, and we have a separate article breaking them down. We won't guarantee a specific final number before the measurement: it depends on the condition of the rotor, the supports, and the foundation.
Related content
On-Site Balancing of Centrifuge Drums and Baskets
We balance the drum or basket on site, in the machine's own bearing housings. But first we answer a different question: is this imbalance at all. On a filtering or sedimentation centrifuge, the dominant vibration more often comes from unevenly settled product than from the rotor. We tell them apart by phase repeatability — the angle by which the vibration is tied to the rotor. We run several starts in a row on a clean, dry basket and check whether the phase of the 1x running-speed component — vibration at the rotation frequency — stays put. If it does, it's mechanical imbalance and weights will help. If it drifts from run to run, weights would only mask the problem for a single cycle. We fit masses at the standard spots on the rim or the reinforcing ring. We don't drill the perforated shell, we don't weld on stainless steel, and we don't put foreign parts in the product zone.
Decanter Centrifuge Balancing: What Can Really Be Done On Site
Honest answer: not always. We balance the decanter bowl on site if the manufacturer permits fitting weights, standard balancing positions exist on the end hubs, and the measurement confirms that the bowl's own 1x running-speed component dominates — that is, vibration at its rotation frequency. The scroll can't be balanced on site as a matter of principle: its correction planes — the spots where balancing weights go — are hidden inside the bowl, and reaching them means a full teardown of the rotor. So some visits to decanters end not with weights but with a measurement that separates out the causes: cake, worn flighting, bearings, the gearbox, or imbalance. You get a report with numbers that lets the conversation with the service shop stay concrete. We usually understand your case before the visit even happens, from the model, photos, and vibration trends.
Multi-bearing rotor balancing: a long shaft on three or more bearings
A multi-bearing rotor is a distinct problem, not an extended two-plane one. For a rigid rotor on two bearings, you're solving a system of two equations with two unknowns, and the influence coefficient matrix is 2×2. Three bearings give a 3×3 system with nine coefficients and a minimum of four runs; four bearings give sixteen coefficients and five runs. You have to measure at every bearing and solve the system as a whole: if you work through the planes one at a time, the vibration will keep running from bearing to bearing, and the process won't converge.
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.